You are on page 1of 1085

Gregory Tsinker

Handbook Of Port And


Harbor Engineering
Geotechnical and structural aspects
HANDBOOK OF PORT AND
HARBOR ENGINEERING
JOIN US ON THE INTERNET
WWW: http://www.thomson.com
EMAIL: findit@kiosk.thomson.com
thomson.com is the on-line portal for the products, services and
resources available from International Thomson Publishing (ITP).
This Internet kiosk gives users immediate access to more than 34 ITP
publishers and over 20,000 products. Through thomson.com Internet
users can search catalogs, examine subject-specific resource centers
and subscribe to electronic discussion lists. You can purchase ITP
products from your local bookseller, or directly through thomson.com.

Visit our Internet Resource Center for information on our new publications, links to
useful sites on the World Wide Web and an opportunity to join our e-mail mailing
list. Point your browser to: http://www.chaphall.com/chaphall.html or
http://www.chaphall.com/chaphall/civeng.html for Civil Engineering


A service of I\!) P
HANDBOOK OF PORT AND
HARBOR ENGINEERING
GEOTECHNICAL AND STRUCTURAL ASPECTS

Gregory P. Tsinker, Ph.D., P.E.

Springer-Science+Business Media, B.V.


Cover Design: Andrea Meyer, emDASH inc., New York, NY
Cover Photo: Courtesy of Han-Padron Associates, New York, NY

Copyright © Springer Science+Business Media Dordrecht 1997


Originally published by Chapman & Hall in 1997

ISBN 978-1-4757-0865-3 ISBN 978-1-4757-0863-9 (eBook)


DOI 10.1007/978-1-4757-0863-9

All rights reserved. No part of this work covered by the copyright hereon may be reproduced or used in any
form or by any means-graphic, electronic, or mechanical, including photocopying, recording, taping, or
information storage and retrieval systems-without the written permission of the publisher.

12345678910XXXOI 009997

Library of Congress Cataloging-in-Publication Data

Tsinker, Gregory P.
Handbook of port and harbor engineering : geotechnical and structural aspects I Gregory
P. Tsinker.
p. em.
Includes bibliographical references and index.
ISBN 978-1-4757-0865-3
I. Harbors-Design and construction. 2. Marine geotechnique. I. Title.
TC205.T747 1996
627' .2-dc20 95-48487
CIP
British Library Cataloguing in Publication Data available

To order this or any other Chapman & Hall book, please contact International Thomson Publishing, 7625
Empire Drive, Florence, KY 41042. Phone: (606) 525-6600 or 1-800-842-3636.
Fax: (606) 525-7778, e-mail: order@chaphall.com.

For a complete listing of Chapman & Hall's titles, send your requests to
Chapman & Hall, Dept. BC, 115 Fifth Avenue, New York, NY 10003.
To Nora

v
contents

Dedication v

Preface xix

Introduction xxv

Contributors xxxvii

1 THE MARINE ENVIRONMENT AND ITS EFFECTS ON PORT DESIGN


AND CONSTRUCTION 1

1.1 Introduction 1
1.1.1 General 1
1.1.2 Seawater and Fouling 4

1.2 Water-level Variations 8

1.3 Weather Factors 10

1.4 Wind 12
1.4.1 General 12

vii
viii Contents

1.4.2 Wind Parameters 14

1.5 Currents 16

1.6 Waves 19
1.6.1 General 19
1.6.2 The Sea State Parameters 26
1.6.3 Wave Theories 31
1.6.4 Design Wave 33

1.7 Ice 37
1.7.1 Introduction 38
1.7.2 Ice Covers 39
1.7.3 Effects of Ice on Port Operations 47
1.7.4 Cold Temperature and Ice Effects on Marine Structures Design 54

References 61

2 PORT (HARBOR) ELEMENTS: DESIGN PRINCIPLES


AND CONSIDERATIONS 69

2.1 General 69
2.1.1 Port Classification 70
2.1.2 Port Details and Definitions 71

2.2 Ships and their Influence on Port Design 73


2.2.1 Ships 73
2.2.2 Ship Influence on Port Design 74

2.3 Access (Navigation) Channel 78


2.3.1 General 78
2.3.2 Navigational and Operational Parameters 79
2.3.3 Environmental Parameters 81
2.3.4 Layout 82
2.3.5 Channel Cross Section 84
2.3.6 Economic Considerations 96
Contents ix

2.4 Port (Harbor) Entrance 98

2.5 Port Water Area (Harbor) 106


2.5.1 Basin Sizes 107

2.6 Location, Orientation, Size, and Shape of the Port 112


2.6.1 Selection of Port Location 112
2.6.2 Size and Orientation of Marine Facilities 115
2.6.3 Harbor Area Requirements 115

2.7 Quay Basin 119

2.8 Offshore Installations 121


2.8.1 Offshore Bottom-Fixed Marine Facilities 121
2.8.2 Single-Point Offshore Moorings 123

2.9 Port-Related Marine Structures 124


2.9.1 Land Requirements 124
2.9.2 Dust and Noise Control 127
2.9.3 Berth Requirements 128
2.9.4 Structures 130
2.9.5 Selection of the Most Cost Effective Structure for Dock Construction 151
2.9.6 Constructability 153

2.10 Structural Materials 154


2.10.1 Structural Concrete 155
2.10.2 Underwater Concreting 172
2.10.3 Precast Concrete 178
2.10.4 Structural Steel in Port Engineering 180
2.10.5 Structural Timber 195

2.11 Breakwaters 198

2.12 In-Harbor Slope Protection 203

2.13 Aids to Navigation 204

2.14 Mooring Accessories 205


x Contents

2.15 Fender Systems 206


2.15.1 Timber Fenders 208
2.15.2 Solid Rubber Fenders 210
2.15.3 Pneumatic Fenders 216
2.15.4 Foam-Filled Fenders 222
2.15.5 Other Fender Systems 222
2.15.6 Fenders Failure 226
2.15.7 General Principles in Fender System Selection and Design 226

References · 232

3 DESIGN LOADS 243

3.1 General 243

3.2 Environmental Loads 244


3.2.1 Wind 245
3.2.2 Currents 248
3.2.3 Waves 251

3.3 Mooring Loads 260


3.3.1 Mooring Lines Arrangement 261
3.3.2 Mooring Line Materials 262
3.3.3 Mooring Forces 262

3.4 Loads From Cargo Handling and Hauling Equipment and Uniform
Distributed Loads 267
3.4.1 General Considerations 267
3.4.2 Design Load Assumptions 269
3.4.3 Uniform Distributed Cargo Loads and Miscellaneous Live Loads 271
3.4.4 Rubber Tire and Crawler Track Mounted Equipment 272
3.4.5 Rail-Mounted Cargo 279
3.4.6 Fixed-Base Equipment 282

3.5 Ship Impact (by M. Shiono in collaboration with G. Tsinker) 283

3.6 Ice Loads 293


3.6.1 General 293
3.6.2 Environmental Driving Forces 294
3.6.3 Ice-Crushing Load 295
3.6.4 Loads Due to Ice Bending Mode of Failure 297
Contents xi

3.6.5 Forces Due to Ice Sheet Adfreeze to the Structure 299


3.6.6 Vertical Loads on Piles or Piers Due to Changes in Water Level 300
3.6.7 Ice Load of Thermal Origin 301
3.6.8 Other Ice-Induced Loads 302

3.7 Seismic Loads (by W. S. Dunbar) 302


3.7.1 Seismic Ground Motion 303
3.7.2 Descriptions of Ground Motion 307
3.7.3 Design Ground Motion Estimation 312
3.7.4 Design Loads 318

3.8 Load Combinations 319

References 320

4 GEOTECHNICAL ASPECTS OF SOIL-STRUCTURE INTERACTION


DESIGN CONSIDERATIONS 331

4.1 General 331

4.2 Subsurface Investigation 333

4.3 Soil Liquefaction and Evaluation of Liquefaction Potential (by G. Tsinker


and W. S. Dunbar) 334

4.4 Basic Design and Construction Considerations 342


4.4.1 Modern Trends 342
4.4.2 Bottom-Fixed Structures 343
4.4.3 Safety Considerations 345
4.4.4 Construction Procedure 347

4.5 Soils and Bedrock 348


4.5.1 Gravel and Sand 348
4.5.2 Silt and Clay 349
4.5.3 Bedrock 350

4.6 Properties and Characteristics of Soils 352


4.6.1 Shear Strength 354
4.6.2 Compressibility (Consolidation) 356
4.6.3 Permeability 357
xii Contents

4.7 Lateral Soil Pressure 358


4.7.1 Active Earth Pressure 359
4. 7.2 Effects of Wall Movement 368
4.7.3 Effects of Time-Dependent Changes in Soil 374
4.7.4 Effect of Ambient Temperature on Earth Pressures 376
4. 7.5 Effects of Backfill Freezing 376
4.7.6 Passive Earth Pressure 376
4. 7. 7 Earth Pressure at Rest 380
4.7.8 Compaction-Induced Pressure 381

4.8 Friction Forces on Walls 381

4.9 Dynamic Soil Pressures 382


4.9.1 Mononobe-Okabe Formulation 383
4.9.2 Effect of Saturated Backfill 385
4.9.3 Hydrodynamic Pressures 385
4.9.4 Effect of Wall Inertia 386
4.9.5 Selection of Ground Motions 387
4.9.6 Effect of Wall Movements 387

References 388

5 GRAVITY-TYPEQUAYWALLS 397

5.1 General 397

5.2 Basic Structural Arrangements 403


5.2.1 Blockwork Structures 403
5.2.2 Quay Walls Composed of Floated-in Concrete Caissons 409
5.2.3 Quay Walls Composed of Large-Diameter Cylinders 428
5.2.4 Cribwork Quay Walls 435
5.2.5 Steel Sheet-Pile Cell Bulkheads 439
5.2.6 Quay Walls 446
5.2.7 Gravity-Type Walls 452

5.3 Basic Design Considerations 461


5.3.1 Loads and Forces Load Combinations 461
5.3.2 Basic Static Principles 464

5.4 Design ofBlockwork Quay Walls 478


5.4.1 Basic Design Principles 478
Contents xiii

5.4.2 Design Phase 1 480


5.4.3 Design Phase 2 483
5.4.4 Design Phase 3 484
5.4.5 Design Phase 4 484

5.5 Design of Quay Walls Comprised of Floated-in Concrete Caissons 485


5.5.1 Basic Design Principles 485
5.5.2 Buoyancy and Buoyant Stability of a Caisson 485
5.5.3 Buoyancy and Stability of a Damaged Caisson 489
5.5.4 Caisson Launch 490
5.5.5 Towing and Sinking 496
5.5.6 Structural Design 497

5.6 Design of Quay Walls Composed of Large-Diameter Cylinders 500

5.7 Design ofL-Shaped Walls 504


5.7.1 Basic Requirements 504
5.7.2 Design of Cantilever Walls 506
5.7.3 Design of Counterfort Wall 507
5.7.4 Design of Wall Constructed from Prefabricated Components with
Internal Anchorage 508
5.7.5 Design of Wall Constructed from Prefabricated Components with
External Anchorage 508

5.8 Design of Cellular-type Steel Sheet-pile Bulkheads 511


5.8.1 Introduction 511
5.8.2 Conventional Design Method 512
5.8.3 Horizontal Shear (Cummings') Method 518
5.8.4 Brinch Hansen Method 519
5.8.5 Seismic Design of Cellular Bulkheads 519
5.8.6 Deflection of Cellular Bulkhead 520
5.8.7 Effects of Concentrated Horizontal Loads on Sheet-Pile Cell 522

5.9 Design ofCribwork-type Quay Walls 522

5.10 Reinforced Earth Quay (by D. Weinreb and P. Wu) 524


5.10.1 General Concept 524
5.10.2 Design of Reinforced Earth Marine Structures 532
5.10.3 Construction of Reinforced Earth Walls Underwater 536

References 542
xiv Contents

6 SHEET-PILE BULKHEADS 549

6.1 Introduction 549


6.1.1 Sheet-Piling-Background 549
6.1.2 Anchoring Systems 552
6.1.3 Sequence of Construction 555

6.2 Sheet-Piling-Structural and Driving Aspects 555


6.2.1 Timber Sheet Piles 556
6.2.2 Steel Sheet Piles 558
6.2.3 Concrete Sheet Piles 561
6.2.4 Selection of Sheet-Pile Section 570

6.3 Anchor Systems 571


6.3.1 Anchor System Comprised of Tie-Rods and Anchorages 572
6.3.2 Anchor System Comprised of Raked Piles 576
6.3.3 Ground (Rock) Anchors 583

6.4 Wall Capping 589

6.5 Construction Methods 591


6.5.1 Construction Sequence 591
6.5.2 Sheet-Pile Driving 592
6.5.3 Pile Jetting 596
6.5.4 Earthwork 602

6.6 Earth Pressures on Flexible Walls: State-of-the-Art Review 606

6.7 Design of Sheet-pile Walls 623


6.7.1 Design Criteria 623
6.7.2 Design of Cantilever Walls 625
6. 7.3 Design of Anchored Bulkheads 630
6.7.4 Design of Sheet-Pile Bulkheads Anchored by Raked Piles 643

6.8 Sheet-Pile Bulkheads Built on Creep Soils 653


6.8.1 Cantilever Sheet-Pile Bulkhead 655
6.8.2 Single-Anchor Sheet-Pile Bulkhead 657
6.8.3 Multianchor Sheet-Pile Bulkhead 661

6.9 Anchorage Design 665


Contents xv

6.9.1 Piled Anchorages 666


6.9.2 Sheet-Pile Anchor Wall 667
6.9.3 Individual Vertical Anchor Piles 669
6.9.4 Deadman (Plate) Anchor 670

6.10 Waling and Tie-Rod Design 672

6.11 Ground (Rock) Anchors 673

6.12 Overall Stability 679

6.13 Seismic Design of Anchored Sheet-Pile Walls (by W. S. Dunbar) 682


6.13.1 Observed Failure Modes 682
6.13.2 Seismic Design Procedure 683
6.13.3 Assumption 683
6.13.4 Factor of Safety Against Failure by Rotation 683
6.13.5 Size and Location of Anchor Block 684
6.13.6 Balanced Design Procedure 685

6.14 Sheet-Pile Wall Failure 686

References 688

7 PILED WATERFRONT STRUCTURES 695

7.1 Introduction 695

7.2 General 697


7.2.1 Structural Schemes and Structural Components 697
7.2.2 Prefabrication 701

7.3 Open Pile Structures With Suspended Decks 701


7.3.1 Open Piled Offshore Piers 702
7.3.2 Piling 710
7.3.3 Suspended Deck Structures for Marginal Wharves 711
7.3.4 Basic Design Principles 712
7.3.5 Suspended Deck Structures Founded on Large-Diameter Cylindrical Piles
718
7.3.6 Protection from Ship Impact 722
xvi Contents

7.3.7 Pile Anchoring in Foundation Soil and the Deck Structure 724

7.4 Relieving Platforms 725

7.5 Structural Elements 734


7.5.1 Pile Foundation 735
7.5.2 Superstructure 776
7.5.3 U nderdeck Slope 782

7.6 Pile-Soil Interaction 795


7.6.1 General 795
7.6.2 Piles Under Axial Static Load 803
7.6.3 Pile Settlement 817

7.7 Laterally Loaded Piles 820


7.7.1 General 820
7.7.2 Conventional Design Methods 822
7.7.3 Broms' Method 826
7. 7.4 Subgrade Reaction Approach 829
7.7.5 Laterally Loaded Socketed Piles 836

7.8 Piled Marine Structures Design Methods 837


7.8.1 Design Criteria 837
7.8.2 Design Methods 838

References 865

8 OFFSHORE DEEP WATER TERMINALS 879

8.2 Layout 881


8.2.1 Dry Bulk Loading/Unloading Facilities 882
8.2.2 Liquid Bulk Loading/Unloading Terminals 886

8.3 Mooring System 888


8.3.1 Basic Structural Concepts 890

8.4 Dolphins and Platforms 893


8.4.1 Breasting Dolphins 893
8.4.2 Piled Breasting Dolphins 895
8.4.3 Gravity-Type Dolphins 896
Contents x:vii

8.4.4 Steel Jacket-Type Structures 898


8.4.5 Fenders 898
8.4.6 Mooring Dolphins 898
8.4.7 Loading/Unloading Platforms 899
8.4.8 Access Trestles and Catwalks 899

8.5 Structural Design 901


8.5.1 Marine Foundation and its Effects on Structural Design 901
8.5.2 Basic Design Procedures 903

References 914

9 MODERNIZATION OF EXISTING MARINE FACILITIES 917

9.1 Introduction 917

9.2 Modernization of Mooring Structures 919


9.2.1 Modernization of Gravity-Type Quay Walls 920
9.2.2 Modernization of Piled Wharves 925
9.2.3 Modernization of Sheet-Pile Bulkheads 925
9.2.4 New Wall Construction 929

9.3 Modernization of Waterfront Structures: Characteristic Examples 930


9.3.1 Gravity-Type Quay Walls 930
9.3.2 Modification of Piled Coal-Loading Pier No. 6 at Norfolk, Virginia 940
9.3.3 Use of Piled Structures and Sheet-Pile Walls for Modernization
of Existing Structures 942
9.3.4 Construction of Brand New Structures 949

References 949

10 BREAKWATER DESIGN (by S. Takahashi) 951

10.1 Historic Development of Breakwaters 952


10.1.1 Structural Types 952
10.1.2 Conditions for Breakwater Selection 956
10.1.3 Comparison of Sloping- and Vertical-Type Breakwaters 956
10.1.4 Historical Development of Breakwaters 957

10.2 Design of Conventional Vertical Breakwaters 977


xviii Contents

10.2.1 Examples of Conventional Vertical Breakwaters 977


10.2.2 Wave Transmission and Reflection by Vertical Walls 978
10.2.3 Wave Forces on Vertical Walls 981
10.2.4 Design of Rubble-Mound Foundation 1001
10.2.5 Rubble-Mound Toe Protection Against Scouring 1005

10.. 3 Design of New Types ofVertical Breakwater 1006


10.3.1 Perforated Wall Breakwater 1007
10.3.2 Inclined Walls 1015

10.4 Design of Horizontally Composite Breakwaters 1020


10.4.1 Wave Transmission and Reflection 1021
10.4.2 Wave and Block Load on a Vertical Wall 1022
10.4.3 Stability of Wave-Dissipating Concrete Blocks 1023

10.5 Design of Rubble-Mound Breakwaters 1024


10.5.1 Wave Transmission and Reflection 1025
10.5.2 Design of Armor Layer 1027
10.5.3 Inner Layers, Core, Toe, and Wave Screen 1034

References 1036

Index 1045
Preface

In past 10 years or so several excellent This book has been written to :fill a niche
books and handbooks on "Port and Harbors in the existing literature on port and harbor
Engineering" and "Coastal and Ocean Engi- engineering and to provide the port design-
neering" have been published in Europe and ers, and particularly those concerned with
in North America. Reference to these works the design of a port- and harbor-related
is made elsewhere in this book. marine structures, with state-of-the-art in-
The authors of the aforementioned works formation and common sense guidelines to
offer a shrewd and comprehensive discus- the design and construction of the basic
sion on the marine environment and its types of marine structures. This book is a
effects on port design, port operation, port companion volume to my earlier work Ma-
hydraulics, coastal geomorphology, littoral rine · Structures Engineering: Specialized
drift and sedimentation, port and shipping Applications published by Chapman & Hall
technology and economics, design and con-
in 1995. That book covers important sub-
struction of a floating port related struc-
jects such as the evaluation of capacity of
tures, and others. However, proportionally
the in-service marine structures and meth-
the geotechnical and structural aspects of
port construction have been given very lit- ods of their remediation and maintenance,
tle attention. This happens, perhaps, be- construction and operation of the marine
cause the subject of marine structures engi- structures in cold regions, design and con-
neering is very broad by itself; it is a blend struction of a shipyard and related marine
that encompasses the array of engineering structures, design of anchored offshore
disciplines, e.g., civil, structural, geotechni- moorings and floating breakwaters, design
cal, hydraulic, strength of materials, corro- and construction of marinas (small craft
sion, naval architecture and others knowl- harbors), and design and construction of
edge of which is required to produce a sound marine structures that are used in naviga-
and economical design of a modern port or ble waterways for protecting the bridge
marine terminal. piers from ship collision. Conversely, this
xix
xx Preface

volume provides the marine structures de- tion principles, and operation of these
signer with basic information on the marine structures have been significantly improved
environment and its effects on port design by the introduction of new and better fend-
and construction (Chapter 1); port elements ering systems and efficient mooring acces-
and their effects on port operation (Chapter sories. New and better structural materials
2); design loads and their combinations that have also been introduced. For example,
are commonly used to design waterfront modern concrete technology now enables the
structures (Chapter 3); information on the engineer to use durable high-strength con-
phenomenon of soil I structure interaction crete, highly resistant to deterioration in
that explains basic principles that affect harsh marine environments. New and bet-
soil lateral thrust against rigid and flexibleter repair procedures and rehabilitation
soil retaining structures (Chapter 4); design techniques for port structures have also
of gravity quay walls, sheet pile bulkheads, been introduced.
and piled marine structures (Chapters 5, 6 Progress in development of new marine
and 7); basic principles of design of the structures and modernization of existing
offshore marine terminals (Chapter 8); structures was based on advances in ana-
modernization of the existing waterfront lytical design methods as well as on result
structure that makes them usable in mod- of numerous scale-model tests and field in-
ern port operations (Chapter 9); and design vestigations conducted all over the world.
and construction of breakwaters (Chapter Today, marine structure design is a unique
10). discipline in the field of civil engineering
In both books each chapter includes a that is based on the use of highly advanced
comprehensive list of relevant cross-refer- methods of soil foundation investigation and
ences intended to help the interested reader thorough understanding of the principles of
to study the subject in additional depth. soiljstructure interaction in the marine en-
In this book I have attempted to provide vironment.
the reader with a clear understanding of Recently sophisticated computational
the phenomenon of interaction between the procedures and mathematical models have
environmental agents such as waves, cur- been developed and used for design of vari-
rents, and wind as well as harbor soils and ous marine structures. It must be stressed,
backfill materials with bottom-fixed marine however, that in many cases the diverse
structures. and complex geology at various port loca-
During my long career as a practicing tions results in a wide variety of geotechni-
waterfront consultant and port engineer cal environments. Such conditions require a
considerable progress has occurred in the careful approach to selection of structure
field of design and construction of port- and type and use of the appropriate design
navigation-related marine structures. method, which should not necessarily be
Progress in port design, and in particular highly sophisticated. It is a misconception
design of waterfront structures, has been that sophisticated computer analysis, with
strongly influenced by the dramatic changes its greater accuracy, will automatically lead
in vessel sizes and in modes of modern to better design. Despite the highly sophis-
terminal operation. Multipurpose ports have ticated analytical methods available today,
been replaced by more specialized termi- the marine structural designer must be
nals, which result in dramatic effects on aware that the design is not merely a stress
both the design of berth structures and lay- analysis process. Use of computers has not
out of the terminal. Furthermore, marine diminished the value of some hand calcula-
structures for various purposes have been tions. In fact, many questions about marine
developed using new design and construe- structure engineering are still best an-
Preface xxi

swered with simple, often empirically based, cialized marine terminals. During the same
but practical formulas. period of time the nature of port traffic has
Computers have revolutionized the pro- changed markedly rendering some older fa-
cess of structural engineering and greatly cilities unsuitable for today's operations.
increased productivity of engineering firms. The obvious example is the emphatic shift
Computer-aided analyses are of great help to containerization of literally all types of
when used in the proper context, for exam- cargoes and transportation of containerized
ple, when modeling of the structure is cor- goods in ever larger vessels. Similar devel-
rect, the real boundary conditions are taken opments have occurred in transporting of
into account and most of all when the out- huge volumes of liquid and dry bulk com-
put is examined and interpreted by an ex- modities; sizes of the vessels that now
perienced engineer. However, some critics transport these cargoes reached 500,000
observed a "rapid deteriorating competency DWT and more. Obviously, these giant ves-
on the part of the engineering community sels require deeper approach channels;
as a consequence of using computers" and larger harbor basins; deep water quays;
found "a serious lack of critical evaluation gantry cranes with greater height clear-
ability in many of the young engineers ... ance, outreach, and lifting capacity; special-
(which) put an inordinate amount of faith ized terminals with appropriate handling
in the computer" (ENR, October 28, 1991). equipment; and so on. This gives obvious
The worrisome trend in the present de- impetus to global modernization of existing
sign and construction practices is that some ports as well as to construction of new ports
inexperienced, however, highly competent and terminals that can accommodate the
in the use of computers, engineers consider modern maritime traffic.
themselves "instant experts" ready to ana- Large capital investments into ports de-
lyze and design anything. velopment have been made in 1960s through
The marine structures designer should 1980s and many experts predict that this
realize that formulation of the mathemati- trend will continue through 1990s and far
cal approach used for structural analysis beyond into the 21st century.
must be practical and compatible with Much of the capital spent on port devel-
available engineering data, for example, opment is allocated for construction, opera-
shearing strength and consolidation charac- tion, and maintenance of its marine facili-
teristics of the foundation and backfill soils, ties. Therefore, economical design of these
environmental and live loads, etc. However, facilities can save a lot of money needed for
sometimes even when analytically correct port development.
results are obtained, the inexperienced en- Successful design of any project is based
gineer can make errors by neglecting some on three pillars; they are
practical aspects related to constructibility,
for example, how to fabricate or how to get • intelligence
the structural component in place. • education
To avoid the disastrous consequences of • experience.
such designs experienced engineers must
spend sufficient time helping their less ex- The latter two assume thorough knowledge
perienced colleagues to prepare the mathe- of a subject matter and the most recent
matical models and review the computer developments in the area of interest. Unfor-
output. tunately, all too frequently good and less
The 1980s and 1990s have become known costly engineering solutions are not used
for global modernization of existing ports because of the lack of familiarity on a part
and construction of new high capacity spe- of the designer. On the other hand, use of a
xxii Preface

"text book" solution to solve the problem This book has been 7 years in prepara-
may also be counterproductive. tion. I have drawn from more than 40 years
The subject of marine engineering is a of my own experience as a marine engineer
field where ingenuity can achieve consider- and scientist involved with research and all
able savings. Every project is site specific practical aspects of structural design, con-
and therefore no engineer should be content struction, and project management. Also,
merely to follow anothers designs but should worldwide experience has been examined
study such designs and use them as a start- and the best of it is included in this work.
ing point for developing his or her own Subsequently, acknowledgments of mate-
ideas that are best suitable for the particu- rial used in this book are given in the ap-
lar site conditions. propriate places in the text and figures. I
AB noted earlier the science and practice wish to extend my deepest gratitude to all
of port and harbor engineering draw from the publishers, authors, and organizations
various disciplines and cover a broad area from whom material for this work has been
of interrelated subjects. This book offers ba- drawn.
sically geotechnical and structural aspects This volume is not a one-man job. I am
of port and harbor engineering, and no at- deeply indebted to many experienced indi-
tempt has been made to include all detailed viduals who have contributed material and
analytical procedures from these interre- comments to this project. In attempting to
lated disciplines, for example, dredging, port make this work most helpful and useful I
operation and maintenance, etc. However, have drawn from sources including the
where relevant, all efforts have been made knowledge and experience of my former col-
to provide the reader with a considerable leagues at Acres International Limited, who
cross-reference on the interrelated subjects; assisted in a variety of ways: Dr. W. S.
this includes the most recently published Dunbar contributed information on
books, papers from the journals of profes- seismic-induced loads, potentials for soil
sional engineering societies, and the pro- liquefaction, soil dynamic loads upon re-
ceedings of specialty conferences. taining structures, design of a soil retaining
The design procedures and guidelines structures for seismic loading and also re-
contained in this book are intended to point viewed several chapters and helped in edit-
out the complexity of the particular prob- ing the book; Mr. R. G. Tanner has reviewed
lem and illustrate factors that should be several chapters and offered useful com-
considered and included in an appropriate ments; special gratitude goes to Mr. D. Pro-
design scenario. They should not be used tulipac who dedicated a great deal of his
indiscriminately and particularly not for the time to editing most of the text.
detailed design, and should always be com- I wish to extend my gratitude and ac-
bined with good engineering judgment. As knowledge a stimulating and enjoyable col-
noted earlier, this work has been conceived laboration with: Mr. M. Shiono, Deputy
as a two-part treatise in which I have at- General Manager-Research and Develop-
tempted to provide marine structures de- ment, Sumitonio Rubber Industries, Kobe,
signers with state-of-the-art information Japan who contributed information on rub-
and common sense guidelines to the design ber fender systems; Dr. S. Takahashi, Chief
of basic types of marine structures associ- of Maritime Structures Laboratory, Port
ated with port activities. and Harbour Research Institute, Yokosuka,
This book is designed to serve as both a Japan, who contributed Chapter 10
guide and a reference for practicing marine "Breakwater Design"; Dr. M. Gurinsky, Pro-
and geotechnical engineers and a text for ject Engineer for Hardesty & Hanover Con-
graduate students and others seeking to sulting Engineers, New York, N.Y., who
enter the field of marine engineering. contributed Section 6.8, "Sheet Pile Bulk-
Preface xxiii

heads Built Upon Creep Soils"; Mr. D. ject and extend my deepest gratitude to Mr.
Weinreb and Mr. P. Wu, both vice presi- M. Shiono and Mr. Ed Patrick of Sumitomo
dents for Reinforced Earth Company, Ltd., Canada for their support. I also wish to
Rexdale, Ontario, who contributed Section thank my publisher, Chapman & Hall, for
5.10, "Reinforced Earth Quay Walls. full cooperation.
I extend my deepest gratitude to my good Finally, special thanks are due to my
friend Roman Glusman for his invaluable wife Nora, for her valuable assistance dur-
help with preparation of some illustrations ing preparation of the manuscript copy, but
and Ms. L. Dunn, who typed the manuscript most important, for patiently tolerating me
and dealt ably with many difficulties in the during the preparation of both volumes.
process. I wish to thank Sumitomo Rubber
Industries, Ltd. for sponsorship of this pro- GREGORY P. TSINKER
Introduction

Ports- Their Past. Present. and Future

Maritime transportation has generally been traffic increased, the existing river ports
the most convenient and least expensive became overcrowded, and in order to permit
means of transporting goods, and this is more ships to berth and at the same time to
why mankind, since ancient times, has been keep the river usable for more ships, piers
steadily extending its activities into this had to be constructed along river banks.
area. This stage may be seen as the beginning
The history of maritime transportation of the development of modern ports. The
and port development dates back to the ever-increasing demand for shipping and
year 3500 B.C. and beyond. Over centuries, port facilities resulted in construction of the
transport of goods by means of water trans- first open-sea ports. Four to five thousand
portation has been evolved in steps with the years ago the Phoenicians established open-
needs of world trade and technical capabili- sea ports along the Mediterranean coast-
ties to build larger ships and ship I cargo line, and the Romans built the famous naval
handling facilities. port near Rome on the Tiber River at Ostia.
Initially, waterborne traffic has existed By the end of the first century A.D. a num-
on a local basis where small ships sailed out ber of large ports had been constructed in
of river ports for other nearby river ports the Mediterranean, the Red Sea, and the
located in the same river system. With ad- Persian Gulf.
vancing navigational skills the merchants Unfortunately, many of these old ports
ventured greater and greater distances. and harbors have disappeared, either being
Thus, larger ships transporting larger destroyed during the wars, buried by earth-
quantities of goods have emerged. As ship quakes, or just through neglect and deca-

XXV
xxvi Ports-Their Past, Present, and Future

dence. Some of these ports are known from methods and technologies for handling and
old documents and others have been discov- hauling of miscellaneous cargoes. However,
ered by archaeologists. As pointed out by in the 1880s many kinds of cargos were still
DuPlat Taylor (1949), these ports have been handled manually as it had been for cen-
well planned and effectively executed. turies. Cargoes that consisted basically of
The description of different methods of bags, bales, bundles, barrels, cases, cartons,
port construction in earlier centuries, drums, pieces of timber, steel, and so forth
specifically in the Roman Empire, that in- have been moved manually in the ship, on
clude the use of tongue and grooved, lami- the quay, in the shed, and in the ware-
nated, and various other types of sheet pil- house, sometimes ''humped" on the back.
ing, and large stone blocks are found in This created a heavy demand for labor
treaties by Roman architects (Leimdorfer, which fluctuated greatly with the arrival
1979). and departure of ships. Thus, if the ships
As ship navigators developed more skill were to be turned round efficiently and eco-
and fears of unknown waters gradually dis- nomically, a big pool of casual laborers was
appeared, merchant mariners, in addition necessary.
to trade between river ports on their own In the late 1880s, ships still continued
coasts, started sailing the high seas, bring- their transition from sail to steam engine.
ing goods from country to country and from The capacity of these vessels was a few
continent to continent. The interchange of thousand tonnes and their draft less than
goods and later of raw materials between 6 m. As ships changed, so did the ports that
countries and continents reached by mar- served them.
itime traffic as well as the development of In many ports, the finger pier was the
powerful navy fleets brought about develop- most characteristic type of berth construc-
ment of large sea ports; this subsequently tion. Typically, goods were stored there in
gave birth to large cities built around these warehouses located in close proximity to
ports. the berth line and were taken in and out of
Many modern cities have been built and port by horse and cart.
expanded around medieval ports located on The shift to mechanized handling of car-
the open sea, bays, estuaries, sounds, and goes in ports began in the early 1900s. This
rivers. Examples are London, Rotterdam, was largely dictated by the growing volume
Hamburg, and many others. However, it of maritime traffic and changing size of
was not until the 1880s that a revival of ships. By the 1920s most of the general
interest in port works reappeared. cargo ships were using onboard booms to
Port developments and their evolution move cargo by the sling-load method. The
has started, motivated by both economic trend toward the growth of ship size coin-
and technological pressures that resulted cided with construction of vessels special-
from the global industrial revolution. At this ized in transporting a certain type of cargo
time, the size, diversity, and complexity of (e.g., general purpose commodities, dry and
ports changed dramatically. To a great ex- liquid bulk cargoes, and others). Naturally,
tent this have been influenced by the new developments in a ship industry in-
changing nature of ships, for example, the evitably brought about innovations in cargo
transition from ships made from wood to handling and hauling technologies, the most
steel and the introduction and rapid devel- radical of which was introduction of a quay
opment of steamboats, and by the demand edge cranes.
that greater volumes of cargo be handled at World War 11-inspired inventiveness took
ports more rapidly. The latter stimulated mechanization of cargo handling one step
the development of more and more efficient forward by the introduction offorklift trucks
Ports-Their Past, Present, and Future xxvii

and pallets that enabled the general-pur- developed and used extensively. This
pose cargoes to be moved faster. In postwar method allows containers, but also cars,
years, pallets have been standardized inter- tracks, trains, and so forth, to roll on ships
nationally by International Standard Orga- via large stern or side ramps.
nization (ISO) (1992). The ISO Committee The advent of containers completely
stipulated that the maximum permissible overshadowed cargo handling on pallets.
width of road vehicles (then mostly flat and The introduction of container systems for
open) must be about 245 em; thus, all stan- transporting goods revolutionized sea and
dard pallets had one dimension, of which land transport and cargo handling methods;
245 em was a multiple, so that they could ship turnaround time was reduced spectac-
be stowed across open vehicles without ularly and speed, efficiency, and safety of
wasting space. The use of forklift vehicles handling all types of containerized cargoes
and pallets was very rapidly developed in increased dramatically. This new technol-
industry all over the world, and palletized ogy drastically changed the approach to port
loads occupied the steadily increasing vol- planning. In most ports, a previously very
ume of ship holds. This, however, changed effective pier system was disused as general
drastically in the not too distant future. cargo operations have been moved to the
The shift to new technologies occurred in usually remote, high-volume container fa-
the 1950s with the introduction of container cilities with their large paved container
ships built to transport large freight con- storage areas and relatively few berths.
tainers. These modern specialized ports and termi-
Containers were soon standardized by nals tie directly into upland staging areas
ISO internationally to 20 or 40 ft in length (marshalling yards) with multimodal links
(6.06 m or 12.19 m) with the outside width to several cities, a region, or the entire
and height being 8 ft (2.44 m). At the pres- country.
ent time, the empty weight of a modern Traditionally, ports have been developed
20-ft container ranges from 19 to 22 kN in natural habors and, as mentioned ear-
with maximum permitted total weight of lier, have formed the nuclei for many cities.
240 kN. The empty weight of a 40-ft con- Today, ports and marine terminals are built
tainer ranges from 28 to 36 kN with a wherever they can be economically justified.
maximum permitted total weight of305 kN. The need for large open areas to accommo-
Initially, containers have been handled date a modern container facility has in-
by conventional quay edge cranes. The first duced ports to move to the periphery of
specially designed container crane was in- cities and often on poor quality land. The
troduced in 1959, and over the last 30 years, latter usually presents a challenge to port
container handling cranes have grown. in designers and has been an area of major
size and handling capacity. The need for controversy related mostly to dredging and
efficient handling of containers stimulated disposal of the contaminated dredged soils.
the development of new equipment, such as Alternatives to dredging have been found in
straddle carriers, heavy lift forklift trucks, constructing offshore island ports and mov-
gantry cranes, special tractors, and others. ing the up-river shallow draft ports down
Older forklift trucks used for handling gen- river, to deeper waters.
eral cargo and pallets had lifting capacities Dramatic changes have also occurred in
of 30-80 kN. In contrast, today's new fork- the handling of liquid and dry bulk cargoes.
lifts for container handling have capacities Movements of liquid bulk petroleum prod-
up to 450 kN. ucts by ship started in the 1880s when
During the last 25 years, the roll-onjroll- special tanks were mounted onto existing
offmethod of handling containers have been vessels. Prior to this, the only means of
xxviii Ports-Their Past, Present, and Future

moving liquids was in barrels, which was needed. In some instances the terminals
not a very efficient way of transporting have been moved as far as 2 km or more
ever-increasing quantities as mankind offshore and have been linked to the shore
moved into the petroleum age. either by a bridgelike trestle, designed to
Since the introduction of the first tank- support pipe lines or conveyor systems and
mounted vessels, the procedure and method to provide access to the terminal for
of liquid bulk handling has not changed in lightweight vehicular traffic or by subma-
principle 100 years later; however, techni- rine pipelines. In some instances, particu-
cal improvements in this area have been larly in heavily populated areas where local
spectacular. The capacity of liquid bulk car- residents object to the construction of con-
riers (tankers) in the 1940s has reached ventional trestles as an unacceptable "vis-
22,000 DWT and at the present time ual pollution," submarine tunnels have been
500,000-DWT tankers ply the oceans. To- constructed as a solution to the problem.
day, several shipyards in Europe and Asia The modern port is developed as an im-
have the capacity to build 1,000,000-DWT portant link in a total transportation sys-
tankers. tem and planners of such multimodal sys-
Similar developments have occurred in tems seek to optimize the total network, not
the transportation of dry bulk materials. just one of its components.
Bulk carriers have lagged but tracked Construction of a new port, or expansion
tanker growth, and, similarly, tankers may or modernization of an existing one, is usu-
also be expected to grow in the future. The ally carried out to increase port capacity
use oflarge and very large deep draft ships and its effectiveness. Traditionally, this has
for transporting liquid and dry bulk materi- been focused on the sea, and, consequently,
als and the material hauling innovations construction of new berths and moderniza-
have changed the nature of the modern tion and expansion of existing ones was the
port. Ports actually become a highly special- prime area of interest. However, as urban
ized terminals able to handle the one spe- coastal areas, particularly in developed
cific cargo at very high rates; for example, countries, have substantially expanded over
loading of up to 20,000 tonnesjh and more the last five decades, while concurrently in-
of dry bulk, and 220,000 m 3 of crude oil per ternational trade has increased and con-
day; thus, annual throughput of tens of mil- tinue to expand, making the world more
lions of tonnes has been achieved. and more economically interdependent, the
Deep draft vessels need deep water ports. port land-side capacity to transfer the cargo
It has been learned, however, that the con- from the wharf to the end user has become
ventional approach to construction of such increasingly critical. In some densely popu-
ports, involving dredging of large quantities lated areas, the available transportation
of sometimes contaminated sediments, can network (e.g., highway and rail) is limited
be prohibitively expensive. The solution has to moving a certain amount of cargo and
been found in the construction of offshore cannot be expanded further. Under these
marine facilities not protected from the ef- conditions there is no logic in increasing the
fects of environmental forces such as waves existing port capacity, unless the land-side
and currents. At these facilities, the low transportation infrastructure is equally ca-
berth occupancy due to rough sea condi- pable of moving the increased volume of
tions has been compensated for by a very cargo through the land-based transporta-
high rate of material handling on calm days. tion network.
These facilities have been constructed far In this respect, to avoid a waterfront con-
enough offshore where sufficiently deep wa- flict, many countries have developed a mas-
ter is found and no maintenance dredging is ter plan for its major ports. For example, in
Ports-Their Past, Present, and Future xxix

Canada, both the Canada Ports Corporation sist of 75% manufactured goods as opposed
(a Canadian Crown Corporation) and indi- to export of raw materials and this trend is
vidual ports have developed land use plans characteristic of most countries in South
and economic impact assessment in cooper- East Asia. Wider, nonstandard container
ation with the cities and local special-inter- are already the reality, and the trend for
est groups which interface with port activi- use of larger containers will continue. New
ties (Gaudreault, 1989). container sizes will inevitably have great
In the view of many experts, moderniza- impact on the design of new containerships
tion of existing ports will continue and many and the container handling technologies.
new ports will be developed in the 1990s The most radical and far-reaching
and beyond due to major expansion of the changes in container terminal technology
world economy. The latter is result of dra- will be their continuing automation which
matic growth of the world population, gen- will lower manpower requirements and op-
eral industrial growth, and growth of erating costs, increase control, and speed
petroleum and mineral material industries. the flow of goods through the ports. Future
The real value of world trade will con- terminals will be more flexible and easily
tinue to grow. The new North American adaptable to changes in the world's econ-
Free Trade Agreement (NAFTA) between omy resulting in cyclic changes in demands.
the United States, Canada, and Mexico and Alternatives to dredging and construc-
new improved GATT treaty arrangements
tion of deep water ports will be design of
are expected to encourage similar agree-
wider vessels with lesser drafts. The impor-
ments elsewhere in the world that will
tant new development in port operations
eventually result in the total world market
that occur about three decades ago was the
free from protectionism and · immune to
virtual disappearance of the true passenger
destabilizing political upheaval in some re-
liners; the cruise industry has emerged to
gions. As pointed out by Barker (1990/1991)
in future trade, replace it. Today, the passenger trade exists
basically on local lines on inland waterways
and between coastal ports .
. . . the exporters will seek to increase the
Cruise vessels, to date, have continued to
added value of their trade, which will tend to
reduce tonnages of raw commodities and in- look like liner vessels, notwithstanding that
crease those in partially or fully processed speed is no longer of paramount impor-
materials. Thus increasingly refined tance. Cushing (1989) predicts that cruise
petroleum products, chemicals, alumina or vessels will change; they will become larger,
aluminum ingots or aluminum products, steel slower moving floating resorts.
products, vehicles, sawn and processed tim- The new ideas in port operations will
ber products, processed agricultural commodi- bring new engineering and construction
ties and such-like will be the cargoes rather ideas in their wake. As pointed out by
than the basic raw commodities. These are
more valuable and readily damaged cargoes
Hochstein (1992), the nontechnical aspects
which require more careful handling and stor- of port performance, such as commercializa-
age. The trend to more specialized vessels tion, liberalization, and privatization, along
such as reefer, parcel tanker, car carriers and with improvements of port administration
roll-onjroll-off will continue and many of the will continue to contribute to the institu-
processed cargoes will end up being handled tional restructuring and drastic improve-
in containers. ments in port operations.
Commercialization gives to the port au-
This trend is already quite visible in Asia. thorities freedom similar to the private sec-
For example, today, Thailand's exports con- tor where decision making is decentralized
xxx Ports-Their Past, Present, and Future

and management is held accountable for erate in the 1990s. Also, today it is most
port performance. apparent in Asia.
Liberalization lessens the port authori- Improvement of port administration en-
ties' monopoly on power by allowing the compasses actions that improve the perfor-
private sector to provide similar services mance of the organization. It may include
and is complementary to commercialization. corporate planning and carrier develop-
Privatization transfers functions previ- ment, as well as installation and constant
ously performed by the port authority modernization of a computerized manage-
(government) to private sector. It may in- ment information systems that enhances
volve transfer of full or partial ownership of management without changing the port's
port facilities, or it may be limited to pri- institutional structure.
vate sector management practices for the Recent experience in ports worldwide
provision of port services via lease and op- suggest that commercialization and privati-
erating contracts. Privatization usually as- zation are the most far-reaching and effec-
sociates with eliminated subsidies and re- tive strategies to achieve the objective of
duced costs of port operation. port effectiveness.
The privatization of ports brings greatly In conclusion to this section it should be
enhanced commercial freedom to port man- noted that the future is not possible with-
agers and is recognized as one way to react out thorough familiarity with the past which
more positively to market opportunities and is a true foundation for new ideas. As Tooth
use human incentives, based on personal . (1989) rightfully said, "The past is not just
gains and improvements, to increase effi- something out of date, it is a record of
human experience-an experience is cer-
ciency of the port operations. This trend is
tainly something which should be used to
natural and will continue both in developed
help shape the future."
and developing countries.
For more information on port develop-
It must be recognized, however, that
ments the reader is referred to Dally (1981),
ports and harbors are built and operate
Cushing (1989), Clearwater (1992), Thom-
within a certain societal framework that
son (1992), and PIANC (1987, 1989, 1990).
includes an array of political, financial, en-
vironmental, and other considerations.
Therefore, the port planner must have a Engineering Advancement in Port
clear understanding of local technical and Design and construction
nontechnical issues. For example, in some The primary construction materials that
developing countries, existing or new ports were used for construction of older marine
are not solely an industrial development facilities were wood and stone. They were
but also enterprises aimed at solving some worked by hand and used for construction
regional, social, or demographical problems. of sheet-pile bulkheads, piled piers, quay
Therefore, a careful approach to port priva- walls, breakwaters, and other structures.
tization is needed in developing countries. Wooden sheet piles and regular piles were
The latter assumes that although commer- driven by using primitive power equipment,
cial spirit there must not be discouraged, and stone was placed from crude construc-
the commercial approach to port operations tion platforms.
in some developing countries should not More than five centuries ago the Phoeni-
pursue a short-term financial gain. cians extensively used wooden sheet piles
In Europe, port privatization has been and piles for construction of their marine
successfully introduced in the 1980s in the facilities. For sheet-piling they used long
United Kingdom where it is likely to accel- planks made from Lebanon cedar. Various
Ports-Their Past, Present, and Future :xxxi

types of timber sheet-piling techniques (e.g., destroy wood. The same is true for wooden
tongue and groove, laminated, and others) piles.
were used. These piles were driven succes- Cast-iron piles became complementary to
sively edge to edge to form a vertical wall timber piling in the early 1800s (Borthwick,
for the purpose of preventing the retained 1936). The earliest reported use of iron sheet
materials from spreading and from being piles was construction of the North Pier of
undermined by the action of waves and cur- Bridlington harbor, the United Kingdom, in
rent. This type of construction was also the early 1820s (Mackley, 1977). Various
known to ancient Egyptians and Romans. types of iron sheet-pile sections were avail-
The Phoenicians also used heavy blocks able at that time, and a considerable
locked together with copper dowels for con- amount of exploratory work was carried out
struction of the open-sea port at Tyre. This in order to develop the most economical
type of construction was also used by the profile. Cast iron, however, as a material
Romans. In the 1800s, both materials still had limitations primarily because of its vul-
played a major role in port construction. nerability to brittle fracture during driving
A great variety of gravity-type walls con- in hard soils. Typically, wrought-iron piles
structed from rubble masonry or heavy were used in a composite riveted form and
granite or limestone blocks have been built were based primarily on the fitting of plates
during the 1880s. The history of heavy between suitable guides or against sup-
blockwork construction is traced back to ports.
ports in Mediterranean, at Marseilles and In 1897 a Danish engineer, Larssen, rev-
Algiers, with much of this pioneering work olutionized the use of iron sheet piles by
being carried out at the Port-of-Bonqie, Al- introducing a new pile section which was
geria where a quay wall composed C!f lime- developed from a rolled trough section plus
stone blocks had been built as far back as a riveted "z" section, to form an interlock;
1840. During the same period of time, wood this shape is very familiar in modern con-
was extensively used for the construction of struction.
piled wharves and piers, as well as for grav- In 1914 Larssen also introduced the first
ity-type quays comprised of floated-in tim- deep-arch section in which interlocks were
ber cribs with or without a masonry super- situated in the neutral axis of the complete
structure. section; thus, their material bulk did not
It should be noted that improved and influence the bending moment to be taken
economically sound blockwork quay walls up.
are still in use. Details are provided in Larssen's inventions and modifications
Chapter 5. Today, timber cribs are used helped to greatly increase the capacity and
where wood is in abundance, and timber effectiveness of sheet piles in their ability to
sheet-pile bulkheads made from treated resist earth and water pressures. Increased
wood have been constructed elsewhere, par- pile stiffness enabled it to be driven without
ticularly in coastal regions as a secondary buckling or springing under the blows of
line of shore defense in ocean-exposed loca- the driving hammer, increased water-tight-
tions and for construction of low-height ma- ness of the sheeting prevented seepage
rine structures in small-craft harbors. through the wall, and, most importantly,
Well-treated timber sheeting is also em- efficient use of rolling mills produced an
ployed in permanent structures where it is economical section with interlocks.
always hidden under water, thus preserved At the beginning of this century, both
from rot, and at locations where there is no wood and iron have been replaced by steel
marine organisms (e.g., borers) which can and reinforced concrete. The first sheet pile
xxxii Ports-Their Past, Present, and Future

ever made from a rolled section was used in crete in the tension zone is thereby largely
Chicago in 1901; it was called the Jackson eliminated and the danger of corrosion of
pile. This was followed by the rather fast reinforcement is decreased.
development of numerous straight or trough The same applies to regular concrete piles
sections of steel piles that were produced that are extensively used in marine applica-
around the world, either with integral lock- tion. The advantage of piled structures is
ing arrangements or with a separate inter- that they enable practically free passage of
locking member. waves, which makes them particularly at-
The first steel "z"-shaped sheet piling tractive for construction of the deep water
known as the Hoesch system was intro- offshore terminals.
duced in Belgium in 1913. In comparison A great variety of concrete and steel piles
with other sections, this type of piling was have been developed and used from the
stiffer and had a higher section modulus at beginning of this century. The wide variety
equal weight with other piling systems. of these piles is discussed in Chapter 7.
Since this time, various combinations of The most significant development in this
different piling systems and various types area is the use of large-diameter (up to
of box piles (H-section) have been intro- 3.0 m and more), very long (60 m and longer)
duced in North America and Europe. At prestressed concrete and steel cylindrical
present, a variety of high-strength sheet- piles. Depending on geotechnical site condi-
pile sections are available from different tions, these piles can be installed by differ-
pile manufacturers. ent methods (e.g., driven by hammer, vibra-
In addition to the above-mentioned piles, tor, hydraulically, or a combination of some
relatively low sectional modulus straight of these). It should be pointed out that the
web steel sheet piles are often used in ma- vibratory hammers introduced in the fifties
rine application for construction of and sixties changed the basic way piles had
cellular-type bulkheads. These piles were been driven since the late 1800s. New hy-
first manufactured and used in the United draulically operated hammers enable the
States in 1908/1909. constructors to drive piles under water.
It should be noted that steel pipe-type Similarly, very large floating and jack-up
sheet piles are now coming into widespread pile-driving equipment was developed and
use for deep water construction where a used for pile installation at exposed off-
sheeting of greater strength is required. shore locations. An array of piles with en-
Reinforced concrete sheet piles have been hanced bearing capacities have been devel-
used in harbor construction since the begin- oped; for example, screw piles of different
ning of this century. They are usually con- designs, prefabricated piles with enlarge-
sidered relatively maintenance-free compo- ments on their shafts, belled piles, and other
nents of a sheet-pile wall. Although many have been s~ccessfully used in port and
different design types have been developed offshore construction.
and used in the past 50 or so years, the High-strength steel and prestressed con-
straight web piling bar provided with a crete allow the port designers and construc-
tongue and groove, similar to that used on tors the flexibility to design for greater
timber piles, is the most commonly used. depth, longer spans and higher capacity.
Since the 1950s, prestressed concrete sheet For example, general cargo wharves are now
piles have replaced almost completely the routinely designed for 5.0 tonnesjm2, up
ones made from regular reinforced concrete. from 2.0 tonnesjm 2 60-70 years ago.
Prestressing of concrete sheet-pile rein- Galvanization and the use of protective
forcement has an advantage, especially in coatings, such as epoxy, which came into
seawater environment, as cracking of con- use in the 1950s increased the longevity of
Ports-Their Past, Present, and Future xxxiii

marine structures. Epoxies are also used by An economical solution to gravity wall
constructors for splicing concrete structural construction in the wet was also achieved
elements in the field. by use of the traditional structures such as
Remarkable progress has been achieved a large-diameter steel sheet-pile cell, a con-
in concrete technology. Today's structural ventional boxlike floated-in concrete cais-
concrete is indeed a mixture of admixtures. son, and blockwork walls of enhanced de-
Products such as superplasticizers, re- signs.
tarders, accelerators, air entrainers, and The economy of gravity wall construction
others allow concrete pours in cold temper- was significantly improved by using innova-
ature or hot weather. A denser and higher tive methods of bed preparation and place-
quality product is obtained by use of silica ment and densification of backfill materials.
fume in a slurry or powdered form. Silica Detailed information on this subject is given
fume can substantially increase strength in Chapter 5.
and density of concrete and make it virtu- Waterfront construction was accelerated
ally impervious to chloride penetration in by the use of prefabricated structural com-
the harsh marine environment. Further- ponents. Prefabrication dramatically re-
more, use of corrosion inhibitors may slow duced the time required for overwater con-
down the potential onset of corrosion in struction and enhanced quality and there-
steel reinforcing bars. fore longevity of marine structures.
Development of huge floating heavy lift The design of any project is a continuous
equipment revolutionized the construction process, which begins with the perception of
a need or opportunity, followed by a feasi-
of gravity-type quay walls and breakwaters.
bility study that usually includes a concep-
These structures, generally built in water
tual design, and embedded by the detail
depths of 6-9 m in the late 1800s, are now
design. The latter is followed by the con-
constructed at depths of 25 m and below.
struction of the project with subsequent
Finally, it should be pointed out that in
commissioning. Furthermore, where re-
the past 30-40 years primitive fenders used quired to support the basic concept of the
for protection of marine structures from ship project (e.g., harbor layout) or permit inno-
impact have been replaced by very efficient vative structural designs to be used with
high-energy-absorbing and low-reaction- confidence, research is undertaken.
force rubber fender systems. At the present In the past, port and its related marine
time, fender units are manufactured from structures have been designed with a high
solid a,nd laminated rubber in different degree of redundancy, largely because of
shapes and sizes. They are also manufac- the relatively rapid deterioration rate of
tured in the form of a low-pressure inflated structural materials in the marine environ-
balloon (pneumatic fenders), or as a closed- ment, but also due to a lack of proper
cell foam filled unit. The pneumatic fender understanding of wave mechanics, mecha-
units have been manufactured up to 4.0 m nisms of ship-structure andjor soil-struc-
in diameter and 12.0 m long. ture interaction. The latter was particularly
In modern marine engineering practice, true in designing a "flexible" structure such
a number of innovative and economical as sheet-pile bulkheads.
gravity-type quay walls have been em- In the past 50 years, substantial progress
ployed. Among them are concrete large- has been achieved in such areas as the
diameter floated-in caissons, bottomless development of new, much stronger and
concrete cylinders, and prefabricated L- more durable structural materials, the in-
shaped retaining walls of miscellaneous de- troduction of better construction tech-
signs. nology, and a better understanding of the
xxxiv Ports-Their Past, Present, and Future

process of soil-structure interaction. Thus, (CAD) open opportunities for transition


research and development directly or indi- from traditional two-dimensional (2-D) de-
rectly became the integral part of the de- sign to three-dimensional (3-D) design. This
sign process. has been made feasible by the rapid ad-
The design of port layout with its related vances in computer-graphics hardware
structures such as breakwaters, piers, and which now permits full 3-D CAD models to
quay walls normally involves a great num- be displayed quickly and effectively, with
ber of parameters that are generally consid- hidden lines removed or with shading; 3-D
ered to be far beyond the ability of purely computer models can be viewed from any
analytical methods to achieve the reliable angle and viewpoint, in orthographic, iso-
solution. This has been overcome by use of metric, perspective, or cutaway views. This
large-scale physical models, which appear helps to make. the project aesthetically more
to be a viable tool in solving a complex acceptable to the community and overcome
multiparameter problem. the public resistance to some projects re-
Today's engineers have abundant analyt- garded as a "visual pollution" to the area.
ical capability supported by computers. The The importance of aesthetic aspects of any
mathematical modeling of a complex phe- project now is fully recognized by the de-
nomenon such as the interaction between signers and developers, and the 3-D ap-
nature and engineering developments has proach to structural and civil design helps
become commonplace in the design process. to bridge the gap between art and science.
It enables the engineer to .closely predict Unfortunately, computerization of the
the behavior of the complex structure in design process has its own drawbacks; it
practice. provides not only the leading edge of tech-
For complex projects, both mathematical nology but have been also a major source of
and physical models are best used in combi- concern. The popularity of computers has
nation. However, despite the prediction of resulted in a flood of software, and on
model studies, the designer should exercise today's software market, there are nearly
the proper level of conservative engineering as many computer programs as there are
judgment which is dictated by the complex- researchers. Unfortunately, the quality of
ity of marine foundations, marine environ- some software presently available on the
ment, ship maneuvers, and so forth. market is questionable, and it would be of
Today, port and harbor engineering has great interest and perhaps shock to some
entered the electronic age. The computer when results of analyses of a certain struc-
vastly enhances an engineer's productivity tures are compared with the same input
and his or her opportunities for innovative data using different programs.
design. In addition to reducing the opportu- The difference in the software output
nities for making errors, the use of comput- happens because some developers of com-
ers drastically enhances engineering judg- puter software blindly rely solely on the
ment. By taking advantage of the speed of mathematical approach to solving the prob-
computer analysis, the engineer can explore lem and are ignorant about current state-
a number of design alternatives in a short of-the-art knowledge.
period of time. Today, computer hardware Sometimes problems with software (e.g.,
and software allow the engineer to see his incorrect sign convention) may cause a com-
project from all perspectives, investigate puter to subtract stresses when it must be
each detail, make changes by shaping as a added can be difficult to detect.
sculptor might, then, when all is finished, Growing reliance on computer-aided
have the calculations and drawings pro- analysis and design without adequate con-
duced. Computer-aided design and drafting trols on misuse can lead to structural fail-
Ports-Their Past, Present, and Future xxxv

ures and, subsequently, to "computer-aided the observation method is a departure from


liabilities," the term used by Backman the traditional design process in geotech-
(1993) in her interesting paper on a subject nicaljfoundation engineering, because it
matter. allows one to make a final decision on foun-
Recognizing this as a potential problem, dation design in the future, both during
the Committee on Practices to Reduce Fail- construction when uncertainty in founda-
ures, under the American Society of Civil tion soils became understood and during
Engineers Technical Council on Forensic facility operation. The latter is particularly
Engineering, is currently preparing a mono- important where long-term changes in
graph titled "Avoiding Failures Caused by soil-structure interaction are expected.
Misuse of Civil Engineering Software." The Several advancements in port engineer-
monograph, scheduled for publication in ing have been pointed out in this section
1996, will examine all computer-related is- and the reader will find much more else-
sues where misuse could result in cata- where in this book.
strophic failures, poor performance of facili- Finally, it should be noted that in the
ties, and poor solutions to problems in civil past in order to reduce the cost of a design,
engineering. attempts to standardize construction of ma-
Due to a worsening legal climate for rine structures have been made. It has been
practicing engineers, the designers are of- proved, however, that standard designs to
ten not willing to accept potential risk asso- meet various site conditions, in general, and
ciated with a more economical or innovative marine facilities, in particular, are not eco-
design. The conservative approach, which nomical. In general, it is because the cost of
limits innovations in design and construc- the waterfront structure is so high that it
tion practices, causing economical prob- would be false economy to attempt to re-
lems, has been especially visible in founda- duce design costs by limiting the scope of
tion engineering. This has been explained design studies.
by the uncertainties in soil-structure inter- In conclusion, it must be said that pro-
action and usually limited information on ducing a good, sound, and effective design
foundation soils available to the designer. is, of course, science; however, it is also an
In the past 35-40 years, this has been im- art. Just as an artist does, the designer
proved by the extensive research into must be imaginative in developing the con-
statistics and risk analysis, as applied to cept of his or her project and, as a scientist,
the field of geotechnical engineering. careful and meticulous in paying attention
Also, the observational (monitored deci- to all details.
sions) method, which provides the designer
with flexibility in the decision-making pro- REFERENCES
cess, has been introduced. BACKMAN, L., 1993. "Computer-Aided Liability."
Statistics provide procedures for obtain- ASCE Civil Engineering, June.
ing information from given quantitative BARKER, J., 1990/1991. "Ports in the 1990's and
measurements, which, in turn, permits Beyond." The Dock & Harbour Authority, De-
analysis of how the aforementioned listed cember/January.
uncertainties of soil and other parameters BORTHWICK, M.A., 1936. ''Memoir on the Use of
involved in soil-structure interaction may Cast Iron in Piling, Particularly at Brunswick
affect the design of the structure; risk anal- Wharf, Blackwall." Transaction Institute Civil
ysis is a set of decision-making procedures Engineers, Vol. 1.
dealing with difficult design circumstances, CLEARWATER, J. L., 1992. "Port Construction
where many components interact such that Since 1885: Evolving to Meet Changing
there is more than one mode of failure; and World." The Dock & Harbour Authority, May.
xxxvi Ports-Their Past, Present, and Future

CusHING, C. R., 1989. ''Vessels of the Future: A LEIMDORFER, R., 1979. "La Saga des Palplanches."
Naval Architect's Viewpoint." PIANC Bulletin PIANC Bulletin No. 34, Vol. III.
No. 66. MACKLEY, F. R., 1977. (Reported by Buckley, P.
DALLY, H. K, 1981. "The Effect of Development J. C.) "The History and Development of Sheet
in Cargo Handling on the Design of Terminal Piling." Proceedings Institution of Civil Engi-
Facilities." PIANC Proceedings XX:Vth neers, Part 1, No. 62, February, London.
Congress, Edinburgh.
PERMANENT INTERNATIONAL AsSOCIATION NAVIGA-
DuPLAT TAYLOR, F. M., 1949. "The Design, Con- TION CONGRESSES (PIANC), 1990. Twenty-
struction and Maintenance of Docks, Wharves Seventh PIANC Congress PCDC Panel Dis-
and Piers." Eyre & Spottiswoode, Ltd., Lon- cussion, Osaka, Japan.
don.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
GAUDREAULT, R., 1989. "Where Have We Been,
GATION CONGRESSES (PIANC), 1989. Panel 3:
Where Are We Going? A Canadian Perspec-
Future Marine Terminal Designs. Five papers
tive." Presentation to ASCE Specialty Confer-
on subject matter. PIANC Bulletin No. 67.
ence PORTS'89, Boston, Massachusetts.
HocHSTEIN, A., 1992. "Implications of Institu- PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
GATION CONGRESSES (PIANC), 1987. "Develop-
tional Changes on Port Performance in Asia
ment of Modem Marine Terminals." Supple-
and Latin America." ASCE Proceedings Spe-
ment to Bulletin No. 56.
cialty Conference PORTS'92, Seattle, Wash-
ington. THOMSON, B., 1992. "A New Era for British
INTERNATIONAL ORGANIZATION FOR STANDARDIZA- Ports." The Dock & Harbour Authority, June.
TION (ISO), Standards Handbook, 2nd edition, TOOTH, E. S., 1989. "A Glimpse of the Past." The
1992. Freight Containers, Switzerland. Dock & Harbour Authority, May.
contributors

Mr. M. Shiono Dr. M. Gurinsky, P.E.


President SRI Marine Servie Co. Ltd. Project Engineer
1-14, 4 chome Isogami-Dori Hardesty & Hanover Consulting Engineers
Chuo-Ku, Kobe 651, Japan 1501 Broadway
New York, N.Y. 10036
Dr. S. Takahashi
Chief Maritime Structures Lab
Port and Harbour Research Institute Dr. W. S. Dunbar
3-1-1 Nagase Yokosuka, Japan 239 Engineering Consultant
925 Leovista Ave.
Mr. P. L. Wu, P.E. and Mr. D. Weinreb, P.E. North Vancouver, British Columbia,
Vice Presidents Reinforced Earth Co. Ltd. Canada V7R 1R1
190 Attwell Dr., Suite 501
Rexdale, Ontario, Canada M9W 6H8

xxxvii
1
The Marine Environment
and its Effects on Port
Design and construction

1.1 INTRODUCTION possible." Most of all, high winds, heavy


waves, and strong currents sometimes make
a marine structural engineer's job a very
1.1.1 General challenging task.
The marine environment significantly
The primary purpose of this chapter is to affects port design and, in particular, its
provide a general background on the princi- layout and related marine structures. The
pal marine environmental factors such as layout of a modern port is usually con-
wind, currents, waves, and ice, and their cerned with geometry, depth, and configu-
potential effects on structures constructed ration of its exterior navigation channel(s),
in a sea, estuary, lagoon, or an inland wa- for example, fairway, approach channel, en-
terway. The information given in this chap- trance, as well as with its entrance chan-
ter has been drawn from the literature with nel(s) and the interior basin(s), turning
the particular aim to provide the reader basin(s), and berths. All the aforementioned
with an overview on the design and con- port elements are greatly affected by local
struction of harbor- (port-) related marine environmental conditions, as well as by the
structures. References to these works are sizes and number of ships calling at port.
provided for the reader interested in addi- This is discussed in detail in Chapter 2.
tional information. To prevent the adverse affects the ma-
The marine environment presents a rine environment has on port operations,
harsh and challenging setting for the port the aforementioned port elements need to
(harbor) designer. Problems associated with be protected against action of short- and
marine engineering, such as the presence of long-period waves, currents, ice, and silta-
soft bottom sediments, the persistently hot, tion. Lack of protection may not only cause
humid, climates responsible for high rates damage to vessels and quay installations,
of steel corrosion, or very cold, ice-infested but, most importantly, interfere with port
waters, sometimes make the construction of operations, causing delays, and economic
marine structures look like a "mission im- losses.

1
2 Port Design and Construction

It must be pointed out that the modern, basin(s), and in some instances by special
usually highly specialized ports, and mod- means, such as pneumatic or hydraulic
ern container, dry and liquid bulk vessels, breakwaters.
operate on a tight schedule; even a few Effective dissipation of the energy from
hours delay may cause a severe economic short waves can also be achieved by the use
loss. Insufficient protection of the harbor of sloping walls or permeable walls, like
against waves and currents can cause rubble mounds or perforated breakwaters.
interruption of vessel loading/uploading There are many examples in which the
operations. It can also bring about such occurrence of long waves severely hampers
inconveniences as breakage of moorings and the port's safety and operation. This is par-
excessive ship movements at the berth. ticularly characteristic of ports located at or
Waves and cross currents may interfere on an open coast.
with the maneuvering of a vessel when it Problems of long-wave activity are usu-
passes through the entrance, in and out of ally much more difficult to handle than that
the interior basin. For example, cross cur- of short waves. Practically, it is very diffi-
rents may be dangerous in an entrance cult, if not impossible, to avoid penetration
where the vessel has to cross these currents of long-period (from 30 s and up) waves into
and enter its bow into a sheltered area, a harbor.
while at the same time the stern is exposed Many rubble-mound breakwaters are in-
to the same current. This exerts a turning effective against the action of very long
moment on the vessel. Additionally, in in- waves; if breakwaters are required in areas
land ports the vessel may be exposed to exposed to long waves, then they should be
strong currents (river flow) in its berthing
as impermeable to wave action as possible.
and mooring procedures necessitating the
Generally, basins in the immediate vicinity
added provisions of the use of mooring
of the breakwater are not able to trap large
buoys, warping dolphins, quay-based
amounts of wave energy. Therefore, the
winches, in combination with an adequate
problem of resonance may also occur in the
fendering system, suitable to the type of
inner basin.
vessel to be handled.
In both cases of short and long waves,
When considering the wave action af-
fecting harbor operations, two different the best solution is usually found through
kinds of waves, the short-period wind wave hydraulic model studies based on actual
(period 3 to 30 s) and the long-period or wave data, as well as on theoretical and
seiche wave (period in excess of 16 min) are numerical methods suitable for investigat-
usually considered. Furthermore, the large ing both short and long waves. In recent
waves caused by a sailing vessel may also years, considerable progress has been
present a problem. It should be pointed out achieved in the field of mathematical mod-
that waves undergo considerable change eling concerning wave action, particularly
when traveling in from deep to shallow regarding long waves, currents, and sedi-
waters. Shoaling changes the wave parame- mentation. Some of the related software is
ters, such as geometry and period. Conse- proprietary; however, most of it is commer-
quently, harbor protection is a very impor- cially available. The success of these studies
tant task that should be carefully evaluated is based on the investigator's ability to ade-
in the port planning process. The adverse quately reproduce field data and properly
effects of short-period waves in harbors can model all hydraulic parameters included in
be controlled by restricting the width of the hydraulic model study. PIANC (1980)
entrance into the harbor, proper layout of strongly recommends that hydraulic model
breakwaters, and proper geometry of studies. based on adequate survey data be
Port Design and Construction 3

undertaken before finalizing any major har- In some particular cases (e.g., offshore
bor project. located terminals), wind, waves, and cur-
The marine environment elements cov- rents can constitute the major source of
ered in this chapter are related basically to loads affecting the marine structure. On the
the marine structures design and particu- other hand, at sheltered locations, effects
larly to the effects they may produce on produced by the aforementioned environ-
structures in terms of loadings and mate- mental sources are of secondary impor-
rial deterioration. tance, whereas operational loads, such as
Winds, waves, currents, and ice consti- the impact produced by the berthing ship,
tute the major sources of environmental mooring loads, and others, will govern the
loads, whereas tides and seasonal water- design of the structure.
level variations affect function and opera- Naturally, in cold climate regions, ice
tion of portjharbor related marine loading can be the major factor to consider
structures. Site-specific environmental con- in the structural design.
ditions may greatly affect the structure The marine structure designer's usual
longevity and durability due to effects of problem is to find the proper balance among
corrosion, fouling, and attack by marine or- environmental factors that affect loadings
ganisms. This, in turn, dictates selection of on the structure, the structure durability,
the appropriate structural materials. and functional and operational require-
Gaythwaite (1978 and 1981) offers a pic- ments, while maintaining safety, reliability,
torial summary of the environmental effects and economy of construction and operation.
on a marine structure (Fig. 1-1). In regions Because environmental conditions acting at
with cold climates, ice effects must also be a given time are usually the result of sev-
included in this picture. eral, interrelational phenomenas, the de-

Wind
veioc•tv ..
Icing
Design wave:
(height and period)
I I
~ Dec!ay Storm tide

---L--t--.---Hr--.~_,,_ ____,____M~·~H~.W~·~
NoJmtJI
tide
range
M.L.W.
Corrosion L.W.O.S.T.
Current Fouling
veioc1ty ..

v
Borer

Loose mud and debris


attrk .• ~position
Firm bottom ~~~cour

Figure 1-1. Effects of marine environment on marine structure. (From


Gaythwaite, 1981.)
4 Port Design and Construction

signer of a marine structure must recognize (1969), Freudenthal and Gaither (1969),
a practical relationship among the various Borgman (1963), and Bee (1973, 1979).
phenomena that may exist during In conclusion to this section it should be
the construction phase, as well as during noted that, recently, PIANC (1986) pro-
operation. posed a list of sea state parameters that are
Because of the probability of extreme en- suggested as a reference list aimed at pro-
vironmental events acting at the same time, moting uniformity in the use of symbols,
the marine structure must be designed to nomenclature and definitions. It includes
possess a sufficient redundancy to prevent a parameters and functions most frequently
catastrophic failure. Bearing in mind the used to define the sea state. However, this
uncertainty associated with the deteriora- proposed list is not intended to reject the
tion rate of a structure in the harsh marine past practice, but rather to establish a com-
environment, caution must be exercised mon base in order to be able to incorporate
when establishing structural parameters new contributions, which are made continu-
such as concrete mix, concrete cover over ally. Some of the parameters included in
reinforced steel, protective coatings, ca- this list represent fairly recent concepts,
thodic protection systems, and others. At whereas others have been used unchanged
present, a better understanding of environ- for many years. The latter ones, in particu-
mental loadings on marine structures, the lar, are essential to the widely accepted
great progress made in the development of design methods and guidelines in coastal
new structural and protective materials, and marine engineering.
and the availability of high-speed comput-
ers to aid analysis enable the engineer to
design the cost-effective marine structures.
Ultimately, the engineer must decide on
1.1.2 seawater and Fouling
whether the structure is adequate for its
Generally, seawater is viewed as a solution
purpose, based on the level of acceptable
containing a great number of elements in
risk versus safety and reliability; engineer-
different proportions. The typical composi-
ing judgment is required for this decision.
tion of seawater is given in Table 1-1. Ele-
While selecting the design values of envi-
ments in solution are found in ions, many of
ronmental loads, the designer must always which combine and precipitate as salts on
be aware that the design of a marine struc- evaporation of the water. Concentrations of
ture is not merely the process of stress these elements are typically given in parts
analysis. The prudent designer should al- per million (ppm) by weight, for example,
ways exercise the commonsense approach equivalent to milligrams per liter (mgjL).
to marine structure design and be on a The combined concentration of these ele-
somewhat conservative side while selecting ments in water or total amount of dissolved
environmental design loads. Practice indi- solids in water is defined as salinity. Salin-
cates that such an approach does not result ity can also be considered as the amount of
in an uneconomical design, but, quite the dissolved solids in a water sample in parts
contrary, it helps keep the structure con- per thousand (%o) by weight. The mean
struction and further operation within a salinity of seawater is approximately 35%o
reasonable budget. and varies typically from 31 %o to 38%o. It
The problems of risk and reliability ver- can be considerably less for nearshore
sus safety and economy has been investi- coastal waters because of the freshwater
gated by many workers and the interested effect, especially in the vicinity oflarge river
reader is referred to works by Marshall mouths.
Port Design and Construction 5

Table 1-1. The composition of seawater at 35%o salinity (in order of ascending
atomic weight)

Micrograms Micrograms
Element per Liter Element per Liter

Hydrogen 1.10 X 10 8 Molybdenum 10


Helium 0.0072 Ruthenium 0.0007
Lithium 170 Rhodium
Beryllium 0.0006 Palladium
Boron 4,450 Silver 0.28
Carbon (inorganic) 28,000 Cadmium 0.11
(dissolved organic) 500 Indium
Nitrogen (dissolved N 2 ) 15,500 Tin 0.81
(as N03, N02, NH!) 670 Antimony 0.33
Oxygen (dissolved 0 2 ) 6,000 Tellurium
(as H 2 0) 8.83 X 10 8 Iodine 64
Fluorine 1,300 Xenon 0.047
Neon 0.120 Cesium 0.30
Sodium 1.08 X 10 7 Barium 21
Magnesium 1.29 X 10 6 Lanthanum 0.0029
Aluminum 1 Cerium 0.0012
Silicon 2,900 Praesodymium 0.00064
Phosphorus 88 Neodymium 0.0028
Sulfur 9.04 x 10 5 Samarium 0.00045
Chlorine 1.94 X 10 7 Europium 0.0013
Argon 450 Gadolinium 0.00070
Potassium 3.92 X 10 5 Terbium 0.00014
Calcium 4.11 X 10 5 Dysprosium 0.00091
Scandium < 0.004 Holmium 0.00022
Titanium 1 Erbium 0.00087
Vanadium 1.9 Thulium 0.00017
Chromium 0.2 Ytterbium 0.00082
Manganese 0.4 Lutetium 0.00015
Iron 3.4 Hafnium < 0.008
Cobalt 0.39 Tantalum < 0.0025
Nickel 6.6 Tungsten < 0.001
Copper 0.9 Rhenium < 0.0084
Zinc 5 Osmium
Gallium 0.03 Iridium
Germanium 0.06 Platinum
Arsenic 2.6 Gold 0.011
Selenium 0.090 Mercury 0.15
Bromine 6.73 X 10 4 Thallium
Krypton 0.21 Lead 0.03
Rubidium 120 Bismuth 0.02
Strontium 8,100 Radium 1 X 10- 7
Yttrium 0.013 Thorium 0.0004
Zirconium 0.026 Protactinium 2 X 10-lO
Niobium 0.015 Uranium 3.2

Source: From Gaythwaite (1981).


6 Port Design and Construction

Seawater contains 11 major elements, of oxidization-reduction potential, and solidi-


which chlorine (1.9 X 10 7 mgjL) in the form fying ability.
of chloride ions (C1-) and sodium (1.08 X
10 7 mgjL) in the form of ions (Na+) are by
far the most abundant. The remaining nine
elements in ionic forms are, in order of where H+ is the concentration of the hydro-
decreasing quantity, sulfur as sulfate gen ion.
(SoJ- ), magnesium (Mg 2 +), calcium (Ca2 +), The H+ concentration of pure water is
potassium (K+), carbon as bicarbonate 10- 7 ; hence, its pH = 7. The pH scale ranges
(HCOg-), bromide (Br-), strontium (Sr+), from 0 to 14; a pH less than 7 is considered
boron (H 3 B0 3 ), and fluoride (F-). The ra- acidic, and greater than 7 is basic. Seawa-
tios of these major ions do not vary signifi- ter is slightly alkaline; the pH of seawater
cantly over the range of salinities normally typically varies from 8.0 to 8.4. Knowledge
observed in the seas. Therefore, by measur- of local pH is important to the marine
ing the concentration of any particular ion, structure designer because it helps him or
the water salinity can be obtained by multi- her assess the degree of biological activity
plying the value of the ion concentration by and potential corrosive effects on struc-
its constant. tures.
The chlorinity, or chloride ion concentra- Oxygen is present in the air immediately
tion, is typically used for determination of adjacent to the seawater-air interface and
water salinity (S). is also present in the water in the form of
entrapped air bubbles and dissolved oxy-
S(%o) = 0.03 + 1.805 chlorinity(%o) (1-1) gen. The latter plays an essential role in the
corrosion of steel in the sea environment,
The density of seawater (approximately whether the steel is exposed, coated, or en-
1025 kgjm 3 ) is not very different from that cased in concrete.
of freshwater (1000 kgjm 3 ). Water density Carbon dioxide (C0 2 ) and hydrogen sul-
typically increases with decreasing temper- fide (H 2 S) that are dissolved in seawater in
ature and with increasing salinity. varying degrees lower the pH of seawater,
Generally, for most engineering pur- depending on its temperature. In addition,
poses, water is considered an incompress- H 2 S may cause hydrogen embrittlement of
ible fluid. The error due to neglecting com- steel.
pressibility in most marine engineering cal- Entrapped bubbles of water vapor, as in
culations is negligible. foam, may collapse suddenly, leading to
The heat capacity of water is among the cavitation and erosion of both steel and con-
highest of all liquids and solids. This prop- crete structures. Silt and clay that are sus-
erty of water explains the absence of the pended in the water, usually in colloidal
extreme temperatures in ·the oceans as it form, render vision difficult because of tur-
helps to maintain uniform body tempera- bidity. Colloidal silt in freshwater will usu-
ture. Although the thermal conductivity of ally drop out of suspension upon encounter-
water is the highest of all liquids, it is still ing seawater due to reduced velocity. Fine
relatively low compared to that of many sand, silts, clays, gravel, and sometimes
other substances. even large stones may also be carried along
The thermal conductivity of seawater is with strong currents, or wave action.
slightly less than that of freshwater. Chem- Marine fouling is also part of the marine
ical reactions in seawater are governed environment. This is an accumulation of
by such chemical factors as acidity and various marine growths and animal organ-
alkalinity as indicated by pH, the isms on immersed and partly immersed
Port Design and Construction 7

surfaces of the marine structure. Fouling


may have a dramatic effect on the structure
performance; it may increase the weight of
the structure and, more importantly, in-
crease drag due to increased surface rough-
ness and projected area of the structural
members; due to increased roughness, the
drag coefficient (CD), used in Morrisin's
equation (consult Chapter 4), is usually in-
creased by 10-20%. In addition, fouling may
increase the rate of corrosion of metal parts
of the structure due to destruction of pro-
tective coatings, and oxygen concentration
cell effects in areas where certain organ-
isms, for example, barnacles, exist.
Fouling creates difficulties for inspection
and maintenance of the structure due to the
presence at times of quite thick and tena-
cious growths. For example, in warm wa-
ters, hard mussel fouling may be 300 mm or
more thick. This will significantly increase
the gravity load and drag force on a struc-
ture. The example of rapid fouling is illus-
trated in Figure 1-2. The flood control
structure depicted in this photograph is
made of steel. It was tested underwater in
Figure 1-2. Fouling of a flood control structure in
the Adriatic Sea (Golfo di Venezia) and was the Adriatic Sea (Golfo di Venezia)
covered by a solid layer of microorganisms
30 to 50 mm thick in only 1 year.
Experimental studies by Blumberg and Fouling density varies from 1.0 tjm 3 (soft
Rigg (1961) show that due to fouling the fouling) to 1.3 to 1.4 tjm 3 (t = metric
drag coefficient may be increased on the tonnes). Generally, information on the types
order of 50-70% for light to medium fouling and extent of fouling for many coastal areas
on a smooth 900-mm-diameter cylinder. of the world can be found in the U.S. Navy
Even in cold water the fouling effect may be World Atlas of Coastal Biological Fouling
substantial. Heaf (1979) presents data on (1970).
fouling of a North Sea oil platform that Marine growth on a structure is influ-
indicates the possibility of a significant in- enced by temperature, oxygen content, pH,
crease in both overall and local loads at- salinity, current, turbidity, and light. Sig-
tributed to a marine growth about 50 mm nificant growth usually takes place on the
thick. Heaf indicates that sometimes the upper part of a structure. The marine struc-
consequences of fouling can be even more ture designer must recognize the fouling
severe than the design wave load. phenomenon when either designing a new
Fortunately, most marine organisms structure or evaluating a load-carrying
have a specific gravity only slightly greater capacity of an in-service structure. ·
than that of the seawater itself; thus, they Some metals are reported to exhibit a
do not add an appreciable mass to the natural resistance to fouling (Efird, 1975)
structure. and some special paints are also reported to
8 Port Design and Construction

be very effective in the prevention of fouling different locations along the ocean coast be-
(Offshore Engineer, 1984). cause of secondary tidal waves or mass of
The marine environment has adverse ef- water moving around the earth. These are
fects on marine structures; most structural also influenced by the depth of shoaling
material deteriorates under attack by mis- water and the configuration of the coast.
cellaneous agents that are present in sea- The highest tides in the world occur in the
water. Of particular importance to the ma- Bay of Fundy, where a rise of approxi-
rine structures designer is the attack of mately 30.5 m has been recorded. Inland
marine organisms on wood. For a detailed and landlocked seas, such as the Mediter-
discussion of the effects of the marine envi- ranean, Black, and the Baltic, have less
ronment on marine structures, the reader is than 30 em of tide. The Great Lakes
referred to the work by Gaythwaite (1981) and the Caspian Sea are not noticeably
and Tsinker (1995). influenced.
Because of water variations, a standard
reference plane is required for the planning
and design of marine structures. Nautical
1.2 WATER-LEVEL VARIATIONS charts usually refer to mean low water
(MLW) which, according to the definition by
The range of water fluctuation is a very NAVFAC DM-26.2 (1982), is the average
important factor in the planning and design height of the low waters over a· 19-year
of marine facilities. The seawater level rises period (metonic cycle). For shorter periods
and falls in response to the gravitational of observations, corrections are applied to
attraction of the moon and sun. In inland eliminate known variations and to reduce
waterways the water level is basically a the results to the equivalent of a mean
function of seasonal water fluctuations. 19-year value. All low-water heights are in-
During new and full moons, when the sun, cluded in the average, where the type of
the earth, and the moon are in a particular tide is either semidiurnal or mixed. Only
line, the highest tide ranges occur; these lower low-water heights are included in the
are called spring tides. When the sun and average, where the type of tide is diurnal.
moon are approximately 90° apart, which is Thus, the mean low water in the latter case
at the first and third quarter of the moon, is the same as the mean lower low water.
the ranges are lower; these are called neap Sometimes MLW is referred to as the aver-
tides. age of all low waters over a tidal epoch for a
Because the lunar month is 1 day shorter semidiurnal tide, as on the U.S. East Coast.
than the solar month, the times of the tidal The mean lower low water (MLLW), ac-
events are constantly changing. Normally, cording to the NAVFAC reference above, is
the tidal cycle moves back by about 50 min the average height of the lower low waters
each day. For example, the time of high tide over a 19-year period. For shorter periods of
will be 50 min later tomorrow then it is observations, corrections are applied to
today. Typically, there are two tidal cycles eliminate known variations and to reduce
which have a significantly greater range the results to the equivalent of a mean
(higher high tide) than the other (lower low 19-year value. The MLLW is frequently ab-
tide). Some areas in the South Pacific have breviated to lower low water as MLW.
a prolonged high tide once each day, fol- Other sources refer to the MLLW as the
lowed by a low tide 12 h later. These tidal average of the lowest lows in areas subject
cycles appear to follow the sun; hence, the to diurnal or missed tides, such as the U.S.
peaks occur about the same time each day. Gulf Coast and West Coast. Local MLLW
Large differences in tidal range occur at references are useful for pilots of surface
Port Design and Construction 9

vessels in navigating nearshore regions, ap- times of slack water do not coincide exactly
proach channels, and harbors. It should be with the peaks of high and low water, be-
noted that all standard navigation charts cause the water continues to flow for some
indicate bathymetry (depth) relative to the period after a peak has been reached. For
local MLLW data. locations on the ocean coast there is usually
Usually the low water datum (LWD) is little difference between the time of high or
used as an approximation to the plane of low water and the beginning of ebb or flood
mean low water that has been adapted as current, but for locations in narrow chan-
the standard reference plane. nels, land-locked harbors, or along tidal
Admiralty charts often refer to the low- rivers, the time of slack water may differ by
est astronomical tides for soundings, which several hours from the time of high or low
is the lowest predicted tide over a long pe- water.
riod of time. Similarly, mean higher high The predicted times of slack water and
water (MHHW or MHW) is the average tidal current velocities are given in tidal
height of the higher high waters over a current tables published by the U.S. Coast
19-year period and the highest high wa- and Geodetic Survey; one for the Atlantic
ter is the highest water of the spring tides coast of North America, and another for the
on record. Pacific coast of North America and Asia.
The highest astronomical tide is des- The tidal cycles described above produce
ignated as HAT. A more useful datum for currents that are discussed further in
predicting tides and meteorologically in- Section 1.5.
duced surges is the local mean sea level Additional changes in the level of the sea
(MSL), which is defined as the arithmetic is caused by storm surges. Storm surges are
mean of all hourly water levels observed caused primarily by the effects of the wind
during the National Tidal Datum Epoch blowing for a long period of time in the
(NTDE). Over such a period, virtually all same direction. They can build up a water
tide-induced variations, and most meteoro- level by up to 1-3 m in height. They can
logically induced variations, are averaged also reduce the water level by the same
out, leaving only the long-term (secular) amount. This effect must also be considered
trend. in determining the design water level. This
In determining design water levels, the level is superimposed on the other tidal
average lowest and highest spring tide lev- levels to determine the required depth of
els averaged over a long time may be cited, navigating area and elevations of marine
designated the MLW and the MHW, respec- structures (e.g., breakwaters, wharves,
tively. piers, etc.). As a note, a secondary, less
For practical use, Tide Tables are readily influential change in the water level is
available for most ports of the world. The caused by barometric pressures. A low-
Admiralty Tide Tables cover major ports in pressure system may raise the water level
the United Kingdom and elsewhere, and, conversely, a high pressure system can
whereas the United States Coast and lower the water level.
Geodetic Survey lists the tides for major In determining the design water levels
ports in the United States and other parts one must be aware of the fact that the
of the world. When using tide data from the measurements taken worldwide indicate the
United States and Geodetic Survey listings overall rise in sea level relative to the land
one should be aware that they give the (U.S. National Research Council, 1987). The
times and heights of high and low waters, studies carried out worldwide indicated that
not the times of flood or slack water. The a minimum sea level by the year 2020 will
latter are periods of little or no flood. The be about 21 em higher than that in 1986;
10 Port Design and Construction

after the year 2000, the rate of rise is likely ciated with EHW, and an extreme low wa-
to increase dramatically because of global ter level is defined as ELW. Selection of the
warming effects, with a maximum increase design water levels usually depends on the
of nearly 60 em possibly by the year 2050. importance of the project.
All the above changes may produce extreme The entire range of design water levels in
water levels and thus an extreme high wa- front of the wharf must be considered in
ter (EHW) and an extreme low water (ELW). determining the seafloor and deck eleva-
Figure 1-3 illustrates the relationship be- tions and in calculating the environmental
tween various tidal datums, storm surge forces and soil pressures acting on the ma-
components, and the design water level rine structure, and mooring and ship
(DWL) for extreme events. impact loads.
Extreme water levels may result in over-
topping and flooding, increased hydrostatic
pressures and buoyancy effects, increased
soil pressures and drainage problems: it 1.3 WEATHER FACTORS
may also result in heavier berthing and
mooring loads due to aggravated approach Weather factors that affect ports and har-
and mooring conditions. bors operation are rain, snow, fog, spray,
A detailed explanation of all the above and atmospheric icing. Heavy rain, snow,
factors is beyond the scope of this book. For and dense fog are primarily a hazard to
detailed information on tides the reader is port and harbor operations because of their
referred to the U.S. Army Corps of Engi- limitations on visibility. At present, how-
neers' Shore Protection Manual (1984), ever, due to widespread use of radar and
Gaythwaite (1981), Reid (1991), and other sophisticated instrumentation, these
Partheniades (1992). (in most cases) no longer constitute as seri-
As noted earlier, water levels in inland ous a constraint as in the past. However,
ports are basically governed by seasonal rain can move quickly and intensely through
water fluctuations. Normally, the water a wide area and an appreciable amount of
level with a 95% return period is consid- water can be poured on port terminal areas
ered as a minimum low water level (MLW), in a very short period of time. If this water
and the water level with a 1% return period is not quickly removed from the terminal
is considered as a maximum high water surface, it may adversely affect port opera-
level (MHW). An extreme flood level is asso- tion. Hence, a proper drainage system must

Extreme HW
MHWS ~

Normill
Tide
Range
-L~~------------~H-
MLWS

II II
II
II II
II
_.U. .u- ll
Figure 1-3. Design water levels. (From Gaythwaite, 1990.)
Port Design and Construction 11

be provided, adequate to remove the rain- in the lower levels of the air mass. The
fall at its maximum rate. resulting temperature inversion inhibits
Snow presents an additional problem. turbulent mixing with the dryer air above.
Blowing snow is the most frequent cause of Beneath the inversion, the air is moist, and
poor visibility in a cold climate region. This when it moves over colder water or drifting
is particularly true when the ground is cov- sea ice, the excess moisture condenses to
ered with fresh, loose snow. At wind speeds form fog or mist.
as low as 10 kmjh, fresh snow begins to Fog is usually widespread over seas dur-
move or drift. At speeds of 20-30 kmjh, it ing the period when ice is melting. As the
is swept into the air in sufficient quantity to season advances, the fog becomes patchy,
reduce the visibility close the the ground. tending to be frequent and dense at the
Low visibility encountered with blowing edges of the drifting ice, and less frequent
snow makes navigation of vessels danger- over ice-free seas and over land.
ous and also inhibits land traffic. Blowing Spray is created when waves break
snow is difficult to forecast and it often against a vessel or a structure. The waves
occurs suddenly. hurl the spray into the air where it is accel-
In regions where snowfall is heavy, erated by the wind. Many tonnes of water
land-based structures must be designed to can be dumped onto a structure or vessel by
carry the design snow load. If not removed this process. Therefore, a proper drainage
expeditiously, it may accumulate and freeze. system is required when spray may be a
This may hamper the operation of cargo problem. It must be remembered, however,
handling and hauling equipment and the that in the regions with cold climates, drains
performance of the wharf fender system. can be effectively plugged by freezing.
Snow can be melted by spraying salt or In regions with a cold climate, spray or
removed by mechanical means. rain may result in ice buildup on vertical
For detailed information on snow effects and horizontal surfaces of a wharf structure
and snow loadings, the reader is referred to on the order of 0.3 m and up to 1.0 m and
Fenco Engineers (1984) and the latest is- more. Ice accretion on a structure can con-
sues of the National Building Codes, such siderably increase the areas exposed to wind
as the U.S. Uniform Building Code, Na- as well as the structure's weight and
tional Building Code of Canada, and others. buoyancy.
Fog reduces visibility and is a serious For structural design, ice density, within
problem. Fog is of two types: the summer- reason, can be estimated as equal to ap-
type and the winter-type fogs. The proximately 890-920 kgjm 3 . Where possi-
summer-type fog occurs when warm air ble, spray or rain (atmospheric) related
passes over colder water in the ocean or icing should be removed from the working
inland. This results in moisture condensa- surfaces as soon as practical. This can be
tion that forms low stratus clouds. The win- done by mechanical means andjor by salt-
ter type is the steam fog which occurs when water jetting. For more information on the
cold air travels over warmer water. The icing phenomenon and its impact on marine
winter fog is usually formed at the water structures the reader is referred to Tsinker
surface. Because the layer of fog tends to be (1995).
quite thick, it can effectively hamper vessel It must be noted that the weather factor
movements within the harbor and inhibit lightning must also be treated seriously.
ground vehicle traffic. Lightning is usually associated with storms
In the arctic and subarctic, evaporation and heavy rainfalls, especially a tropical
from areas of open water and saturated rain squall. The port related structures as
ground surfaces produces a further cooling well as cargo handling equipment, such as
12 Port Design and Construction

portal cranes and similar equipment, must Some remote areas, however, may suffer
be reliably grounded. from the lack of observation stations.
During the brief summer in the Arctic
Ocean the interaction between the cold air
mass over pack ice and the warmer air of
1.4 WIND the adjacent land mass can spin off a series
of intense, however localized, cyclonic
storms, creating strong winds and short
steep waves in open water.
1.4.1 General It should be pointed out that local topog-
raphy may modify wind patterns, resulting
Of all the factors affecting the port and in localized wind jets. At some sites, partic-
harbor operation, the wind is usually con- ularly in proximity to a high hillside or
sidered the most significant; it affects ves- canyons created by high mountains, local
sel entry into the port and berthing/depar- tunneling of the wind may be experienced.
ting maneuvers. This is particularly true This can impose direct wind velocity forces
for large vessels sailing high in the ballast against structures, cargo handling and
condition, container ships, and large ferries, hauling equipment, and ships. The latter
all with large areas exposed to wind. may obviously have a profound influence on
Wind, by definition, is a movement (pre- port operations.
dominantly horizontal) of air in response to Nearly every location in the world is sub-
differences in air pressure that are caused ject to what is termed the prevailing wind,
primarily by differential heating and cool- that is, a wind blowing from one general
ing. direction of the compass for a major portion
The ocean winds circulate clockwise of the year. Prevailing winds are not neces-
around permanent high-pressure areas in sarily the strongest winds. Very often, winds
the Northern Hemisphere and counter- of greater intensity, but that occur less fre-
clockwise in the Southern Hemisphere. In quently, come from other directions. The
the region south of the major continents, direction, frequency, and intensity of the
that is, the Indian Ocean and the South average winds at a particular location over
Pacific, the pressures are lower and thus a period of time are represented graphically
strong westerly winds prevail. In the tropi- by a wind rose (Fig. 1-4). A wind rose indi-
cal and subtropical zones, the extreme heat cates the percent frequency of wind of a
and the interface between atmosphere and given force from a given direction, for a
ocean create deep low pressures. This re- given month, season, year, and so on. This
sults in violent storms which, depending on information is very useful and is an impor-
their geographic localities, are known as tant consideration for port design and oper-
tropical cyclones, hurricanes, or typhoons. ation. It is, perhaps, less important for the
The occurrence of such storms is usually construction phase.
seasonal and, fortunately, somewhat infre- It should be pointed out that the nautical
quent. They are easily spotted by local ob- terminology for wind directions is some-
servation and sibilates, and their routes are what contrary to the wind rose; when a
usually predicted with reasonable accuracy. mariner speaks of a north wind, it means
Thus, storm alert reports to ships at sea, as that the wind is blowing from the north,
well as to the port and harbor operators, and when he speaks of an onshore or off-
are usually issued when the storm first de- shore wind, it means the wind is blowing
velops. Now, storm forecasting services are respectively from the sea toward the land or
available for all principal localities on earth. from the land toward the sea.
Port Design and Construction 13

Percent
calms-- 1

1-3
4-6
Mean 7-10
velocity 11-16
in knots 17-21
22-27
s >27

Figure 1-4. Wind rose.

The side of the structure or vessel facing and land-based structures must be de-
the direction from which the wind comes is signed to withstand these unusually heavy
called the windward side, and the opposite forces. Heavy wind may also affect water
side is called the leeward side. Generally levels in the harbor, raising or lowering the
the term lee is the sheltered side, away water level substantially in a matter of a
from the wind. However, in offshore termi- few hours. The rise and fall of the water
nology, a lee shore is not a sheltered shore level due to a heavy wind condition can be
but the shore toward which the wind is computed by using relevant guidelines or
blowing and, hence, a dangerous shore. by examining historical records for the
Seasonal winds which blow in one direc- area (Gaythwaite, 1981; Reid, 1991;
tion over part of the year and in the oppo- Partheniades, 1992). In hurricane areas,
site direction the remainder of the year are where destructive wind forces can be be-
termed monsoons. These winds are basi- yond practical design factors of safety, some
cally prevalent in the western Pacific. structures of secondary importance could be
It should be pointed out that wind is not designed as "sacrificial," or with sacrificial
steady, but blows in gusts. For example, the elements, for example, doors andjor win-
3-s gust may have one-third to one-half dows, to avoid complete collapse.
greater speed than the same storm wind Other wind effects, such as the genera-
averaged over 1 h. The force of a wind is tion of waves or the movement of sand in
classified in accordance with the Beaufort dune areas that may shoal the basin or the
Scale. entrance to the marina, must be considered.
In most regions, wind data are readily The former usually requires the construc-
available and wind loads on structures are tion of breakwaters, whereas the latter re-
specified by local building codes. Where quires the implementation of some sand
wind can reach hurricane strengths, water- stabilization measures such as land stabi-
14 Port Design and Construction

lization by planting grass or the construc- pose; for example, permanent structures
tion of sand fences. designed for long-term service should be
designed for extreme wind speeds associ-
ated with longer return periods, such as the
50- and 100-year storms. Temporary struc-
1.4.2 Wind Parameters tures can be designed for storms of much
shorter return periods.
The basic wind parameters that affect the Figure 1-5 illustrates an example of the
design of a port and port related structures long-term probability distributions of wind
are wind speed, gust factor, and force. speeds over the open ocean for several ar-
Wind Speed: The wind usually exhibits a eas of offshore oil drilling interest. In these
marked variation in speed. Hence, for de- curves, the maximum sustained wind is de-
sign purposes an appropriate averaging fined as the average of the maximum mea-
time interval is selected so that the wind sured wind speeds over a 1-minute interval.
data can be analyzed on a consistent basis. Near the coast and within harbors, extreme
The averaging interval must be long wind speeds are generally less than those
enough so that the recorded speed extends at offshore locations. Normally, the extreme
over the length of the structure and dis- wind speed can be found in airport meteoro-
counts peak velocities that are of shorter logical records, as well as in National Build-
duration than that structure's response ing Codes and other relevant sources.
time, and be short enough to obtain a real-
Recommendations for collecting and in-
istic maximum velocity from which forces
terpreting wind data at remote sites where
can be computed. Meteorologists usually
record data are scarce are given in PIANC
consider data collected over 30 years to be
(1979).
sufficient to arrive at representative aver-
Table 1-2 illustrates approximate ranges
ages. In many practical cases, however, the
of threshold wind velocities as they affect
30-year-span data are not available. In these
marine terminal operations.
cases, shorter periods of data collection are
used. Three-year wind observation data can Gust Factor: The term "gust" is usually
provide some valuable information, whereas assumed as a shorter averaging period of
1 year of data collection is regarded as an wind speed. The term also includes the con-
absolute minimum. Available data are usu- cept of a "squall," which is usually used for
ally analyzed on monthly or seasonal basis. mean wind speeds of approximately 1-5
Wind observation is usually carried out mins, while "gust" is commonly used for
within 30° intervals (i.e., a total of 12 sec- periods of a few seconds.
tors). The data obtained are statistically The gust factor ( C 2 ) is defined as the
manipulated and plotted as a wind rose ratio of short-period wind speed to the mean
(Fig. 1-4). Because wind velocity increases wind speed. It depends on temperature gra-
with height above ground owing to retarda- dient and altitude and increases with in-
tion of the lower layers by friction, a stan- creased elevation, roughly in accordance
dard reference height, which is 10 m above with the one-twelfth power of the altitude
ground, is normally considered. Reference [e.g., Cz/C10 = (Zj10)lf 12 , where Cz =gust
data on wind speed distribution over the speed at an elevation of Z (in meters) and
U.S. territory for 10-, 20-, 25-, 50-, and 100- C 10 = gust speed at an elevation of 10m].
year return periods in the form of charts Furthermore, the gust factor is depen-
are found in Thorn (1968). Selection of the dent on the "roughness" of the surface and
design return period depends on the impor- the stability of air masses. Generally, in
tance of the project, as well as on its pur- marine environments short-period gusts
Port Design and Construction 15

over the sea are associated with the wind at


higher elevations and atmospheric stability.
Bretschneider (1969) summarizes gust
0.4 r---.---.---.-~--.-----.----.
factors for various durations and various
mean hourly wind speeds as follows: For
mean hourly speeds from 20 to 80 knots
0.2 "•\ \

1\ \\\"-.~
o.q-- \
---ATLANTIC COAST
- · - NORTH SEA
- - - - GULF OF ALASKA 10
(approximately 10-40 mjs), the 1 min gust
\ •1
\ \
- - GULF OF MEXICO
averages 1.25 times the mean hourly speed,
\ \ ~\ whereas the 5- and 0.5-s (instantaneous)
\ \ .....
\ \\\. 25 average gust factors are 1.48 and 1.61, re-
\ \\ ..... spectively. It should be pointed out that
\\\ \ ..
\\\ ·,.. 50 short-duration gusts of 0.5-5 s are within
\" \ ·.
~\ ·•·.•. the range of the natural harmonic periods
,,\ ·····.... 100
0.01
of most structures. In most practical de-
signs for calculating the effects of wind
0.00460'~-=.,---::,00::-----,:::,:--~ ..:-.---,,:::,.----:-:,.,::----'
pressures, an average gust factor in a range
MAXIMUM SUSTAINED WINO IN KNOTS from 1.25 to 1.45 is considered. Detailed
information on gust and gust effects on
Figure 1-5. Long-term wind speed versus return
period for selected offshore regions. (From OCS Oil and
structure response to gusty winds is found
Gas-An Environmental Assessment, 1974.) in Davenport (1977) and other relevant lit-
erature.
Last but not least, the possibility of verti-
cal gusts must not be overlooked. At heights

Table 1-2. Wind speed versus operational criteria

Effect on Operations•
Beaufort Scale I Wind Speed
Seaman's Descriptions (knots) Vessel Facilities

0 Calm 0-1
1 Light air 1-3
2 Light breeze 4-6
3 Gentle breeze 7-10
4 Moderate breeze 11-16
5 Fresh breeze 17-21 i
6 Strong breeze 22-27 Berthing limit i
! Crane operations
cease!
7 Near gale 28-33 Tugboat
limit i
8 Fresh gale 34-40 Ferry operations Loading arms
cease disconnected
9 Strong gale 41-47 Emergency mooring lines !
10 Whole gale 48-55 Larger vessels Facilities
put to sea secured,
cranes lashed, etc.
11 Storm 56-63
12 Hurricane 64-71
• Because of wind alone, wave action at exposed locations may result in greater limitations.
Source: After Myers et a!. (1969).
16 Port Design and Construction

of 30 m or more above the surface, vertical potential for this condition must be care-
gusts can be of the same order of magnitude fully reviewed.
as horizontal gusts.
Drag (Shape) Coefficient: The total wind
force acting on a structure is associated
with its shape andjor orientation in rela- 1.5 CURRENTS
tion to wind direction. Drag coefficients vary
with aspect ratio, that is, the ratio of width Currents are essentially horizontal move-
to length, and Reynolds number NR = ments of water which even when small in
VD 1 v, where V is the wind velocity, D is magnitude may have a sigirificant impact
some characteristic dimension, for example on port construction and operation. Cur-
diameter, and v is the kinematic viscosity. rents influence vessel movement, may
Drag coefficients for standard structural change the wave characteristics, and create
shapes are found in National Building eddy patterns around structure, which, in
Codes, NAVFAC DM-2 (1970), ASCE (1961), turn, may lead to scour and erosion of soil.
the work of Simin and Scanlan (1978), and Currents may also cause vibration of free-
other literature. standing piles and result in ice and flotsam
For important structures and those (floating debris) movements around the area
where the consequences and potential for to hamper portjharhor operations. Strong
disaster are great, the drag coefficient is currents can also create an uplift force as
usually obtained from a wind tunnel test on they flow over a submerged structure, or
a scale model. produce a downpull load when they are
In general, to estimate the wind force on trapped beneath the base of a caisson dur-
a moored vessel the following approximate ing its installation. This may also result in
values of drag coefficient Cn can be consid- a scour of caisson natural on prepared bed.
ered: Furthermore, strong currents may scour a
seafloor around a pipeline while also sub-
jecting it to the uplift or downpull forces.
• Vessel hull 1.0
There are several different types of cur-
• Flat sides of houses and bridge 1.2
0.6-1.2
rents such as large-scale ocean surface cur-
• Booms, stacks, and rigging
rents caused by major wind systems, deep-
running sea bottom currents associated with
In most practical cases the shape factor oceanic circulation, more localized currents
K = 1.3 is used to account for drag and caused by tides, waves, and storm related
suction increase on the leeward side. wind stress, currents due to nearby river
Dynamic Effects: Slender and flexible discharge, underwater discharge from ma-
structures can be highly susceptible to aero- jor water retaining structures, and so forth.
dynamic forces generated by wind. The It should be pointed out that river cur-
forces are basically dependent on wind ve- rents, especially those of great rivers, ex-
locity, shape of the object, and the relation tend far out into the sea. Because the den-
of the structure's natural period in a partic- sity of the freshwater is less, and perhaps
ular mode of oscillation to the periodicity of because of the presence of silt, the water
the exciting force and the degree of damp- masses tend to persist for a long period
ing in the system. without mixing; thus, substantial surface
It should be noted that, in general, port currents may reach to considerable distance
related marine structures are not vulner- from shore. These currents may combine
able to wind related resonant conditions. with tidal currents to produce higher veloci-
However, in certain special situations, the ties on ebb and reduce velocities on flood.
Port Design and Construction 17

The worldwide ocean circulation pro- extended some distance offshore when
duces such currents as the Gulf Stream and channeled by subsurface reefs or a particu-
others with relatively well-defined flow di- lar type of bathymetry. Due to the nature of
rections and velocities. These major cur- tides, the tidal currents generally change
rents may also, on occasion, develop strong four times a day; in other words, their ve-
currents and spin eddies within local coastal locity and direction are constantly chang-
configurations. The value of current velocity ing. Naturally, this may have a profound
ranges from a few tenths of a knot for major effect on port operation when the vessel
ocean currents to several knots for localized approaches the port and while moored at
tidal currents. The port designer can get the berth.
general information on tidal current veloci- The direction in which a current flows is
ties for many localities in the world from referred to as the set, and its average speed
the tidal current tables published annually as the drift. The time of maximum revers-
by the National Oceanic and Atmospheric ing-type current is generally near midtide.
Administration (NOAA, the U.S. Depart- The tidal currents may, however, lag or
ment of Commerce). precede water levels so that water move-
Major ocean currents are of primary in- ment may be significant even when the tide
terest to oceanographers and meteorolo- itself is changing. As with tides, the daily
gists, as they represent a massive transport tidal current predictions are published for
of water and exert a major influence on the many localities (NOAA, annual); current
earth's climate. velocities are usually reported only in ship
These currents are not usually of partic- channels and at selected reference stations.
ular interest to the port designer and there- A special case of tidal current is observed
fore are not discussed further in this work. in channels connecting two bodies of water
In general, the port marine structure de- (e.g., ocean and a harbor). The speed of this
signer does not need a deep knowledge· of current can be quite high and should be
the theory of ocean currents but should be carefully addressed during the port plan-
aware of some particular aspects of their ning phase. Similar phenomenon is ob-
behavior. served in channels connecting two tidal
Of all the aforementioned types of cur- bodies of water subjected to different ranges
rents, the tidal currents, storm related wind and times of high tide; notable examples
stress currents, and littoral currents are of are the Cape Cod Canal in Massachusetts
greater concern to the port designer. and the East River entering New York har-
Tidal Currents: These are currents that bor. The speed of currents there varies
associate with the vertical rise and fall of nearly as the square root of the difference
the tides. They are usually stratified verti- in tide heights. These tides also exhibit
cally, so that the lower waters may be flow- shorter periods of slack water than normal
ing in while the upper waters are flowing tides (Brown, 1932; Wilcox, 1958).
out. This is particularly noticeable where Wind-Stress Currents: These are the re-
tidal currents are combined with river cur- sult of a wind blowing over the surface of a
rents or where relatively fresh water of body of water. This effect tends to drag the
lower density overlies heavier salt water. surface layers of water along with the wind
Tidal currents are usually rotary in nature. due to frictional shear stresses. Naturally,
This can be complicated by the effects of the higher the wind velocities that are sus-
tidal waves when the current velocities are tained for longer periods of time, the more
superimposed on the orbital particle veloci- significant are the surface currents set up.
ties of the wave. Tidal currents are usually These currents can be added vectorially to
stronger in proximity to shore, but may be the prevailing tidal currents. Surface wind-
18 Port Design and Construction

stress currents may attain velocities of Currents and Navigation: Usually a ship
around 1-5% of the sustained wind speed enters the harbor with a speed that is suf-
(U.S. Navy, 1966). ficient to maintain her maneuverability. The
Bruun (1989) states that the currents set passage of the harbor entrance becomes
up by shear stress acting on the water sur- more difficult in the presence of cross-
face may be in the order of 1/20 to 1/40 of currents and cross-winds. Obviously, more
the wind speed. This same author also pro- predictable loads from currents affect most
vides some information on expected current fully laden ships at their maximum draft,
velocities for various wind speeds and dura- whereas vessels in ballast condition and
tions in the open sea. Obviously, shallow those with large areas exposed to wind (e.g.,
depths and the proximity of land can sig- large ships in light condition, ferry, and
nificantly modify the current's parameters, container ships) are most vulnerable to the
such as velocity and direction. These are effects of gusty winds. Consequently, to
discussed by Ekman (1905). NOAA also pro- avoid the effects of strong cross-currents, or
vides guidelines for estimating wind-stress strong cross-winds at the entrance to the
currents and include some observed values port, these vessels should be oriented, as
of relative direction and velocity with re- close as practical, parallel to the predomi-
spect to the wind direction and speed, for nant currents or winds. In river ports and
some typical reference stations. tidal channels, the docking structure is usu-
After the wind driving force stops, wind ally placed parallel to the river flow or
generated current continues for some time strong tidal currents. There the ship is usu-
ally maneuvered to the berth into the cur-
as an inertial current until it is dissipated
rent direction. However, the effects of a
by friction.
steady, high wind on ship berthing maneu-
Littoral Currents: Littoral currents, and
vers must always be carefully examined.
their associated longshore currents, are
In conclusion of this section it should be
usually generated by waves breaking at an
noted that in most practical cases, particu-
angle to the shoreline; t;hey move primarily
larly where ports are located at protected
in the breaker zone, parallel to the shore.
locations, the effects of currents acting on
Longshore currents are set up inside the
waterfront structures are of secondary im-
breaker zone parallel to the shore. Littoral portance, relative to other loads. However,
currents along with waves cause the sedi- where relevant, the selection of the design
mentary material moored in the littoral current parameters must be given proper
zone. The latter is usually identified as an attention. The most important single factor
indefinite zone extending seaward from the in designing for current loads is the selec-
shoreline to just beyond the breaker zone. tion of the maximum design velocity; where
Wind-stress currents may also be signifi- practical, it must be based on actual mea-
cant in the along shore direction. surements. It may also be selected from
Bretschneiger (1966) developed the bathys- reliable current data, with due considera-
trophic technique for predicting such long- tion given to a seasonal flood or storm surge
shore currents. effects and superposition of various inter-
Littoral currents and related movements acting sources (i.e., tidal, wind stress, river
of sedimentary material are of great impor- runoff, etc.).
tance in shoreline protection and construc- For detailed information on currents and
tion and operation of shore-based marine their effects on marine structures and in-
facilities. The U.S. Army Shore Protection teraction with the related marine environ-
Manual (1984) provides a thorough treat- ments, the interested reader is referred to
ment of littoral transport problems. Gaythwaite (1981), the U.S. Army Shore
Port Design and Construction 19

Protection Manual (1984), Reid (1991), and the passing of a ship, and the discharge of
Partheniades (1992). water in large quantities from hydraulic
structures such as navigation locks, spill-
ways, and others. The primary cause of
waves, however, is the wind. Waves cause
1.6 WAVES
direct structural loadings and violent mo-
tions of water.
Gravity waves, which are still being
1.6.1 General formed by the action of the wind within the
generating area, and hence in the develop-
In general, the wave is defined by PIANC ment stage, are called waves, or seas,
(1980) as a ridge, deformation, or undula- whereas when these same waves have been
tion of the surface of a liquid. Wind waves transmitted beyond the wind affected zone
exist at the air-water interface as the wind by distance or time, they are termed swells.
begins to blow. A light air movement can Swells can travel hundreds and even thou-
cause small ripples on the sea surface, and sands of kilometers. For example, swells
as wind speed increases, larger gravity generated by cyclone activities around
waves begin to develop. The exact mecha- the Antarctic continent can reach the
nism of wave growth is still not completely equator and beyond. The long, high-energy
understood; however, it is clear that the swells that arrive at the coast of southern
wave size depends on the fetch, or the dis- California and Mexico in May are generated
tance over which the wind blows, the wind by tropical hurricanes in the South Pacific.
velocity, and the duration of time for which The swells eventually decay due to the en-
the wind blows. Thus, for a given steady ergy lost to internal friction, air resistance,
wind speed, the development of waves may dispersion, and lateral spreading.
be limited by the fetch, or the duration. If, Other wave classifications include ship
however, the wind blows over a sufficient generated waves, astronomical tidal waves,
distance for a sufficient length of time, a storm surges, harbor seiches, tsunamis,
more-or-less steady-state condition, where capillary waves, and interval waves.
the average wave heights do not change, Again, the wind is the primary cause of
will occur. Periodic undulations of the body waves and it is these waves, and their asso-
of water caused by waves affect construc- ciated swells, that are considered in the
tion, operation, and maintenance of marine design of coastal and port related marine
structures. Waves impose highly variable installations. The period for these wind gen-
loadings on a marine structure; they may erated gravity-type waves range from 1 to
:::ause fatigue-type cyclic loading, induce im- 30 s; waves typically have significant peri-
pact-type dynamic loading, cause scouring ods of 5-15 s, and swells may develop peri-
and erosion around the structure, greatly ods as great as 30 s and more. Typical
affect stability of a floating mooring struc- deepwater waves are illustrated in Figure
ture, and produce many other adverse ef- 1-6a.
fects. Water motion in the sea can be initi- The energy of swells is proportional to
ated by a variety of natural or man-made their length. Thus, even relatively low
causes, such as wind in the atmosphere, swells can cause severe forces on moored
which, through friction, transmits energy vessels and structures.
from the wind into wave energy, the gravi- As noted in previous sections, the wave
tational attraction between the earth and pattern can be altered by strong currents.
moon, the eruption of submarine volcanos, When seas or swells meet a strong current,
earthquakes, marine landslides, explosions, significant alterations to the wave length,
20 Port Design and Construction

(a )

(b )

(c)

Figure 1-6. Gravity waves. (a) Typical deep water gravity waves; (b) irregular (confused)
waves; (c) long-crested waves (Pacific Ocean at Lima, Peru).
Port Design and Construction 21

steepness, andjor height may occur. A cur- Bottom friction is also a contributing factor
rent that follows a wave may increase the that causes waves and swells to slow down,
wavelength and decrease its height, retract around, and break toward the shore.
whereas an opposing current decreases the Generally, waves can begin to break when
length and increases the height of the wave. their steepness exceeds 1/10, approaching
When seas or swell meet a strong current or the theoretical limit 1/7, or when they en-
wind at an angle, the resulting waves be- ter water with a depth less than approxi-
come more complex, shorter and steeper, mately 1.28H. The limiting depth for break-
less predictable, thus more hazardous for ing waves is most important, as the maxi-
operation of offshore terminals. mum wave height for a given site may be
Seas may also be superimposed by local governed by the water depth.
wind waves and swells coming from differ- Breaking waves exhibit rapid forward
ent directions. The result is complex or con- motion at the crest while the lower part of
fused seas with occasional crests and the wave is moving slowly in the opposite
troughs. An example of such confused waves direction. Unlike the wave travel of deep or
is illustrated in Figure 1-6b. Typical waves shallow water waves, in form only, the
with some exceptions (Fig. 1-6c) are not breaking wave causes a net forward dis-
"long crested"; the crest length of wind placement of a mass of water near shore
waves usually averages 1.5-2.0 times the because the orbital velocity at the surface
wavelength. has exceeded the velocity of wave travel.
The celerity of waves in deep water is The most dangerous stage of a breaking
governed almost solely by wavelength, but wave is the plunge when the wave crest
in shallow water, the bottom friction con- travels faster than the wave as a whole.
trols the celerity. The arbitrary definitions The front of the wave then begins to fall
for deep and shallow waters are as follows: and finally develops a jet of water which
the former is defined as such for which strikes the base of the water, entrapping a
d > L j2, and the latter is when d < L j2, pocket of air and throwing up a splash which
where d is the depth of the water. typically rises as high as the crest elevation
Typically, the deep water unbroken before plunging. The splash also has a for-
swells are waves of oscillation with a crest ward speed which carries it some distance
length that is usually 3-4 times the wave- toward the shore before it touches down in
length. These crests are not necessarily all the surf (Galvin, 1969).
oriented parallel to one another. Note that When breaking waves impact against a
when waves, traveling in deep water, reach vessel or structure, they exert very high
a steepness (HjL) between 1/17 and 1/13, local pressures that can reach 0.2-0.3 MPa;
they usually break. Here, H is the wave these pressures, although short in duration,
height and L is the wavelength. Michell may control some design features of the
(1893) and later Havelock (1918) demon- offshore terminal.
strate that the theoretical limit for maxi- Breaking of a wave directly on a marine
mum wave steepness in deep water can structure produces impact (shock) pres-
reach approximately 1/7. sures which, as mentioned earlier, can be
In deep water, the broken wave will usu- extremely high in magnitude and short in
ally reform itself into an oscillatory wave; duration. As pointed out by Kirkgoz (1992),
however, of reduced height. As waves and when the water depth in front of the wall is
swells move from deep water into shallow approximately equal to the breaking depth
water their length shortens and their height of a plunging breaker, the whole of the
increases; however, their periods remain es- breaker front can make contact with the
sentially unchanged. This leads to steepen- wall at the same instant, resulting in se-
ing of the wave until it eventually breaks. vere blows upon the wall. On the basis of
22 Port Design and Construction

extensive laboratory testing on breaking the transformation of the wave height from
waves impacting (by plunging) directly on a deep water through the breaking zone.
vertical, 10°, and 30° backward inclined More information on the wave breaking
walls with a 1/10 foreshore slope, Kirkgoz force on vertical and inclined walls is given
concludes that, within a range of water in Chapters 3 and 10.
depth conditions used in his investigation, Wave-Induced Scour: Waves and the
the most frequent location of the maximum associated wave generated currents may
impact pressure for all three walls tested cause major sediment transport, eroding
remained almost at the still-water level. beaches, and causing rapid scour around
This subject of a wave breaking onto the marine structures. This may be aggravated
marine structure is of great importance. Re- further when the wave-induced current is
search into wave dynamics has increased superimposed on the nature current.
considerably in recent years, mostly be- The scour commonly appears as local soil
cause of dramatically increased level of ac- erosion in front of the structure or around
tivities concerned with offshore oil and gas piles and accretion of a moved material at
exploration and production. some distance. This must not be overlooked
Important developments in the under- during the design process because the for-
standing of wave dynamics, and the inter- mer may eventually lead to failure of a
action between the wave and the structure, structure, and the latter may result in dock
result from extensive field investigation operational problems due to reduction in
programs, as well as from wave simulation the underkeel clearance.
studies in laboratories. The latter, in partic- The accurate prediction of the scour pro-
ular, allows the measurement of wave pa- cess requires a detailed study of the near-
rameters usually not easily detectable at structure hydrodynamics and the seafloor
sea. response to it.
Available information can help the port The present knowledge of waves and cur-
and marine structures designer produce rents effects on structures, such as slope
better quality and more reliable designs. protection, is discussed in PIANC (1987 and
The interested reader is referred to recent 1992) and a simple prediction method is
important works in the area of wave- given in the U.S. Army Shore Protection
structure interaction as follows: the tempo- Manual (1984). Discussion on scour around
ral variation and spatial distribution of the piles is provided in Chapter 7.
impact pressures from breaking waves on An example of the effect of wave-induced
vertical walls have been investigated by currents on an offshore structure is illus-
Mogridge and Jamieson (1980), Witte trated in Figure 1-7. The offshore portion
(1988), and Kirkgoz (1990), who carried out of this recently constructed dry bulk mate-
model tests, and by Blackmore and Hewson rial loading facility comprises five dolphins,
(1984) and Partensky (1987), who reported spaced at 70 m center to center. Struc-
wave pressure measurements on proto- turally, all these dolphins are identical; they
types; Stive (1984), FiihrbOter (1986), and have been designed and constructed in a
Kirkgoz (1991) investigated the wave im- form of floated-in concrete caissons in-
pact pressures due to breaking waves on stalled on stone bedding. The central cais-
steep slopes of 1/3 and 1/4; and, most son, approximately 19.5 X 24.0 m in cross
recently, extensive laboratory experiments section and 18.5 m deep, and four other
on impact wave pressures were conducted caissons, approximately 19.5 X 19.5 m in
by Hattori et al. (1994), Hattori (1994), cross section and 18.1 m deep, have been
Oumeraci et al. (1994), and others. Kam- installed on stone bedding, constructed from
phuis (1994) analysed the phenomenon of coarse granular material (300-mm-diameter
Port Design and Construction 23

Figure 1-7. The Offshore Dry Bulk Loading Tenninal, Newfoundland, Canada. All five dolphins included
in this facility are displaced both vertically and horizontally due to wave-induced scour of the stone bedding.

and less stone) overlaid by a leveling course The surfaces of the antiscour slabs and
(approximately 50-rom-diameter stone), 400 exposed bottom portion of all caissons show
mm thick. The approximately 2.5-m-wide significant abrasion resulting from violent
berm around all caissons was protected by movements of wave-borne coarse aggre-
precast concrete slabs attached to the cais- gates removed from the mattress and from
son foundation slab by means of marine the seafloor. The walls of one caisson also
chains. The design mean depth of water is show evidence of damage consisting of ag-
15.0 m . gregate and rebar exposure in some areas,
After installation on bedding, all caissons presumably the result of impact from the
were filled with granular material, and the antiscour slabs and abrasion by wave-borne
central caisson, which supports the mate- stones. Significant abrasion of concrete was
rial loading machine, was capped with a also evident on the adjacent concrete piers.
concrete slab. All the above resulted in a rotation as
A heavy storm that allegedly produced well as substantial uneven vertical and hor-
heavy waves approximately 15.0 m high, izontal displacements of all five caissons.
breaking at an approximately 30° angle to This is quite evident from Figure 1-7. As it
the berthing line of this terminal, and a is also seen from this figure, the granular
subsequent smaller storm both generated ballast material was washed away from all
very strong wave-induced currents. This has uncapped caissons by waves that over-
resulted in substantial damage sustained topped these caissons. Naturally, the facil-
by the facility that has taken it out of ser- ity was out of service and required substan-
vice. In some cases, the antiscour concrete tial remediation work.
slabs have been displaced to the extent that Data on wave climates for the various
some of the interconnecting chains have oceans are published by a number of gov-
been broken. Movements of antiscour slabs ernmental organizations. In the United
resulted in severe scour of the mattress States, the National Oceanic and Atmo-
material, basically at the exposed corners, spheric Administration (NOAA) publishes a
and in deposition of the bed material on the complete set of weather tables entitled
back side of the caissons and elsewhere. "Summaries of Synoptic Meteorological Ob-
24 Port Design and Construction

servations" based on data compiled from by Tobiasson and Kollmeyer (1991) and
ship observations on ocean data buoys. Gaythwaite (1981).
However, one must be aware that the pub- The vessel-induced waves are classified
lished data sometimes ~derestimate wave as bow waves, transverse stern waves,
height and period and may not provide and secondary waves. The mechanics of
sufficient information on the persistence of these waves are the same as that for wind
the wave climate. Knowledge of the persis- waves. The bow wave propagates at an an-
tence wave environment is of great impor- gle from the sides of the vessel; the shape of
tance to port construction and, specifically, the bow affects both the magnitude and
to construction and operation of deep propagation of these waves. In a confined
water offshore terminals. waterway, the magnitude of the bow wave
Ship-Induced Waves: Within harbors and is generally of a lesser magnitude than the
along relatively narrow inland waterways, transverse stern wave. The bow waves may
waves produced by the wakes of the deep coincide with the stern wave forming inci-
draft vessels, or barges traveling at rela- dental peaks with amplified wave heights.
tively high speeds, may be very damaging. The new waves that are formed term as a
These waves may induce substantial mo- secondary wave. The transverse stern
tion, particularly in smaller moored ships, wave is the resulting wave formed as water,
interfere with operation of small-craft har- displaced by the vessel, flows around the
bors (marinas), and cause erosion of a hull to the stern. In a narrow waterway or
shoreline. near a bank, the transverse stern wave
The height of waves generated by moving moves in the same direction as the vessel
and is generally perpendicular to the bank.
vessels is dependent on vessel speed, draft,
Here, this wave can resemble a moving hy-
shape of the hull, depth of water, and block-
draulic jump. This wave, related to draw-
age ratio of ship to channel cross section.
down, sometimes is referred to as the
The effect of ship-induced waves will de-
water-level depression, which is essentially
pend on the height of the wave generated
a drop in the water level alongside the ves-
and the distance between the ship and the
sel caused as it is moving forward. The
project site. As a rule of thumb it can be water is displaced from the bow to the.stern.
assumed that the wave height is equal to The downdrawn accompanies the return
twice the amount of vessel squat. The wave currents. The magnitude of a drawdown is a
height at the riverbank is computed using function of a vessel's traveling speed, size,
refraction and diffraction techniques. The draft, block coefficient, and last but not least
wavelength is equal to approximately one- the waterway geometry. As mentioned ear-
third of the vessel length (U.S. Army Corps lier, secondary waves are those correspond-
of Engineers, 1984b). If ship generated ing to the interference peaks formed by the
waves are considered to be the design wave, transverse stern and bow waves, or by other
model tests or prototype measurements are diverging waves propagating away from the
needed to verify or adjust the predictions. vessel. The parameters of secondary waves
For a detailed discussion on ship generated are discussed by Verhey and Bogaerts
waves, the reader is referred to Comstock (1989).
(1967), Sorensen and Weggel (1984), and Quantification of vessel-induced waves
Weggel and Sorensen (1986). Additional in- has been a subject investigated during the
formation on the possible impact of vessel past 30-40 years by researchers in many
wakes may be obtained from Camfield et al. countries. The experience gained worldwide
(1980) and Kurata and Oda (1984); a useful is summarized in PIANC (1987). Results of
discussion on the subject is provided the most recent research on effects of navi-
Table 1-8. Selected ship generated wave heights

Distance from Sailing


Line, in feet (meters)
Speed, in 100 500 1,000
Displace- Water knots (30.5) (152.4) (304.8)
Length, Beam, Draft, ment, in depth, (meters Hmax Hma:<
in feet in feet in feet tons in feet Hmax
per in feet in feet in feet
Vessel type (meters) (meters) (meters) (kilograms) (meters) second) (meters) (meters) (meters)
Cabin cruiser 23 8.3 1.7 3 40 6 0.7 0.4
(7.0) (2.5) (0.5) (2.722) (12.2) (3.1) (0.2) (0.1)
10 1.2 0.8
(5.1) (0.4) (0.2)
Coast Guard 40 10 3.5 10 38 6 0.6 1.0
cutter (12.2) (3.0) (1.1) (9.072) (11.6) (3.1) (0.2) (0.3)
10 1.5
(5.1) (0.5)
14 2.4°
(7.2) (0.7)
Tugboat 45 13 6 29 37 6 0.6 0.3
(13.7) (4.0) (1.8) (26.309) (11.3) (3.1) (0.2) (0.1)
10 1.5 0.9
(5.1) (0.5) (0.3)
Reconverted air- 64 12.8 3 35 40 6 0.3
sea rescue vessel (19.5) (3.9) (0.9) (31,752) (12.2) (3.1) (0.1)
10 1.4 0.8
(5.1) (0.4) (0.2)
14 2.0° 1.1° "d
0
(7.2) (0.6) (0.3) ~
Fireboat 100 28 11 343 39 6 0.4 0.2 t:l
ct>
(reconverted tug) (30.5) (8.5) (3.4) (311,170) (11.9) (3.1) (0.1) (0.1) f!J,
10 1.7 1.0 ~
(5.1) (0.5) P>
(0.3) .,
p..
14 3.1 2.6
0
(7.2) (0.9) (0.8) 0
.,
Barge 263 55 14 5,420 42 10 1.4 0.7 0.3
(80.2) (16.8) (4.3) (4,917,000) (12.8) (5.1) (0.4) (0.2) (0.1) ~
Moore dry dock 504 66 28 18,800 56 14 1.5 1.1 !+
::;·
tanker (153.6) (20.1) (8.5) (17.1 X 10 6 ) (17.1) (7.2) (0.5) (0.3) .,
18 5.2 4.7
(9.3) (1.6) (1.4)
to.:)
01
Source: After Sorenson (1973).
26 Port Design and Construction

gation by typical commercial tow boats and described with mathematical precision; the
barges operating in the U.S. inland water- wave parameters such as height, length,
ways have been reported by Martin and and period are not necessarily equal, and
Maynard (1992). Table 1-3 presents a sum- even in a given wave, the crest is not con-
mary for ship wave data for various vessel tinuous along its own length. In practice,
types. For more discussions on navigation- waves invariability create a very random
induced ·waves the reader is referred to situation when crests of successive waves
Chapter 7. are not all oriented parallel to one another
Naturally, port sites need to be protected but have a directional spread (Fig. 1-6b).
from adverse wave effects. Some sites have However, despite the fact that it is impossi-
naturally protected entrances from the main ble to give a precise description of the sea
body. This protection may be provided by surface at any. given time by observing a
one or more islands that shield the en- certain area of the sea, it is possible to
trance from waves by reducing wave height, observe a certain characteristic pattern.
or by shoals where the waves break. If the An early formula to estimate the maxi-
site does not have natural protection against mum wave height took into account either
wave action, then the construction of break- fetch length (F) or wind speed (V) (U.S.
waters or other wave energy dissipating de- Naval Oceanographic Office, 1951). Useful
vices at the entrance or inside the port empirical formulas for estimating the maxi-
basin must be considered to reduce waves mum wave height (Hmax) and maximum
to an acceptable height. period (Tmax) for short fetches and high wind
Although detailed in-depth knowledge of
speeds, such as found in hurricanes and
the wave phenomenon is part of the exper-
other intense storms, have been suggested
tise of a hydraulic engineer, or oceanogra- by Bretschneider (1957).
pher, the marine structure engineer must
A general rule of thumb is that the high-
have a sufficient body of knowledge and a
est waves in feet will be approximately 80%
good understanding of the wave-structure
of the wind speed in knots, and that a swell
interaction phenomenon in order to be able
realistically to evaluate wave forces and will lose approximately one-third its height
their potential effects on marine structures. each time it travels a distance in miles
Marine engineers must also be capable of equal to its length in feet (Gaythwaite,
making a preliminary assessment of poten- 1981).
tial waves at a given site and suggesting The typical idealized wave of translation
whether a more detailed theoretical or labo- can be approximated as a solitary wave,
ratory study of wave climate is warranted. which is a single crest of water above the
In the following sections the basic infor- still-water level, traveling without change
mation on wave climate and wave parame- of a form at a constant speed, with a net
ters is presented, and the reader interested displacement of water in the direction of
in in-depth information will be referred to wave travel. The wave of this type is illus-
the appropriate sources. The wave forces trated in Figure 1-8.
are discussed in detail in Chapters 3 Referring to this figure, the wave height
and 10. (H) is defined as the vertical distance be-
tween the crest and successive trough; the
wavelength (L) is the distance between two
1.6.2 The Sea State Para·meters successive crests; the wave period (T) is the
time required for the passage of successive
As follows from the previous discussion, the crests; and the wave celerity (or phase ve-
sea surface at any given time is in a very locity) (C) is the velocity of propagation of
complex condition and therefore cannot be the wave, which is equal to L/T.
Port Design and Construction 27

WAVE CELERITY
DIRECTION OF PROPAGATION

L= WAVELENGTH

NOTE r. al77=a cos(Ziix-zifr)


L T
d =DEPTH bl FOR GIVEN ORIGIN b=ol WAVE PROFILE
IS SHOWN FORt= 3T/4, 7TI4, IIT/4 ••••.•
cJ7l =a= H/2 AT WAVE CREST
77=-a=-H/2 AT WAVE TROUGH
dlt= T1ME

BOTTOM, z=-d

Figure 1-8. Wave definition sketch.

The various wave characteristics are The wave amplitude (a) above the SWL
roughly related by the following equations: is usually greater then below it (a > -a);
the wave crest elevation above the SWL is
C = L = ( gL ) o.5 gT usually equal to 0.55H to 0.75H which
T 27T 27T should be considered when figuring the dock
elevation, as well as the underside clear-
27TC 2 gT 2 ance of fixed piled piers and platforms.
L=--=-
g 27T It should be pointed out that sometimes
the maximum wave height has been defined
(1-2) as the difference between the highest and
the lowest water elevation anywhere within
a 20-min record, and the wave direction is
where g is the gravity acceleration. If one often given in two different ways, either as
characteristic is known, the others can be the direction from which the waves are
computed by substituting numerical values coming or to which the waves are going. In
for the constant values of 7T and g. Equa- the former case it agrees with the conven-
tions (1-2) are then transformed into the tional definition of the wind direction, and
following: in the latter it follows the common practice
of defining the direction of a current.
C = 2.26(£) " = 5.12T
0 5 As pointed out earlier, waves travel in
groups. The wave group velocity (Cg) is
L = 0.195C 2 = 5.12T 2
commonly defined as the velocity at which
T = 0.442(£) 0 "5 = 0.195C (1-3) the wave group propagates.
In practice it is difficult to define any
Note that in Figure 1-8 the quantity (d) is uniform wave parameters such as height,
the water depth and (7]) is the elevation of length, and period. A typical sea surface
the water surface at a given point relative record is shown in Figure 1-9. From this
to the still-water level (SWL). figure it is readily apparent that the wave
28 Port Design and Construction

based on the scientifically best choice, but


rather on trying to readily achieve a wide
acceptance. It should also be emphasized
that definitions or estimates of some pa-
rameters in the PIANC list of sea state
parameters are based on the assumption
that the sea surface can be described as a
Gaussian stochastic process. However, some
(a)
parameters are specifically defined to deter-
mine the non-Gaussian stochastic charac-
teristics of the sea surface.
Furthermore, PIANC (1986) points out


that some confusion exists as a result of a
method of wave data analysis in the time
domain, commonly referred to as "zero
crossing analysis." The transitions of the
water surface elevation from a level above
(b)
the mean to a level below is generally known
as "zero downcrossing," whereas the transi-
tion in the reverse sense is known as a
"zero upcrossing." As is seen from Figure


1-9 there is no ambiguity in the definition
of a zero crossing crest or a zero crossing
trough. However, the definition of a wave
height depends entirely on the choice of the
trough occurring before or after the crest.
(c)
PIANC (1973) state,s that a wave should be
Figure 1-9. Typical sea surface record (from PIANC, identified as an event between two succes-
1986): (a) zero upcrossing waves, PIANC; (b) zero sive downcrossings. Wave periods are ac-
downcrossing waves, IAHR Recommendation; (c) zero cordingly defined by two successive down-
upcrossing waves, IAHR.
crossings. However, although the wave
height parameter recommended by PIANC
parameters have to be selected on the basis can indeed be identified by two successive
of selecting uniform sea state parameters. downcrossings, it is referred to as the zero
Permanent International Association of upcrossing wave height. Lately, most insti-
Navigation Congresses (PIANC) concludes tutions, for example, the International As-
that the existing approach to defining wave sociation for Hydraulic Research (IAHR),
parameters should be modified to some de- refer to this wave height as the zero down-
gree. This results in a new list of sea state crossing wave height because of its logical
parameters, as recommended by PIANC identification by two successive downcross-
(1986), that include uniform parameters and ings, analogous to the wave period parame-
functions most frequently used to define the ter. Figures 1-9a and 1-9b show that the
sea state. This list, however, does not reject PIANC zero upcrossing wave height is the
the past list of definitions, symbols, and same as the IAHR zero downcrossing wave
nomenclature, but rather establishes a com- height. PIANC (1986) indicates that, be-
mon base in order to be able to incorporate cause there are also some data sets which
new contributions being made continually. are based on a zero upcrossing wave height,
However, PIANC (1986) admits that not use of the IAHR approach can lead to a
all specified recommendations made are different parameter (Fig. 1-9c).
Port Design and Construction 29

Although all three conventions have a stacles (e.g., island or man-made breakwa-
substantial following, for sake of uniform- ters), the waves propagate into the lee of
ity, PIANC (1986) · recommends the zero the obstacle by wave diffraction. Waves can
downcrossing definition as indicated in also reflect off beaches, shorelines, and
Figure 1-9b. structures. The movement of a wave corre-
It has been observed that for a given sponds to the transmission of its energy.
location and sea state, the probability of However, it is the wave form that is moving
occurrence of a given wave height roughly and not the water surface itself. As the
follows the Rayleigh wave distribution. wave form is passing a point on the surface
From this distribution, a mean height, most an individual water particle undergoes a
probable height, highest of a given per- circular orbital motion. Figure 1-10 shows
centile, and so forth can be obtained. the oscillatory wave form and its character-
Oceanographers define the wave height as istics as its propagates towards the coast.
the average of the highest one-third of sig- To the observer it may appear that the
nificant wave height (H5 ). This statistical propagation of the wave form, and its crest
term is related to one-third of the highest in particular, is due to a net displacement
waves of a given wave group and defined by of the water in the direction of the wave
the average of their heights and periods. travel. However, this is just an illusion. The
The composition of the higher waves de- simple test indicates the fact that a floating
pend on the extent to which the lower waves object in deep water will rise and fall with
are considered. Experience indicates that the undulations of the wave, but will not
an experienced observer who attempts to move horizontally unless it is moved by
visually establish the character of the wind or current, except for a small back and
higher waves will record values which ap- forth motion caused by the orbital motion of
proximately fit the definition of the signifi- the water particles.
cant wave. Based on the cumulative In deep water, each water particle on the
Rayleigh probability distribution, the other wave surface describes a circle, the radius
statistical heights are related to the signif- of which is one-half the wave height (H j2)
icant wave height approximately as follows: about its normal center, midway between
the crest and trough of the wave. The cen-
• Average height 0.64H. terline of rotation is elevated above the
• Average highest 10% 1.29H. still-waterlevel by the height h 0 = a - Hj2
• Average of highest 1% 1.68H. because, as pointed out earlier, the crest is
• Highest 1.87H. at a greater distance above still water than
is the trough below it. The difference de-
As pointed out earlier, when waves propa- pends on the wave steepness; for a very
gate from deep water into intermediate and steep wave the proportion is about two-
shallow waters, their properties are trans- thirds above and one-third below the still-
formed. Usually, the wave period is as- water level.
sumed to remain constant during these The amplitude of waves in deep water
transformations. In the process of transfor- decreases rapidly with depth, but the wave-
mation, the wave height first decreases rel- length normally remains the same. The
ative to the deep water wave height, then radii of the generating circles decrease with
increases rapidly with a decrease in water the depth in a geometrical progression, of
depth until breaking occurs. The change in which the common ratio is 1je 0·5 , where e
wave height as a function of water depth is is the base of the natural (Naperian) loga-
termed "wave shoaling." Waves also change rithms. The orbital velocities are propor-
height and direction of propagation by wave tioned to the orbital radii; they are largest
refraction. Upon encountering natural ob- at the surface and decrease rapidly with
30 Port Design and Construction

-1 J-- Mass transport


/--..._
t.....__./\
Deep water particle orbit
---E- L decreasing
H increasing

Figure 1-10. Idealized sketch of wave propagating toward coast. (From Gaythwaite,
1981.)

depth, becoming negligible at an approxi- are usually presented in terms of the pre-
mate depth of either Lj2 or 2T 2 • sent frequency of occurrence via histograms
As noted earlier, the wave motion in andjor cumulative distributions expressing
shallow water (d < Ljs) is affected by the yearly averages for each month. Long-term
sea bottom; there the normally circular mo- wave statistics are usually given in terms of
tion of the water particles becomes nearly maximum wave height versus statistical re-
elliptical with the major axis being close to turn period. Where the effect of cumulative
horizontal. The ratio of the major axis to damage can be related to the number of
the minor axis becomes greater with in- wave cycles of a given wave height, as is
creased depth until the orbital motion is required in fatigue analysis, plots of wave
almost entirely horizontal near the bottom. height versus cycles over a specified period
Unlike the circular orbital motion in deep may be prepared, as discussed in detail in
water, in which the horizontal and vertical American Petroleum Institute (1987).
velocities are equal, the orbital velocity in Another way of defining the wave climate
shallow water is greater in the horizontal for design purposes is in the form of sea
than in the vertical direction. The velocity spectra or the wave height frequency spec-
of wave propagation is slow in shallow wa- trum (Bretschneider, 1959; Pierson and
ter and is correlated to the ratio of the Moscowitz, 1963; Wiegel, 1981; Sarpkaya
vertical axis (b) to horizontal one (a) by and Isaacson, 1981; Goda, 1985). The U.S.
coefficient c = (bja) 0·5 • The value of c is Army Corps of Engineers' Shore Protection
given in Table 1-4. Note that when d = Manual (1984) provides guidelines and
L j2, c ""' 1.0 and therefore the wave propa- graphs for predicting wave heights under
gation velocity (C) is essentially that of a specific conditions. It must be stressed,
deep water wave. however, that for the design of a major port
Most often, wave climate data are com- and harbor related facilities exposed to wave
piled from forecasting methods using wind action, the wave climate prediction should
record data and knowledge of the local to- be carried out by a qualified hydraulic engi-
pography and bathymetry. Short-term data neer or oceanographer.

Table 1-4. Coefficient c for velocity of wave propagation in shallow water

djL 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
c 0.552 0.746 0.858 0.922 0.958 0.977 0.988 0.994 0.997 0.998

Note: L = wavelength, d = depth of water.


Port Design and Construction 31

1.6.3 wave Theories the wave amplitude (a) is small compared


to the wavelength (L) and water depth (d).
The theoretical treatment of sea motion has However, this is true only for very small
a long history dating back to the past cen- waves. In general, Airy's small-wave theory
tury and typically involves rigorous mathe- predicts quite correctly the water particles'
matical analysis. The scope of this work velocities acceleration, and the surface pro-
does not provide a detailed review of the file and pressure at any location, and, in
existing wave theories. The interested general, completely describes the water
reader, where appropriate, will be referred particles' motion. This theory, however, does
to the relevant source of information. For a not accurately predict the properties of
rigorous theoretical treatment of free- waves near breaking. The ·Airy theory is
surface hydrodynamics the reader is re- usually referred to as the linear, or first
ferred to the classic work of Lamb (1932); order of approximation, small-amplitude
the in-depth treatment of water wave theo- wave theory.
A summary of the linear (Airy) wave
ries with engineering applications is found
equations is found in Gaythwaite (1981). In
in Wiegel (1964), Eagleson and Dean (1966),
the opinion of most wave experts, the Airy
McCormick (1973), and Mei (1991); regions
theory is sufficient for most engineering
of validity of various wave theories are dis-
purposes in deep water. However, the
cussed by LeMehaute (1969).
higher-order Stokes' theories give a better
Originally it was thought that the wave
representation of steeper waves.
form was cycloidal; that is, the particles of
Stokes (1880) was the first to develop
water in their circular orbits moved in the
equations for waves of "finite amplitud~" by
path of a given point on the perimeter of a
considering terms higher than first order in
circle with a diameter equal to the wave solving the Laplace equation. He pointed
height. Gerstner (1809) first developed the out some discrepancies in Gerstner's rea-
trochoidal wave theory; in fact, by his pio- soning and, eliminating them, developed
neering work he laid the foundation for formulas which take into account the fact
modern wave theory. He formulated the ge- that the wave crest rises higher above the
ometrical relationship between the different still-water level than the trough falls below
parameters of a wave and gave the first it.
theoretical equations of wave motion. He The Stokes' waves, of successively higher
assumed circular orbits (the trochoidal wave order, give wave surface profiles that are
form) with decreasing diameters at increas- steeper in the crests and flatter in the
ing depths. As pointed out earlier, this is trough than those given by small-amplitude
true only for deep water waves. He also wave theory; they more closely resemble
assumed that all particles have a constant waves that are actually observed in the
angular velocity and the surface of the wa- ocean. It should be pointed out that the
ter, and all other streamlines with constant linear (Airy) theory is a special case of
pressure, are trochoids or curves described Stokes' theory, namely in the case of very
by a point on the spoke of a wheel while the small waves.
wheel is rolling along a straight line. Another important aspect of Stokes'
Airy (1845) took a similar two- waves is that the water particle orbits are
dimensional approach but went beyond the not closed, as assumed by linear theory,
pure geometrical treatment of wave motion. and thus the phenomenon of mass trans-
He used elliptical orbits where necessary, port is accounted for; such waves being
but assumed that the crest is as high above known as progressive waves.
still-water level as the trough is below it Stokes' equations have been modified by
(a= -a= H/2). Airy also assumed that Skjelbreia and Hendrickson (1962), who
32 Port Design and Construction

have also prepared tables that made their predicting the wave profile and particle ve-
equations readily available for practical use. locities, the conidal and solitary waves
Dean (1965) developed a numerical ap- solution that involves the use of elliptic
proximation to the solution of hydrody- functions was developed. These solutions,
namic equations, which was developed fur- however, are difficult to apply. To simplify
ther by Dean (1967) and Monkmeyer (1970). application of the above theories and expe-
The theory developed is. called the Stream dite their practical application, Masch and
Function Theory. This is a nonlinear the- Wiegel (1961) developed tables of conidal
ory, similar to the higher-order Stokes' (elliptic cosine) functions.
waves, with a relatively wide range of ap- AB can be concluded from the aforemen-
plications; it better describes the actual tioned, there are several basic theories on
wave phenomenon over a wider range of wave parameters, each of which provides
relative water depths than the other theo- the basic hydrodynamic solution to the wave
ries. The practical application of this theory phenomenon with respect to the assumed
for determining wave forces on structures boundary conditions, and all of which de-
can be done with the help of tables and pend on a problem to be solved. The relative
graphs, which have been developed by Dean range of applications of some of the wave
(1974). theories are illustrated in Figure 1-11. In
Because, in progressively shallower wa- this figure, deep water is defined by the
ter, Stokes' waves become less accurate at ratio djL > 0.5, and shallow water by

?d =.05 r2 = 2.55
.!!..

Stokes 3rd
Breaking

Stokes 2nd

Linear (airy) theory

I
. _ __ _ _L . __ __,___ _ _ _ _ __ . __ _ _ _ _ ___.__, 0.001
0.01 0.1 1.0 10.0

~ (ft/sec 2 )
T

Figure 1-11. Regions of validity of various wave theories. (After LeMe-


haute, 1969.)
Port Design and Construction 33

djL < 0.04, leaving a transitional zone be- ally given to structure type or function (U.S.
tween these limits. As follows from this ·Army Corps of Engineers, 1984). For exam-
figure, in shallow water as defined, the ple, in the design of breakwaters and sea-
choice is mostly limited to the conoidal or walls some overtopping by very high waves
solitary wave theory, and in deep water the may be acceptable. Also, rubble-mound
linear (Airy) theory is probably sufficient to breakwaters allow for some damage that
fit most engineering purposes; the higher- can be quickly and economically repaired.
order Stokes' theories give a better repre- For details, consult Chapter 10.
sentation of steeper waves. Note that in The cost and extent of repairs of the
Figure 1-11 a line representing L 2 Hjd 3 = structure, as well as potential delays in port
26 is plotted. It represents the value known operation, must be evaluated in detail, and
as the Ursell parameter (Ursell, 1953). This the subsequent economics of construction,
parameter is a line of demarcation designed maintenance and repair in relationship to
to aid in the selection of the appropriate the selection of design waves should be
wave theory. evaluated. This should also be supported by
the appropriate risk analysis.
The current practice is to first obtain a
long-term probability of wave distribution
1.6.4 Design wave for the site under consideration, and the
design wave is likely to be presented in
In general, marine structures are designed terms of the significant wave height. The
to withstand the highest wave-induced force probability of other heights occurring, in
over their economic life. However, in the relation to the significant wave height, are
design of any structure subjected to nonde- calculated through the application of the
terministic and probabilistic environmental Rayleigh distribution, as discussed earlier
loads, such as waves, the ultimate design in Section 1.6.2. The long-term distribution
condition must invariably be weighed can be obtained from hindcasting of histori-
against economics that include capital (first) cal storm records andjor can be predicted
cost, maintenance and repair costs, and the on the probability of the occurrence of mete-
consequence of structural damage with re- orological conditions required for develop-
gard to the anticipated type of failure. Obvi- ment of a maximum wave. The wave height
ously, the possibility of a sudden catas- and period can also be obtained through
trophic failure that may involve loss of analysis of wave observations. When possi-
human life as a result of a singie high-wave ble, both procedures should be used and the
action must be of a serious concern. This is potential differences between results should
why the engineer, when considering the be evaluated to ensure a more reliable out-
problems involving wave action in the de- come from the study. Both procedures are
sign of ports and marine structures, is al- given in detail in the U.S. Army Corps of
ways confronted with a dilemma of the Engineers' Shore Protection Manual (1984).
selection of the design wave. Wave forecasting is usually carried out
In general, the maximum wave that can by a specialist oceanographer or meteorolo-
occur at the site is dependent on many gist familiar with the local climatic condi-
factors, among which the most important tions.
are depth of water, site sheltering, and In general, a 20-25-year design wave,
underwater topography. coupled with an annual extreme water level,
In the design of marine structures ex- is considered by some authorities as appro-
posed to wave action, consideration is usu- priate for design of small to medium sized
34 Port Design and Construction

projects. Again, selection of larger waves world oceans and many years of wave ob-
having a recurrence interval more than 25 servation, summarizes extreme wave condi-
years should be related to the economics of tions for different locations around the
construction operation. The selection of the world (Table 1-6). American Petroleum In-
design wave (design conditions) for large stitute (1989) provides wave parameters for
and important projects requires a more de- ten areas in the U.S. coastal waters (Table
tailed evaluation of all previously men- 1-7). The reference level heights given in
tioned conditions. this table are intended to be applied to a
Actual storm wave data are quite scarce; specific set of conditions to result in a given
hence, actual data reported are of particu- level of design force and overturning mo-
lar interest. The highest deep water signif- ment on offshore platforms.
icant wave heights for selected hurricanes The economics of considering the smaller
occurring off the U.S. East Coast are sum- design wave (e.g., a wave with a recurrence
marized in Table 1-5. In this table the interval of 25 years versus a wave having a
forecasted wave heights for the actual storm 30-year recurrence interval) must be evalu-
waves observed are compared with heights ated, and the physical and economic factors,
that would have occurred for a stationary such as design wave parameters versuf'\ the
storm of the same intensity; they are also capital cost and costs of repair and mainte-
compared with a storm with critical for- nance, must be optimized.
ward speed resulting in maximum wave The principle of optimization is schemat-
generation. ically shown in Figure 1-12 where the an-
Sellars (1978), whose work covers most nual cost (first cost plus maintenance cost)

Table 1-5. Summary of highest deep water significant waves predicted for 13 selected hurricanes off the
eastern coast of the united states

Data for
Data for Storm Moving Data for Storm
Stationary At Actual Moving at Critical
Storm Speed Speed Ratio

No. Date H 0 ,(ft) T8 ,(s) Hac>(ft) T.c,(s) vcr>(knots) Hcr>(ft) Tcr,(s) Hcr/Ho Vcr!T.
1 10/15/54 37.9 13.0 59 16 24.8 59.8 16.4 1.58 1.91
2 9/19/55 34.2 12.4 42 13.5 24.1 56.4 15.9 1.65 1.94
3 8/12/55 31.0 11.8 37 13.0 22.7 50.0 15.1 1.61 1.92
4 12/2/25 28.7 11.3 40.5 13.5 22.0 47.7 14.6 1.66 1.95
5 9/3/13 26.0 10.8 37.5 13.0 20.7 42.1 13.7 1.62 1.92
6 8/17/55 23.9 10.3 35.0 12.5 20.0 39.0 13.2 1.63 1.94
7 9/17/06 23.5 10.2 35.5 12.5 19.3 37.2 12.8 1.58 1.89
8 8/11/40 22.6 10.1 31.5 12.0 18.7 34.7 12.4 1.54 1.85
9 8/28/11 22.4 9.9 28.0 11.0 18.6 34.7 12.3 1.55 1.88
10 8/14/53 20.8 9.7 29.0 11.5 18.5 29.5 12.2 1.42 1.91
11 10/15/47 19.5 9.4 28.0 11.0 17.2 28.2 11.4 1.45 1.83
12 8/1/44 19.1 9.3 28.0 11.0 17.5 29.5 11.5 1.55 1.88
13 8j30j52 18.5 9.2 25.0 10.5 17.1 27.6 11.3 1.49 1.86
Average 1.56 1.90

Source: After Bretschneider (1957).


Port Design and Construction 35

Table 1-6. Maximum wave heights recorded

Height Period
Location (ft) (s) Steepness Comment
Gulf of Mexico 71.5 12.0 0.097 Hurricane Camille
(21.8 m) wind speed = 120 mph
(109 knots) waves
over top of wave staff
North Atlantic 67.0 Occurred 12 Sept.
Station Juliett (20.4 m) 1961 Tucker gauge
Antarctic 81.7 11.0 0.131 Data from sterophoto
(24.9 m)
Pacific 80.0 16.2 0.060 Data from motion
(24.4 m) picture film, occurred
December 1, 1969
North Sea 61.0 13.3 0.067 Occurred 23 November
(18.6 m) 1969 Tucker gauge
North Atlantic 65.0 12.5 0.081 Occurred February 16
Station India (19.8 m) 1962 in 44-knot wind
Tucker gauge

Source: After Sellars (1978).

is plotted as a function of design wave lated to an annual cost, which is accom-


height. This plot is made by designing the plished by amortizing the first cost using an
structure for a range of wave heights. appropriate interest rate and time period.
Obviously, as the design wave height in- On the other hand, the annual maintenance
creases, the first cost of the structure also cost will decrease if the structure is de-
increases. As pointed out by NAVFAC signed for a larger wave. This is usually
DM-26.2 (1982) the first cost must be re- based on some arbitrary assumptions made

Table 1-7. Wave parameters for 10 areas in U.S. waters

Wave Height
Reference Reference Level
Level Guideline Deck Clearance•
ft m ft m Wave Steepness ft m

Offshore Gulf of Mexico 70 21.3 1/12 48 14.6


Offshore Alaska
1. Lower Cook Inlet 60 18.3 50-70 15-21 1/13 56 17.1
2. Icy-Bay Gulf of Alaska 100 30.5 90-120 27-37 1/15 80 24.4
3. Kodiak Shelf-Gulf of Alaska 90 27.4 80-110 24-34 1/15 72 21.9
4. Bering Sea/Bristol Bay 85 25.9 75-95 23-29 1/13 63 19.2
Offshore California
1. Santa Barbara Channel 45 13.7 40-50 12-15 1/16 38 11.6
2. Outer Banks 60 18.3 55-70 17-21 1/15 44 13.4
Offshore Atlantic Coast
1. Georges Bank 85 25.9 75-95 23-29 1/12 59 18.0
2. Baltimore Canyon 90 27.4 80-100 24-30 1/12 62 18.9
3. Georgia Embayment 75 22.9 65-85 20-26 1/12 53 16.2

• Above MLW in Atlantic; above MLLW in the Pacific.


Source: From American Petroleum Institute (1989).
36 Port Design and Construction

ANNUAL COST (FIRST COST


1-
(/)
PLUS MAINTENANCE COST)
0
u
...J
~ 5%
z
z
<(

ANNUAL
ANNUAL MAINTENANCE COST

RANGE

DESIGN WAVE HEIGHT, H

Figure 1-12. Selection of optimum design wave height. (From NAVFAC (DM-26.2, 1982.)

on a part of how many times in the life of harbor) generated waves, vessel wakes,
the structure repairs will be carried out and ocean swells penetrating the harbor and
how much it will cost. The experienced en- long-period waves resulting from miscella-
gineer can produce reasonable assumptions neous disturbance forces (e.g., an offshore
which enables him or her to select the de- earthquake). The latter waves are known as
sign wave by adding the annual mainte- tsunamis. They are of very long length and
nance cost and the annual first cost to pro- travel at very high speeds at sea. When this
duce a curve which represents the annual wave strikes a coastline, the run-up of wa-
cost of the structure. The engineer can then ter can obtain a high elevation and thus
identify the wave height which represents impinge on fixed objects such as port struc-
the least annual cost. Because some of the tures. This can be devastating; the destruc-
assumptions involved in arriving at the "op- tive potentials of tsunamics are discussed
timum" design wave height are arbitrary by Camfield (1980) and Wiegel (1964).
and not based on accurate data regarding The wave climate may be transformed
the cost of maintenance, some latitude is further due to the effects of refraction,
permitted in selection of the design wave shoaling, diffraction, reflection, and so on.
height when the above optimization proce- The general problem of defining the design
dure is used. If the cost varies by 5-10%, wave climate for a harbor has been ad-
the optimum design wave would have a dressed by Seymour (1977), Bruun (1989),
range of heights. The designer should use Gaythwaite (1990), and others.
other factors such as environmental, opera- As pointed out earlier, most wave-
tional, and maintenance considerations to transformation studies calculate the signif-
help select the proper design condition. icant wave height, H 8 , at the project site.
Normally, the wave climate within a har- However, the typical wave system has
bor is a result of the interaction of waves heights that exceed H 8 • Therefore, the abil-
from several sources such as locally (within ity of the structure to withstand the force
Port Design and Construction 37

generated by an occasional larger wave Honsinger, 1962; Maddox, 1974; American


must be evaluated. In general, nonrigid Petroleum Institute, 1989). The latter is
structures, such as rubble-mound breakwa- particularly important for design of deep
ters and revetments or rather flexible pile water piled piers. Aspects of dynamic re-
supported structures, can tolerate forces sponse of offshore platforms to waves are
from waves in excess of H 8 • On the other discussed by Wirsching and Prasthofer
hand, rigid or semirigid structures, such as (1976), Nolan and Honsinger (1962), Nath
cellular sheet-pile bulkheads, sheet-pile and Harleman (1969), Brannon et al. (1974),
walls, and similar structures are usually Malhotra and Penzien (1970), Thompson et
designed to absorb forces by waves that al. (1982), and other investigators.
exceed H 8 • It must be noted that although, in some
Table 1-8 summarizes general guidance cases, a single "maximum" design wave may
for selecting the appropriate wave height. It not be critical, the concept of a "design
should be pointed out that usually the ap- wave," despite some differences in philoso-
plication of design wave parameters to the phy of design wave determination, remains
direct calculation of forces and moments is central to the design of port related marine
a process of static analysis. This is appro- structures.
priate in most practical cases for the design
of a port related marine structure. How-
ever, it must be understood that although
the structure may not fail due to effects of 1.7 ICE 1
the highest incident wave, it may fail due to
cyclic wave loading that will induce fatigue The following is a concise discussion on ice
stresses in the structural elements, or due and its effect on port related marine struc-
to a resonant condition set up by some wave tures. For detailed treatment of this subject
trains of considerably less height than matter the reader is referred to Tsinker
the maximum design wave (Nolan and (1995).

1 Material included in this section is borrowed in part from Tsinker (1995).

Table 1-8. General factors for wave height selection

Type of Structure Wave Height Example


Nonrigid: minor damage Rubble-mound breakwaters
to armor units can be and revetments; pile-
tolerated without supported structures
threat to the function of
the structure
Semirigid: the structure Cellular sheet-pile walls
can absorb some excessive
wave force without
catastrophic failure
Rigid: damage may cause Cantilever sheet-pile
complete failure if the walls; braced sheet-pile
design wave is slightly walls
exceeded

Note: H, =significant wave; H 10 = 1.29H8 =average of highest 10%; H 1 = 1.68H, =average


of highest 1%.
Source: From NAVFAC DM-26.2 (1982).
38 Port Design and Construction

1.7.1 Introduction 5. In-harbor large areas have to be kept open


to provide ships with sufficient maneuver-
The navigation and port and harbor opera- ing room into and away from the docks.
tions in cold regions are generally associ- 6. Ice forming in berthing areas must be
ated with the ice cover which hinders or removed in order to permit the ship safe
sometimes prohibits navigation. mooring operation.
In North America, during the winter sea-
son and early spring, the ice cover extends The principal solutions to the above prob-
as far south as Chesapeake Bay, through lems include a set of methods usually re-
all the Great Lakes, the Upper Mississippi ferred to as "ice control" andjor "ice man-
River, and its principal tributaries-the agement."
Ohio, Illinois, and Missouri rivers, and the "Ice control" and ice management by
St. Lawrence Seaway-and hinders naviga- definition comprise of a set of methods for
tion along the east coast of the United States reducing or eliminating the growth of ice in
and Canada, in the Cook Inlet and the navigational areas and on berthing struc-
Baring Sea of Alaska, as well as throughout tures.
the U.S. and Canadian arctic regions. The behavior of marine structures in
As pointed out by PIANC (1984), some of ice-affected waters is also influenced by ice
the common problems featured in the ice- action. Depending on the geographical area,
affected waterways and harbors are as ice can be present in many forms, ranging
follows: from sheet ice to icebergs. The motion of ice
against a structure leads to the generation
1. Ship traffic is restricted to certain lanes, of substantial forces which depend on the
both offshore and in the channel. The structure's geometry, ice parameters, and
lanes are typically kept open by icebreak- other environmental forces such as wind,
ers. However, the icebreaking itself accel- waves, and currents.
erates ice formation in that large amounts The effects of global and local ice loads
of broken and refrozen ice may accumu- must be of interest to the structural marine
late after repeated passages. engineer; global ice forces in some cases can
2. Occasionally during the winter season, control the general stability of a structure,
wind-driven landfast or offshore ice out- and the local loads can determine the re-
side the open lanes, or ice from the river quired strength of some structural compo-
upstream, may break up and enter the nents.
ship lanes. In order to properly understand and
3. Navigation aids must be able to with- specify ice loads on marine structures, an
stand heavy ice forces. Furthermore, buoy engineer has to understand how ice as a
icing may start when freezing spray builds material acts under stress.
into a heavy ice accumulation on the buoy
Although formal studies in ice engineer-
cage. This accumulation may eventually
capsize the buoy, or at least cause it to ing date back to the turn of the century, the
tilt excessively, thereby reducing its visi- field of ice mechanics has not yet arrived at
bility. a mature stage; a good deal of judgment
4. Ice growth and its adherence to struc-
and experience is still required to achieve
tures such as navigation. locks and docks satisfactory results in the determination of
may hamper and disrupt the port/water- ice loads on marine structures.
way operation. Usually it is necessary to In recent years, significant activity relat-
remove this ice. In addition, ships may ing to the arctic offshore oil and gas explo-
transport broken ice from the channel or ration has led to a surge in research and
harbor area into the lock or the dock area. engineering practices relating to ice me-
Port Design and Construction 39

chanics and ice loads exerted on marine 1.7.2.1 Ice Microstructure


structures. As a result some improved theo- and Morphology
ries on ice action against marine structures
The many processes that can occur dur-
have been introduced (Kry, 1978, 1980;
ing the formation and growth of seawater
Croasdale, 1980). The recent state of the
or freshwater ice cause it to assume a vari-
art has been described in works by
ety of crystalline structures with a hexago-
Chakraburtty et al. (1984), Cammaert and
Muggeridge (1988), Croasdale (1985), nal crystallographic symmetry.
The typical size and shape of an ice grain
Hallam and Sanderson (1987), Maattanen
which actually is an individual crystal'
(1987), Metge et al. (1981), Sinha et al.
(1987), Sodhi and Cox (1987), Sodhi (1995), varies greatly. The ice grain may range fro~
and Watt (1982). Hager and Klein (1990) 1 mm to several centimeters, and its shape
can be tubular, granular, or columnar. The
summarize a European experience and dis-
axis of molecular symmetry of an ice crystal
cuss the basic principles of ice formation in
is parallel to the basal plane. Different types
inland waterways. The dimensioning of ma-
of ice are classified and described by many
rine structures exposed to ice loading is
investigators. For a convenient summary,
included in the Recommendations of the
consult Tsinker (1995).
Committee for Waterfront Structures (EAU,
1990). Most recently Richter-Menge (1992) The number of bonds to be broken by
deformation parallel to the basal plane are
compiled major investigations in ice me-
less than in other directions. Therefore, the
chanics conducted in the United States in
deformation and strength of ice, at least on
the period from 1987 to 1990.
a microscale, are significantly affected by
The major port components that can be
the direction of the applied stress in rela-
affected by the cold region environment are
tion to crystal orientation.
harbor areas, breakwaters, wharf protec-
Broken (and then refrozen) ice will gen-
tion, dock structures and their fender sys-
erally contain randomly arranged columnar
tems, all waterside and dockside equip-
ice p~eces so that, on a larger scale, for
ment, and aids to navigation.
practical purposes the ice can be considered
In navigable waterways it may also in-
isotropic.
clude navigation locks and bridge protec-
The structure of ice may be complicated
tion structures. For efficient winter opera-
further by the presence of salts and air.
tion (ice navigation) all of the above compo-
During ice growth these inclusions are ex-
nents must be designed to function reliably
pelled to the plate boundaries. At the usual
under severe winter conditions.
temperature of sea ice, these inclusions or
brine pockets may remain unfrozen and
hence have an effect on the strength of ice.
1.7.2 Ice covers This fact results in sea ice being weaker
than freshwater ice at the same tempera-
ture.
The purpose of this section is to provide a
very concise discussion on the formation of
ice and its various forms, conditions, and
1.7.2.2 Ice Formation
properties to facilitate a better understand- Under calm water surface conditions and
ing of the issues of ice loads upon marine relatively low temperature gradients, a very
structures. thin, supercooled water layer may appear
For detailed information on each particu- at the water surface. The temperature of
lar subject the reader will be referred to an this layer will actually be several degrees
appropriate reference. lower than the freezing temperature of the
40 Port Design and Construction

water. In such calm, supercooled conditions, The growth of a sea ice cover is difficult
ice crystals will grow in the form of needles to predict, as it requires accurate informa-
with vertical axes at random orientation to tion on climatological factors, such as solar
the water surface. At small temperature radiation and snowfall, and the thermal
gradients, crystallization will· proceed at a properties of ice and snow.
slow pace, and at large temperature gradi- Nakawo and Sinha (1981) developed a
ents the solidification of the ice cover will numerical method for calculation of ice
occur more rapidly, with the initial needles thickness that accommodates variations in
interlocking. snow conditions and the thermal properties
In most cases, sea ice will begin to form of ice and snow.
when wind and waves agitate the surface The air temperature is the principal fac-
layer; supercooling will then extend to tor ~that governs the ice melt rate, although
deeper levels. Natural nucleation will then snow-ice formation, snow-cover depth, solar
cause the formation offrazil particles in the radiation, currents, and wind also play a
form of small discoids. As these grow, they role (Bilello, 1980).
will adhere to each other first to form slush, Sea ice is generally classified as either
only to refreeze quickly and form the first first-year or multiyear ice. The characteris-
layer of ice (primary ice). The crystal orien- tics of each type of ice change significantly
tation in this layer will usually be random. in time and are also highly variable within
If relatively undisturbed, the smooth a given region.
sheet ice grows continuously throughout the Much of the sheet ice in the Arctic Ocean
winter season. It can vary from a few cen- survives more than one season. During the
timeters in the southern regions of North cycles of warming and cooling, the brine is
America to 1.0 and 2.5 m in the Bering Sea gradually expelled so that this multiyear
area and Canadian Arctic Islands, respec- ice is stronger than annual or first-year sea
tively. ice. Furthermore, first-year sea ice can be
As previously noted, a layer of primary overridden by the ice sheets of the same or
sheet ice first forms when ice crystals begin varying thicknesses which tend to consoli-
to concentrate on the surface. Because sea date quickly into a single, thicker sheet. In
surfaces are almost always in motion, these addition, primary ice can frequently be su-
crystals form a soupy layer. Such ice can perimposed by the ice caused by flooding
coagulate and form a continuous sheet, or, through ice cracks, refreezing of melted ice,
if there is strong wind and wave action,. it or the formation of snow-ice. All the above
can agglomerate into "pancakes" of ice that factors may result in a consolidated multi-
can reach substantial dimensions. Such ice year ice sheet about 4-5 m thick. The struc-
is usually called frazil ice. ture of multiyear ice is, however, greatly
Any ice that remains frozen to the shore- affected by seasonal surface melting, and a
line for the remainder of the winter is la- layered structure often results (Weeks,
beled fast ice. In most areas, however, ice 1982).
is affected by movements resulting from Much of the sea ice is present in the form
winds, waves, tides, currents, or thermal of ridges and rubble fields. Ice pressure
changes. Continuous, large-scale move- ridges are formed by ice pressures and ap-
ments of ice can be caused by major current pear as ridges on the surface of the ice. The
systems, such as those in Cook Inlet, Alaska. ice pressure is largely the result of wind
Movements of a shorter duration can occur drag acting over the ice cover, causing the
from wind stresses or wave action. ice to move.
Port Design and Construction 41

First-year ridges are initially a loose ac- varies from 910 to 920 kgjm 3 • Examples of
cumulation of ice blocks held together by sea ice are illustrated in Figure 1-13.
gravity and buoyancy forces. As the winter
progresses, the "core" of the ridge consoli- 1.7.2.3 sea Ice Parameters
dates as the water between the ice blocks of Importance
near the waterline freezes. Typically, the
ridge keel (below the waterline portion) These include grain size and crystal ori-
is 4.5-6 times greater than the sail entation, brine volume, and stress and
height (above the waterline portion). The strain rate. Grain size and orientation are
total thickness of a ridge can reach about the characteristics defining homogeneity (or
20-30 m. inhomogeneity) and isotropy (or anisotropy)
First-year ridges gradually consolidate of ice. Most available test results for salt-
and freeze into multiyear ridges, or ridges water ice are for specimens of columnar ice
that have survived a number of melt sea- which represents best the average proper-
sons. In the process the pore water between ties of ice sheets, as far as horizontal
the ice blocks continues to freeze, and melt strength is concerned, because it usually
water produced during the summer months forms the thickest zone.
drains down into the ridge core and re- Due to the columnar structure of most
freezes quickly; the accumulated ice will ice sheets, uniaxial strength properties vary
then become a solid, resistant mass. widely if specimens are loaded parallel or
Ridges are generally considered to be lin- perpendicular to the plane of the ice sheet.
ear features. When they become areal in The grain boundaries and basal planes of
extent, because of pressure acting from sev- individual crystals, which are the planes
where brine pockets are located, represent
eral directions, they are often referred to as
weak zones in the ice where failure occurs
ice rubble fields. Ice rubble fields also form
more easily when loaded perpendicular to
around grounded objects. If ice rubble fields
the growth direction. Typically, the strength
survive and become multiyear features, they
of the vertical direction is about three times
are usually referred to as hummock fields.
that in the horizontal direction.
Such extreme ice features may control the
The relationship between brine volume
design of offshore structures of any kind in and strength is explained by Weeks and
the Arctic Ocean. Assur (1967) and Schwarz and Weeks
The internal structure of rubble fields is (1977).
usually composed of a consolidated ice layer The deformational behavior, failure
at the waterline; under this is a randomly mechanism, and strength of ice are highly
mixed distribution of solid angular blocks, affected by the strain rate. Ranges of strain
voids, and soft or slush ice. rate may be classified as ductile, transi-
Because ports and marine terminals are tional, and brittle. At very high rates of
usually located in protected harbors, the loading, stresses are affected by strain rate
effects of ice ridges are seldom considered effects, the impact velocity, and drainage
in port related marine structures design. effects.
Ice in any form floats with most of its vol- When subjected to low rates of loading,
ume below the water surface, depending on or sustained loads, ice creeps. As strains
the relative density of the ice and water. are slowly increased, a ductile failure is
The density of sea ice usually varies from reached; the maximum stress that is at-
830 to 920 kgjm 3 and the density of fresh- tained is referred to as the yield strength.
water ice depending on air temperature The behavior in this zone is controlled by
42 Port Design and Construction

(a )

(b )

Figure 1-13. Sea ice. (a) Landfast; (b) first year.


Port Design and Construction 43

the number of mobile dislocations occurring The following is a brief review of the
along the basal planes in the ice crystals. state of the art on the mechanical proper-
For low rates of compressive strain in a ties of ice.
constant strain rate test, the dependence of
the yield stress on the strain is in good 1. 7.2.4.1 Sea ice
agreement with the dependence of sec-
ondary creep rate on the stress in a con- COMPRESSIVE STRENGTH. The strength of
stant load test. ice in compression is of fundamental impor-
Rapidly applied loads cause a sudden, tance in almost all aspects of ice mechanics.
brittle failure. At high strain rates, disloca- The compressive strength of sea ice, which
tion velocities are too slow to allow ductile varies with brine volume, strain rate, tem-
behavior so that the mechanisms of fracture perature, porosity, and grain orientation, is
initiation and propagation control the not affected by the ratio of grain size to
strength in this range of strain rates. specimen size as well as the scale effect
At strain rates intermediate between the between specimen size and prototype (Iyer,
ductile and brittle zones lies a so-called 1983).
transition zone where the mode of failure is Butkovich (1959) obtained compressive
unstable, and either ductile or brittle fail- strength values ranging from 7.6 MPa at
ures may be encountered. The range of - 5°C to roughly 12.0 MPa at - 16°C from
strain rates corresponding to the transition vertical cores ·of sea ice. He found average
zone is temperature dependent for both values of 2.1 MPa and 4.2 MPa for horizon-
compressive and tensile strength. Further- tal cores having the same two tempera-
more, strain or stress ratio is affected by tures.
the temperature effects and confinement. In tests performed by Imperial Oil Ltd.
The effect of temperature on the mechan- (Croasdale, 1974) ice crashing strength val-
ical properties of sea ice is usually ac- ues from 4 to 6 MPa were measured.
counted for by the effect of .. the brine It should be noted that the comparison of
volume. Temperature, however, has a negli- miscellaneous unconfined test data indi-
gible effect on tensile strength. Tempera- cates that the strength of sea ice loaded
ture as well as grain size are important vertically is approximately four times
parameters in the prediction of brittle com- higher than the strength of the same ice
pressive strength. For a complete list of loaded horizontally.
references consult Tsinker (1995). The following empirical values of "ice-
pressure resistance" for the German coastal
1.7.2.4 Mechanical Properties region were proposed for inclusion into the
of Ice EAU-90 (Hager and Klein, 1990):
These include compressive, tensile, flex-
ural, shear, and adhesion strengths of ice, North Sea ice 1.5 MPa
and also its deformation characteristics such Baltic Sea ice 1.8 MPa
as elastic modulus and Poisson's ratio.
Freshwater ice 2.5 MPa
By definition, the strength of ice is the
maximum stress that an ice specimen can
support. Numerous investigators have con- Weeks and Assur (1967) propose the fol-
ducted research on ice strength and its de- lowing relation between compressive
formation characteristics. strength, o;,, and brine volume, vb, of sea
44 Port Design and Construction

ice: FLEXURAL STRENGTH. The flexural


strength may be calculated from

O"c = 1.65 [ 1 - (
vb
275
)0.5] (1-4)
(1-8)

where CTc is in MPa and vb is in %o. where ur is the flexural strength, P is the
For a given creep strain rate, Wang failure load, L is the span, b is the width,
(1979) suggests the following equation for and h is the thickness of the ice beam.
the determination of the compressive Ice flexural strength has been measured
strength of granular ice, uc, in MPa: in several test programs involving both lab-
oratory and field tests. Results on ice flexu-
ral strength tests have been reported by
many workers.
where €. is the strain rate (s- 1 ). Nadreau and Michel (1984) developed the
following equation for determination of
TENSILE STRENGTH. Direct-tension test flexural strength of sea ice as a function of
results have been reported by Peyton (1966). brine volume ( vg· 5 < 0.33):
Dykins (1970) obtained tensile strength re-
sults for both vertical and horizontal direc-
tions as a function of the brine content.
Dykins' results show that the ice is two to
three times stronger when the tension is
where ur is in MPa and vb is the brine
applied in the vertical direction rather than
volume (%o).
in the horizontal.
Schwarz and Weeks (1977) analyzed the ELASTIC (YOUNG'S) MODULUS (E). This is
results from Dykins and proposed the fol- usually obtained by direct tests on ice sam-
lowing relations for ice tensile strength: ples. The results of some tests, summarized
by Voitkovskii (1960), show considerable
scatter in the value of Young's modulus
which ranges from 0.3 to 10 GPa for static
measurements, and 6 to 10 GPa for dy-
a 1(horizontal) 0.82 [ 1 - (
vb )0.5] (1-6)
namic measurements.
=
142 The comprehensive theoretical model of
Young's modulus of ice is yet to be devel-
oped. It is, however, very difficult, because
where the tensile strength, u 1 , is in MPa as indicated by current experimental obser-
and the brine volume, vb, in per thousand vations, the realistic model for the modulus
(%o).
of sea ice has to consider a variety of vari-
Saeki et al. (1978) recommend the follow- able factors such as brine volume, crystal
ing equation for tensile strength as a func- orientation, and porosity, which in addition
tion of temperature when the load is ap- includes a random distribution of pores.
plied perpendicular to the grain structure: Porosity and brine volume specifically
have a substantial effect on the variation of
at= 95 +50( -T) (1-7) the modulus of sea ice.
POISSON'S RATIO ( jL). As long as ice de-
where ut is in kPa and T is in degrees
formation remains elastic, Poisson's ratio
Celsius.
can be considered constant with a value of
Port Design and Construction 45

0.3. With lateral confinement and plastic strength and developed an analytical ap-
deformation, Poisson's ratio increases to a proach for calculating adfreeze loads on
value of 0.8. conical structures. Parameswaran (1987)
Weeks and Assur (1967) propose the fol- reports results on adfreezing strength of
lowing equation based on Lin'kov's (1958) freshwater ice to model piles made of vari-
data: ous materials such as wood, steel, and alu-
minum. According to Croasdale (1985), a
f.Ln = 0.333 + 6.105 X 10- 2 exp(- _!__)
5.48
conservative adfreeze strength of 700 kPa
at high strain rates corresponds to a value
(1-10) of 100 kPa at low strain rates.
FRICTION COEFFICIENTS. Static and dy-
where namic friction coefficients for different ma-
f.Ln = the dynamic Poisson's ratio terials in contact with sea ice have been
T = ice temperature (in °C) reported by Saeki et al. (1979). Table 1-9
summarizes some of this available data.
At very low stress rates, the effective From this table it can be seen that the
values of Poisson's ratio approach the limit static-friction coefficients are appreciably
of 0.5, as predicted by Mellor (1983) from larger than the dynamic coefficients.
the equation Static coefficients are relatively indepen-
dent of surface pressure, whereas dynamic
JL = 0.5 - 0.167( ~) (1-11)
coefficients decrease rapidly with increases
in surface pressure up to 0.5 kPa and then
usually remain constant, with values be-
where tween 0.04 and 0.11.
JL = the effective Poisson's ratio THERMAL EXPANSION. The coefficient of
E =the effective Young's modulus thermal expansion of freshwater ice de-
E 0 = the true Young's modulus for zero porosity creases slightly with decreasing tempera-
SHEAR STRENGTH. In many engineering
ture, the averag~ value being 5 X 10- 5
(ac- 5 ). The thermal expansion coefficient of
problems there is a need to know the shear
strength of ice. Shear is an ice property saltwater ice is similar to that of freshwater
characterized by lateral movement within a ice, although much more variable and com-
material, that is, angular distortion or plex as a result of the variability of crystal-
change in shape. Test results on sea ice lographic brine entrapment at variable
shear strength obtained by many workers freezing rates combined with brine crystal-
suggest that sea ice shear strength may lization at various temperatures.
vary in a range from 0.3 to about 3.4 MPa.
It basically depends on ice salinity and am- 1.7.2.4.2 Freshwater and
bient temperature. /ow-salinity ice
characteristics
ADHESION STRENGTH. Adhesion forces
primarily occur when an ice cover adfreezes COMPRESSIVE STRENGTH. Many tests
to a structure and by horizontal and verti- have been conducted to measure the uniax-
cal movement of the ice field causes consid- ial compressive strength of freshwater ice.
erable forces to be transferred to the struc- From these studies it was concluded that,
ture. in general, the uniaxial compressive
Cammaert et al. (1986) have reviewed strength of freshwater and of low-salinity
the published information on adhesive ice types varies with strain rate tempera-
46 Port Design and Construction

Table 1-9. Friction coefficients

Test Material Static Kinetic Temperature


Brackish ice (0.9%o)
Enkvist (1972)
Smooth steel 0.025-0.045 -5°C
Rough steel 0.11
Snow
Smooth steel 0.09-0.19
Rough steel 0.31
West snow
Smooth steel 0.03-0.10
Rough steel 0.14
Sea ice (Finke, 1974)
Steel 0.045-0.065 -4.5°
Sea ice (Airaksinnen, 1974)
Steel 0.4-0.7 0.07-0.25
Sea ice (Grothnes-Spork, 1974)
Steel 0.3-0.5 0.12-0.23 -7°C

Source: From Frederking, 1983.

ture and porosity. It varies for nonsaline ice - 7°C, the high rate of tensile strength is
at -5° to - 10°C by three orders of magni- about 2 MPa, with the compressive strength
tude (0.01-10 MPa) as strain rate varies around 10 MPa.
from about 10- 11 to 10- 2 s- 1 . Gammon et The effect of temperature on tensile
al. (1983a) found that iceberg ice at - 5°C strength is the same as its effect on com-
has uniaxial compressive strength averag- pressive strength at low strain rates. By
ing 5.3 MPa at a strain rate averaging contrast, the lack of sensitivity to strain
10- 3 s- 1 . Sinha (1984) found that the com- rate at very high rates leads to the expecta-
pressive strength of multiyear ice averages tion that there will be a corresponding in-
between 2 and 4 MPa at strain rates be- sensitivity to temperatures at that range.
tween 10- 5 and 10- 4 s- 1 when loaded in Similar to compressive strength, tensile
the horizontal plane. strength decreases with decreasing density
According to Weeks and Mellor (1983), at (increased air porosity) at high strain rates.
high strain rates, compressive strength de- More coarse-grained nonsaline material
creases as porosity increases (density de- that is encountered in glacier ice and lake
creases). For low strain rates, the effect of ice will usually have a tensile strength much
porosity is more indirectly related. lower than 2 MPa (typically 1 MPa or less).
FLEXURAL STRENGTH. Gow (1977) con-
TENSILE STRENGTH. At low strain rates
(< 10- 7 s- 1 ),
there is little difference be- ducted flexural strength tests on freshwater
ice and found that the lower temperature,
tween tensile and compressive strengths for
structurally unmodified ice under simple
isotropic ice. In both cases the ice yields by
supported conditions and center loaded
shearing, and the difference in normal
modes, averages from 0.5 to 1.5 MPa.
stress does not seem to have much effect.
At high strain rates (;::: 10- 5 s- 1 ), the SHEAR STRENGTH. Roggensack (1975)
tensile strength tends to a limiting value, carried out direct shear tests on columnar
whereas the compressive strength contin- grained ice with shear plane surfaces per-
ues to increase. For fine-grained ice at pendicular to and parallel to the growth
Port Design and Construction 47

direction. He found that the shear stress water ice are summarized in Table 1-10. It
(Fu) in MPa can be described by the follow- can be seen that ice adhesion to low-
ing expression: contact-angle materials (i.e., wood, metals,
concrete) is higher than that to high-
fv = 0.7 + 0.47fn (1-12) contact-angle materials such as polymers.
In general, adhesion is greater for ice with
minimal or no salinity than for ice with a
where fn is the normal stress in the range
high brine porosity. Adhesion also increases
0.5-1.4 MPa. Note, extrapolating Equation
with increasing loading rates.
(1-12) to the case of zero normal stress, a
shear stress of 0.7 MPa is obtained.
ELASTIC MODULUS. For polycrystalline 1.7.3 Effects of Ice on Port
low-saline and nonsaline ice of low porosity Operations
(density approaching 0.917 mgjm 3 ), high-
frequency dynamic measurements of The restraints imposed on port operations
Young's modulus (E) give values of approxi- by the ice conditions during the winter sea-
mately 9.0-9.5 GPa in the temperature son dictate more stringent and demanding
range - 5°C to -lOoC (Hobbs, 1974; design considerations than would normally
Gammon et al., 1983b). As temperatures be required for just seasonal operation. De-
decrease, E increases nonlinearly pending on the local condition of the harbor
(Chakraburtty et al., 1984) but the effect is and the ice regimen, the problems experi-
small for a "true" Young's modulus (as op- enced can be due to one or all of the follow-
posed to "effective" values of E which in- ing:
clude creep effects). Porosity, n, which can
be expressed altematively as bulk density 1. The accelerated buildup of ice on the dock
p, has a significant effect on E. It is inter- structure due to tidal action.
esting to note that E drops sharply below 2. The interaction of the dock structure with
the density that represents a close packing ice driven by environmental forces.
of grains ( p :::::: 0.55 mgjm 3 ). 3. The buildup of an ice "active zone" in
POISSON'S RATIO. Poisson's ratio J.L, as front of the dock due to tidal action.
measured by dynamic tests, has values close 4. The accelerated buildup of ice in the
to 0.3 for nonsaline ice of low porosity berthing area from repeated passages of
(Mellor, 1974). There is not much variation vessels due to the freezing of water that is
repeatedly exposed to the atmosphere.
with porosity over the range where the ma-
terial is regarded as "ice" rather than 5. The ice buildup on the dock faces both
vertical and horizontal due to wind or
"snow" ( p > 0.8 mgjm 3 ).
wave generated icings.
ADHESION AND FRICTION. The coefficient 6. The ice-caused abrasion of structural
of friction for freshwater ice can be in the members in the tidal zone.
range 0.01-0.1, depending on the above-
mentioned factors. Some results on friction The composite diagram of ice forces acting
of sea ice for several materials, as summa- upon marine pier is depicted in Figure 1-14.
rized by Frederking (1983), are presented As pointed out earlier, ports and marine
in Table 1-9. These results show that, gen- terminals in most cases are built at shel-
erally, the friction coefficient for sea ice is tered locations, either natural or man-made
higher than that for freshwater ice. Addi- protected harbors. However, sheltering usu-
tionally, adhesion measurements for a ally increases the length of the •ice season
number of different materials against fresh- by cutting down on the effects of waves and
48 Port Design and Construction

Table 1-10. Adhesion strength of some materials to ice

Temperature Strength
Material <•c) (kPa) Adhesion Source
PVC (1981) -2.5 to -3 65 Frederking and Karri
(1981, 1983)
PE -2 to -3 59
Concrete -1 to -3.5 440
Wood -2 to -4 470
Steel -1 to - 3.5 480
Wood -6 1380 Parameswaran (1981)
Concrete -6 840

Source: From Frederking (1983).

Increase
wind area

ttmber and
concrete)
LW I
¥

Figure 1-14. Composite diagram of ice forces. (From Gaythwaite, 1981.)


F0 = weight of ice accumulated upon the dock structure; FH =horizontal thrust
due to wind or current effects on ice sheet or due to ice thermal expansion;
Fg = dynamic impact by floating ice features, such as ice sheets, small bergs and
so forth; Fu = force due ice adhesion to piles; Fv = vertical force due to weight of
ice trapped between pile bracings.

currents which break the ice formed in calm is well demonstrated at the port of
water andjor retard freezing by replacing Montreal where the harbor basin some-
the cool surface water with warmer water times acts as a storage for ice produced
from below. At some locations, harbors have upstream. Shore harbors are filled less reg-
a tendency to accumulate drifting ice. This ularly than harbors located in inland water-
Port Design and Construction 49

ways. The unpleasant fact is that nature Accessibility to the harbor will be deter-
tends to bring ice more readily into the mined by ice conditions outside the harbor.
harbor than out of it. If large areas of high-concentration multi-
Ice problems in harbors can be reduced year ice exists outside the harbor, accessi-
considerably if advantage is taken of the bility will be more difficult than if only
sun and thermal energy sources. For exam- first-year fast or pack ice exists. Likewise,
ple, in straits and river mouths, the incom- the occurrence of ridges will hinder access
ing flow may reduce the accelerated growth to the harbor.
of brash ice because it will be mixed and Because of the ice problems that can re-
contain some thermal energy. Furthermore, sult from tidal effects, where possible the
winter dock operators are quite familiar harbor should be located in an area of
with icing of the dock wall faces; it tends to low tidal range. Otherwise, considerable
hamper the berthing and mooring opera- amounts of ice could accumulate on the dock
tions. Dock walls covered with ice bulges structure which could lead to berthing prob-
may also be a hazard because the ice may lems as well as result in increased loads on
fall off and injure people working on a ship. the dock structure and the foundation. This
Effects of each of the above problems can problem would be manifested more in a
be mitigated by proper siting and layout seasonal rather than a year-round opera-
design of the port (terminal) as well as by tion.
employing miscellaneous ice control tech- Where possible, the port should be lo-
niques (Tsinker, 1995). cated in an area that has a natural source
of ice control. This could be in an area that
1.7.3.1 Effects of Ice on Site shows a thermal stratification in the water
Selection column with warmer water at depth, or is
near a freshwater lake whose bottom water
In considering the effects of ice on port could be transported to the harbor and used
siting, it is clear that maximum protection as a means of ice control. If the port is to
from moving ice is desirable, and at the serve an industrial complex, it should be
same time maximum accessibility to the located as close as possible to the plant to
harbor must be provided. Protection is best utilize any waste heat that may be avail-
provided by locating the harbor in landfast able from the plant. However, a thorough
ice with the entrance to the harbor oriented assessment of the impact this may have on
away from the direction of the prevailing environmental factors should be considered
wind. This will prevent wind-driven ice from in the design.
penetrating the harbor and also encourage In general, ice control is made easier
ice floes to move out of the harbor during when there is good water circulation in the
summer breakup. harbor. In this regard, it is not favorable to
If the harbor is exposed to moving ice, surround a harbor with breakwaters or to
particularly multiyear ice, the ice loads on locate it in a well-protected bay as far as ice
the dock structure could be severe and port control is concerned.
operations could be significantly hindered.
Hence, where ports must be located in mov-
1.7.3.2 Layout
ing ice, as for example in the case of off-
shore loading terminals, provision should Port layout plays a major role in the
be made in the design to minimize ice inter- ability of ships to reach and be berthed at a
action with the dock structures and inter- dock. In port layout design it is important
ference with vessel operations. that ice navigation aspects be taken into
50 Port Design and Construction

account as early as possible in planning a be required. This area will depend on the
new harbor. Experiences from existing har- vessel design and the degree of tug assis-
bors with ice navigation can provide a valu- tance provided. Without tug assistance, it
able contribution to the planning work. has been estimated that the turning diame-
In a well-planned harbor the use of dif- ter of a 300 000 DWT vessel in arctic ice
ferent ice-controi methods should be opti- is approximately 12 km (Griesbach and
mized to minimize potential problems. Kremer, 1973). The turning diameter of the
The siting and layout of port components icebreaking-type LNG carrier of 234,000
should be carried out so that the ice prob- displacement tonnes designed for the Arctic
lems are minimized. The dock orientation Pilot Project was estimated to be 4 km (Gill
should be such that movement of the ice et al., 1983). In such cases, the dock and
cover under the effects of environmental other harbor facilities should be sited so
forces such as wind, current, and waves as to allow the maximum area for vessel
would result in minimum loads on the dock maneuvering.
structures and minimum interference with In general, the minimum turning radius
harbor operations. of a large ship traveling under its own power
The connection between the open sea and in a track broken in a continuous ice cover
the harbor should be as short, wide, and by the vessel itself will depend on the width
straight as possible. The entry tract to a of the track which has attained its maxi-
harbormust not be the same as the depar- mum natural growth thickness and the
ture tract. Separate tracts eliminate large shape of the vessel. For preliminary design,
turning areas where large amounts of bro- assuming that the track is about 3-4 m
ken ice can hinder maneuvering vessels. wider than the vessel's beam and that the
The layout is particularly important in a critical points of contact are at the quarter
river port where moving ice can have a points of the vessel length, the maximum
serious effect on dock operations if the dock turning radius in meters (RmaJ may be
orientation is such that the ice is blocked determined as Rmax = L 2 j29.5, where L is
during its downstream movement. the vessel length in meters.
The possibility of loads being exerted on
the dock structure by moving ice during
1.7.3.3 Effects of vessel
breakup and freeze-up should be accommo-
Operation on Ice Growth
dated in the design of the structure. How-
in the Ship Track
ever, in some cases, particularly from an
operational point of view, it might be desir- The passage of vessels along the same
able to design the layout in a way to pre- track in a continuous ice sheet results in an
vent the moving ice from impacting the dock accelerated formulation of ice through the
or entering the berthing area where it could repeated exposure of water to the atmo-
interfere with vessel berthing. In these cases sphere. Mter a certain number of passages,
it is necessary to contain or divert the ice the accumulation of ice may reach the point
floes, and the appropriate ice control tech- where further passages will be precluded.
niques will be required to do this. At this time the track may have to be aban-
The proper layout design is also impor- doned in favor of an alternative route. The
tant from the point of view of vessel maneu- same ice buildup will occur in the docking
vering in ice, particularly if large vessels area and if the ice accumulation there
such as very large crude carriers or liquid proves to be excessive, ice control tech-
natural gas (LNG) carriers are being con- niques will be required to permit proper
sidered. If the vessel is to make a turning berthing of vessels. Mathematical models
circle in the harbor, a considerable area will for this type of ice buildup have been pro-
Port Design and Construction 51

posed by Michel and Berenger (1975) and normal undisturbed ice growth is also
Ashton (197 4). shown for comparative purposes.
Sandkvist (1981) states that average field Ettema and Hung-Pin Huang (1990) and
data observed at Lulea Harbor in the Gulf Ettema (1991) produce the most recent
of Bothnia corresponds reasonably well to state-of-the-art review on ice formation in
the results obtained by the Michel and frequently transited navigation channels
Berenger formulation. and also report results of extensive labora-
According to Ashton's observations, sev- tory experiments conducted in ice tanks us-
eral passages of a vessel usually results in a ing model vessels. A conclusion based on
more or less uniform coverage of the chan- some field observations and ice tank experi-
nel area. The percentage of open water sur- ments tentatively drawn by the above in-
face area is in the order of 20-30%. As a vestigators is that the ice formation in navi-
result of this open water area, the produc- gation channels is significantly affected by
tion of ice tends to increase. hull form and the length of transiting
Computer programs developed for the vessels, the channel geometry, and the
above methods were used to generate the presence of significant water currents. For
example accelerated ice buildup curves example, quite different patterns of ice for-
based on an assumed number of freezing mation may occur in river channels that are
degree-days and vessel frequency (Acres plied by bluff-bowed, flat-bottomed tows or
Consulting Services, Ltd., 1983). The re- have significant water currents that occur
sults are illustrated in Figure 1-15. The in harbor and coastal channels, which are

4.S

4.0

3.11

3.0

..."'"'z Z.ll
"'
u
....% z.o
...
~
1.11

1.0

0.5

o.o
0 10 60 IZO 180 210 240

DAYS SINCE FIRST ICE GROWTH

Figure 1-15. Theoretical determination of ice growth in a ship track with repeated passages from a trip
once every 15 days. Practical example: 1. Natural ice growth, 2. Ashton's (1974) model, 3. Michel and
Berenger's (1975) algorithm. (From Acres Consulting Services Ltd.," 1983.)
52 Port Design and Construction

plied by ship-form hulls and are subjected grows adjacent to the dock. The exact pro-
to comparatively weak currents. The essen- cess depends on the dock configuration and
tial difference in the patterns is that tran- tidal extent.
siting tows and water currents may cause In some cases, the tides cause the ice
broken and brush ice tD be conveyed along formed at the dock to effectively form an
channels and can, consequently, lead to the active zone that responds to the tides and
markedly nonuniform ice formation. acts as a hinge between the outer, larger ice
The latter investigators also conclude sheet and the dock structure. For very low
that frequent transit of ice-covered naviga- vessel frequencies, this active zone may
tion channels result in increased volume of build outward to a distance of several me-
ice growth, but the increase may not be as ters from the dock as well as increase in
dramatic as might be expected based on thickness to several times the regular ice
existing numerical models. thickness due to repeated floodings. This ice
As fewer icebreaking transits result in growth may severely hamper ship docking
less ice growth, then convoying appears at- operations. For details the reader is re-
tractive because it reduces the number of ferred to Tsinker (1995).
icebreaking transits. In their studies,
Ettema and Hung-Pin Huang found that a
problem peculiar to tow transit of ice- 1.7.3.5 Effect of Harbor
covered channels is the accumulation of Operation
broken and brush ice beneath and along the From an operational point of view, mini-
flat bottom of tows. Ettema and Hung-Pin mizing the ice problem in harbors would
Huang also produced an extensive list of expedite the normal procedures related to
references on the subject of the effects of
ship and harbor operations. The intended
vessel operation on ice growth in ship
operation of the harbor is obviously a basic
tracks. The interested reader is referred to
input to the overall harbor and dock design,
these papers for additional details.
and in that sense, the harborI dock system
should be designed to mitigate ice interfer-
1.7.3.4 Effects of tee Buildup ence with harbor and vessel operations. The
Due to Tidal Action degree of tolerance of ice interference will
In tidal harbors, marine structures can depend to some extent on the use of the
accumulate exaggerated ice growth due to harbor. For example, if a vessel needs a
the freezing of water on vertical surfaces berth immediately adjacent to the dock for
following a high tide. Such growth creates a loading andjor unloading, it would be es-
buildup of ice that may be sufficient to ham- sential not to have ice jammed between the
per ship berthing. In addition, this ice accu- vessel and dock. On the other hand, if the
mulation must be considered during design vessel uses loading arms it might be able to
to ensure that the dock and the foundation tolerate some ice in that area, and the ice-
can accommodate the added loads. control requirements might be lessened.
The initial layer of ice on the piles or The degree of the ice problem will also
columns forms during falling tide when a depend on whether the operation is sup-
film of water is left on the pile. This layer ported by a dedicated icebreaking tug in the
gradually becomes built up with each tidal harbor. If icebreaking tugs are available to
cycle and forms an ice "bustle." Depending aid in berthing, the ice problems facing the
on the dock structure, these "bustles" may berthing vessel can be reduced. On the other
eventually join and form a lattice of ice ribs hand, consideration would have to be given
between the piles or they may become in- to the effect of the extra volumes of ice
corporated into the larger ice sheet as it generated (Section 1.7.3.3).
Port Design and Construction 53

1.7.3.6 Other Effects of Ice long as man has navigated cold sea areas. A
more recent problem is the accumulation of
These include ice-caused abrasion of
ice on marine structures in cold sea areas.
structural members; ice accretion upon the An example is shown in Figure 1-16.
structure due to icing; continual freezing Icing can cause considerable loads on
and thawing of water trapped in pockets of horizontal and vertical surfaces of a marine
porous concrete, cracks, or joints; removal structure and inconvenience the berth oper-
of smaller stones from rubble mounds; and ation. Asymmetrical ice accretion on a
others. Last but not least, the presence of structure above the waterline can affect the
ice in frozen soils (permafrost) must be con- center of gravity of the structure and in-
sidered where appropriate. crease its exposure to wind area.
Ice-Caused Abrasion: This normally oc- Typically, icing due to a wind generated
curs in the tidal zone. Moving ice can abrade spray that is the result of a direct whipping
concrete cover over the reinforcing steel, of wave crests by the wind, and icing due to
thus exposing it to seawater. Ice may cause spray generated by waves hitting a struc-
concrete spalling, destroy protective coat- ture is taken into consideration.
ings and films, and increase the rate of In strong winds and low temperatures
steel corrosion. Timber piles are also sus- the spray freezes on a structure if the air
ceptible to abrasion by ice. Sometimes the temperature is lower than the freezing point
ice may carry abrasive materials within it of water.
(e.g., rock, sand, and gravel) that may sub- Shehtman (1968) found that rapid ice
stantially increase the rate of abrasion. accumulation may occur from sea sprays
Thus, the structure's tidal zone should be with air temperatures as low as - 2°C and
considered as highly susceptible to abrasion wind speeds of 10 mjs. Unfortunately, the
by moving ice and therefore given the exact mechanism of spray cloud generation
required protection. on wave impact with a structure is un-
Concrete covers over reinforcement known and there is at present no theory
should be similar to those recommended for that can be used to estimate the spray liq-
marine concrete structures operated in a uid water content and its variations with
general marine environment. However, height under various wave impact condi-
where severe scouring action is expected tions. Quantitatively, it is obvious that the
to cause concrete abrasion, the minimum amount of spray water caused by splash
concrete strength should be increased to depends mainly on the wave energy and the
45 MPa. The concrete's resistance to abra- geometry of the impact surface. Hence, the
sion can be substantially enhanced by mean liquid water content at any height
adding reinforcing fibers to concrete mix. above sea level should be a function of the
For details, consult Chapter 2 and Tsinker significant wave height, group velocity,
(1995). Additional protection can be steepness and period of the waves, and the
achieved by using concrete aggregates hav- shape of the structure. Unfortunately, there
ing equal or higher hardness than the are very limited data available to derive the
scouring material. Where appropriate, relationship among the aforementioned pa-
epoxy resins may be utilized for water- rameters. Various parameterizations for the
proofing, sealing construction joints, repair- liquid water content of wave generated
ing cracks, and other similar usages. spray have been proposed, basically, based
Icing: When droplets generated from on scarce data from ship observations.
seawater fly in cold air, cool, and hit an At present, a modified Stallabrass (1980)
object, spray ice will form. Problems caused icing model is usually used for computing
by spray ice for vessels have been known as accretion of ice on a structure due to splash-
54 Port Design and Construction

Figure 1-16. Ice buildup on a shiploader and the supporting structure.

ing a structure with wave impact generated on icing is given in Zakrzewski (1987) and
spray. Input into this model requires infor- Makkonen (1989), and useful information is
mation on air temperatures, sea-surface found in publications by the U.S. Army
temperature, wind speed, wave height and Corps of Engineers CRREL edited by Minsk
period, and sea surface salinity. The model (1983). Miscellaneous deicing and anti-icing
computes icing rates on a surface by consid- techniques that include the use of steam,
ering the primary heat transfer terms, and hot water, ice-phobic coatings, chemical
assuming that icing is a continuous, quasi- freezing-point depressants, electric heat
steady-state process. tracing, infrared radiation, and diteren me-
The equilibrium surface temperature T8 chanical methods are discussed by Lousdale
can be solved by the Tabata et al. (1968)
and Norrby (1985).
empirical relationship:

T. = (1 + n)Tr (1-13)

where 1.7.4 Cold Temperature and


Tr = freezing temperature of seawater Ice Effects on Marine
n = freezing fraction (can be determined from structures Design
heat transfer terms)
Major port elements that may be affected
For more information on the icing phe- by the cold environment are the harbor
nomenon the reader is referred to Tsinker water area, breakwaters, shore protection,
(1995). A comprehensive list of references wharf structures, miscellaneous mooring
Port Design and Construction 55

(floating and bottom fixed) facilities, and Construction also can be facilitated
aids to navigation. greatly by using modular technology. The
In most cases, ports built in cold regions advantages and disadvantages of this type
are located in harbors protected from ice of construction are discussed by Buslov
action, available through natural geograph- (1985).
ical features or by man-made structures Marine structures built in cold regions
such as breakwaters or ice booms. If a are typically designed to withstand all oper-
protected area is formed by man-made ational and environmental forces, ice forces
structures, then it is essential to assess the included. Depending on the location, in most
impact of these structures on the natural cases the ice forces are the governing factor.
environmental conditions of the site. Of They may exceed other environmental and
particular interest should be impacts on operational locales by several magnitudes.
future ice regimens in the area where site Basically, this is why in most practical
conditions require construction of breakwa- cases of marine structure construction in
ters or installation of ice booms. ice-affected waters, the bottom-fixed struc-
In some cases, construction of port ma- tures represent the most practical solution.
rine facilities will substantially modify ex- Depending on the character of ice loading
isting environmental conditions at the port's and geotechnical conditions, these struc-
location, ice regimen included. tures are basically designed and built in the
The wharf structure is the prime physi- form of ice-resistant, gravity-type struc-
cal element providing an interface between tures, such as steel sheet-pile cells,
the vessel and terminal. In general, and floating-in concrete caissons, or timber cribs.
especially under ice conditions, it must pro-
Sheet-pile bulkheads, steel jacket-type
vide safe approach, berthing, and departure
structures, and piled structures have also
of the ship under site-specific environmen-
been used. In the latter case, piles usually
tal conditions. Therefore, when siting the
have been protected from direct impact by
port or marine terminal in cold regions, the
key elements that must be considered care- ice features (Tsinker, 1995). This is usually
fully are navigational aspects, structure de- accomplished by the installation of ice-
sign, constructability, and operational as- resistant large-diameter steel piles in front
pects. The latter must consider problems of the conventional piling, or by driving
associated with ice control in the harbor sheet piles in front of conventional piling
and ice management in the berth area. Fur- with subsequent filling of the space be-
thermore, particular attention should be tween the sheet piles and conventional piles
paid to the selection, operation, and mainte- with granular material.
nance of the fendering systems which may Piled and jacket-type structures have
be highly susceptible to cold temperature been used in moderately heavy ice condi-
effects, icing, and ice damage. tions such as one that is constructed at
Cook Inlet, Alaska. However, it is doubtful
that piled structures, even those with piles
1.7.4.1 structures well protected from direct contact with ice
Because ports and terminals in cold re- features, can survive heavy ice loads such
gions are typically located in remote areas, as those that exist in arctic or subarctic
most, if not all, construction materials must regions.
be brought to the site, and often from far The presence of permafrost may preclude
away. This requirement speaks to the ad- the use of piles on sheet piles. However, in
vantage of the use of prefabricated compo- some cases it may be possible to drive heavy
nents that are easy for assembly at the site. section H-piles into permafrost.
56 Port Design and Construction

To prevent excessive ice accumulation on ice features having crushing strengths up


piled or jacket-type structures, nonvertical to 7 MPa.
elements, such as brackets and braces that In some cases, marine structures in ice-
can be affected by the tidal range, should be affected waters have been built in the form
avoided. of circular or boxlike floated-in concrete
The structure itself should be designed caissons. These structures are typically built
for adequate buoyant forces and weight of at shipyards or other protected locations,
built-up ice on structural members. In some then launched and towed to the site of de-
cases, platforms placed on large-diameter ployment and finally sunk on a prepared
steel or concrete caissons, designed to take pad and filled with granular material. In
heavy ice static and dynamic loads, can some cases, compaction of the pad is re-
perform well under heavy ice conditions. quired to prevent large settlement.
To date, circular steel sheet-pile cells Timber cribs have been used at remote
have been the most popular type of marine locations where wood is a locally available
structure constructed in ice-affected waters construction material. The crib is usually
with large-diameter steel sheet-pile cells completed with a concrete superstructure
being built at many locations (Fenco Engi- extending from the lowest water level to the
neers, 1984). Recently, steel sheet-pile final grade. The wooden portion of such a
cells were used for the construction of a structure is usually placed below the lowest
91.5-m-long X 30.5-m-wide dock (depth of water level to protect the wood from dam-
water about 6.5 m) at Nome, Alaska (Anon- age by ice and from petrification. In cold
ymous, 1991). This type of construction can climate regions, timber cribs are typically
resist large lateral loads by its large mass constructed in relatively shallow water.
andjor by embedment into the seafloor. Timber cribs can sustain large vertical and
Natural freezing may greatly strengthen the horizontal movements and can withstand
cell. It is difficult, however, to predict the large horizontal forces from the direction of
extent to which freezing may take place. In the sea because of resistance from the gran-
this regard, freezing of a granular fill below ular fill behind the cribs.
the water level and seasonal ice levels would Permanently installed floating mooring
not be expected. However, freezing of a systems, such as navigational aids or float-
granular fill above the water level, and to ing docks, are basically used at protected
some extent within the tidal zone, if signif- locations with no substantial ice move-
icant, would be expected in areas where the ments. At unprotected locations, where sub-
average annual air temperature is below stantial ice movement is expected, a re-
- 1ac. Growth of permafrost into the steel tractable mooring buoy system is usually
sheet-pile cells from the land side and from considered and floating docks are usually
underlying permafrost is also a considera- not practical.
tion. Thermal monitoring of actual steel
sheet-pile cells is considered to be impor- 1.7.4.2 Dock Fendering
tant for adequate comprehension of the
mechanisms involved. The ship impact force during berthing
Sometimes the use of frozen granular operation is perhaps the second (after ice
material as fill may turn water inside the loads) largest horizontal force imposed on a
cell into ice. This would hamper fill consoli- dock structure. To protect both the ship and
dation with the possibility of substantial the structure from being damaged, an ap-
settlement in the future due to thaw. propriate fender system should be installed
In general, steel sheet-pile cells with at the dock face.
well-compacted fill can effectively resist lo- In the process of selection of the fender
cal buckling of sheet piles in collision with system for installation on structures operat-
Port Design and Construction 57

ing in cold regions, consideration should be as such fenders are subject to ice damage
given to the effects of cold temperature and and would likely be costly to maintain.
ice action on fender performance. The cold In areas where tides are small it may be
temperature stiffens the rubber, which re- possible to mount the fender units above
sults in much larger reaction forces during the ice-affected zone high enough to prevent
ship-fender interaction. In addition to cold severe icing. In areas with large tides it
temperatures, ice buildup on individual may be practical to maintain removable or
fender units can severely hamper their retractable fender systems that can be
effectiveness. readily placed in position for the duration of
If incorrectly designed andjor installed, the navigation season or, in extreme condi-
the fenders may be torn off the face of the tions, can be positioned for each specific
dock by moving ice. In this respect the pile- berthing. The most effective systems for op-
supported fenders, or units mounted within eration under cold temperatures are pneu-
tidal or ice-affected zones, should be avoided matic fenders, both fixed to the dock face

Figure 1-17. Ice buildup on a fender system.


58 Port Design and Construction

and floating. An example of ice buildup on a technique for ice control in areas where
fender system is illustrated in Figure 1-17. y{l.ar-round shipping is conducted. However,
Detailed discussion on cold temperature this technique alone is often not adequate
effects on rubber fenders is given in Chap- for year-round portjharbor operation where
ter 2. accelerated buildup of ice in the berthing
area is taking place as a result of repeated
1.7.4.3 Ice control passages of vessels or the icebreakers them-
Most of the ice problems in a port can be selves. In this case, the use of icebreakers
mitigated by the introduction of an effective in combination with ice removal techniques
ice-control technique suitable for a particu- could provide practical solutions for con-
lar location. ducting year-round berthing operations.
By definition, ice-control techniques are
methods for reducing the growth of ice in 1. 7.4.3.2 Ice suppression
navigational areas and on structures in Ice suppression is used to prevent ice
navigation channels and in harbors and formation and includes such techniques as
methods for breaking ice and keeping the thermal discharge, bubblers, use of chemi-
broken ice away from navigation channels. cals, and ice dusting.
In general, these methods can be classified
THERMAL DISCHARGE. Where waste heat
as mechanical, thermal, or chemical.
Typical solutions for ice control fall from the cooling system of industrial plants
within the following categories: such as nuclear power stations, thermal
power stations, gas liquefaction plants, or
1. Ice breaking, which involves any possible other industrial processes is available, it
means for ice fragmentation to ease and may be a valuable source of heat used for
facilitate berthing maneuvers. and to inhibit the growth of ice.
2. Ice suppression, which inhibits the forma· This may be accomplished by discharging
tion of ice. warm water directly into the ice-control area
3. Ice diversion, where the moving ice is through a system of diffusers or nozzles.
deflected in order to prevent collision with To avoid heat concentrations in small
a vessel or mooring structure. This is areas, with heat losses and ice fog prob-
practiced primarily in dynamic ice situa- lems, the outlet water must be spread over
tions involving movement of large ice large areas and have an initial temperature
masses. of just a few degrees above the freezing
4. Ice removal and disposal, which involves point. Many outlet points may be used. The
the physical displacement of ice to mini· heat losses are reduced considerably when
mize vessel interference.
a thin ice cover is kept on the water surface.
If warm water is discharged on the bot-
1.7.4.3.1 Jcebreaking
tom of a harbor entrance area or in a ship
This method involves bending, breaking, channel, means have to be provided to
crushing, and shearing of ice and includes transfer this water to the surface. This can
the use of equipment such as conventional be accomplished with an air bubbler sys-
icebreakers, air cushion vehicles, explo- tem. The air bubbler line should be at-
sives, and mechanical, thermal, and chemi- tached or placed close to the warm water
cal ice cutters that allow the passage of a discharge pipe.
vessel. Probably the most efficient way to use
Ice breaking is commonly accomplished by heat is to eject the warm water vertically
various classes of icebreakers. Icebreakers and into the immediate vicinity of the ice
provide by far the most popular and proven cover (Gill et al., 1983). By keeping the
Port Design and Construction 59

warm water discharge arrangement in close ICE DUSTING. Both laboratory and field
proximity to the ice cover, the heat loss to experiments have shown that dusting of an
the ambient water would be minimized and ice surface with coal dust or dark granular
the temperature of the water contacting the material can increase the absorption of so-
ice surface would be maintained at a higher lar radiation and so increase the rate of
level. melting of the ice.
BUBBLER SYSTEM. Air bubbler systems Air temperature and the amount of sun-
suppress ice formation and thereby allow shine during the early stages of breakup
navigation in harbors, ports, and water- are critical factors in ice dusting operations.
ways during periods when thick ice would Ice dusting is not very effective if the daily
otherwise halt navigation. They are also minimum air temperature falls much below
used to suppress ice in and around dock 0°C. If the air temperature is too low, heat
structures and recreation vessels. They sup- losses from the surface by convection, evap-
press ice growth when the bubble plume oration, and long-wave radiation will offset
entrains the warmer bottom water and the increased solar energy absorbed by the
brings it to the surface. darkened ice surface.
The most important requirement for the
CHEMICAL METHODS. Reduced ice
successful performance of an air bubbler
strength at the time of icebreaking is one
system is a supply of warm water.
benefit achieved when chemicals are used.
The release of air bubbles at depth below
the water level from a system of pipes cre- The chemicals, for example, salt, can be
ates an upward circulation of the warmer spread on an existing ice cover or can be
water layer to replace cooler water at the mixed with the water before freeze-up.
surface. This heat, in addition to turbulence Due to negative environmental effects,
caused by the surfacing bubbles, retards the use of chemicals must be carefully local-
the formation of an ice cover in the immedi- ized; for example, piles can be protected
ate area. from ice-lifting forces by salting the adhe-
Typical air discharge rates used in field sive ice around them.
installations have been on the order of Chemicals along with electric heating can
2.1 X 10 3 m 3 js per 30 m (PIANC, 1984). be used for preventing ice from adhering to
Several types of bubbler arrangements wharf or lock walls.
have been used for the retarding of ice
growths, but typically a system consists 1. 7.4.3.3 Ice diversion
simply of perforated pipes that are placed
on the bed or at predetermined depths of Ice diversion is the deflection of moving
water and connected to an air compressor. ice floes to prevent collisions with harbor
The application of air bubblers to create structures. Movement of ice in harbors and
ice-free conditions in freshwater is common, waterways can be controlled by numerous
but installations in seawater generally have methods. Among them ice booms, small is-
been unsuccessful, although there are iso- lands, breakwaters, large cells, dolphins,
lated examples of limited success. clusters of piles, and flow developers of mis-
Baddour (1989) developed a computer cellaneous construction are only a few. The
model simulation of ice control with a ther- most common of these structures is the ice
mal bubbler plume. This technique may be boom.
carried out to determine the optimal mix- As reported by PIANC (1984), the most
ture of warm water and air that would common boom · section in use today is
provide a desired degree of ice control made from a number of 0.3 X 0.6 X 6.0-m
under given atmospheric conditions. Douglas Fir timbers.
60 Port Design and Cunstruction

Ice booms are designed to allow ice to 1.7.4.3.6 Environmental aspects


override them when the ice load becomes of ice-control
excessive. A typical single boom has a load management
capacity before override of approximately
7.0 kN jm (70.0 kN jm for a double pontoon The environmental aspects of an ice-
boom). Detail discussion on boom design is control/management system implemented
given in Kennedy (1957). at the harbor that have to be taken into
Harbors that are subject to large, mas- account depend on the type of a port activ-
sive ice runs will require a structure other ity and the site environmental conditions.
than a typical ice boom. In some areas it For example, ice navigation itself changes
might be feasible to consider the use of the ice conditions; several ship passages,
structures such as artificial islands, break- with breakups and refreezing periods, cause
waters, and berms to modify ice movement an accelerated growth of brash ice. Ice boom
in an ice-control zone. arrangements forming a stable ice cover in
a river may change the risk of ice jamming
1.7.4.3.4 Ice removal and and flooding at breakup.
disposal Erosion problems in narrow, shallow sec-
A number of possible techniques may be tions in dredged channels and under docks
addressed for the removal and disposal of may be caused by currents generated by
ice in critical ice-control areas. It may in- propellers of powerful vessels. Furthermore,
clude ice lifting, towing, or pushing, and the the extra propeller thrust often used for
use of conveyors and slurry pipelines for ice maneuvering ships in ice may cause severe
disposal. However, feasibility and practical- erosion and damage the foundations of piers
ity of the above techniques have yet to be and other structures.
established. Thermal systems have an effect on the
environment. The effects of warm water
1.7.4.3.5 Ice management in a outlets depend on the local hydrographical
berthing zone conditions. Subsequently, the expected
spreading. and mixing of outlet water have
Previous studies for year-round naviga-
to be examined to avoid "thermal ponds."
tion in cold climate regions, and particu-
Ineffective mixing of warm water may
larly in the Arctic, indicate that an effective
ice management system is a key component generate an open, ice-free area near the
in a successful dock operation. An impor- outlet point. The heat losses from such
tant function of an ice management system areas will be large, and during cold days fog
is to alleviate the problems associated with will be generated, reducing visibility for
ship movement into the dock berthing area navigation. The fog may also cause severe
and the reduction of any kind of forces icing on equipment and structures.
(static or dynamic) imposed by ice on the In harbors at river mouths, a warm wa-
dock structure. ter outlet may hinder migrating fish on their
In addition to the previously discussed way upstream. Environmental aspects have
ice-control techniques, ice management in a to balance the ice-reducing aspects. On the
berthing zone may include the use of floats other hand, some ice-control measures can
and pontoons of miscellaneous design placed increase the oxygen content in the water
in front of the dock structure, and utiliza- and, in this respect, improve conditions for
tion of special "hold tanks" of warm water fish. For example, an air bubbler system
where previously crushed ice can be effec- will reduce stratified conditions in a water
tively melted. body by mixing.
Port Design and Construction 61

For additional details on ice control, the BLUMBERG R. and RIGG, A. M., 1961. "Hydrody-
reader is referred to Tsinker (1995). namic Drag at Supercritical Reynolds Num-
bers." ASME, Petroleum Session.
BoRGMAN, L. E., 1963. "Risk Criteria." Proceed-
REFERENCES ings ASCE WW-3, Vol. 89, August.
ACRES CONSULTING SERVICES LTD., 1983. "Ice BRANNON, H. R., LOFTIN, T. D., and WHITFIELD,
Control for Arctic Ports and Harbours." Final J. H., 1974. "Deepwater Platform Design."
Report, Vol. 1. Transportation Development Proceedings Offshore Technology Conference,
Centre, Transport Canada. Montreal and Paper No. 2120. Houston, TX.
Canadian Marine Transportation Administra- BRETSCHNEIDER, C. L., 1957. "Hurricane Design
tion, Transport Canada, Ottawa. Wave Practices." Proceedings ASCE, WW-2,
AIRY, G. B., 1845. On Tides and Waves. Encyclo- Vol. 83, May.
pedia Metropolitania, London. BRETSCHNEIDER, C. L., 1959. "Wave Variability
AMERICAN PETROLEUM INSTITUTE, 1989. "Recom- and Wave Spectra for Wind Generated Grav-
mended Practice for Planning, Designing and ity Waves." TM-118, U.S. Army Beach Ero-
Constructing Fixed Offshore Platforms." sion Board.
RP-2A. Washington, DC. BRETSCHNEIDER, C. L., 1966. "On Wind, Tide and
ANONYMOUS, 1991. "Nome Dock Withstands Cli- Longshore Currents Over the Continental
mate, Helps Commerce." Civil Engineering, Shelf Due to Winds, Blowing at an Angle to
Vol. 61, No. 10. the Coast." National Engineering Science
Company, Report to Office of Naval Research,
ASCE, 1961. "Task Committee on Wind Forces Washington, DC, December.
'Wind Forces on Structures'." Transaction
BRETSCHNEIDER, C. L., 1969. "Overwater Wind
ASCE, Vol. 126, No. 3269.
and Wind Forces." Handbook of Ocean and
AsHTON, G. D., 1974. "Evaluation of Ice Manage- Underwater Engineering. McGraw-Hill, New
ment Problems." U.S. Army CRREL Report. York.
Hanover, NH. BROWN, 1932. "Flow of Water in Tidal Canals."
BADDOUR, R. F., 1989. "Computer Simulation of Transaction ASCE, Vol. 96.
Ice Control with Thermal Bubbler Plumes: BRUUN, P., 1989. Port Engineering, 4th ed. Gulf
Point Source Configuration." Proceedings Publishing House, Houston, TX.
ASCE Specialty Conference PORTS'89. Buswv, V. M., 1985. "Modular Construction
Boston. Technology for Arctic Islands." Proceedings
BEE, R. G., 1973. "Selection of Environmental ASCE Conference Archive '85, San Francisco.
Criteria for Offshore Platform Design." Pro- BUTKOVICH, T. R., 1959. On the Mechanical Prop-
ceedings Annual Offshore Technology Confer- erties of Sea Ice. Thule, Greenland, 1957. U.S.
ence. Paper No. 1839, May. Houston, TX. Army Snow, Ice and Permafrost ResearchEs-
BEE, R. G., 1979. "Earthquake and Wave Design tablishment (U.S.A SIPRE), Research Report
Criteria for Offshore Platforms." Proceedings 54.
ASTE, St-2, Vol. 105, February. CAMMAERT, A. B. and MUGGERIDGE, D. B., 1988.
BILELLO, M. A., 1980. "Maximum Thickness and Ice Interaction with Offshore Structures. Van
Subsequent Decay of Lake, River, and Fast Nostrand Reinhold, New York.
Sea Ice in Canada and Alaska." CRREL CAMMAERT, A. B., KIMuRA, T., NOMA, N., YASHIMA,
Report 80-6, Hanover, NH. N., YANO, S., and MATSUSHIMA, Y., 1986. "Ad-
BLACKMORE, P. A. and HEWSON, P. J., 1984. "Ex- freeze Forces on Offshore Platforms." Pro-
periments on Full-Scale Wave Impact Pres- ceedings International OMAE Symposium,
sures." Coastal Engineering, Vol. 8. Vol. 4, Tokyo.
Blumberg, R. and Rigg, A. M., 1961. "Hydro- CAMFIELD, F. E., 1980. "Tsunami Engineering."
dynamic Drag at Supercritical Reynolds USACE, CERC, SR-6, Fort Belvoir, VA.
Numbers." American Society of Mechanical CAMFIELD, F. E., RAY, R. E., and ECKERT, J. W.,
Engineers. Petroleum Session. 1980. "The Possible Impact of Vessel Wakes.
62 Port Design and Construction

on Bank Erosion." Report No. CG-W-1-80, U.S. namics. lppen, A. T. (ed.). McGraw-Hill, New
Coast Guard, U.S. Department of Transporta- York.
tion, Washington, DC. EAU, 1990. Recommendations of the Committee
CHAKRABUR'ITY, S. et al., 1984. Arctic Marine for Waterfront Structures, 6th ed. Ernst and
Design and Construction Handbook, Vol. 1. John, Berlin.
Prepared for Transportation Development
EFIRD, K. D., 1975. "The Interrelation of Corro-
Center. Transport Canada, Montreal.
sion and Fouling of Metals in Seawater."
COMSTOCK, J.P. (ed.), 1967. Principles of Naval National Association of Corrosion Engineers.
Architecture. Society of Naval Architects and Proceedings Corrosion Conference, Toronto.
Marine Engineers, New York.
EKMAN, V. W., 1905. "On the Influence of the
CROASDALE, K. R., 1974. The Crushing Strength Earth's Rotation on Ocean Currents." Ark.
of Arctic Ice. The Coast and Shelf of the Mat. Astron. Fys., Stockholm.
Beaufort Sea. Reed, J. D. and Sater, J. E.
(eds.). Arctic Institute of North America. ETTEMA, R., 1991. "Ice Formations in Navigation
CROASDALE, K. R., 1980. "Ice Forces on Fixed Channels." PIANC Bulletin No. 74, Brussels.
Rigid Structures." Working Group on Ice ETTEMA, R. and HUNG-PIN HUANG, 1990. "Ice
Forces on Structures. A State-of-the-Art Re- Formation in Frequently Transited Naviga-
port. CRREL Special Report 80-26, Hanover, tion Channels." The U.S. Army CRRL, Report
NH. 90-40, December.
CROASDALE, K. R., 1985. "Recent Developments FENCO ENGINEERS, 1984. "Arctic Marine Design
in Ice Mechanics and Ice Loads." Behaviour of and Construction Handbook." Transport
Offshore Structures. Elsevier Science Publish- Canada Report No. TP 572E, Vol. 1, Montreal.
ers, Amsterdam. FREDERKING, R., 1983. "Ice Engineering I." Off-
DAVENPORT, A. G., 1977. "Wind Structure and shore Mechanics and Cold Region Engineer-
Wind Climate. The Prediction of the Response ing Symposium, University of Calgary.
of Structures in Gusty Winds." International Calgary, Alberta.
Research Seminar on Safety of Structures
FREUDENTHAL, A.M. and GAITHER, W. S., 1969.
Under Dynamic Loading. The Norwegian
"Design Criteria for Fixed Offshore Struc-
Institute of Technology.
tures." Proceedings Annual Offshore Technol-
DEAN, R. G., 1965. "Stream Function Represen- ogy Conference, Paper No. 1058, Houston, TX.
tation of Non-Linear Ocean Waves." Journal
of Geophysical Research, Vol. 70. No. 18. FDHRBOTER, A., 1986. "Model and Prototype Tests
for Wave Impact and Run-Up on a Uniform
DEAN, R. G., 1967. "Relative Validities of Water
1:4 Slope." Coastal Engineering, Vol. 10.
Wave Theories." Proceedings ASCE Confer-
ence Engineering in the Oceans I, September. GALVIN, C. J., 1969. "Breaker Travel and Choice
DEAN, R. G., 1974. "Evaluation and Develop- of Design Wave Height." ASCE Journal of
ment of Water Wave Theories for Engineering Waterways and Harbor Division, Vol. 95
(WW2).
Applications," 2 vols. U.S. Army Corps of
Engineers, CERC, S.R. No. 1, November. GAMMON, P. H., GAGNON, R. E., BOBBY, W., and
DYKINs, J. E., 1970. "Ice Engineering Tensile RUSSELL, W. E., 1983a. "Physical and Me-
Properties of Sea Ice Grown in a Confined chanical Properties of Icebergs." Proceedings
System." Naval Civil Engineering Laboratory of OTC, Vol. 1, Houston, TX.
Technical Report R689. Port Hueneme. GAMMON, P. H., KrEFTE, H., CLOUTER, M. J., and
DYKINs, J. E., 1970. "Ice Engineering-Material DENNER, W. W., 1983b. "Elastic Constants
Properties of Saline Ice for a Limited Range of Artificial and Natural Ice Samples by
of Conditions." U.S. Naval Civil Laboratory Brillouin Spectroscopy." Journal of Glaciol-
Technical Report R720, Port Hueneme. ogy, Vol. 29, No. 103.
EAGLESON, P. S. and DEAN, R. G., 1966. "Small GAYT,HWAITE, J. W., 1978. "Structural Design
Amplitude Wave Theory'' and "Finite Ampli- Considerations in the Marine Environment."
tude Waves." Estuary and Coastline Hydrody- ASCE Boston Section.
Port Design and Construction 63

GAYTHWAITE, J. W., 1981. The Marine Environ- HEAF, H. J., 1979. "The Effects ofMarine Growth
ment and Structural Design. Van Nostrand on the Performance of Fixed Offshore Plat-
Reinhold, New York. forms in the North Sea." Offshore Technology
GAYTHWAITE, J. W., 1990. Design of Marine Conference, Paper No. 3386, Houston, TX.
Facilities for Berthing, Mooring and Repair HoBBS, P. V., 1974. Ice Physics. Clarendon Press,
of Vessels. Van Nostrand Reinhold, New York. Oxford.
GERSTNER, F. J., 1809. "Theorie die Wellen." IYER, S. H., 1983. "Size Effects on Ice and Their
Annalen Physik, Vol. 32. Influence on the Structural Design of Off-
shore Structures." Proceedings of Interna-
GILL, R. J., TSINKER, G. P., and CAMMAERT, A. B.,
tional Conference POAC, Vol. 2, Helsinki.
1983. "Design Criteria for Arctic Ports and
Harbours." Proceedings International Confer- KAMPHUis, J. W., 1994. "Wave Height From Deep
ence on Port and Ocean Engineering Under Water Trough Breaking Zone." ASCE Journal
Conditions, POAC, Helsinki. of Waterway; Port, Coastal and Ocean Engi-
neering, Vol. 120, No. 4.
GODA, Y., 1985. Random Seas and Design of
Maritime Structures. University of Tokyo KENNEDY, B. J., 1957. "Forces Involved in Pulp-
Press, Tokyo. wood Holding Grounds." Lecture read at the
71st Annual General and Professional Meet-
Gow, A. J., 1977. "Flexural Strength of Ice on ing of Engineering Institute of Canada, Banff,
Temperate Lakes." Journal of Glaciology. Vol. Alberta.
19, No. 81.
KlRKGOZ, M.S., 1990. "An Experimental Investi-
GRIESBACH, R. J. and KREMER, H., 1973. "Con- gation of a Vertical Wall Response to Break-
ceptual Study of Marine Terminal Facilities ing Wave Impact." Ocean Engineering, Vol.
in Arctic," Proceedings 23rd Congress, 17, No.4.
PIANC, Ottawa.
KIRKGOZ, M. S., 1991. "Impact Pressure of
HAGER. M., 1990. Part 1-"Principles for the Breaking Waves on Vertical and Sloping
Dimensioning of Structures Against Ice Load," Walls." Ocean Engineering, Vol. 18, No. 1/2.
PIANC Proceedings of the 27th Congress, KlRKGOZ, M.S., 1992. "Influence of Water Depth
Section 1, Subject 2, Osaka, Japan. on the Breaking Wave Impact on Vertical and
HAGER, M. and KLEIN, E., 1990. "Part !-Princi- Sloping Walls." Coastal Engineering, Vol. 18.
ples for the Dimensioning of Structures KRY, P. R., 1978. "A Statistical Prediction of
Against Ice Loads." Part 2-The Struggle Effective Ice Crushing Stresses on Wide
Against Ice Using the Elbe River as an Exam- Structures." Proceedings IAHR Symposium on
ple." Proceedings PIANC 27th Navigation Ice Problems. Part 1, University of Lulea,
Congress. Osaka, Japan. Sweden.
HALLAM, S. D. and SANDERSON, T. J. 0., 1987. KRY, P. R., 1980. "Ice Forces on Wide Structures."
"Advances in Ice Mechanics in the United Canadian Geotechnical Journal, Vol. 17,
Kingdom." ASME Applied Mechanics, Vol. 40, No.1.
No.9. KURATA, K. and ODA, K., 1984. "Ship Waves in
HATTORI, M., 1994. "Wave Impact Pressures on Shallow Water and Their Effects on Moored
Vertical Walls and Resulting Wall Deflec- Small Vessel." ASCE, Proceedings Coastal
tions." Proceedings of International Workshop Engineering Conference, Chapter 218.
on Wave Barriers in Deep Waters, Yokosuka, LAMB, SIR HORACE, 1932. Hydrodynamics.
Japan. Cambridge University Press, Cambridge.
HATTORI, M., ARAMI, A., and YUI, T., 1994. "Wave LEMEHAUTE, B., 1969. "An Introduction to Hy-
Pressure on Vertical Walls Under Breaking drodynamics and Water Waves, Water Wave
Waves of Various Types." Coastal Engineer- Theories," Vol. II, TRERL 118-POL-3-2.
ing, Vol. 22. U.S. Department of Commerce, ESSA,
HAVELOCK, T. H., 1918. "Periodic Irrotational Washington, DC.
Waves of Finite Height." Proceedings Royal' LIN'KOV, E. M., 1958. "Izuchenie Uprigikh Svoistv
Society, London. Ledianogo Pokrova v Arktike" (Study of Elas-
64 Port Design and Construction

tic Properties of an Ice Cover in the Arctic). MICHEL, W. H., 1968. Marine Tech11rology,
Vestnik, Leningradskogo Univ. Vol. 13. SNAME.
LOUSDALE, J. T. and NORRBY, T., 1985. "Electric MICHELL, J. H., 1893. "On the Highest Waves in
Heat Tracing Designed to Prevent Icing." Water." Philosophical Magazine, Vol. 36.
Offshore, November. MICHEL, D. and BERENGER, D., 1975. "Algorithm
MAATTANEN, M., 1987. "Advance in Ice Mechan- for Accelerated Growths of Ice in a Ship's
ics in Finland." ASME. Applied Mechanics, Track." Third International Symposium on Ice
Vol. 40, No. 9. Problems, Hanover, NH.
MADDOX, N. R., 1974. "Fatigue, Analysis for MINSK, L. D. (ed.), 1983. Proceedings of First
Deep-Water Fixed Bottom Platforms." Pro- International Workshop on Atmospheric Icing
ceedings Offshore Technology Conference, of Structures. U.S. Army Corps of Engineers
Paper No. 2051. Houston, TX. CRRL, Hanover, NH.
MAKKONEN, L., 1989. "Formation of Spray Ice on MODRIDGE, G. R. and JAMIESON, W. W., 1980.
Offshore Structures." Fourth Special State-of- "Wave Impact Pressures on Composite
the-Art Report 89-5. U.S. Army Corps of Engi- Breakwaters." Proceedings of the 17th Con-
neers CRREL, Hanover, NH. ference on Coastal Engineering, Vol. 2.
MALHOTRA, A. K. and PENZIEN, J., 1970. "Re- Sydney.
sponse of Offshore Structures to Random MONKMEYER, P. L., 1970. "Higher Order Theory
Wave Forces." Proceedings ASCE, ST-10, Vol. for Symmetrical Gravity Waves." Proceedings
96, October. ASCE 12th Conference on Coastal Engineer-
MARsHAL, P. W., 1969. "Risk Evaluation for Off- ing.
shore Structures." Proceedings ASCE ST-12, MYERS, J. J., HoLM, C. H., and McALLISTER (ed.),
Vol. 95, December. 1969. Handbook of Ocean and Underwater
MARTIN, S. K. and MAYNARD, S. T., 1992. Riprap Engineering. McGraw-Hill, New York.
Design in Marine Terminals." ASCE Proceed- NADREAU, J. P. and MICHEL, B., 1984. "Ice Prop-
ings Specialty Conference, PORTS '92, erties in Relation to Ice Forces." Proceedings
Seattle, WA Second State-of-the-Art IAHR Working Group
MAscH, F. D. and WIEGEL, R. L., 1961. Conidal on Ice Forces, Vol. 4, Chap. 1.
Waves, Tables of Functions. Council on Wave NAKAWO, M. and SINHA, N. K., 1981. "Growth
Research, The Engineering Foundation. Rate and Salinity Profile of First-Year Sea Ice
Richmond. CA. in the High Arctic." Journal of Glaciology,
McCORMICK, M. E., 1973. Ocean Engineering Vol. 27, No. 96.
Wave Mechanics. John Wiley & Sons, New NATH, J. H. and HARLEMAN, D. R. F., 1969. "Dy-
York. namics of Fixed Towers in Deepwater Ran-
MEl, C. C., 1991. "Basic Gravity Wave Theory," dom Waves." Proceedings ASCE, WW-4, Vol.
Handbook of Coastal and Ocean Engineering 95, November.
Vol. 2. Berbich, J. B. (ed.). Gulf Publishing NAVFAC DM-2, 1970. U.S. Navy Design Man-
House. Houston, TX. ual: Structural Engineering, October. U.S.
MELLOR, M., 1974. "A Review of Basic Snow Navy Facilities Engineering Command,
Mechanics." Snow Mechanics Symposium. In- Washington, DC.
ternational Association of Hydraulic Research NAVFAC DM-26.2, 1982. Coastal Protection De-
IAHR, Grindlwald, Switzerland. sign Manual 26.2, April. Department of the
MELLOR, M., 1983. Mechanical Behavior of Sea Navy Naval Facilities Engineering Com-
Ice. CRREL Monograph 83-1, Hanover, NH. mand, Alexandra, VA
METGE, M., DANIELEWICZ, B. W., and HOARE, NOAA (Annual). Tidal Current Tables. U.S.
R. D., 1981. "On Measuring Large Scale Ice Department of Commerce, Washington, DC.
Forces, Hans Island 1980." Proceedings of the NOLAN, W. C. and HONSINGER, V. C., 1962. Wave
6th POAC Conference, Laval University, Induced Vibrations on Offshore Structures,
Quebec City, Quebec. SNAME, New York.
Port Design and Construction 65

OCS OIL AND GAS-An Environmental Assess- lines for the Design and Construction of Flex-
ment, 1974. A Report to the President by the ible Revetments Incorporating Geotextiles for
Council on Environmental Quality, Executive Inland Waterways." Report of Working Group
Office of the President, Vol. 1. Washington, 4, Supplement to Bulletin No. 57, Brussels.
DC. PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
OFFSHORE ENGINEER, 1984. "Anti-Fouling Paint GATION CONGRESSES (PIANC), 1992. "Guide-
Makes Barnacles Lose their Grip," July. lines for the Design and Construction of Flex-
0UMERACI, H., KLAMMER, P., and KORTENHAUS, ible Revetments Incorporating Geotextiles in
A., 1994. "Impact Loading and Dynamic Re- Marine Environment." Report of Working
sponse of Vertical Breakwaters-Review of Group No. 21, Supplement to Bulletin No.
Experimental Results." Proceedings of Inter- 78/79, Brussels.
national Workshop on Wave Barriers in Deep PEYTON, H. R., 1966. "Sea Ice Strength-Effects
Waters, Yokosuka, Japan. of Load Rates and Salt Reinforcement." Arctic
PARAMESWARAN, V. R., 1987. "Adfreezing Drifting Station Conference, Warrent, VA.
Strength of Ice to Model Piles." Canadian PIERSON, W. J. and MOSCOWITZ, 1963. "A Pro-
Geotechnical Journal, Vol. 24. posed Spectral Form for Fully Developed Wind
PARTENSKY, H. W., 1987. "Neue Bemessung kri- Seas Based on the Similarity Theory of S. A.
terieu fiir senkrechte und geschiiffete Wellen- Kitaigorodski." U.S. Navy Oceanographic
brecher." Jahrbuch Hafenbautechnischeu Office, T.R., Contract No. 62306-1042.
Geselshaft, Vol. 41. REID, R. 0., 1991. "Tides and Storm Surges."
PARTHENIADES E., 1992. "Tides and Currents." Handbook of Coastal and Ocean Engineering
Handbook of Coastal and Ocean Engineering Vol. 2, Herbich, J. B. (ed.). Gulf Publishing
Vol. 3, Herbich, J. B. (ed.). Gulf Publishing House, Houston, TX.
House, Houston, TX. RICHTER-MENGE, J. A., 1992. U.S. Research in
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- Ice Mechanics-1987-1990. Elsevier Science
GATION CONGRESSES (PIANC), 1973. "Final Publishers, Amsterdam.
Report of the International Commission on ROGGENSACK, J. R., 1975. "Large Scale Labora-
the Study of Waves." Bulletin No. 15, Vol. II, tory Direct Shear Tests on Ice." Canadian
Brussels. Geotechnical Journal, Vol. 12, No.2.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- SAEKI, H., NoMURA, R., and OZAKI, A., 1978.
GATION CONGRESSES (PIANC), 1979. "Report "Experimental Study on the Testing Methods
of Working Group I; Methods for Analyzing of Strength and Mechanical Properties for Sea
Wind, Wave and Swell Data (ICORELS)." Ice." Proceedings IAHR Ice Symposium.
Supplement to Bulletin No. 32, Vol. 1, SAEKI, H., ONO, T., and OZAKI, A., 1979. "Experi-
Brussels. mental Study on Ice Forces on a Cone-Shaped
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- and an Inclined Pile Structure." Proceedings
GATION CONGRESSES (PIANC), 1980. "Final International Conference, POAC, Vol. 2,
Report of the 3rd International Commission Trondheim, Norway.
for the Study of Waves." Supplement to Bul- SANDKVIST, J., 1981. "Conditions in Brash Ice-
letin No. 36, Brussels. Covered Channels with Repeated Passages."
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- Proceedings International Conference
GATION CONGRESSES (PIANC), 1984. "Ice Nav- POAC'81, Quebec City, Quebec.
igation." Supplement to Bulletin No. 46, SARPKAYA, T. and ISAAESON, M., 1981. Mechanics
Brussels. of Wave Forces on Offshore Structures. Van
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- Nostrand Reinhold, New York.
GATION CONGRESSES (PIANC), 1986. "List of SCHWARZ, J. and WEEKS, W. F., 1977. Engineer-
Sea State Parameters, Summary." Supple- ing Properties of Sea Ice. Journal of Glaciol-
ment to Bulletin No. 52, Brussels. ogy, Vol. 19, No. 81.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- SELLARS, F., 1978. "Maximum Wave Statistics
GATION CONGRESSES (PIAN C), 1987. "Guide- for Design." T & R Bulletin No. 1-37, SNAME.
66 Port Design and Construction

SEYMOUR, R. J., 1977. "Defining the Wave Cli- ies of Ice Accumulation on Ships, Part 1."
mate for a Harbor." ASCE Proceedings Spe- Translated by E. R. Hope. NRC Canada,
ciality Conference PORTS'77, Long Beach, Technical Translation 1318.
CA. THORN, H. C. S., 1968. "New Distributions of
SHEHTMAN, A N., 1968. "Hydrometeorological Extreme Winds in the U.S." Proceedings
Conditions in Icing Up of Vessels at Sea." ASCE, ST-7, Vol. 94, July.
Translated by Naval Scientific and Technical
THOMPSEN, C., NELSON, A R., and SEDIVY, D. G.,
Center (U.K), Translation No. 2311/71, 1971.
1982. "Wave Group Anatomy of Ocean Wave
SIMIN, E. and SCANLAN, R. H., 1978. Wind Effects Spectra." Proceedings 17th Coastal Engineer-
on Structures: An Introduction to Wind Engi- ing Conference, Cape Town, South Mrica.
neering. John Wiley & Sons, New York.
TOBIASSON, B. 0. and KOLLMEYER, R. C., 1991.
SINHA, N. K, 1984. "Uniaxial Compressive
"Marinas and Small Craft Harbors." Gulf
Strength of First-Year and Multi-Year Sea
Publishing Company, Houston, TX.
Ice." Canadian Journal of Civil Engineering,
Vol. 11, No. 1. TSINKER, G. P., 1994. "Assessment of Physical
SINHA, M. K, TIMco, G. W., and FREDERKING, R., Condition of the Dock Structure at XXXXXX."
1987. "Present Advances in Ice Mechanics in Unpublished report, Acres International
Canada." ASME, Applied Mechanics. Vol. 40, Limited.
No.9. TSINKER, G. P., 1995. Marine Structures Engi-
SKJELBREIA, L. and HENDRIKSON, J. A, 1962. neering: Specialized Applications. Chapman
Fifth Order Gravity Wave Theory and Tables & Hall, New York.
of Functions. National Engineering Science URSELL, F., 1953. "Mass Transport in Gravity
Company, Washington, DC. Waves." Proceedings Cambridge Philosophi-
SODHI, D. S., 1995. "Breakthrough Loads of cal Society, Vol. 49, January.
Floating Ice Sheets." ASCE Journal of Design U.S. ARMY CORPS OF ENGINEERS, 1984. Shore
Engineering, Vol. 9, No. 1, March. Protection Manual. Water Experimental Sta-
SODHI, D. S. and Cox, F. N., 1987. "Advances in tion, Vicksburg, MS.
Sea Ice Mechanics in the U.S.A." ASME, U.S. ARMY CORPS OF ENGINEERS, 1984. Hy-
Applied Mechanics, Vol. 40, No.9. draulic Design of Small Boat Harbors. Engi-
SORENSON, R. M., 1973. "Water Waves Produced neering Manual No. 1110-2-1615. U.S. Army
by Ships." Proceedings ASCE, WW-2, Vol. 99. Corps of Engineers, Washington, DC.
SORENSON, R. M. and WEGGEL, J. R., 1984. "De- U.S. NATIONAL RESEARCH COUNCIL, 1987. Re-
velopment of Ship Wave Design Information." sponding to Changes in Sea Level-Engineer-
ASCE Proceedings Coastal Engineering Con- ing Implications. National Academy Press,
ference. Washington, DC.
STALLABRASS, J. R., 1980. "Trawler Icing." A
U.S. NAVAL OCEANOGRAPHIC OFFICE, 1951. Tech-
Compilation of Work Done at NRC. National
niques for Forecasting Wind, Waves and
Research Council Canada, Mechanical Engi-
Swell, H.O. No. 604. U.S. Naval Oceano-
neering Report ND-56, NRC No. 19372,
graphic Office, Washington, DC.
Ottawa.
STIVE, R. J. H., 1984. "Wave Impact on Uniform U.S. NAVY, 1966. Ocean Thermal Structure Fore-
Steep Slopes at Approximately Prototype casting. USNOO, ASWEPS Manual Series,
Scale." Proceedings Symposium on Scale Ef- Vol. 5, SP-105, U.S. Navy, Washington, DC.
fects in Modeling Hydraulic Structure, Kobus, U.S. NAVY WoRLD ATLAS oF CoASTAL BIOLOGICAL
H. (ed.), IAHR. FOULING, 1970. USNOO, I.R. No. 70-51,
STOKES, G. G., 1880. "On the Theory of Oscilla- September.
tory Waves," Mathematical and Physical VERHEY, H. J. and BOGAERTS, M. P., 1989. Ship-
Papers, Vol. 1. Cambridge University Press. waves and the Stability of Armor Layers Pro-
Cambridge. tecting Slopes. Publication No. 428. Delft Hy-
TABATA, T., IWATA, S., and ONo, N., 1968. "Stud- draulic Laboratory, Delft, The Netherlands.
Port Design and Construction 67

VOITKOVSKII, K. F., 1960. "The Mechanical Prop- Wave Prediction for Port and Channel De-
erties of Ice." Idatel'stvo Akademii Nauk sign." ASCE Proceedings Specialty Confer-
USSR, Moscow (Translation: Air Force Cam- ence PORTS '86.
bridge Research Laboratory. WIEGEL, R. L., 1964. Oceanographical Engineer-
WANG, Y. S., 1979. "Crystallographic Studies and ing. Prentice-Hall, Englewood Cliffs, NJ.
Strength Tests on Field Ice in the Alaskan
WIEGEL, R. L. (ed.), 1981. "Directional Wave
Beaufort Sea." Proceedings 6th International
Spectra Applications." Proceedings of ASCE
Conference POAC '81, Vol. 1. Trondheim,
Conference at U.C.L.A., Berkeley, CA.
Norway.
WATT, B. J., 1982. "Hydrocarbon Extraction in WILCOX, 1958. "Tidal Movement in the Cape Cod
Arctic Fr.ontiers." Proceedings of the Third Canal." Proceedings ASCE, Journal Hy-
International Conference on the Behavior of draulic Division, April.
Offshore Structures, MIT, Cambridge, MA. WIRSCHING, P. H. and PRAsTHOFER, P. H., 1976.
WEEKS, W. F., 1982. "Physical Properties of the "Preliminary Dynamic Assessment of Deep-
Ice Cover of the Greenland Sea." CRREL Spe- water Platforms." Proceedings ASCE, ST-7,
cial Report 82-28, Hanover, NH. Vol. 102, July.
WEEKS, W. F. and AssUR, A., 1967. Mechanical WITTE, H. H., 1988. "Druckschlagbelastung
Properties of Sea Ice. CRREL Monograph IIC3, durch Wellen in Deterministischer und
CRREL, Hanover, NH. Stochastischer Betrachtung." Thesis for
WEEKS, W. F. and MELLOR, M., 1983. "Mechani- the Ph.D. degree, Technical University of
ca] Properties of Ice in the Arctic . Seas." Brauuschweig, Germany.
Proceedings Arctic Technology and Policy. ZAKRzEWSKI, W. P., 1987. "Splashing a Ship with
Hemisphere Publishing, New York. Collision-Generated Spray." Elsevier Science
WEGGEL, J. R. and SORENSEN, R. M., 1986. "Ship Publishers, Amsterdam.
2
Port (Harbor> Elements:
Design Principles and
considerations

2.1 GENERAL As pointed out in the introduction to this


book, in the past 50 years there has been
As a part of a total transportation system, steady and very
significant progress made
ports play a vital role in the overall system in the area of cargo handling that greatly
efficiency. They provide a sufficient storage affected design and construction of the port's
capacity in that a ship need not wait for its loading, unloading, and storage facilities.
cargo from a train and, similarly, a train Use of high-capacity unit trains as well as
need not wait for a ship in order to have a ships have steadily increased, resulting in a
storage area to unload its contents. In other need for higher-capacity, efficient ship load-
words, the port serves as a buffer between ing and unloading technologies. The later
the two transportation modes. resulted in increase in port's capacity to
The port extends its influence through handle the cargo, and in lower unit opera-
the whole chain of the transportation sys- tion costs.
tem and its planning must be well thought Port capacity, which is commonly ex-
off. For example, an underdesigned port pressed in terms of the amount of cargo
which processes ships or trains too slowly throughput, and its efficiency, which is the
or has inadequate storage capacity can be- ability to handle cargo with minimum over-
gin a chain reaction of delays and interrup- all costs, are primarily determined by num-
tions in the system. On the other hand, an ber and size of its berths, storage, cargo
overdesigned port, although able to process handling and hauling equipment, and the
the cargo normally, represents excess in- size of ships able to call at the port.
vestment in the facilities. Thus, whereas in Every port is site-specific and its func-
the former case the interruptions and de- tioning is governed by specific operating
lays may far overweigh the intended sav- conditions and rules that affect port opera-
ings, in the latter case the excess invest- tion. Environmental conditions (e.g., wind,
ment may result in a high cost per tonne of waves, swells, tides, fog, storms, or night-
throughput. time restrictions for some classes of ships)

69
70 Design Principles and Considerations of Port Elements

also affect port capacity and effectiveness of basin(s), and usually specialized, but some-
its operations. times general-purpose terminals. The latter
The approach channels can also have a typically include ship handling facilities
significant effect on port operation due to with miscellaneous cargo handling equip-
the limitation to ship(s) sailing through its ment, different kinds of specialized stor-
bends or passing each other in two-way ages, and the means for cargo transfer
channels. Sometimes limitations in turning between ship and storage, Naturally, any
circles or in the area around the berth can port must have sufficient and convenient
result in reduced port efficiency. The land- connections to the local land transportation
side transportation link (e.g., railway, high- system to provide for effective transfer of
way, pipeline, etc.) may also be a factor cargo to and from the port.
affecting port operations.
The port is an important link of a total
transportation chain. It is built to serve
maritime transportation in coordination 2.1.1 Port Classification
with the inland transportation system.
By definition, a port is a sheltered place In general terms, ports may be classified
where ships can be safely loaded, unloaded, according to their location or function. Ac-
refueled, or repaired. It is also a place where cording to their location, ports can gener-
ships transfer cargo which can be temporar- ally be classified as sea ports, inland ports
ily stored and preserved before being loaded or mixed seajinland ports.
on a ship or transferred by land elsewhere. Sea ports are typically located in artifi-
The term harbor is generally used to cially or naturally protected coastal areas,
define any protected water areas affording in natural harbors (bays), tidal estuaries,
a place of safety for vessels, whereas the and river mouths. Usually these ports are
term port is usually used to define a harbor accessible by large oceangoing vessels. For
which includes terminal facilities that are instance, although some of the large Great
used to transfer cargoes from ships to the Lake ports such as Buffalo, Chicago,
land and vice versa (in some instances, for Cleveland, and Thunder Bay, as well
embarking and disembarking passengers). as the ports of Philadelphia, Baltimore, New
In general, to distinguish from the term Orleans, and many others, are located many
harbor, a port involves some degree of de- kilometers from the ocean, they are also
velopment for the purpose of commerce. considered as sea ports.
Physically, if there is no marked indenta- Inland ports are located on navigable
tions of shorelines, ports may exist without rivers, canals, and lakes. These ports are
harbors (e.g., in inland waterways). Pro- typically served by river boats and barges.
tected or sheltered water areas adequate Mixed sea and inland ports typically
for safe mooring of a vessel can be obtained comprise specialized areas to serve seago-
either by man-made barriers (e.g., moles ing vessels of different sizes andjor river
and breakwaters) or by combination of nat- boats and are convenient for transshipment
ural and man-made barriers. of goods from one port area to another.
The modern port complex includes a According to their principal function, all
number of elements to provide for safe and ports can be either commercial, military or
efficient operation. These elements typically naval bases, fishing ports, small-craft har-
are approach channel(s), breakwater(s), in- bors, or ports of refuge.
harbor and offshore anchorage(s), turning Commercial ports are those which pro-
Design Principles and Considerations of Port Elements 71

vide all the necessary services for handling Small-craft marinas are harbors de-
commercial cargo vessels and handling and signed to accommodate local and transient
hauling of miscellaneous cargoes passing pleasure craft. They provide berthing space
through the port. and miscellaneous services such as fuel,
Commercial ports may be multipurpose food, and a fresh water supply to boaters.
ports to handle different kinds of miscella- They may also have eating, sleeping, and
neous cargoes ranging from general to liq- recreation facilities available. A typical
uid or dry bulk, chemicals, or they can be small-craft marina is furnished with boat
highly specialized to handle only one type of launching and removing equipment, and in
cargo, such as oil, dry bulk (coal, iron, ore, most cases provides boat storage and ser-
pellets), or containers. Also, commercial vice facilities.
ports may include special terminals to serve Ports of refuge by definition are ports
passenger traffic andjor the tourist indus- which are built with the intention to pro-
try. Often in large commercial ports, ship- vide a sheltered area for vessels caught by a
yards offer all kinds of basic ship repair and sudden heavy storm. Essentially, any shel-
maintenance services. tered port can serve as a port or refuge.
There is no uniform internationally Usually, larger ships leave the port during
adopted criteria which define commercial a heavy storm and ride at anchor in the
ports according to size. For example in the open sea.
United Kingdom, the port category is usu- A port may consist of a single marine
ally established according to its share in the terminal with a loading/unloading facility,
total of the country's sea port cargo or it may include several terminals and
turnaround; in Norway the port category is ship service facilities.
typically established on the basis of the size
of the largest vessel which can be accommo-
dated there; Poland's standard suggests
that the port category should be defined on 2.1.2 Port Details
the basis of the ratio of the cargo through- and Definitions
put to ship traffic.
Military ports or naval bases are those Any typical port usually comprises several
which in most cases are used exclusively to specialized zones, some of which belong to
accommodate and serve a fleet of naval ves- an offshore portion and some to an inshore
sels. Typically, they are self-sufficient and part of the port.
have all kinds of ship supply, repair, and The offshore part of the port normally
maintenance facilities. includes exterior access channel(s), a basin
Fishing ports are those which specialize with an interior access channel naturally or
in accommodating different commercial artificially protected from storms, anchor-
fishing vessels such as trawlers and drifters age(s), and maneuvering (turning) and
and a great variety of others. These ports berthing area(s). (Fig. 2-17).
can be relatively small, furnished with fuel, Depending on siting, a port may include
water, and food supply facilities, as well as all of or a portion of the above elements.
with fish handling equipment, or if fairly Functionally, interior access channels pro-
large, they are furnished with a refrigera- vide the transition between open sea and
tion plant, cold-storage, and fish processing protected water environments. Maneuver-
facilities. Fishing ports can be located as ing (turning) basin(s) provide area(s) for a
required, but usually they are a specialized ship to maneuver while approaching its final
part of regular commercial ports. terminus, either alongside berths or in open
72 Design Principles and Considerations of Port Elements

mooring areas. The sizing of water area water, enabling mooring of ships at an angle
dimensions is related to both capacity and to the shoreline. A pier may have one or more
operational requirements. If the protected berths on one or both sides. It may abut land
basin is small enough to accommodate an over its full width or be linked to land by a
anchorage area, then offshore moorings can trestle, causeway, or bridge. It may also be
be provided outside of the protected area. parallel to the shore or land in which cases it
is referred toasT-head pier, or L-shaped pier.
The inshore part of the port normally
Both piers and wharves may be either open
includes marine cargo terminal(s), which pile, or gravity type construction.
provide docking facilities, cargo handling
and hauling equipment, and short-term Anchorage: portion of a harbor (or designated
area(s) outside of harbor) in which ships are
storage facilities; passenger terminal(s),
permitted to lie at anchor.
which provide facilities for passengers to
embark and disembark and for handling of Offshore mooring: a place where a ship can be
safely moored while waiting her turn to dock.
baggage and small miscellaneous cargo; ship
It may consist of ground-secured cables or
repair and maintenance facilities such as
special features for attaching the ship moor-
dry (graving or floating) dock, shiplift, and ing lines. Offshore moorings may be arranged
marine rail or shipway. A railway station is in a way to take just a bow line, or located so
very often an integral portion of the inshore as to take the bow and stem, as well as the
part of the port. breasting lines.
The specific terms which will be used in Mooring dolphins: marine structures for
this chapter and elsewhere in this work are mooring ships. They may be used in combina-
defined as follows: tion with breasting dolphins, piers, or conven-
tional dock structures. Depending on the type
Marine terminal: an area designated for cargo of berth, mooring dolphins may be located
exchange, including areas for vessel mooring, along berthing line or behind it, about 45° off
storage, and other related facilities. In mod- the bow and stem, so that the mooring lines
em practice, terminals are mostly specialized will be no less that 60 m nor more than 120 m
facilities with a limited purpose, such as long. The number of mooring dolphins re-
handling oil, dry bulk, containers, grain, quired is usually dependent on the size of the
passenger traffic, and so forth. ship. Mooring dolphins are provided with
mooring accessories such as bollards, quick-
Berth or dock: a general term to describe a
marine structure for safe mooring of a ship, release hooks, and capstans to handle heavy
mooring lines.
accommodating cargo handling equipment,
loading or unloading cargo, or to embark or Breasting dolphins: designed to absorb the
disembark passengers, or for the purpose of impact of a ship when docking and to hold her
repair or maintenance. against a broadside wind. They are provided
A berthing facility may be land based or with a fender system to absorb ship impact
constructed offshore. In some cases of offshore and to protect the ship from damage. They
construction, a berthing facility may include are also provided with bollards or quick-
only moorings for ships or may be constructed release hooks to handle ship spring lines,
as a fixed berth including a cargo loading/ which hold the ship against alongside cur-
unloading platform and breasting and moor- rents.
ing dolphins. The line along the edge of the Breakwater: a structure protecting a harbor,
berth is called the berthing line. port basin (shore area), or anchorage from
Wharf or quay: one or several bordering docks waves; land connected solid-type breakwater
(berths) placed continuously parallel to the is termed as a mole.
shore. The latter is usually termed a marginal Protected water: sheltered water area(s) pro-
wharf. tected either by natural or manmade wave
Pier or jetty: a structure projected into the barriers (e.g. breakwater(s)), which are ade-
Design Principles and Considerations of Port Elements 73

quate for a safe mooring and safe operation of beam, and depth of hull and draft. These
a ship. parameters are affected by ship type (e.g.,
general cargo ship, bulk-carrier, oil tanker,
containership, passenger ship, navy vessel,
2.2 SHIPS AND THEIR etc.), and ship capacity or tonnage.
For practical design purposes the vessel's
INFLUENCE ON PORT DESIGN
dimensions are defined as follows:

Length: The overall ship length is the length


2.2.1 Ships between the forward and aft vertical struc-
tures. For a ship with a bulbous bow this will
In general terms all vessels may be defined be the overall length between the forward
as follows: perpendicular at the bulb and aft. The length
between perpendiculars is the length between
the forward and aft vertical structures of the
• Merchant
ship's hull at the waterline (loaded line) when
• Passenger
the ship is fully loaded.
• Naval
• Special purpose Beam: This is the maximum width of the ship
• Service hull usually occurring at the ship's mid-
• Small commercial and pleasure crafts section.
Depth of hull: This is the depth usually me-
asured at midship from the bottom of the keel
Merchant vessels are engaged in trade to the top of the main deck.
and transportation of miscellaneous cargoes
Draft: This is the distance from the ship's
such as general (break-bulk), containers, waterline to the lowest point of the bottom of
dry and liquid bulk, heavy lifts, and others. her keel.
All kinds of general cargo ships, container
ships, roll-onjroll-off vessels, different
kinds of liquid and dry bulk carriers, vehi- The following ship characteristics are
cle carriers, ferries, and barges fall in the typically used for the design of miscella-
category of merchant vessels. neous port facilities:
Passenger vessels transport people and Deadweight tonnage (DWT) is the carrying
also provide onboard services in the form of capacity of the ship which includes the total
short- and long-term vacation cruises. weight of cargo, fuels, stores, fresh water,
Naval vessel include a wide variety of and so on. The DWT traditionally was given
warships which varies from small patrol in long tons (1long ton= 1016 kg), but now
vessels to the mammoth aircraft carriers. it is usually given in metric tm:ines. It rep-
Special-purpose vessels are those in- resents the difference between light dis-
tended for use in specific tasks (e.g., placement and displacement when the ship
dredgers, exploration ships, pipe- or cable- is loaded to the load line (the definition for
laying vessels, research vessels, fishing ves- the load line is given in the "displacement
sels, refrigerators, and many others. tonnage" paragraph). The DWT varies with
Service vessels include tugboats, fire- latitude and season as well as with change
boats, pilot, and line-holding boats. in buoyancy due to change in water salinity
Small commercial and pleasure crafts in- (density) in which the ship operates.
clude small fishing boats, local tour boats, Gross tonnage (Gr) is the entire intemal
yachts, and small motor boats. cubic capacity of a ship. It was typically
Ship parameters as they relate to design expressed in units of 100 ft 3 (2.83 m 3 ). Now
of port marine structures include length, it is often given in cubic meters. The Gr
74 Design Principles and Considerations of Port Elements

less space provided for the crew, engine TO


room, machinery, and fuel represents ship's
net tonnage (NT).
Displacement tonnage is the actual weight
TF
or mass of the ship, or weight of water she
F
displaces while loaded or in "light" condi-

---
. . ._ _T
tion. When fully loaded with cargo, the ship

...._
.___s
sits in the water at the load line, which is
also referred to as the Plimsoll mark (Fig. w
2-1). It is visibly marked on the hull of a WNA
ship. This line usually designates the ship's
minimum free board in different bodies of Figure 2-1. Load-line mark (Plimsoll mark), TF-
water during various seasons of the year. tropical fresh water allowance; F-fresh water al-
lowance; T-load line in tropical zones; S-summer
The fully loaded displacement of a freighter
load line; W-winter load line; WNA-winter North
can be roughly estimated by multiplying Atlantic load line; TD-top of the deck line.
the gross tonnage (Gr) by 2.1; the bulk
carrier and tanker displacement is about
1.2-1.4 times deadweight; passenger ship as length, beam, and draft may vary by as
displacement can be taken as equal to about much as 10-15%. Detailed information
1-1.2 times the gross registered tonnage. on the world fleet may be obtained from
The displacement of a ship has been tradi- the latest issue of the Lloyd's Register of
tionally measured in long tons. However, it Shipping.
is now described in metric tonnes. A ship's The United Nation's UNCTAD (1978)
"light" displacement designated her mass recommends, as a rule of thumb, the follow-
without cargo, fuel, and stores. An oceango- ing approximate relationship for the deter-
ing ship would almost never sail in a light mination of full loaded draft (D) of cargo
condition because of her high exposure to ships and bulk carriers at zero speed in
the wind. In this case, ballast water would mean summer saltwater carriers:
normally be added in the holds or ballast
compartments to improve ship stability and DWT ) 0 ·5
D = ( 1000 +5 (m) (2-1)
manueuverability. The ballast displacement
is usually about 20-50% of full displace-
ment, depending on the weather and sea Discussions on today's and future vessels
conditions. Sometimes when required for that will influence the design of the future
safety reasons the ship may approach the ports and effect the modernization of the
dock structure or be moored while in light existing ones are found in Dickson (1969),
condition. Roseman and Jakobsen (1986), Cushing
(1989), Eyre (1989), Haskell (1989), Jones
If ship dimensions are not known, the (1989), and Vickerman (1992).
average data presented in Tables 2-1 and
2-2 may be useful for preliminary design.
Additional information on ship dimensions 2.2.2 Ship Influence on
is found in Quinn (1972), Brunn (1989), Port Design
Agerschou et al. (1983), Thoresen (1988),
and Gaythwaite (1990). Ship parameters have a great influence on
Depending on the country of ship con- port design and particularly on the design
struction, the basic ship dimensions, such of marine structures. In general, the length
Design Principles and Considerations of Port Elements 75

Table2-1. Typical vessel dimensions of cargo and passenger boats, ore carriers and tankers

Fully
Loaded
Tonnage• Length Width Depth Draft Displacement
(m) (m) (m) (m) (t)

Cargo Boats
700 52 8.3 3.8 3.6 900
1,000 60 9.3 4.4 4.1 1,300
2,000 77 11.5 5.8 5.1 2,700
3,000 90 13.1 6.8 5.7 4,000
4,000 100 14.3 7.7 6.3 5,300
5,000 109 15.3 8.4 6.7 6,700
6,000 117 16.2 9.0 7.1 8,000
7,000 124 17.0 9.6 7.5 9,300
8,000 130 17.7 10.1 7.8 10,700
9,000 136 18.4 10.6 8.1 12,000
10,000 142 19.0 11.1 8.3 13,300
12,000 152 20.1 11.9 8.8 16,000
15,000 165 21.6 13.0 9.5 20,000
17,000 173 22.4 13.7 9.8 22,700
20,000 184 23.6 14.6 10.3 26,700
Passenger Boats
500 50 8.2 4.5 4.0 500
1,000 65 10.0 5.3 4.5 1,000
2,000 82 12.0 6.4 5.2 2,000
3,000 95 13.5 7.3 5.7 3,000
4,000 105 14.8 8.0 6.3 4,000
5,000 113 15.8 8.8 6.8 5,000
6,000 121 16.7 9.5 7.2 6,000
7,000 127 17.5 10.2 7.6 7,000
8,000 135 18.2 10.8 8.0 8,000
10,000 145 19.2 12.0 8.5 10,000
15,000 165 21.5 13.0 8.8 15,000
20,000 180 23.0 13.8 9.0 20,000
30,000 210 26.5 15.5 9.5 30,000
50,000 245 30.5 18.0 10.5 50,000
80,000 290 36.0 21.0 11.7 80,000
Ore Carriers
1,000 61 8.9 4.8 3.3 1,300
2,000 77 11.1 6.0 5.1 2,700
3,000 88 12.7 6.8 5.7 4,000
4,000 96 13.9 7.5 6.1 5,300
5,000 104 14.9 8.1 6.5 6,700
15,000 149 21.3 11.5 8.6 20,000
20,000 164 23.4 12.7 9.2 26,700
25,000 176 25.1 13.6 9.8 33,300
30,000 187 26.6 14.4 10.3 40,000
40,000 206 29.2 15.9 11.0 53,300
50,000 222 31.4 17.1 11.7 66,700
60,000 235 33.3 18.1 12.3 80,000
70,000 248 35.0 19.0 12.8 93.300
80,000 259 36.6 19.9 13.2 106,700
100,000 278 39.3 21.4 14.0 133,300
(Table 2-1 continues on next page)
76 Design Principles and Considerations of Port Elements

Table 2-1. (Continued)

Fully
Loaded
Tonnage• Length Width Depth Draft Displacement
(m) (m) (m) (m) (t)

Tankers
300 37 7.0 3.3 3.0 400
500 43 7.8 3.8 3.5 700
700 54 7.9 4.0 3.8 900
1,000 61 8.9 4.5 4.2 1,300
2,000 76 11.2 5.7 5.1 2,700
3,000 87 12.8 6.5 5.7 4,000
4,000 96 14.0 7.2 6.2 5,300
5,000 103 15.1 7.8 6.5 6,700
6,000 110 16.0 8.2 6.9 8,000
7,000 116 16.8 8.7 7.2 9,300
20,000 164 23.7 12.3 9.5 26,700
25,000 176 25.5 13.3 10.1 33.300
30,000 187 27.1 14.1 10.6 40,000
35,000 197 28.5 14.8 11.1 46,700
40,000 206 29.7 15.5 11.5 53,300
50,000 222 32.0 16.7 12.2 66,700
60,000 236 34.0 17.8 12.8 80,000
70,000 248 35.7 18.7 13.4 93,300
80,000 260 37.3 19.6 13.9 106,700
100,000 280 40.1 21.1 14.8 133,300
120,000 297 42.6 22.4 15.5 160,000

• Tonnages of passenger boats are given in gross tons; all other tonnages are given in deadweight
tons.

of a ship influences the width of an access port elements. It depends on ship size, in-
channel, the size of turning basin, and the stalled power, presence of side thrusters,
length of a berth; the beam governs the and local hydraulic conditions. The ability
reach of cargo handling equipment and also of a ship to maneuver will start to decrease
affects the width of the access channel and with the reduction of her underkeel clear-
the size of port basis. The fully loaded draft ance below a certain critical value called
determines the depth of the channel and the "maneuverability margin."
basin and the height of mooring structures. A considerable amount of judgement is
The mass of a design ship determines the usually required to define a ship's maneu-
docking forces which also are a function of a verability. For example, a ship that pos-
ship approach maneuver and speed. sesses good maneuverability in the open sea
Ship maneuverability, or her ability to (e.g., maintaining a stable course at her
perform the maneuvers intended by the ship service speed) is not necessarily course-
master or pilot without assistance of tugs, stable while navigating around tight bends
is an important characteristic which has to in the approaches to a port. PIANC (1995)
be considered in the design of miscellaneous suggests the following rough guide for a
Design Principles and Considerations of Port Elements 77

Table2-2. Typical vessel dimensions of motor, sailing, ferry, and fishing boats, and of barges

Fully
Loaded
Length Width Depth Draft
Vessel Tonnage• (m) (m) (m) (m)

Motor and Sailing Boats


Wooden 100 21.0 6.3 2.6 2.4
Steel 100 25.0 5.3 2.5 2.3
Wooden 200 29.0 7.4 3.4 3.0
Steel 200 33.0 6.6 3.3 3.0
Wooden 300 32.0 8.0 4.0 3.5
Steel 300 38.5 7.2 3.6 3.2
Ferry Boats 50 20 6.0 2.3 2.0
100 25 7.5 2.7 2.5
200 35 9.0 3.2 2.6
300 42 10.0 3.5 3.0
500 50 11.5 3.9 3.2
1,000 64 13.0 4.4 3.4
Fishing Boats
Crabpacking vessel 10,000 162.2 20.7 12.0 10.0
17,000 189.5 23.6 12.7 10.8
20,000 178 22.8 17.4 13.8
Whaling vessel 400 53.8 8.3 5.0 4.7
800 62.7 9.4 5.5 5.6
1,000 68.3 10.2 6.0 5.8
Trawl boat 400 53.8 7.9 6.2 5.8
800 67.2 10.2 6.5 6.1
1,000 76.2 10.7 6.6 6.2
2,000 87.4 13.1 7.0 6.5
3,000 98.6 14.2 7.3 6.7
Barge 50 18.0 5.0 1.5 1.1
100 20.5 5.5 1.8 1.3
150 22.5 6.3 2.1 1.5
200 25.0 6.6 2.2 1.6
300 30.0 6.9 2.6 2.0

• Tonnages of barges are given in deadweight tons; all other tonnages are given in gross tons.

general classification of the inherent ma- 4. Single-screw jsingle-rudder ships will


neuverability of ships: maneuvre quite well, but will experience
screw bias (an offset due to lateral move-
ment of the stern induced by the propeller
1. Long slender ships (LsfB > 6.5) are more
necessitating counter-rudder).
directionally stable than short (L.!B < 6)
ones; the latter, however, will be able to 5. Ships with single controllable-pitch
navigate around tight bends more easily. screws may experience screw bias, even
when the propeller pitch is set for low or
2. In shallow water (hjT s; 1.5) all ships zero thrust.
will turn less readily.
6. Twin-screw /twin-rudder ship generally
3. Low-speed maneuverability may be quite have good maneuverability and control at
different from that at the service speed all speeds.
for which the ship was designed. 7. Twin-screw /twin-rudder ships may have
78 Design Principles and Considerations of Port Elements

good maneuverability at service speed, but 2.3 ACCESS (NAVIGATION)


poor maneuverability at low speed.
CHANNEL
8. Ships fitted with adequate low or other
thrusters may have very good low-speed
maneuverability. Ships with omni-
directional thrusters will generally have 2.3.1 General
excellent low-speed maneuverability.
The access channel is a buoyed unob-
In the above L s is length of the design structed waterway of suitable depth, width
ship, B is the design ship beam, h is water and alignment of which the function is to
depth, and T is ship draught. provide for the safe and convenient naviga-
For more information on ship maneuver- tion in an out of the port harbor. In most
ability the interested reader is referred to cases, but not always, channels are con-
PIANC (1992a, 1982/1983), Heikkila structed for two-way traffic and their
(1990), Herlich (1992), Eden (1971), Kray design is dominated by the largest and the
(1972, 1973). deepest-draft vessel expected to use the
As indicated by PIANC (1985a), if the channel. Sometimes, however, the design
actual underkeel clearance of a large ship is ship may be the one with poor inherent
less than 1 m, a return current between maneuverability or excessive windage, or
ship and channel bottom can no longer ex- one that carries a particularly hazardous
ist. Under such conditions, the resistance to (dangerous) cargoes. To date, there is no
ship movement increases and the propeller sta~dard procedure for channel design and
efficiency decreases significantly. Conse- each design is strictly site-specific. Certain
quently, the turning ability of large ships of guidelines, however, are available. They are
all types is hampered. The negative effect of developed by different national and inter-
reduced underkeel clearance on the turning national bodies, (e.g., PIANC (1995), U.S.
ability of containerships and gas carriers is Army Corps of Engineers EM-1613 (USCE,
even more pronounced (PIANC, 1985b). In 1983), Canadian Coast Guard TERMPOL
most cases, a large ship traveling at a speed code (1983), United Kingdom's National
of about 4 knots (2mjs) with an actual Maritime Institute (NMI, 1981), Navfac
underkeel clearance of about 2 m will re- DM-26.1 (1981), PHRI (1980) and others)
main maneuverable. and modified from time to time.
It should be pointed out that when large The permanent International Association
and very large ships are mentioned in this of Navigation Congresses has been involved
book the following definitions are assumed: in several efforts to review various national
studies and developed several guidelines
(PIANC 1980, 1982/1983, 1985b, 1992,
(a) Oil tankers, ore carriers, and combina- 1995). These reports are essentially sum-
tion carriers-100,000 to 200,000 DWT
maries of relevant information along with
and more
recommendations on design methods and
(b) Liquid gas carriers-transport capacity practical design criteria. Reference to these
of 125,000 m 3 and over reports are made elsewhere in this book as
(c) Container carriers-overall length of applicable.
250m or more The navigation channel is usually classi-
(d) Ferries and roll-onjroll-off (RojRo) ves- fied as restricted or unrestricted. The for-
sels-overall length of 90 m or more, or mer is typically fairly narrow and bounded
45,000 DWT and more by shoreline banks on both sides, and the
Design Principles and Considerations of Port Elements 79

latter is a dredged or naturally deep chan- to providing an adequate distance between


nel bounded by open water of less than the passing ships using the channel in both
required navigation depth. directions and to providing a sufficient ship
In channel design the following factors to bank clearance. These provisions mini-
are to be considered and dealt with accord- mize the risk associated with the ship los-
ingly: ing steerage thereby inducing a risk of a
collision of grounding.
(a) Design ship characteristics such as size The design of the channel cross section,
and maneuverability. In order to ensure with respect to ship grounding risks, re-
that the channel design not be rendered quires an account of the design ship maneu-
obsolete shortly after construction, the vering lane, the width of which depends on
design ship's draft, beam, and length the ship's course-keeping characteristics,
must consider the general trends in ship
and the pilot's experience and capability to
development.
keep the ship as close as possible to the
(b) Environmental conditions, such as credi- centerline of the maneuvering lane track.
ble wind data, visibility, tidal informa-
The proximity of channel banks can ad-
tion, currents, salinity, and littoral pro-
cesses should be considered. It should be
versely influence the ship's course-keeping
remembered that prevailing environ- performance, and therefore adequate clear-
mental conditions, rather than "the ance between banks and the ship's maneu-
worst" conditions, should be considered vering lane is an essential feature of a navi-
in channel design. gationally safe channel.
(c) Navigational, operational, and mainte- An empirical approach to channel design,
nance aspects. This includes the effects which is based on existing practice and
of human factors, such as pilot perfor- takes into account potential navigational
mance during nighttime operations, and and operational problems, is essential.
the skill and knowledge required to However, it is a good practice to submit this
take into account a combination of many channel design to further simulation tests
environmental factors.
in order to establish that the various design
features incorporated are fully compatible
The design process also involves coordi- with the safety requirements of the design
nation with relevant reviewing agencies ship intended to use the channel.
[e.g., local coast guard, port authorities, U.S. Accident statistics show that the major-
Corps of Engineers (in the United States), ity of ship accidents occur in port or in
environmental agencies, and other relevant confined waterways and approaches
organizations]. (PIANC, 1985a). Therefore, safety in the
access channel is a ground for concern, es-
pecially when the channel is used by ships
2.3.2 Navigational and transporting dangerous goods (e.g., liquid
operational Parameters gases, crude oil, and explosives). In modern
ports with heavy traffic a vessel traffic ser-
Safety of navigation should always be the vice (VTS) usually operates in order to con-
main concern of the port designer. Typi- trol vessel movement and to reduce the risk
cally, navigational and operational aspects of collision (Knott, 1989).
will dictate channel parameters. Accord- The VTS is typically responsible for the
ingly, for safe navigation the access channel following:
should have convenient layout and suffi-
cient depth and width. This mainly relates • Exchange of information with ships on rel-
80 Design Principles and Considerations of Port Elements

evant traffic situations and on the ship's U.S. National Research Council, 1992;
condition PIANC, 1992a; Daggett, 1995; Webb, 1995).
• Providing pilotage and search and rescue Input data on a particular area which in-
if required clude visibility, height of tide, tidal current
• Assistance to ships in difficult naviga-
rate and direction, wind speed, gusts and
tional or meteorological circumstances,
direction, tug forces at bow and stern, fog
warning her of obstacles and providing in-
formation on alternative routing signals from other ships in the area, as well
• Scheduling of all ship movements as ship and ship related characteristics (e.g.,
speed of sailing, draft and windward area,
underkeel clearance, beam, and length), are
In the case of channel use by a ship stored in a computerized database. These
carrying dangerous goods, additional spe- data, together with ship maneuvering char-
cial precautions must be adopted. The most acteristics, can then be used by the pilot
basic of these are compulsory piltage, re- in a matter of seconds, to assist in ship
striction to daylight operation, priority navigation.
status, installation of time slots for PIANC (1992a) provides an information
entrance according to tide and strongest on the capabilities of ship maneuvering
cross-currents at the port entrance, prohibi- simulation models which are being recently
tion to enter in excessive wind speeds, or used in the design of approach channels
when visibility is low, prohibitance of any and fairways to and in harbors. PIANC has
overtaking maneuver involving a ship car- concluded that mathematical models can be
rying dangerous goods, and single-lane use an indispensable tool in the design of chan-
in a two-way channel. Additional protective nels, as well as harbors and fairways.
measures would include tugboats sailing in Compulsory pilotage can greatly reduce
standby, port police high-speed boat escort, the risk of ship collision or grounding inci-
and the use of a reliable warning system. dents and the consequent potential danger
Human error due to possible confusion to environment and disruption of port oper-
between controlling personnel and the ations. Pilots to whom the safety of a ship is
bridge personnel at times may be the cause entrusted must be well trained to use simu-
of ship collision or grounding. Therefore, lators adapted to the particular ship ma-
factors such as availability of experienced neuvering characteristics.
pilots, along with limited human ability to It should be noted that a pilot's involve-
react especially under difficult circum- ment in channel design is highly desirable;
stances and comprehend several types of experienced pilot judgement and insight
information having a high repetition rate, should be considered and integrated in
should be taken into consideration. For ac- channel design.
curate and safe maneuvering of large ships Finally, for economy of construction the
in restricted areas such as the access chan- operational limits for channel transit should
nel, it is essential that shipmasters and be chosen such that the time that the chan-
ship pilots be aware of ship and channel nel is closed to navigation to certain ships is
particulars, and therefore be able to antici- minimized, whereas the tidal windows
pate any tendency to deviate from course. should be as large as possible (Vlemmix et
Ship navigator experience along with ship al., 1985). For the same reason, an economic
simulators on a bridge usually help to safely trade-off has to be made between allowable
navigate a ship. draft and the dredged water depth in the
Simulators are extremely helpful for both approach channel and other· port navigable
channel design and ship navigation (McGee areas. However, safety consideration12
and Vann, 1984; Barber and Douglas, 1985; should never be compromised, and the de
Design Principles and Considerations of Port Elements 81

signed maximum draft allowed must be porarily increasing draft. Ship oscillation
controlled to avoid ship grounding. If there will depend on wind strength, angle of en-
are two or more channels, then, again for counter, area exposed to wind, and hull
the sake of economy, one of them may be shape.
designed for use by fully loaded large ships The cross-wind affects the ship at all
and the other (or others) by smaller ships, speeds; however, it produces it greatest ef-
or large ships in light condition. fect on ship when she sail& at low speed.
Wind causes the ship to drift sideways or to
take up an angle of leeway, both of which
affect the required width of channel's ma-
2.3.3 Environmental Parameters neuvering lane. Cross-wind effects depend
on the windage area of the vessel and wind
These primarily are waves, wind, currents, direction. Wind causes less drift at small
fog, and ice. They have a direct effect on underkeel clearances (PIANC, 1995). A
channel geometry; for example, current- and judgment as to the wind speed and direc-
wind-induced drift of a ship directly influ- tion to be used in design of channel's ma-
ence channel width, just as tide and waves neuverity should be based on exceedance
influence channel depth. Variations over curves deduced from wind data collected
space and time in environmental parame- over a larger period. Useful information on
ters have important effects on a ship mov- a subject matter is given in Sohugeu (1995).
ing in a channel. For example, current di- Wave effects. When a ship sails in
rection and intensity may change radically waves, the actual underkeel clearance may
in different part so the channel due to tide become smaller. However, duration of this
and wind conditions. Furthermore, the underkeel reduction is small-being of the
presence of large ship(s) in a narrow or order of just a few seconds. Although waves
shallow channel can also modifY the water may not have a significant effect on ship
flow and thus the currents. maneuverability particularly when she is
All the above can produce destabilizing sailing on a straight course, under certain
effects on a large ship and eventually may conditions, the wave generated drift forces,
cause her to start yawing. In the evaluation may have a significance influence on ship
of available data on environmental parame- maneuverability (Delft Hydraulics Labora-
ters it must be understood that channel and tories, 1984).
breakwater construction may also have a A ship in motion experiences a lowering
pronounced effect on current, waves, and of the water level around her which in-
littoral processes in the area. Because of the creases the required depth of channel. The
complexity of interaction of the above pa- water depth decreases around the ship de-
rameters., physical and numerical model pending on her speed, the relation between
studies are usually conducted during the the ship beam and draft to the width and
design or important ports in order to pre- depth of channel, and ship location with
dict the current, waves, and littoral drift respect to the channel centerline. If the
pattern to be expected. When the port is wave fronts move across the channel, they
constructed, typical new current data must can cause transient effects on the ships yaw
be provided to pilots must be taken at the and also result in some drift in the wave
times of tide reversal when dangerous cou- direction.
pling effects, particularly on large ships, Current effects. Cross-current affects
can be experienced. the ship in a way similar to that induced by
Wind effects. The wind acting on a ship wind. Under the·aforementioned navigation
can cause her to oscillate, thereby tern- condition, the current forces, particularly
82 Design Principles and Considerations of Port Elements

under a small underkeel conditions can be tween n;lvigation and hydraulic aspects
substantial (PIANC, 1979). In most cases, with regard to tide, waves, currents, wind,
the current may have an adverse effect on ice, siltation, visibility, and geotechnical
ship maneuverability. However, if the cur- condition of the seafloor.
rent is from ahead, it may have a favorable Wherever possible, in order to reduce cost
effect on ship maneuvering ability even with of construction and maintenance, an access
a small underkeel clearance. The effect of channel is located in an area where dredg-
the above factors on ship maneuverability ing and potential for siltation are minimal.
is usually considered as a combined effect. At selected locations, the orientation of the
In some ports, the currents may be too channel would preferably be perpendicular
strong at certain states of the tide to allow to the depth contours, provided that this
certain ships to navigate with safety. This does not result in increases in dredging of
may cause their arrivals and departures to hard materials and maintenance dredging,
be restricted to certain time periods in the caused by waves andjor current action.
tidal cycle ("current windows"). This im- Channels with straight alignments coin-
plies times for which the channel will ciding with the deepest troughs are to be
not be available for such ships and the preferred. The ideal channel will be the
decision regarding acceptable downtime shortest possible, to best allow for safety of
levels will be based mainly on economic navigation. However, where curves are un-
considerations. avoidable, their number should be mini-
The lack of thorough knowledge of pre- mized, and they should not be sharp, to
vailing environmental conditions can result permit safe radial steering. PIANC (1995)
in uncertainties in ship maneuverability recommends avoiding bends close to port
during passage through the channel. In entrances. It also suggests that the edge of
some instances, the consequences of the the channel be no less than 2.5 times the
various uncertainties in environmental pa- beam of the design ship from any moored
rameters may have a cumulative effect. For vessel or the edge of the swinging circle of a
example, the decrease of salinity or silta- ship at anchor.
tion will result in a reduction in underkeel In some cases, depending on the design
clearance, which, in turn, may result in conditions, the individual legs and curves
increasing squat of the ship. The additional may have different width and depth, and be
draft in freshwater (density 1000 kmjm 3 ) navigated at different speeds. In all cases, a
is usually taken as 2-3% of the saltwater single curve is better (and preferred) than a
draft (density about 1025 kgjm 3 ). There- sequence of smaller curves at close inter-
fore, the same ship drawing 15.0 m in salt vals. At bends, a channel should be widened
water would draw about 0.30-0.45 m more to account for the fact that the path of a
in freshwater. ship in a bend is wider than in straight
section. PIANC (1995) suggests that the in-
crease in channel width at bends (LlW)
should not be less than L~j8R, where L 8 is
2.3.4 Layout lengths of the design ship and R is bend
radius. For more details consult Herbich
When choosing a channel layout from sev- (1992).
eral alternatives, the one which offers the The criteria for designing a channel curve
more economical solution and allows the radius depend on the angle of deflection,
easiest navigation under most stable and speed of ship, channel characteristics, aids
clearly defined conditions should be se- to navigation, as well as the demands placed
lected. This, typically, is a compromise be- on the pilot. NAVFAC DM-26.1 (1981)
Design Principles and Considerations of Port Elements 83

recommended the following set of rules lengths in deep water, increasing to per-
governing the radius, R, of a channel bend haps 2.8 or more ship lengths at a
(Fig. 2-2): depthjdraft ratio of 1: 1; bends subject to
cross-currents, winds, and waves require
1. Based on angle of deflection a additional width. The way a ship turns de-
a< 25° R = 3L. pends very much on her depths to draft
25° <a< 35° R = 5L. ratio. According to Dand (1981) the ship
a > 35° R = 10£. "sideslip", as ship turns in a bend, can reach
2. Based on ship length (L.) 30-40% of ship beam at a deathjdraft ratio
L. <150m Rmin = 1200 m of 1 : 1, to 100-160% of the beam in deep
L. = 150m Rmin = 2100 m water, depending on the depth of
150 m < L. < 210m water.
Rmm = 1200-3000 m In the past, a curve radius of 1000-2000
m was used when ships were to negotiate
On the basis of the ship traveling speed access channels without tug assistance. In
(VmaJ, Smirnov et al. (1979), recommend the more recent port projects, a radius of
the following criteria for bend radius: 3000 m, which corresponds to (8-10)£8 or
more has been used.
Vmax ~ 5 mjs R ~ 4L., Sea bottom geotechnical characteristics
Vmax > 5 mjs R = 8-10 times L •. are important and should be considered in
the layout design. For example, a hard bot-
Smirnov et al. (1979) do not recommend the tom is more likely to cause damage to the
usage of an angle of deflection greater than ship than a soft one. Therefore, a rocky
30° for ships traveling without tug assist- seafloor may requires a larger underkeel
ance. Dand (1981) suggests that in calm clearance. On the other hand, a soft bed
water with no wind, a hard-over turn may may result in the transportation of solids
be accomplished by an unaided-by-tugs ship and suspended sediments, causing a reduc-
having average-to-good maneuverability tion in channel depth. This may result in
with radius of approximately 1.8 to 2.0 ship frequent maintenance dredging. Therefore,
channel construction in unstable soft soils,
and in particular silty soils, should be
avoided where possible. High susceptibility
of these soils to movement by water flows
generated by environmental sources and
ship-propeller-induced jets is of prime
concern.
When planning a channel course, it is
. desirable that the currents and predomi-
nant winds be met head on. However, where
this is not practical, it is preferable that the
ship drift angle in cross-currents andjor
winds do not exceed 10°-15°, at least at
locations where difficult navigational condi-
tions are to be expected. The channel, par-
ticularly at narrow passages, should have a
Figure 2-2. Channel curve. a = deflection angle;
well-marked straight steering line on both
!::..B =widening on inside of bend; R = curve radius; sides. At narrow passages, such as at
B = width of channel; L. = length of largest ship. bridges or port entrances, the marked
84 Design Principles and Considerations of Port Elements

straight portion should be equal to at least separation scheme. This involves separat-
three to five times the length of the largest ing passing operations by a predetermined
ships negotiating the channel. minimum distance and width of a separa-
The channel layout depends on its opera- tion zone (PIANC, 1980).
tional characteristics such as volume of An inland waterway channel will usually
traffic, speed of ships negotiating the chan- follow a natural river course with cutoffs, as
nel, and associated environmental condi- necessary, to eliminate sharp bends. Ade-
tions. Operational conditions usually deter- quate straight segments between bends are
mine the number of ship maneuvering lanes. required to allow a large ship or tows the
A one-way channel is used when the theo- time to obtain proper alignment for passing
retical channel occupancy (t) equal to t = through the next bend. PIANC (1985) rec-
Lc!Vmax, is less than the average passing ommends that a straight leg between two
time interval required, where Lc is the bends should not be shorter than 5 lengths
length of the channel and vmax is the maxi- of a design ship. River training structures
mum speed of a ship passing the channel. are usually needed in erodible rivers to
The latter is expressed as 24/Qt> where Qt maintain channel dimensions and align-
is the average 24-h volume of traffic. ment.
In the case where Lc!Vmax > 24/Qt> a In all difficult cases of channel design,
two-way channel has to be considered. If physical or mathematical models can help
dictated by economy, a compromise solution to estimate shoaling rates for various chan-
may be adopted which may be a one-way nel layouts.
channel with designated passing areas. In
this case, the maximum distance between
designated passing areas should not exceed 2.3.5 Channel cross section
24Vmax/Qt.
It should be noted that in most cases, The channel cross section may be fully or
vmax is a function of channel geometry (e.g., semirestricted (Fig. 2-3) and should satisfY
width and depth); vmax is greater in wider the following basic requirements: provide
and deeper channels with gentle slopes than
in narrow channels with steep slopes. How-
ever, bearing in mind that under vmax > 5 (a)

mjs conditions, a ship's squat sharply in-


creases, ship movement in channels should
be limited to 5 mjs from the point of view ;:::::;:;:;:::::::::::::;:;.. ··::?::::::::::::::::::::·:·:
of economy of construction and mainten-
ance. Note that, in general, to keep a ship ·-:,:::::::::::::::::::::::::::::;:::;:;:::;:;::;:;::;:::;:::::::::::::::::::::::::::::::::::::::;:::;::;::::::::;::::::::::::::::::::::::t'' ' ' ':::··

under control she has to maintain at least


vmin > 1-1.5 mjs.
The safety of navigation in a channel is (b)
to a large extent related to the traffic den-
sity as well as to diversity of ship type (e.g.,
large or small, fast or slow, etc.) which tran-
sit the channel. The volume of traffic deter-
mines the frequency of passing and over-
taking maneuvers in the channel. The risk
of accidents during such maneuvers can be Figure 2-3. Typical cross section of a navigation
greatly reduced if navigation adheres to a channel. (a) Semirestricted channel; (b) fully restricted
one-way system by establishing a traffic channel.
Design Principles and Considerations of Port Elements 85

for safe and convenient navigation, as well


as economy and minimum maintenance.
- · - - - --+- - ..--- -
O.T. L.
The channel cross section normally is
governed by the size of the largest ship T,

(e.g., maximum design draft and length) - r,


T

W. R. l .
designed to use the channel. As a rule
of thumb, the channel minimum cross-
0
sectional area should be equal to about four 1-
to five times the submerged cross-sectional
area of the largest ship or ships passing in z, s
the channel at the same time; when ex- -
posed to a cross-current, waves, and wind z,
- ~Z i
locations, the factor may be more than 5. z,
In most cases, the access channel has a
H
- z.
trapezoidal cross section. The channel cross
-
Zo
-+--N.C~
section is characterized by its depth (or un- - --- z.
derkeel clearance) and width. - z, 8
-
2.3.5.1 Channel Depth
The channel is the minimum depth of
Figure 2-4. Factor determining the required under-
water in the channel which should satisfy keel clearance. D.T.L.-design tidal level;
the following safety criteria with respect to W.R.L.- water reference level; N.C.B.L-nominal
ship underkeel clearance (Van de·Kaa, 1977, channel bed level; C.D.L.- channel dredge level. T1 =
1984). tidal change during ship transit and maneuvering;
T 2 = allowance for unfavorable meteorological condi-
tions and lack of precise water-level dat a; D = ship
• There should be no accidents involving static draft; E Z i = gross underkeel clearance; Z 1 =
ships, such as collision and grounding due allowance for static draft uncertainties, Z 2 = change
to r educed maneuverability associated in water density, Z 3 = squat, Z 4 = wave response al-
with the decrease of underkeel clearance. lowance, Z 5 = net underkeel clearance; Z 6 = allowance
• The probability that the ship will hit the for bed-level uncertainties; z7 = allowance for bottom
bottom of the channel, which may result in change between regular channel maintenance dredg-
a total loss of a ship or a large scale con- ings; Z 8 = dredging execution tolerance. [From PIANC
tamination of the marine environment, (1985a)]
should be virtually zero.

The required depth of the channel depends water-Level Factors (T l


on different factors related to the ship, to
the bottom, and to the water level. The These include tidal changes during ship
minimal requirement in the determination transit and maneuvering (T1), of which
of the channel depth is that it is passable either may be increased due to unfavorable
by the design ship with the largest draft meteorological conditions (T2 ). The refer-
during high tide. ence depth of the channel at the upper end
PIANC (1985a) recommends that the is the "water reference level" (W.R.L.). This
minimum design depth of a channel should is the water level for the design tidal. condi-
be determined considering water-level fac- tion, including any reductions resulting
tors (T), ship related factors (S) and bottom from the following:
related factors (B) (Fig. 2-4). These are
defined below. (i) Tidal change during ship transit and
86 Design Principles and Considerations of Port Elements

maneuvering and uncertainty of the will increase when the ship is entering an
time and duration of transit of the ship.
estuary which has a lower water density.
Sometimes, in order to maintain a better
(ii) Lack of precise water-level data in the
area. steering ability, a ship is not loaded to an
even keel. This may cause her to set down
(iii) unfavorable meteorological conditions.
at the stern about 0.3-0.6 m. According to
It should be kept in mind that this water Eisiminger's (1953) observations, this may
reference level is different from the chart cause ship set down at the stern approxi-
datum, which may be considerably lower mately 25 mm per 10 m of her length.
than the water reference level when the Furthermore, salinity (density) of the water
ship travels with the tide. The design water may change with the tide and this may
reference level for regular traffic, which may result in some increase in a ship's draft.
not have any time restrictions for usage of Obviously, the aforementioned effects must
the channel, will be at the lowest low water be considered in the determination of a
level. loaded ship's draft. Potential errors in de-
termination of a ship's draft are discussed
Ship related factors fSJ in details by Kimon (1982).
Squat is defined by PIANC (1985a) as
Ship related factors which determine a
the supplementary sinkage of a ship, rela-
nominal channel bed level, include a num-
tive to the calm water level, which is caused
ber of factors which affect ship immersion
by ship movement at a given speed.
in the water: hydrodynamic forces induced
Bruun (1989) explains the squat phe-
by her own movement through the water,
nomenon as the depression of the surface of
interaction with the elements obstructing
water associated with a ship entering shal-
water flow, such as the channel bottom,
low water. This results in a rapid increases
banks, other ships, and by waves and wing.
in the height of waves produced by the ship.
They operate simultaneously and their in-
Accompanying this increase in the wave
teraction is quite complex.
height is an average decrease in the water
Ship factors include the following:
surface along the profile of the ship relative
to the still-water level.
• Static loaded draft in sea water (D)
• Allowance for static draft uncertainties
Numerous theoretical and experimental
(Zl) studies have been carried out to determine
• Change in water density (Z 2 ) squat. An extensive list of relevant refer-
• Squat (Z 3 ) ences is found in Permanent International
• Wave response allowance (Z4 ) Association of Navigation Congresses (1980,
• Net underkeel clearance (Z5 ) 1985a). Substantial contributions to the
study of squat phenomena have been made
These factors are briefly discussed below. by Thiele (1901), Schijf (1949), Constantine
Loaded draft is the ship immersion as (1960), Hooft (1974), Tuck, (1965), Tuck and
measured when she is stopped in calm Taylor (1972), McNown (1976), Van de Kaa
water (usually in the port waiting area). It (1977), Huuska (1976), Barrass (1978),
depends on water density (salinity) and is Sorensen (1966), and others.
typically measured fore, aft, and midship to The following basic phenomena have been
determine the maximum draft, stationary established:
trim, and the amount of hagging and
sagging. • Squat increases rapidly with speed.
The draft and trim change during the • Squat increases with the reduction of un-
voyage due to fuel consumption. The draft derkeel clearance.
Design Principles and Considerations of Port Elements 87

• Modern full-form ships have the tendency Autonome de Marseilles, 1976) with four
to sail "down by the head"; in other words, different size ships varying from 250,000
the bow goes deeper than the stern, al- to 300,000 DWT. It is also in agreement
though ~treamline ships, having a block with model test results and full-scale
coefficient less than 0.7, have the opposite measurements made in the United King-
tendency.
dom (Barrass, 1978; Dand and Ferguson,
• When a ship is sailing into an area where
the water depth decreases, the part of the 1973).
ship with the greatest draft goes even More recently, Ribadeau-Dumas (1982)
deeper. and the French Lighthouses and Naviga-
tion Authority reviewed the state of the art
Hooft (1974) proposed the following at the time and adopted formula (2-3), pro-
equation, based on two-dimensional flow posed by Barrass. It has been generally
theory, to estimate the bow squat of a ship concluded that this formula gives an ade-
of conventional shape which sails in a water quate safety margin and is simple to use:
of constant depth:

v p2
c (-s- )2/3 y2.08
z 3 =~ (2-3)
z3 = 2.4 -2- -------,-~ (2-2) 30 1- s
LPP (1- p2)o.s
where
where C8 =block coefficient C8 = VjLP 0 BD, where B,
2.4 = dimensionless empirical coefficient D, V, V, and LPP are the same as in Eq.
V = volume of displacement (m 3 ); V = (2-2). Representative typical value of block
CBLppBD coefficients for selected vessel types are
given in Table 2-3
LPP = length of the ship between perpendiculars
(m) S = channel blockage factorS =AMIAc, where
AM is the midship section submerged area
C8 = block coefficient
AM::::: BD (m 2 ) and Ac is the channel cross
B =beam (m) section (m 2 ).
D = static loaded draft (m)
F = Froude number F = V j(gh) 0 ·5 , where V = The ratio S /(1 - S) actually represents
ship speed (mjs), g = acceleration due to the channel confinement factors. However it
gravity (mjs 2 ), and h = depth of water
does not take into account the possibility of
(m)
bank effects.
Equation (2-2) will produce the same
Note that F is a nondimensional parameter
result for a given channel whether it is deep
related to ship behavior in shallow water.
and narrow, or wide and shallow. It also
At F = 1 a ship is moving at "critical" speed
of the wave of translation for the depth of
water, and will make extremely large trans- Table 2-3. Block coefficients for selected vessel types
verse waves and will experience a massive Block Coefficients (C8 )
Vessel Type
increase in resistance. Few commercial dis-
placement ships have sufficient power to General cargo 0.55-0.8
attain such speeds in shallow water, and Tanker 0.8-0.9
LNGandLPG 0.7-0.75
operate in the subcritical regime. Bulk carrier 0.7-0.85
Squat values calculated by Eq. (2-2) have Great Lake orejcoal carrier 0.9
been confirmed by experiments conducted Container ship and 0.55-0.65
by Huuska (1976) and by tests carried out passenger liner
by the Marseilles Port Authority (Port Ferry 0.45-0.5
88 Design Principles and Considerations of Port Elements

does not consider the ship's location relative should be noted that on the basis of model
to the bank. Model tests by the U.K. De- tests, full-scale trials, and theoretical stud-
partment of Transport (1982) showed that a ies, Kimon (1982) produced squat and squat
vertical bank in a channel of depth H, at a variance curves that are recommended by
distance of one-half beam from ship side, the U.K. Department of Transport (1982).
may produce an additional sinkage of about According to Kimon the squat values pre-
47% for HjD = 1.1 and about 25% for dicted by these curves are in good agree-
HjD = 1.2, when the ship speed corre- ment with mean values of squat predicted
sponds to F = 0.3. An increase of bank by Eqs. (2-2) and (2-3).
clearance to one beam may decrease the For preliminary design PIANC (1995)
sinkage for HjD = 1/1 and HjD = 1.2 to suggests that the safe minimum depth of
24% and 16%, respectively. The influence of navigation channel should conform to a
the bank does not exist when bank clear- minimum depth/draft ratio equal to 1.1 in
ance is increased to three beams. sheltered waters, 1.3 in waves up to 1.0 m
Additional sinkage may occur when a in height, and 1.5 in higher waves with
ship travels too close to another ship. Ex- unfavorable periods and directions. The
periments conducted by Finland Ships Froude Depth Number must be less than
Hydrodynamic Laboratory reported by 0.7.
Haatainen (1978) indicated that a rapid Useful information on the effects of hy-
change in the depth modifies the water flow. drodynamic forces acting on a ship navigat-
For example, a ship traveling at 6-8 knots ing through a restricted channel is found in
(3-4 mjs) can suffer an additional10-20% Fuehrer and Romish (1982-1983).
sinkage if the channel depth to ship draft More recently Headland (1995a) pro-
ratio decreases from HjD = 1.7 to HjD = posed methodology that accounts for vessel
1.1, or from HjD = 1.8 to HjD = 1.24. motions due to waves and statistical un-
Therefore, if these values are practical, certainties, associated with inaccuracies in
then in channel design this additional sink- input data, used for calculating depth of
age should be added to the design squat navigation channel.
value corresponding to the lower depth. Wave Action. The greatest motion of
Blaauw and van der Knaap (1983), in the ship in waves appears when the fre-
their review of the state of the art in squat quency of the wave encounter coincides with
prediction methods, note that most squat the natural frequency of the ship. Natural
prediction methods are generally applicable periods for heave, pitch, and roll motions of
to large ships such as tankers and bulk large ships are generally over 8 s. In this
carriers. However, due to a lack of sufficient regard, PIANC (1980, 1985a) stated that for
information for smaller vessels such as con- large ships only long-period waves (greater
tainerships and ferries, the reviewers con- than 9 s) need to be addressed when consid-
clude that squat prediction methods are not ering an increase in draft due to wave re-
applicable in these cases. sponse (Z4 ): Shorter waves essentially have
Schijrt (1949), Constantine (1960), Tuck no significant effects.
(1965), Ribadeau-Dumas (1982), and other In general, the magnitude of a ship's ver-
investigators point out that an uncontrolled tical motion (i.e., heave, pitch, and roll) due
phenomenon of a rapid increase in squat to wave action depends on wave parameters
occurs when the Froude number ap- such as height, period, and direction, depth
proaches 1. Hence, care should be exercised of water, and the ship related characteris-
in the channel design process when this tics such as hull shape, draft, and speed.
situation may be expected, such as the case Ship motions may be characterized by a
with fast sailing ships. Furthermore, it response amplitude operator which gives
Design Principles and Considerations of Port Elements 89

the ratio of motion to wave amplitude for a ship striking the bottom. Net underkeel
each wave frequency. This ratio has a peak clearance is usually taken as equal to
near the resonance frequency for pitch 0.5-1.0 m (Tsinker, 1986); the lesser values
andjor roll and can reach a value of about being for sand bottom and slow ship speeds
5 (Van Wijk, 1982). When the natural fre- and the higher values for hard (rocky) bot-
quencies for pitch and roll are nearly equal, toms and fast speeds. In channels that are
the ratio has a high value. This results in subjected to intense shoaling, the net un-
substantial vertical motion. In shallow derkeel clearance may be increased to 1.2 m
water, a ship's motion is less pronounced, or more, depending on local conditions.
and, consequently, a smaller underkeel Gross Underkeel Clearance. The sum
depth tends to dampen the ship's motion Z 1 + Z 2 + Z 3 + Z 4 + Z 5 is, by definition,
according to the change in added masses. the theoretical margin between the keel of
However, the roll motion in shallow water the ship traveling with maximum design
may not be reduced when the wave period draft and the nominal channel bed level. As
is close to the ship's own roll period. recommended by the PIANC (1980), at the
It is quite apparent that the head waves preliminary stage of port planning, the gross
will dominate pitch motion, and beam waves underkeel clearance above the nominal
the roll motion. In general, the resultant seabed level for the design ship can be taken
maximum ship motion is due to oblique and as a percentage of maximum draft as
beam waves. As reported by Campbell and follows:
Zwamborn (1984), the maximum sinkage of
a large vessel sailing in very long waves
• At or near the open sea, exposed to strong
may reach 0.5H8 -2H8 , where H 8 is the
and long stem or quarter swell, where
significant wave. speed may be high-20% of the maximum
Quinn (1972) and Tsinker (1986) sug- draft
gested that half of the design wave height • At a channel exposed to strong and long
should be considered as the amplitude of swell-15% of the maximum draft
ship's pitch, and Smirnov et al. (1979) rec- • At a channel less exposed to strong and
ommended an empirical value of Z 4 = long swell-10% of the maximum draft
0.3Hd = Z 5 , where Hd is the height of de-
sign wave and Z 5 is the net underkeel
Channel Bottom related factors (B)
clearance. In the case of a negative values
These are as follows:
of Z 4 , it must be taken as equal to zero.
It must be emphasized that in channel
design the effect of ship pitch and roll oc- • Allowance of bed-level uncertainties [e.g.,
curring under the influence of waves must sounding (Z6 )]
be given appropriate attention. For exam- • Allowance for bottom changes between
ple, a 1o pitch angle or 5° roll angle for a maintenance dredgings (Z 7 )
• Dredging execution tolerances (Z8 )
ship having a length of 300 m and beam of
46 m would increase the draft by about 2. 7
and 2.0 m, respectively. All of the above channel bottom allowances
For more information the reader is re- are factors in determining channel dredge
ferred to Bruun (1989), Herbich (1992), and levels. Consideration is also given to the
PIANC (1995). construction equipment used for the chan-
Net Underkeel Clearance (Z 5 ) is largely nel dredging. The accuracy of sounding (Z 6 )
an empirical factor used in addition to those is typically equal to 0.15-0.2 m.
factors previously discussed. It is consid- Because underwater excavation is a very
ered as an additional safety margin against costly item, all factors involved in the deter-
90 Design Principles and Considerations of Port Elements

mination of channel depth must be care- Finally, the possibilities of human re-
fully evaluated. lated errors must be considered. The first
If frequent maintenance dredging is re- step in channel design is the determination
quired due to silt accumulation or littoral of the width ofthe maneuvering lane, which
drift, the dredging depth must be deeper is defined as the portion of the channel
than the planned bottom by the amount within which the ship can safely maneuver
determined by local experience. It may vary without encroaching on the bank clearance,
from 0.5 to 2.0 m, depending on local condi- or without approaching dangerously close
tions. It must be noted that the minimum to a passing ship so that interference be-
depth of dredging is usually determined by tween the ships can occur. In the case of
the effective use of available dredging two-way channels where traffic involves
equipment. An excess depth 0.25-0.3 m be- large vessels, a ship clearance lane must be
low the channel design level is generally provided between the two maneuvering
justified without special calculations. lanes.
An access channel constructed in a hard The width of the maneuvering lane (Bm)
bottom should be thoroughly swept to make (Fig. 2-5) may be determined by Eq. (2- 4):
sure that all obstructions, such as boulders,
rocky pinnacles, and so forth, above the
required depth are removed. Bm = Ls sin {3 + Bmax cos {3 + a (2-4)
For more useful practical information the
reader is referred to PIANC (1989), Herbich
(1992), Eryuzlu et al. (1994) and TERMPOL
(1983).

2.3.5.2 Channel Width


This is the nominal width at the bed of
the channel over which the nominal chan-
nel depth is ascertained. The width of a
channel is determined from the maximum
beam and maneuvering characteristics of
the design ship, the volume and nature of
traffic (e.g., regular or involving dangerous
cargos, one- or two-way) channel cross-
section geometry (e.g., fully restricted or
semirestricted), current and wave action,
and winds that will cause the vessel to yaw.
Furthermore, a safety margin is usually
added to ensure a ship safe travel through
the channel.
It should be pointed out that quite often
during the design process adequate and re-
liable data on ship maneuvering character-
istics are usually not available. The size of
B
the design ship must therefore be treated
with caution; the bigger the ship, the more
predominant becomes the influence of iner-
tia and the greater the necessary channel Figure 2-5. Plan for passing ships in two-way navi-
width. gation channel.
Design Principles and Considerations of Port Elements 91

where The bank clearance (BbJ is provided to


reduce the bank suction effect which may
L. = the length of the largest ship
occur due to asymmetrical flow of water
Bmax = beam of the largest ship around the ship. Bank suction is caused by
f3 = maximum drift angle due to current and the contraction of water between the bank
wind; usually f3 = 5°-10° (Thoresen, and the bow of a ship, causing a current in
1988) but generally does not exceed a this constricted area. When the ship runs
value of 25° (Smirnov et al., 1979) too close to the bank, the propeller draws
a = the allowance for yaw due to wind ef- water from between the ship and the bank,
fects; a = t(sin a)Vmax> where t is the producing lateral asymmetric hydrody-
yawing time, a is the angle of yawing, namic forces which cause the stern to drift
and vmax is the maximum speed of the into the bank and direct the bow away from
ship traveling. the bank. This attraction, called "bank suc-
tion," increases rapidly from forward to aft.
Common sense suggests that the influ- To reduce the yaw moment and keep the
ence of various parameters involved in Eq. ship in equilibrium and on course, an in-
(2-4) by their very nature precludes the crease in rudder angle is required (Kray,
determination of Bm with mathematical 1973).
precision and therefore the calculated value Bank clearance is a function of the equi-
must be treated with caution. With consid- librating rudder angle, the width and depth
eration given to the potential disaster and of the channel, and the speed of the ship
economic loss due to ship collision or (Wicker, 1965; Herriman and Herbich,
1977). The equilibrium rudder angle is de-
grounding and where the ship is known to
fined as the angle that causes the course of
have poor controllability, Herriman and
the ship (not the longitudinal centerline of
Herbich (1977), NAVFAC DM-26.1 (1981),
the ship) to be parallel with the channel
and Thoresen (1988) recommend Bm = (1.6
bank. Accordingly, a wider bank clearance
to 1.8)Bmax. In the case where ships with a is required for the 5° angle than for the 10°
very high superstructure such as container
angle. Also, lateral forces acting on a ship
carriers, passenger and ferry ships, and gas, are much lower for a ship located further
oil, and bulk carriers in ballast condition away from the banks and in deeper water.
are used, or where the vessel is known to Additionally, for a given rudder angle, a
have poor controllability the lane width wider clearance is required for ships travel-
might be increased to 2Bmax- Where strong ing at higher speeds than at lower speeds.
yawing forces are expected, the width of the An empirical value of bank clearance
maneuvering lane, according to Wicker Bbc = (1 to 2)Bmax is generally used in
(1965), should be additionally increased by channel design. Smaller values of Bbc are
up to 30-60 m, depending on the character- used for banks with gentle slopes, and larger
istics of the design ship. values used for steeper slopes.
For good, moderate, and poor ship ma- Herbich (1992) recommends Bbc =
neuverabilities PIANC (1995) recommends 1.5Bmax and PIANC (1995) recommends
values of Bm equal, respectively, to 1.3, 1.5, bank clearance values as functions of bank
and 1.8 times Bmax. This document also steepness, channel location (e.g., outer
recommends increase in channel width from channel exposed to open water or inner
0.1 to 3.0 times Bmax depending on various channel, located in protected water) and
values of vessel speed, prevailing wind vessel speed. The bank clearance is defined
and currents, significant wave height and so that a ship, when close to the edge of its
length, aids to navigation, channel depth, maneuvering lane, will experience bank
and level of cargo hazard. suction effects that are at a controllable
92 Design Principles and Considerations of Port Elements

minimum. For example, in the case of slop- Table 2-4. Underwater slope in access channels
ing channel edges and shoals for vessels Type of Soil mo
that sail at fast, moderate, and slow speeds
the bank clearance (Bbc) is equal to 0.7Bmax• Mud and silt 15-30
Silt and shells 10-15
0.5Bmax• and 0.3Bmax accordingly. Simi- 7-10
Plastic silt
larly, for steep and hard embankments or Sand
structures the above values of bank clear- Loose 7-9
ance are equal to 1.3Bmax• l.OBmax• and Medium dense 5-7
0.5Bmax· Dense 3-5
It is recommended that bank clearance Sandy clay 2
Firm clay 1.5
also be supplemented by a t1Bbc value, Stiff clay 1
which is an allowance for possible long-term Rock Nearly vertical
dynamic instability of the channel slope
(Fig. 2-6):
t..Bbc = 0.5H(m - m 0 ) (2-5) relatively reliable method of predicting both
initial and flattened slopes. A successful test
where . may result in the design of a more economi-
m0 = slope immediately after dredging cal channel cross section. Furthermore, the
m = flattened slope. observation of the sedimentation process in
the test channel can produce additional data
The initial underwater slope (m 0 ) of for schedule of maintenance dredging.
dredged channels may be used as indicated For more information on the subject mat-
in Table 2-4. The flattened slope (m) is ter, the reader is referred to PIANC (1984b,
generally dependent on the degree of expo- 1989) and Truijens et al. (1980).
sure to wave and current effects as well as For better control of a ship passing an-
on expected frequency of maintenance other ship in a two-way channel, adequate
dredging. clearance between maneuvering lanes is re-
Reliable slope related data may be ob- quired.
tained by carrying out test dredging. Test Passing problems are particularly pro-
dredging, although expensive, provides a nounced in restricted channels, where

i channel

Bm/2

Figure 2-6. Change in channel bank slope due to long-term


stability (A 1 == A 2 ). !-Dredged (short-term) slope; 2-final slope.
Design Principles and Considerations of Port Elements 93

two-way traffic is faced with decreased con- of one another in a two-way channel, they
trollability during meeting situations, com- tend to spread apart because of pressure
bined with bank suction. When two ships buildup between them. During this phase,
meet in a channel, asymmetric pressure is bank suction plays a positive role by op-
developed on both sides of both ships with a posing this tendency.
tendency of diverting these ships from the 2. When the bow of one ship is near the
stern of the other, lower water at the
path they followed before meeting. Kray
stern draws the bow toward the stern of
(1970, 1973) described this situation as fol- this ship. During this phase, bank suction
lows (Fig. 2-7). will reinforce the yaw of the bows toward
low water to the port of each ship.
1. When the bows of two ships come abreast 3. When the sterns of both ships approach

s
-- K~-=-==-==-=
T \_.I
___)
np
P

t
( : _ _?3--
s

k-::-' =-)
--
-- p
p

~ -- - =~
'-s--v t

- [:(_:n _s

p
:--~
---__,
~

Figure 2-7. Interaction of two ships in two-way navigation channel. P-port;


S-starboard.
94 Design Principles and Considerations of Port Elements

each other, they yaw to port. During this sity and ship sailing conditions, e.g., is she
phase, the bank suction will decrease thisnavigating the outer channel exposed to
dangerous tendency. However, if the ships open water, or is sailing in inner channel,
are close enough, then their sterns will located in protected water. For this matter
have a tendency to collide in the de-
in the case of a fast sailing ship (Vmax = 12
pressed water, created by the passage of
these ships.
knots) in outer channel Be= 2Bmax• when
speed is moderate (Vmax = 8-12 knots)
Be = 1.6Bmax• and for a slow sailing ships
To provide for safe navigation, maneuver- (Vmax ==5-8 knots) Be= 1.2Bmax· When a
ing lanes in a two-way channel must be ship is sailing in the inner channel at mod-
separated by a clearance lane. PIANC (1980) erate speed Be = 1.4Bmax and for a slow
recommends that in a two-way channel the moving ship Be= Bmax· Furthermore, for
passing of large ships is not dangerous if moderate and heavy traffic the above values
the distance apart is at least twice the beam of Be should be increased by 0.2Bmax and
of the larger ship, but the limited accuracy 0.5Bmax, respectively, in the case of the
of the passing maneuver has to be taken outer channels and by 0.2Bmax and 0.4Bmax
into account. in the case of inner channels.
Some authorities (Thoresen, 1988; The design width of a navigation channel
Berriman and Herbich, 1977; Wicker, 1965; is expressed by the following equations
Herbich, 1992) recommend the minimum (Figs. 2-5 and 2-8):
width of clearance lane (B) to be equal to
the beam of the larger ship (i.e., Be 2. Bmax). One-way channel:
PIANC (1995) recommends values of Be
as a function of vessel's speed, traffic den- (2- 6)

(a )
B

B,.

(b ) B

Bm t•.

Figure 2-8. Typical cross section of navigation channel. (a) One-way channel; (b) two-way
channel.
Design Principles and Considerations of Port Elements 95

Two-way channel: The disadvantages are as follows:

1. The contracted time scale. Froude scaling


is used, which implies that the time scale
A review of various one-way channels by
of the model is that of the ship multiplied
PIANC (1980) show that their widths vary by the square root of the scale ratio.
between 4, 8, and even 10 times the beam of Therefore, the further the ship model is
the largest ship: a result of local conditions removed in size from the real ship, the
and circumstances. more contracted the time scale and the
On these bases, PIANC (1980) has of- greater the possible effect on human fac-
fered the only general recommendation con- tors (Wagenaar and Mickon, 1968).
cerning channel width. It has been sug- 2. Scale effects the viscous forces that may
gested that the nominal width of a one-way be incorrect. This may be less pronounced
channel should not be smaller than 5Bmax. when the ship speed is low.
Under adverse conditions, such as the pres- 3. Because distorted dimensional scaling
ence of strong cross-currents, the width cannot be used, the investigator must al-
should be increased accordingly. The width ways compromise between the size of a
of a two-way channel should be as recom- ship model and an unacceptably large and
mended for a one-way channel but aug- expensive waterway model.
mented by 3Bmax to 5Bmax plus the drift
effect due to current andjor wind. Where For more details on physical model tests
the latter cannot be achieved, then simulta- and mathematical (numerical) modeling,
neous two-way traffic for larger ships should the reader is referred to Tumer (1984),
not be allowed. Smallman (1986), PIANC (1992a) and
In some cases, scale models are used to Thevenot (1992).
determine the safe width of the channel. Finally, it should be noted that parame-
However, there are difficult problems in- ters recommended for one- and two-way
volved in scale modeling, such as difficulties channels, as described in this section, may
of simulating shallow water and low-speed be used as a basis for the determination of
navigation, problems of scale and naviga- channel width for busy channels, especially
tional similarity, insufficient knowledge of in sea straits, having cross-traffic. Possible
the currents in port approaches, and, last channel arrangements are shown in Figure
but not least, uncertainties in the human 2-9.
element involved in ship navigation. In a curved section, the width of the
There are some advantages and disad- channel must be greater than that used in a
vantages of the physical scale models in straight channel. The increase in channel
comparison to mathematical models. width (!l.B) is a function of length of a large
The advantages are as follows:

1. They allow direct observation and mea- (a) (b) (CJ (d) (e) (f)

surement of the behavior of ship models


exposed to all local factors that may affect
vessel motion (e.g., seabed irregularities,
presence of cross-currents, waves, wind,
and so forth). Figure 2-9. Type of navigation channels; (a) one-way
channel; (b) two-way channel (passing ships); (c) chan-
2. They allow a large body of information to nel suitable for overtaking ships; (d) channel suitable
be collected in a short time because the for passing and overtaking ships; (e) two separated
time scale of the model experiments is of one-way channels; (f) two separated channels suitable
necessity contracted. for overtaken ships.
96 Design Principles and Considerations of Port Elements

ship negotiating the channel and the radius cated at about 100m from the main struc-
of channel curve (R). It may be taken as ture. At the port entrance, the channel may
equal to approximately L~j8R, where L 5 is be protected by jetties, sheet piling, or other
the length of the ship and R is the radius of structures projecting into the sea. For mis-
a curve (PIANC, 1980, 1985). The value of cellaneous examples of such arrangements,
A.B obtained for the one-way channel should see the U.S. Army Corps of Engineers, Shore
be doubled for the two-way channel. Protection Manual (1984).
It is also generally accepted that the
widening of the inside of the curve serves
the purpose better (Fig. 2-2). However, if
dictated by local conditions, it may be split 2.3.6 Economic considerations
on the inside and outside of the curves
equally or unequally. If widening cannot be Naturally, the wider and deeper channel
achieved due to existing soil conditions or allows for more efficient use of vessels; it
structures, tugs must be used to assist ships. also provides safety benefits by reducing
As a minimum requirement, the ends of the possibility of accidents in the channel
zones having different widths should be and reduces the water transportation costs
joined together by straight lines; the alter- of commodities. However, all technical de-
ation of width should not exceed 10 m for a sign decisions as they apply to channel de-
stretch of 100 m (PIANC, 1980). If more sign must be balanced against their eco-
than one curve is required, then a straight nomic consequences and a compromise is
section between these curves should be pro- usually required between adequate naviga-
vided. Thoresen (1988) recommends that the tional safety, adequate channel capacity,
straight section should be equal to 200 m or acceptable construction and maintenance
twice the length of the design ship, costs, and consequently, harbor (port) dues.
whichever is greater. For the same purpose, The economic value or benefits from nav-
PIANC (1980) recommends that, where pos- igation channel construction are typically
sible, the straight section should be at least generated from the reduction in transporta-
10 times the length of the largest ship and tion costs, which in turn depend on the
PIANC (1995) recommends that a strait leg number of ships that navigate the channel
between two bends should not be shorter and their cargo. A ratio of the estimated
than 5 lengths of a design ship. More useful channel construction costs to that of the
information on curve design can be found average number of ships and their cargo
in Smirnov et al. (1979), Kray (1973), per day can be a useful indicator of the
Berriman and Herbich (1977), Bruun (1989), economics of channel construction (PIANC,
the U.S. Army Corps of Engineers Design 1995). The latter implies that a high value
Manual (1983), PIANC (1995), Herbich ship and cargo may more readily justify a
(1992) and other sources. high channel construction cost. The average
If economically justifiable, the channel ship and cargo values per day should be
may be protected against excessive sedi- obtained from all ships likely to use the
mentation by cutting catchments parallel to channel for the extended period of time, say
the channel (Fig. 2-10). These catchments one year, implying that a higher traffic den-
may be located on one or both sides of the sity will justify a greater channel construc-
channel, depending on sediment movement. tion cost. This will require an iterative pro-
The side catchments not only accumulate cedure, .once the initial construction cost
sediments but also absorb some wave en- estimates have been made, should these
ergy, thus diminishing damage to the main prove to be unacceptably high. It is possible
structure. In silty soils, these cuts are lo- to reduce width of the channel by reducing
Design Principles and Considerations of Port Elements 97

(a)

(b)

( C)

(d)

Figure 2-10. Protection of navigation channel against siltation. 1- channel,


2-sediment catchment cut.

speed of a vessel, increasing the depth/ draft realigning the channel, if possible, andjor
ratio, reducing the cross-current, etc. Some reducing the channel's depth andjor width.
or all of these will have an effect on the The former can produce some savings; how-
operation of the channel (by, for example, ever, if the initial alignment has been prop-
r educing the size andjor speed of the, ac- erly chosen, such savings are likely to be
ceptable ship or reducing the operational insignificant. Reducing channel depth and
windows) and the commercial!economic width may result in restricted use of the
consequences of such measures would have channel, because the design ship may not
to be balanced against any savings in con- be able to pass along it at all states of the
struction cost. The latter can be reduced by tide and other environmental parameters.
98 Design Principles and Considerations of Port Elements

The commercial cost of the down-time asso- breakwaters, or as in most cases by a com-
ciated with reduced tidal windows would bination of the two.
have to be balanced against any reduction Breakwaters are built to protect the har-
in dredging and maintenance costs. Such bor (port basin) from waves and create calm
trade-off studies are usually carried out as water in the basin area. They should be
an integral part of the channel design pro- arranged in a way to provide safe ship navi-
cess. This results in several iterations be- gation in and out, safe mooring, and effi-
fore a satisfactory conclusion is achieved. cient handling of ships at miscellaneous port
The accepted channel alignment and its facilities. Accordingly, the entrance width
cross-section(s) must not compromise safety depends on density of ship traffic, the size
of navigation. Accidents or casualties gener-
of the ships calling at port, and wave and
ate costs in the form of vessel damage and
current conditions at the e:p.trance location.
vessel delays, and increase in the risks of
It should not be wider than necessary and
these events increases the expected damage
is usually a compromise between the safe
and delay costs. Hence, changes in channel
configuration may affect costs of trans- navigation requirements and the degree of
portation and therefore economics of the harbor protection. It may also be affected by
channel (Haneman, 1994). For useful infor- some particular local conditions.
mation on economic values of risk reduc- Minikin (1963) and Thoresen (1988) rec-
tions in navigation channels, the reader is ommend that, in general, the entrance
referred to Moser et al. (1995). width should be equal to about 0. 7-1.0 times
the length of the largest ship expected to
call at port. Quinn (1972) states that the
entrance width should be in proportion to
2.4 PORT (HARBOR) ENTRANCE the size of the harbor and the ship using it.
In general, Quinn found that widths of
An access channel would normally lead to a about 90 m for small harbors, 120-150 m
port (harbor) entrance. In most cases, one for medium harbors, and 150-240 m for
entrance into a commercial harbor is suffi- large harbors may be considered satisfac-
cient for adequate functioning of a port. tory.
However, in some cases, particularly at Bruun (1989) suggested that the re-
coastal ports stretched along the coast line, quired entrance width should conform to
two and sometimes three entrances are the same criteria used for the access chan-
used. nel design, assuming a reasonable yawing
Ideally a port entrance should be located angle equal to about 10°. Bruun also recom-
on the lee side of the harbor. If it must be mends that an allowance must be made for
located on the windward end of the harbor, the effective entrance width in case a ship's
then adequate overlap of the breakwaters approach is not at a right angle to the
should be provided so that the harbor inte- entrance. Smirnov et al. (1979) recommend
rior is protected from wave action, and the
the following equation be used to determine
ship is able to pass through the restricted
the entrance width (BE):
entrance and be free to turn with the wind
before it is hit broadside by waves. How-
ever, it is usually very difficult to com-
pletely protect the harbor interior from the BE= Bmax + ( ; : )Ls + vmint sin {3 +ABE
effect of waves. Typically, a port entrance is
(2-8)
formed by shore-connected or offshore
Design Principles and Considerations of Port Elements 99

where and normal component of ship drift under


Bmax = beam off the largest ship (m) wind
v"n = normal component of speed drifting out V.,:' = 1.0 knot (0.5 mjs).
of course under the action of current and
wind (mjs)
Hence, Van= 0.75 + 0.5 = 1.25 mjs.
vmin =minimum speed of ship travel when she
enters the port protected area (m/s)
3. Ship yawing condition
L. = length of the design ship (m)
t = time of ship yawing (s) (usually t is taken t = 60 s and f3 = 10°
as equal to 60 s)
f3 = yawing angle f3 = 3o to 10° depending on 4. Width allowance
wind/ current effect on the ship
!::.BE= width allowance to prevent the ship from !::.BE = 1.5Bmax = 45 m.
colliding with breakwater(s) or channel
slopes (m) 5. Ship travels at Vmin = 3 knots (1.5 mjs).

It should be noted that the value of 6.BE Hence, According to Eq. (2-8)
depends on the amount of entrance protec-
BE= 30 + (1.25/1.5)220 + 1.5
tion. It can be taken as Bmax for well-
protected entrances and 2 Bmax for less pro- X 60 X sinl0° + 45=274m
tected ones.
The drifting speed of the ship, which oc- Note that the calculated width of entrance
curs due to wind and current action (Van), is can be reduced if the largest design ship
given by will be assisted by tugs when entering the
harbor.
(2-9) An additional allowance in the design of
an entrance's width should be considered
for ports operating during the winter sea-
where
son when presence of ice requires the use of
yen =normal component of cross-current acting icebreakers. For detailed information on the
on the ship (mjs)
ice effects on harbor operation, the reader is
V.,:' =the speed of the ship's drift under the referred to Tsinker (1995).
normal component of the strongest wind
acting on the windward side (mjs) To limit wave penetration into the har-
bor area, the entrance must be as narrow as
practical. However, it must be wide enough
to prevent development of strong currents
Design Example in the entrance area associated with the
tidal regime in the area. According to
Determine the width of port entrance for Cornick (1958), Grothaus and Ripley (1958),
the following design conditions: and Quinn (1972) to prevent possible scour
at the base of adjacent breakwaters and to
1. Design ship length L. =220m and beam allow safe navigation, the tidal velocity at
Bmax =30m.
the entrance to tidal harbors should not
2. Normal component of current at entrance
exceed 2.5-3.5 knots (1.25-1. 75 mjs). If the
location
current velocity exceeds these values, then
the entrance geometry should be adjusted.
Vt = 1.5 knot (0.75 mjs) Only under special circumstances should
100 Design Principles and Considerations of Port Elements

. .·.:~'l',?'.''~·.··i·',i.'!, ··.·~··:". •::;.: ..•'i:··, 1• • • • \: ..

:~:::=.

cNrRANCE 8 AA; I N :~;k~~


~rr;;,;s:~,::'i!l))';');;);.,;¢~ :;;;(i;!j;:1:~·;;;,:;V/·i:~t!i.\$)'
.. ,.,
jf~f;

p L A N

P R 0 F L E
Figure 2-11. Basin with nonconstricted entrance. [From Sorenson (1977).]
Design Principles and Considerations of Port Elements 101

the current exceed 3.5 knots, in which case (b) The basin water area is relatively
the ships calling at port must be assisted by constant.
tugs. (c) Freshwater inflow in minimal.
The following procedure is recommended (d) The ocean tide is approximately
in NAVFAC DM-26.1 (1981) to determine sinusoidal.
tidal currents in entrance channels:
Then a good approximation for the cur-
A If the entrance is not constrictive (Fig. rent velocity (Vc) can be determined by
2-11), the following conditions (a through Eq. (2-11)
d) are fulfilled:
(a) The basin is relatively short and deep (2-11)
and must conform to condition de-
scribed by Eq. (2-10):

Lbj(gd) 0 "5 T ~ 0.05 (2-10) where


Vc = average cross-section velocity at
where: maximum tidal flow (mjs)
L b = basin length (m) T = period of tide (s)
d = average basin depth (m) Ab = surface area of basin (m 2 )
g = gravity acceleration, g = 9.81 Ac = cross-sectional area of entrance at
mjs 2 mean tide level (m 2 )

PROFILE
Figure 2-12. Basin with constricted entrance. [From Sorenson (1977).]
102 Design Principles and Considerations of Port Elements

a. = one-half the range of the ocean tide The coefficients K 1 and K 2 are defined
(m) as follows:

Note: The entrance not being constric- (2-13)


tive, together with condition (a),
implies that the water surface in
the bay fluctuates uniformly and (2-14)
equals the ocean tide.

B. If the entrance is constrictive (Fig. 2-12), where


thereby reducing the tidal range in the Lc = entrance channel length (in)
basin, Eq. (2-11) will overestimate the Fe= ken+ kex + fLc/4R in which ken is
tidal currents. Provided that the four con- the entrance-loss coefficient (ap-
ditions in A are satisfied, the maximum proximately 0.1) and kex is the
current in the constrictive entrance can exit-loss coefficient (approximately
be determined from Eq. (2-12):12 1.0)
f = Darcy-Weisback friction factor (ap-
proximately 0.03)
(2-12) R =hydraulic radius of inlet channel (m)
K 1 represents the ratio of the magnitude
of the friction forces and inertia forces.
where e is the dimensionless factor which K 2 is a measure of the magnitude of the
depends on coefficients K 1 and K 2 . inertia forces relative to the magnitude of
(Figure 2-13.) the pressure (water-level) gradient.

--
1.2

1.0

..............
--
' '.........
......
......
... ~
.......
r--.., --- ..............
-........11' 10

K1: 30........
r-......
-r-- ""'~"- K1= 3

' ' ' ' ' 'r--.


0.8
~~·100.......... r-....
' \ ...
'
{/1300"' ~

" "'
e 0.6

' , I~ i'.
.... r-.r-..
.....
~t 1000"' ~
0.4
.... ... ~<t= 3ooC I'!'-. ....
'
............... r--...
........... ......... r--.....,
0.2 ......... I'-..
..... !'-.
-.......... ... ......... ...... ~~
"""' """"
0
0.01 1.0

Figure 2-13. Dimensionless maximum velocity (e) versus K 1 and K 2 • [From Sorenson (1977).]
Design Principles and Considerations of Port Elements 103

Because the entrance is constrictive, the where:


amplitude of the bay tide will differ from R = average hydraulic radius of channel (m)
the amplitude of the tide range in the ocean. Ac = average cross-sectional area of the channel
The bay tide amplitude may be determined at mean tide level (m 2 )
from R,. = hydraulic radius at each of n sections of
equal length !lX,. (m)
(2-15) A = cross-sectional area of the channel at each
" of n sections of length !lX,. (m2 )

where: In general, an entrance should be located as


far as possible from the shoreline. When
E = a dimensionless factor which depends on selecting an entrance location and its orien-
the coefficients K 1 and K 2 . (Figure 2-14) tation, safe navigation requirements as well
ab = one-half the range of the bay tide (m) as wave penetration effects into the basin
must be considered. The entrance must be
It is noted that for values of K 1 , which designed in such a way that a ship shall not
indicates large inertia forces, the value of € need to make any maneuvers at the en-
is greater than 1. For irregular entrance trance. The ship can start her maneuvers
channels, an effective channel length I.:c, only after she passes the entrance and en-
can be used in place of Lc: ters the protected area. To avoid danger of
having a ship grounded under the effects of
heavy wind, waves, and current, the en-
(2-16) trance channel shall not be placed parallel
to the shoreline. The angle between the

1.2
"-
;'!.;" ~ .... ,K1• 3
" 1.0 ....... ....... ~1(1:10
\U
F': r---....
F;;;
~~-.
~
~

"'\.
IJJ
0 '-..... ~ f\.K 1= 30 ~
i: 0.8 '\ '\

' 1\.
~
..J ~ 1 =3P~ I w ·JO"
f"ii~ ~
'
Q.
~ 0.6
..J
~ ~K,· 300" ~ r"\
<
"'
' "',...
0 """~ K 1• t,ooo""".,
i= 0. 4
~ ~ ~
< ....... .......

"'
IJJ
"-...
""""
Cl'l ~

r--..... f""'-o.. ,......


oO. 2 ..........
1- ~-....
!""'- ~ ......
>
<
al
~.01 0.1 I JJ

Figure 2-14. Ratio of bay to sea tidal amplitude versus K 1 and K 2 • [From Sorenson (1977).]
104 Design Principles and Considerations of Port Elements

(a)

(b)

Figure 2-15. Port entrance orientation: (a) entrance formed by two shore-
connected breakwaters; (b) entrance formed by two shore-connected breakwaters
and offshore breakwaters. !-shore-connected breakwater; 2-offshore breakwater:
3-access channel; 4-ship; 5-direction of major storms; 6-line parallel to the
shoreline.

channel axis and shoreline a 2 (Fig. 2-15b) gle, a 2 is usually taken as equal to 30°-35°.
shall not exceed about 30°. On the other If shore-connected breakwaters cannot pro-
hand" from the standpoint of navigational vide sufficient protection, then an offshore
safety, it is desirable to keep the angle a 1 breakwater shall be placed to mitigate the
(Fig. 2-15a) as small as practical. This helps wave situation (Fig. 2-15b). An offshore
to prevent ship collisions with breakwaters breakwater can also improve ship entrance
under heavy wind and wave action. into the basin.
In practice, for safe navigation and bet- Finally, where applicable, the entrance
ter protection of the basin from waves an- layout must provide for protection against
Design Principles and Considerations of Port Elements 105

wind-driven ice penetration into the harbor harbor located within protected water areas
area; at the same time, it must not hinder may have reduced depths.
ice movement out of the basin. The water Typical entrance cross-sectional geome-
depth at the entrance is usually taken as tries are shown in Figure 2-16. In the case
equal to the depth of the access channel. It of an entrance shaped as is shown in Fig-
should be noted that an entrance is usually ures 2-16c and 2-16d, the entrance width
exposed to larger waves than those found in should be marked by navigation buoys.
a port basin. Therefore, those parts of a For more useful information on subject

. . ;L;,;.,w; ;,;·,;;,L.,..,...._-+~---.---J:~-.·: ~ ::·:: ·,


: ::~.
::.. -_.: .. .
i.._ •• • ••
·.. .·· I -~.

·;. _:
:-·~

··. ·.
...:
...:: :;
i
I
.:- . .· . .

Ia> (h)

.. L.W. L.
.· .·..- . -'. :::
·~ · ! ' : .:.·.-... ·:.
:;·· ;~- ;.. ·.. :
I
~
•• _.- : . . , . 0

DE
-~
-~ ...
·...

--+--- 1
(C)

DE

(d)

Figure 2-16. ·Port entrance: (a) and (b) entrance formed by vertical wall breakwaters; (c)
dredged entrance protected by vertical wall breakwaters; (d) entrance formed by sloped
(rubble-mound) breakwaters.
106 Design Principles and Considerations of Port Elements

matter the reader is referred to U.S. Na- mining the wave disturbance inside the
tional Research Council (1981) and Herbich harbor resulting from storm conditions out-
(1992). side the basin. To be effective, these models
must realistically represent the local envi-
ronmental conditions, bathymetry, and lay-
out geometry.
2.5 PORT WATER AREA
Several good numerical wave prediction
(HARBOR) models have been developed. Two examples
are the SEYWAVE and HINDWAVE mod-
A port water area or harbor by definition is els (Hawkes, 1985). The SEYWAVE model
a sheltered water area, bounded by natural uses the Seymour (1977) method which
features, man-made structures, or a combi- takes into account the fact that the wind
nation of both, where ships may seek refuge, generates waves at an angle to its own
transfer cargo, obtain services, or undergo direction. It is particularly useful for pre-
repair or maintenance. dicting waves in areas with relatively long
The ideal harbor will be completely shel- approaches from the open sea. The input to
tered from open-sea waves, have minimum this model is in the form of an idealized
tidal range, moderate currents, and free- wind time history with a steady wind speed
dom from long-wave agitation (seiche), ice, persisting for a given duration. Alterna-
and fog. Additionally, it will be large enough
tively, the HINDWAVE model, which is a
to provide for safe entrance and departure
wave hindcasting model, uses recorded wind
of ships, safe passage of other ships either
data in the form available from anemo-
in motion or under loading/unloading oper-
graph stations.
ations, and mitigate possible wave action
The use of ship simulators can greatly
against moored ships. The harbor should be
assist in the port design process by predict-
divided into areas where different func-
tions, such as ship maneuvering, turning ing the ship dynamic behavior under dif-
areas, anchorage area, and berthing area, ferent kinds of operational conditions and
can occur. reduced underkeel clearance. However, as
In designing the port layout in general, with any mathematical model, it is neces-
and its basin in particular, a number of sary to have a clear idea and understanding
different factors, such as hydraulic criteria, of its range of applicability and to be aware
practicalities of different types of construc- of how successful the model represents
tion, limitations of existing structures, as physical processes described herein.
well as short- and long-term requirements, It is appropriate to mention that the re-
need to be considered. cent popularity of personal computers has
An appraisal of these constraints will resulted in a flood of problem-solving soft-
normally allow the port designer to arrive ware. It was quite rightfully noted by some
at a number of alternative designs. Once experts that the quality of the software
this process is completed, the preferred so- presently available ranges from very good
lution, or a number of options, can be se- to unacceptable. Therefore, the users of
lected for further scrutiny. mathematical models must gain a full un-
The selected alternative may be further derstanding of the software they use,
optimized by using physical or mathemati- through either a software documentation
cal modeling of the port or port basin. Mod- study or through the use of relevant consul-
els, both mathematical and physical, are tant services.
used to examine the degree of shelter pro- Verification of the mathematical model
vided by the harbor. This is done by deter- validity may be achieved by conducting
Design Principles and Considerations of Port Elements 107

physical modeling studies or by prototype breakwaters and port entrance structures.


measurements. It is possible to conduct a comprehensive
A random wave physical model of a port investigation if the model is at a sufficiently
is the most comprehensive type of model large scale (Owen and Allsop, 1983).
available to the engineer involved in the
design of port layout. The scale of such
models is typically within the range of 1 : 60 2.5.1 Basin Sizes
to 1 : 120, with smaller harbors being repre-
sented at a larger scale. Random waves Basin sizes, such as length, width, and
have been used in physical models for over depth must be sufficient to provide for safe
a decade and have been found to give and convenient ship entrance into the basin
results which are more realistic than and approach to the terminal, being either
those obtained from regular wave tests self-propelled or assisted by tugs.
(Smallman, 1986). In random wave model- A typical port water (offshore) area in-
ing, the number of individual tests can be cludes the ship stopping area, maneuvering
significantly reduced because a single ran- area, turning basin, anchorage area,
dom wave test covers a wide range of inci- berthing area, and sometimes a special-
dent wave frequencies. The random wave purpose area (Fig. 2-17)
physical modeling technique is therefore
less costly than regular wave tests.
2.5.1.1 stopping Length
In addition to wave distributions inside
the basin, which may affect the ship's ma- From the viewpoint of a safe ship en-
neuverability and vertical response to trance, the length of the basin adjacent to
waves, the engineer must also be concerned the entrance and immediately inside the
with stability and hydraulic performance of basin should be sufficient to allow a ship

en Sea Environment Protected Area

,--·~
· - · -·~· ---,
"""'"• "".. "''K'"'' '"""' i // ... --- \I i
- ------ ~, 0·-----4. i
~- --~-- .\, //! i
7 LT{~~
Figure 2-17. Port water (offshore) area: 1- access channel; 2- interior channel (stopping
area); 3- maneuveringj turning area; 4- anchorage area; 5- berthing area; 6-special-purpose
area; 7- breakwater.
108 Design Principles and Considerations of Port Elements

entering at a reasonable speed to stop component coming from astern should not
safely. A preliminary ship stopping length exceed 0.15 mjs.
(Lst) may be determined by Eq. (2-17) The port designer should appreciate the
(Dzhunkovski et al., 1964): vital importance of the ship stopping dis-
tance during the planning stage. There is a
0.27V~D wealth of experience to show that a ship-
Lst N (2-17)
= master or pilot, when navigating an ap-
proach channel to a confined harbor, must
where bear in mind that large ships should be
V 0 = speed of ship (knots) navigated at the lowest possible speed. The
D = displacement (tonnes) use of tugs to assist the entering ship may
N = installed power (hp) influence the length of the stopping dis-
tance.
In general, the stopping distance of a ship In the modern science of port engineer-
depends on such factors as ship mass ing, a rational stopping length is usually
(displacement), traveling speed, installed obtained by thorough planning and evalua-
engine power, and local environmental con- tion of a ship maneuvering into the basin.
ditions. No general rules exist for determin- The latter is best obtained by either testing
ing the length of this portion of the basin. It physical models or using computer simula-
is usually taken as equal to 3.5 to 8 times tion models.
the length of the longest ship expected to
enter the port, subject to ship mass and 2.5.1.2 Maneuvering Area
speed of entree.
Table 2-5 shows some practical data on Before a ship starts the actual berthing
ship stopping distances which are required procedure it usually stops in a maneuvering
when using full astern power from full speed area, or a turning circle.
and from 4 knots (2 mjs). The stopping The maneuvering area usually includes
maneuvers should be carried out preferably the necessary area to allow a ship to reduce
in areas sheltered from currents or, if this speed and the necessary swinging area. In
is not possible, with the ship stemming the general, the maneuvering area is located in
current. If stemming the current is not pos- a protected water area. Under such condi-
sible, an area at least partly sheltered from tions the ship's motion due to swell arid
significant cross current is required. As rec- squat can be neglected. Consequently, the
ommended by PIANC (1980) both the cross depth of the maneuvering area can be de-
current component and the longitudinal termined on the basis defined in Section 2.3
with values of Z 3 and Z 4 that are equal to
zero. The length of the water area required
Table 2-5. Stopping distance, with full astern
. to bring a ship to a full stop includes a
power from full speed and from 4 knots
speed decreasing length, in which the speed
Stopping Distance; Engines of a ship decreases to less than 4 knots (2.0
Deadweight Going Full Speed Astern mjs), and a stopping distance, where the
Tonnage Length From Full From speed reduces to zero.
(Long Term) (m) Speed 4Knots The width of the necessary water area
18,000 170 1,600 240 beyond the port entrance, where the ship
50,000 230 2,400 250 enters a calm water area and is no longer
110,000 265 2,600 350 affected by possible cross-current and
210,000 330 4,000 520 waves, is less than required width in the
Source: From Dickson (1969). approach channel. However, in a transi-
Design Principles and Considerations of Port Elements 109

tional water area, where the ship bow en- be noted that usually large ships approach
ters calm water at the end of channel but the berthing structure with tug assistance
the stern is still subjected to cross-current only. Before the berth approaching maneu-
and/ or swell action, the width of the water ver, she stops parallel to the dock structure
area must be sufficient to account for the about 50-100 m away, then slowly ap-
steering to which the ship is subjected. Fur- proaches a dock structure while being con-
thermore, the width of the maneuvering trolled by tugs.
water area must account for the drift dur-
ing the stopping maneuver due to the en- 2.5.1.4 Anchorage Area
gine of a single-screw ship running full
speed astern. A large natural harbor may offer suffi-
cient water area for anchorages, where ships
2.5.1.3 Turning Basin can be safely anchored while waiting their
turn for berthing in favorable weather con-
The water area needed for the movement ditions, or for any other reasons. If economi-
of a ship during the turning maneuver is cally feasible, an artificial harbor should
defined as a turning basin. A turning basin provide space for an anchorage for small or
may be of various shapes, depending on the relatively small ships, whereas the large
size of the port-protected water area, the ships can be anchored out in the open sea,
size and maneuverability of the ship, and keeping out of the path of the main port
the layout of berthing facilities. traffic.
It is usually considered that during the The water depth at the anchorages should
turning operation the current in the turn- be no less than required for regular traffic
ing area is less than 0.1 mjs and that if the in the area, but preferably should not ex-
turning ship is in a light condition, then the ceed 50-60 m because of the limited length
wind speed should be less than 10 mjs. If of the anchor chain on a ship. Hard bottom
these conditions are not met for any reason, conditions should be avoided because the
the ship needs to be assisted by tugs ade- anchor would not be able to penetrate
quate in number and power. An example is the hard soil and could be dragged along
shown in Figure 2-18. the bottom.
If the available water area permits the In the unlimited space of the open sea,
largest design ship to turn a complete circle the water area required for ship anchorage
without the aid of tugs or side thrusters, is equal to a circle that enables a ship to
then the minimum diameter of this circle swing clear of any adjacent obstacles. In the
should be a factor of 3.5 to 4 times the latter case, the ship will swing at anchor
length of the ship, depending on ship ma- and be generally parallel with the wind or
neuvering characteristics (Fig. 2-19). Where current. The radius of the circle may be
the ship has tug assistance, the turning estimated as equal to the ship length plus
diameter could be reduced to two times the 6-10 times the water depth at the anchor-
ship length or even less. If the ship is turned age location plus 20-50 m of clear space
by warping around a dolphin, or around the between the ship and adjacent obstacles.
head end of a pier, the turning diameter The radius of the swing can be reduced by
could be a minimum of 1.2 times the length mooring ships to a mooring buoy and by
of the ship (Figs. 2-24 and 2-26). In con- adding weight (sinker), such as concrete
fined areas where the ship cannot complete blocks or special metal features, for holding
the circle and requires assistance of tugs, between the buoy and anchor down to the
the minimum maneuvering radius should seafloor (Fig. 2-21). If required, additional
be R = 2.5L 8 to 3.5L 8 (Fig. 2-20). It should offshore moorings may be provided to limit
110 Design Principles and Considerations of Port Elements

horizontal movement of the ship. In this Naturally, the number of mooring points
case, the ship can be moored with her own depends on the size of the ship, or the
anchors to a buoy or group of buoys, or by number of ships in one group, as well as on
any combinations of her own anchors and the environmental ·forces. Typical compo-
buoys. Some of these combinations are nents of a mooring buoy unit are shown in
shown in Figure 2-22. It should be noted Figure 2-21.
that ships also can be moored in groups as AB a general rule, port service vessels are
shown in Figure 2-23. available to assist a ship with mooring op-
In the most common arrangement, the erations at anchorage areas.
ship's anchors are placed at the angle of 30° Size and orientation of an anchorage area
and stern lines are secured to mooring buoys depend on the required capacity and method
at the angle of 45°. Where strong winds of operation and must take into account an
may occur broadside to the ship, breasting independent ship approach to each individ-
lines are added to the ship mooring system. ual anchorage.

Figure 2-18. Tug-assisted very large crude carrier. (Courtesy Port


Autonome de Marseille.)
Design Principles and Considerations of Port Elements 111

In a typical offshore anchorage, the ship


will be approaching an anchorage area
heading into the wind, whereas in inland
ports, typically, the ship will be heading
upstream. Inland waterway anchorages are
typically located upstream (for departing
ships or tows) and downstream (for ar-
rivals) off the port location. Hence, ships or
tows arriving from the upstream locations
must pass the anchorage first, then make a
full turn in order to approach an anchorage
heading upstream.
At inland waterway anchorages, ships are
usually anchored in line, or several lines
Figure 2-19. Turning basin in unrestricted port with clear space between ships in line of not
water area: L.t-ship stopping distance; !-turning less than 50 m. From a safety standpoint,
water area; 2-maneuvering lane; 3- breakwater; anchorages for oil or gas carriers must al-
4-ship.

Figure 2-20. Ship maneuvers in confined port water area: R = 2.5L, to 3.5L,; 1- ship;
2- tugboat; 3-pier; 4-breakwater.
112 Design Principles and Considerations of Port Elements

growing shift toward specialization, which


was brought about by the enormous growth
in cargo volumes and also by a dramatic
change in the size of ships. This changed
the nature of ports as well as the material
handling systems and terminal docking fa-
cilities. The need for large water areas to
handle large ships, along with the require-
ments for large open land space to accom-
modate container facilities, oil storage, or
storage for raw materials, have induced
I ports to move to the periphery of cities,
I
I often onto marshlands or contaminated land
I
I which previously served as a dump for vari-
}-------0 ous kinds of hazardous waste (Farr, 1986).
I
I Naturally, this type of location requires ex-
tensive geotechnical investigations to deter-
mine the extent and the economics of the
soil improvement required. In general, con-
temporary specialized ports are not re-
stricted to natural harbors only, but they
are built whenever they can be economi-
cally justified. This contrasts with the tradi-
tional location of ports in natural harbors
near urban areas. Despite this, selection of
port location and port operation are done
within a social framework that, besides eco-
nomical considerations, includes an array of
Figure 2-21. Typical offshore mooring system: political, environmental, and other consid-
1-mooring ring; 2-chain shackle; 3-peg-type buoy; erations. A major area of controversy has
4-bridle chain; 5-main chain; 6-swivel; 7-ground been concerned with environmental prob-
chain; 8-sinker chain; 9-anchor; 10-sinker.
lems, such as that inherent in erosion,
sedimentation, disposal of dredge wastes,
ways be located downstream from the port. transportation, handling and storage of oil,
Tankers must be anchored in a staggered liquified natural gas, coal, and chemicals,
pattern with clear space between ships in a pollution, and other possible damages to
line of no less than 70 m and no less than flora and fauna.
five times the beam of the largest ship be- These are the problems the port designer
tween adjacent ships. must usually deal with when selecting a
location for new port construction or ex-
panding an existing port.
2.6 LOCATION, ORIENTATION,
SIZE, AND SHAPE OF 2.6.1 Selection of
THE PORT Port Location
Traditionally, ports have been developed in In principle, selecting the best location for a
natural harbors. In contrast to tradition, port is a· process which includes collecting
the current trend in port development is a and processing all relevant data (e.g., topo-
Design Principles and Considerations of Port Elements 113

(C)

t f
I I
I I
I

............. ,~
--/
...........
~
I I
I I
I I
t ~
(h) (d )

Figure 2-22. Various arrangements of offshore moorings: (a) two-buoy mooring, (b) and (c)
four-buoy mooring, and (d) six-buoy mooring. 1-ship; 2-ship anchor; 3-buoy; 4-buoy anchor
system. Note : a and f3 and number of buoys depend on environmental conditions, size of the
ship, and sea bottom conditions.

graphic and hydrographic conditions, and ditions, a lengthy period of time is usually
environmental data and information) for needed to complete the full environmental
orientation and design of such port ele- and social impact assessment, to conduct
ments as breakwaters, different kind of the public hearings and to obtain the re-
basins, and water- and land-based facilities. quired approvals by loca:J, state, and federal
Usually, the above process involves investi- authorities.
gating alternative port locations and lay- In the process of site selection, the poten-
outs in conjunction with designing the most tials of existing ports and new alternative
economical port related structures. The lat- sites are evaluated. A survey should then
ter involve selection of the site with mini- be conducted in order to study existing port
mum required dredging and with the most potentials by the evaluation of harbor po-
favorable bottom geotechnical conditions. tentials to accommodate larger ships and
Locating a suitable site for a new port the evaluation of potentials of the available
construction, especially in a heavily popu- land for storage of miscellaneous cargoes;
lated areas, is a difficult task generally be- the need for rehabilitation or modernization
cause of public concerns over dust, noise, of existing marine facilities to accommodate
and impact on the surrounding community larger ships and to support potentially
or natural habitat. This is complicated by heavier cargo handling and hauling eguip-
the fact that often the only land available ment must also be investigated.
for port development is on a land that has A common problem associated with the
high environmental value [e.g., tidal estu- redevelopment or expansion of existing
ary or saltwater marsh (wetland)]. Often ports if the lack of available water andjor
the site development requires some dredg- land space, and environmental restrictions,
ing or landfill operations. Under these con- as mentioned previously.
.....
........
t:l
(I)

250 .0 250,0 "'



'1:1

(")
c

.;:;·
~
"'
§
p..
C":l
0
0
ci ~
0 i5.:
... C!)

~.
0
~
g,
'1:1
0
:;:1.
t':!
~
(I)

5
~
0
ci ~~ 7
0
N L 6 ~ C ,, I
/~CE:] \
/ \
/ \
/ \
y ~

Figure 2-23. Practical example. Part of over winter anchorage area for drilling fleet, built in Beaufort Sea.
a-h: mooring buoys; 1-4: drill ships, Lmax =150m; 5-7: dredgers, Lmax =140m; 8-17: barges; 18: supply
vessel; 19: anchor chain: 20-anchor.
Design Principles and Considerations of Port Elements 115

Consequently, a new site, typically out- ways or in areas where substantial currents
side the city limits, is selected. Ideally, this may hamper berthing operation, the marine
site will be located in areas such as a pro- facilities are typically oriented parallel to
tected deep natural bay, sheltered lagoons, the direction of dominant current. This
or behind an island or stable shoal. Unless helps to reduce berthing impact and moor-
the site is fixed by special requirements, the ing forces. However, it must be noted that
location selection should be based on the in some cases where the berth is exposed to
following major factors: strong parallel currents, the moored ship
can be pushed away from the berth, as the
1. Sufficient water area for safe ship opera- result of considerable hydrostatic (stand-off)
tions forces developed between the ship and a
2. Required water depth mooring structure (Khana and Sorrensen,
3. Geotechnical conditions 1980; Wood et al., 1981). Details are pro-
4. Protection from open-sea exposure vided in Chapter 3.
5. Prevailing wave, wind, current, and ice
conditions
6. Sufficiency of land suitable for develop- 2.6.3 Harbor Area Requirements
ment of the terminal facilities
The minimum harbor area is the space re-
quired to accommodate one ship berth with
2.6.2 Size and Orientation of a turning basin, where the ship is turned by
Marine Facilities warping her around the turning dolphin or
around the end of the pier. Depending on
In general, the number, type, and size of the local conditions, the minimum harbor
ships calling at port as well as the mode of length can be taken as equal to five to six
handling these ships will determine the size lengths of the largest ship, with a width of
and orientation of the port marine facilities. two to three lengths of the largest ship.
These factors also dictate the sizes and The small restricted harbor depicted in
orientation of the local port basins. For ex- Figure 2-24 typically requires dredging of
ample, large fully loaded tankers and bulk the access channel and basin and the con-
carriers with small underkeel clearance are struction of entrance-protecting jetties,
vulnerable to broadside current action. In warping dolphins, and berthing facilities.
contrast, ships with a very high superstruc- Flexible warping dolphins, which also pro-
ture, such as containerships, car carriers, tect the pier head from ship impact, are
passenger ships, ferries, tankers, and bulk shown in Figure 2-25.
carriers in light condition, are sensitive to A typical layout of a harbor with lesser
broadside wind and wave exposure. There- restrictions is shown in Figure 2-26a. Here,
fore, if not properly oriented, the port the harbor is protected by breakwaters and
marine facilities may be a cause of consider- can accommodate three ships at a time. It
able forces imposed on ships, resulting in also has an anchorage area capable of ac-
problems in ship handling. commodating three more ships awaiting
Typically, port waterfront structures are their turn to dock. The ships can be assisted
oriented to minimize mooring forces, and by harbor tugs, if required. The harbor has
this is usually accomplished by aligning the two openings: one for entering and another
berth, located in the harbor area, protected for leaving.
from waves, parallel to the prevailing wind The breakwater structure itself very of-
direction. In ports located on inland water- ten is used as the facility for mooring of
116 Design Principles and Considerations of Port Elements

ships awaiting their turn to the dock or as


the specialized terminal (dock). The rele-
vant examples are illustrated in Figures
2-26b and 2-26c.
A layout of the typical middle-sized
multipurpose port is shown in Figure
2-27. A layout for a large multipurpose
artificial harbor is shown in Figure 2-28.
Here, the harbor is enclosed between two

~·Il
breakwaters and has an area and depth
sufficient enough to accommodate oceango-
ing large passenger, container, and general

.JI·
cargo vessels. The entrance is completed
with a ship turning circle. The harbor also

.
...
C>
/I comprises general cargo, container, and
passenger terminals, an oil dock and ship-
yard, complete with dry dock and outfitting

I
I
-!!!. facilities. On the shipyard side, there is a
:fishing harbor enclosed by two more break-
waters, which provide a sheltered water
area for two piers. The basic harbor shown
in Figure 2-28 provides limited sheltered
space to anchor ships that are awaiting
their turn to dock or to take refuge from
storms, without interfering with traffic to
(2 -3) L 1
and from the docks. A basic anchorage area
Figure 2-24. Typical layout for a small, restricted is located outside the harbor. The available
dredged harbor: 1- ship; 2- pier; 3- warping dolphin; land used for the port construction was too
4-breakwater; 5- port entrance; 6-coastalline. low and therefore had to be sufficiently

Figure 2-25. Flexible warping dolphins with floating fenders.


Design Principles and Considerations of Port Elements 117

"-- ___ ,
<---~

(a)

(b)

Figure 2-26. Port layout and some details. (a) Typical layout of a partly restricted coastal
medium-size artificial harbor: 1- berthing area; 2- anchorage area; 3- small-craft basin;
4- waterfront structure; 5- mooring dolphin; 6-breakwater; 7-coastalline. (b) and (c)
Port-of-Gibraltar: outer breakwater that is used for mooring of the vessels awaiting their turn to
the dock, and the mole (breakwater that is linked to the land) used as a passenger and the navy
facility.
118 Design Principles and Considerations of Port Elements

(c)

Figure 2-26. Continued.

Figure 2-27. Port-of-Canaveral, Florida. (Photo by H. Kyle.)


Design Principles and Considerations of Port Elements 119

0 200 400 600 800


Scale

Figure 2-28. Layout of a large multipurpose artificial harbor: !-general cargo terminal;
2-container terminal; 3-passenger terminal; 4-oil berth; 5-fishing harbor; 6-anchorage
area; 7-ship repair area; 8-dry dock; 9-turning circle; 10-mooring dolphin; 11-break-
water; 12-tugboat berth; 13-coastalline.

raised above the high water level in order A number of excellent examples of exist-
to avoid inundation by high tides and storm ing conventional and specialized ports and
waves. The dredged overburden from the harbors are described by Gornik (1969),
harbor and access channel provided suit- Quinn (1972), and Bruun (1989).
able material for this fill.
In general, the economy of using dredged
material for landfill depends on the dis-
tance and the method of dredged spoil de- 2.7 QUAY BASIN
livery and the severity of exposure of the
shoreline to wave action. The latter may not Generally, port quays may be placed in an
be a problem in naturally sheltered harbors unlimited number of ways. However, in con-
or in harbors where breakwaters provided ventional ports, two principle systems are
protection against wave action. traditionally used: the parallel quay system
An example of a very successful port built and the pier system. A combination of these
on reclaimed land is shown in Figure 2-29. two (e.g., L-shaped or T-shaped piers) is
Built in 1967, this oil-receiving port com- also used (Fig. 2-48). The length of the
prises a 1,882,000 m 2 area fully reclaimed basin in a parallel quay system is equal to
from the sea with five deepwater terminals. the length of the quay itself; its width is
These terminals can receive oil tankers up usually taken as being equal to the length
to 500,000 DWT. The 54 storage tanks can of the largest ship as shown in Figure 2-28.
store up to 100,000 kl of crude oil. Addition- The basin width can be reduced if the
ally, there are water treatment facilities largest ship calling at port approaches the
and special docks for tugboats and fire- berth assisted by tugs, whereas, in general,
fighting vessels. the length of the basin in a pier system may
120 Design Principles and Considerations of Port Elements

Figure 2-29. Oil-receiving Port-Kiire, Japan. (From Nippon Oil Co. Ltd. commercial brochure.)

be as long as required, provided that the With respect to basin width, the follow-
basin width is sufficient to allow safe and ing criteria are recommended for basins
convenient passage of the largest ship call- with two, three, and four berths (Fig. 2- 30):
ing at port (with or without tug assistance)
alongside other ship(s) sitting at the dock(s). Two-berth basin W = 2Bmax +30m
Agerschou et al. (1983) suggested that the
Three-berth basin W = 2Bmax + 40 m
basin length due to land traffic congestion
should not exceed about 10 average ship Four-berth basin W = 2Bmax + 50 m
lengths. However, the minimum length of a
pier should be sufficient to provide berth for where Bmax is the beam of the largest
the longest ship expected to arrive. ship (m).

n
28"'..... •o.o 28m••,. SO.Om

Bm1.1.
I I

lr"l
s ....
~ ~ ~ ,...,

..__
~ ~ l'~
v ~ ~ ~ ~
Figure 2-30. Width of port basin between piers.
~~
v
Design. Principles and Considerations of Port Elements 121

For long basins, with quays at both sides,


Fugl-Meyer (1957) recommends the follow-
ing approach:

1. The width of the basin is determined as


equal to the beams of the maximum-sized
ships located at both sides of the basin
with two rows of lighters on the outer side
of each ship and a fairway twice the beam
- (ma.x .)
+

(I)
N


..
i...

-
of the largest ship between moored ships.
~
(I)
2. The average ship size located at both sides
of the basin with one row of lighters on
"'"
the outer side of each ship and a fairway
four times the beam of the average ship,
so that two average are able to pass one
another.

For a similar situation, Agerschou et al.


Figure 2-31. Port basin width for two-way traffic.
(1983) recommend that the basin width
should be sufficient to allow passage of one
ship (with or without tug assistance) be-
tween two ships alongside, with one or two
2.8 OFFSHORE INSTALLATIONS
lighters one their side, all of maximum beam
for their respective categories.
Agerschou et al. (1983) specifically rec- 2.8.1 Offshore Bottom-Fixed
ommend that the basin width should be Marine Facilities
equal to the width of moored ships with an
added space, equivalent to one-half the ship Over the past four decades a dramatic shift
beam added to each side of the moving ship in the development of large merchant ships
and tug(s). The width has to be increased by and bulk carriers has had a significant
another two ship beams to allow for two-way affect on port development. Aside from
traffic, thus providing space equivalent to general-purpose conventional ports, new
the beam between traveling ships and one- specialized marine facilities , often in
half of the beam between traveling and remote locations, have been built to serve
moored ships (Fig. 2- 31). It should be noted the very large ships (100,000 DWT and
that two-way traffic under normal operat- more) that transport enormous volumes of
ing conditions may not be required, How- petroleum, raw materials, and bulk com-
ever, in a variety of situations the wider modities. Examples of such installations are
basin may not be more costly than a narrow described by Soros et al. (1986), Wright et
basin. al. (1987), Yaron and Shimoni (1982), Gruy
Naturally, all the above recommenda- and Kiely (1977), Reed (1980), Webb (1982),
tions are approximate, and in real design, and Youdale and Shrivastave (1986).
local conditions, such as exposure to wind, The ports and terminals used in the
wave, and currents, tug requirements movement of millions of tonnes of liquid
and availability, and so forth, must be and dry bulk materials are tending to be-
considered. come large, highly specialized installations,
122 Design Principles and Considerations of Port Elements

capable of handling very large ships with tives, namely the movement of very large
sizes in the range of 150,000 to 750,000 volumes of bulk cargo (dry or liquid), can be
DWT. Also, loading and unloading of cargo achieved by proper design of the offshore
is conducted at the highest possible rates vessel handling facilities, projected into
(e.g., loading of up to 20,000 tonnesjh of open sea to the extent that allows for safe
dry bulk and 220,000 m 3 of crude oil per approach of a fully loaded or in ballast large
day) with annual throughput of tens of mil- ship, and installation of high-capacity ma-
lions of tonnes. terial handling equipment. Under these
The conventional approach to creating an conditions a lower berth occupancy and the
artificial deep water harbor involves loss of time due to weather conditions, can
dredging and breakwater construction. Con- be sufficiently compensated for by high
structing such installations becomes pro- berth capacity.
hibitively expensive. A feasible and practi- In the case of dry bulk terminals, these
cal option for loading and unloading of very facilities are typically designed and built in
large vessels was found in the construction the form of piers, or loading/unloading
of terminals in areas unprotected from the platform, in combination with breasting and
open sea. It was also found that port objec- mooring dolphins (Figs. 2-32 and 2-33). In

(a) (b)

\ •

Figure 2-32. Typical offshore dry bulk terminal: 1-large bulk carrier; 2-pier; 3-
approach trestle; 4-mooring dolphin; 5-breasting dolphin; 6-loadingjunloading
platform; 7-deep water contour; 8-shoreline.
Design Principles and Considerations of Port Elements 123

some cases, these facilities have been pro- 2.8.2 Single-Point Offshore
vided with extra dolphins to allow for vari- Moorings
able ship orientation due to a change in
prevailing waves (Fig. 2-34). Single-point offshore moorings (SPM) are
In all of the above cases, berthing facili- currently used to provide safe, efficient, and
ties are connected to the shore by a trestle. economical accommodation for mammoth
In some instances, it has been found that a tankers (Fig. 2-36). The largest installation
better alternative to a long trestle is con- of this kind known to date can accommo-
struction of an off-shore island, which is date a supertanker of 750,000 DWT (Gruy
used as transitional storage for the mate- and Kiely, 1977). SPMs have proved to
rial delivered to or from the island by be extremely efficient for the following
smaller ships, rather than being trans- reasons:
ported directly to or from a large ship. For
this purpose, islands are build in a way to
1. They can be prefabricated and readily
provide protection for a marginal wharf placed for offshore use in deep water to
used by smaller ships. The berthing facility accommodate deep draft tankers.
for the large ship is extended offshore and
2. An SPM can operate efficiently in rough
joined with the island by a relatively short
seas and are not sensitive to directional
trestle (Fig. 2-35). changes of wind, waves, and currents.
Naturally, such facilities are designed to With the tanker moored via bow lines to
survive environmental loads, such as waves, an SPM only, the tanker is free to swing
wind current, ice associated with long re- around the SPM and always stay head on
turn periods, and the berthing and mooring to the waves, wind, current, or any combi-
forces of design vessels under maximum nation of them, thus reducing mooring
operating conditions. Detailed discussion forces and increasing terminal utilization.
on offshore marine terminals is given in 3. An SPM reduces the operational depend-
Chapter 8. ency on tugs and other support vessels.

Figure 2-33. Offshore coal unloading terminal, Hadera, Israel. (Courtesy Yaron-
Shimoni-Shacham, Tel Aviv, Israel.)
124 Design Principles and Considerations of Port Elements

Figure 2-34. Offshore dry bulk loading terminal with variable berth
orientation: 1- platform for the slewing loader with retractable boom;
2-loader supporting trestle; 3-breasting dolphin; 4-approach trestle;
5-large bulk carrier; 6-deep water contour.

4. In general, an SPM can be more economi- 2.9.1 Land Requirements


cal than any other type of liquid bulk
loading/unloading facility.
Port marine and land-based facilities pro-
vide ship accommodations and transfer
The SPM is typically linked to the shore by
andjor short-term storage of cargo trans-
submarine line.
ported to and from the port area by rail,
road, pipeline, and waterway, or any combi-
nation thereof. Also, port land-based
2.9 PORT-RELATED facilities may provide space for port
MARINE STRUCTURES related industries such as passenger termi-
nals, fishing and small-boat harbor facili-
The purpose of this section is to provide the ties, or shipyards. In general, considerable
port designer with basic insight information space is provided for the operation and ma-
on principles for design of port-related neuvering of cargo handling and hauling
bottom-fixed marine structures. Informa- equipment. The latter includes portal, con-
tion on port land requirements, layout, tainer, and mobile cranes, dry and liquid
selection of the relevant type of waterfront bulk loading/unloading equipment, heavy
structure for a variety of site conditions, weight lifts, and different kinds of conveyer
and properties of basic structural materials systems, cars, and trucks.
are included. The amount of port land area depends on
Design Principles and Considerations of Port Elements 125

usually provides for the highest possible


rate of the ship loading/unloading.
Because the arrival and removal of cargo
to and from storage areas by inland carriers
usually takes place at a much slower rate
than the ship loading/unloading rate, the
area for storage is typically designed to
compensate for the cargo handling rate
differences.
Storage facilities usually charge high
rates for keeping cargo beyond "free" time.
Therefore, cargo owners endeavor to move

'
the cargo as fast as is practical. For this
reason, in modern commercial ports, cargo

~
moves much faster and the size of storage
areas is reduced compared to older ports,
which were designed to serve infrequent
sea and land traffic.
The amount of land space required for

__..U.__.._,c;::>---'L;..,.__
port development is dependent on the type
and quantity of cargo to be handled. For
9
example, container terminals required large
areas for marshaling and storing of contain-
ers. Bulk material requires backland that is
Figure 2-35. Typical offshore dry bulk terminal with not less than one shipload, and the general
transshipment island. !-conventional port; 2- cargo area, as noted before, must be suffi-
artiticial offshore transshipment island; cient to compensate for the difference in
3-loadingjunloading facility; 4-trestle; 5-
marginal wharf; 6- large bulk carrier; 7-lighter; ship loading/unloading rates and cargo
8- deep water contour; 9- shoreline. handling of storages. Ferry terminals re-
quire direct highway or railway linkage.
Additional warehouse or open storage areas
would be required in case of infrequent in-
the specific objectives of the port's opera-
land or offshore services resulting in inter-
tion. According to PIANC (1987), for a mod- ruption of the cargo handling operation or
ern general cargo marine terminal a depth in the case of damage to breakwater or
of a least 200 m behind the dock apron is waterfront structures caused by environ-
required, although 400-500 m is preferred, mental forces or ship operation. Last but
and full container terminal would require not least, the selection of the land space for
as much as 700-800 m. port development should take into consider-
In principle, an efficient conventional ation the possibility of future expansion.
port, handling homogeneous general cargo Ideally, the dock facilities provided for
traffic, can achieve cargo handling rates up handling of a design cargo volume should
to 4000 tjday and berth throughput of more be designed for the specialized ships that
than 300,000 tjyear. However, this perfor- will use them. Unfortunately, this is rarely
mance is not typical. The usual rate of gen- possible, except in the case of a transport
eral cargo handling is less. In busy ports, industry chain operating under unified
due to the high cost of idle ships the num- management, controlling the shipping fleet
ber and capacity of the installed equipment and the land distribution system.
126 Design Principles and Considerations of Port Elements

Figure 2-36. Offshore single-point mooring to accommodate VLCC.

The area occupied by the nonspecialized effect on the area of land required. For
cargo port is usually zoned according to the example, the area where coal or other
different kinds of operations and services it "dusty" materials that may cause an air-
provides, such as handling of miscellaneous bone pollution are handled must be sepa-
cargoes, passenger services, or ship repair rated from that for timber products by at
and maintenance. Within cargo handling least 100m; general cargo areas by a mini-
areas there are specialized zones usually mum of 200 m, from chemical, grain, sugar,
provided to handle one particular kind of and perishable cargo by at least 300 m; and
cargo (e.g., containers, general cargo, or dry from passenger terminals by a minimum of
or liquid bulk). 400 m. Where dust is suppressed by water
Within passenger service areas there can spray or "dusty" material is treated with a
be areas for cruise services, local trans- chemical that binds free dust (Goldstein,
portation, or (which is not common today) 1983), the distance between specialized
for transocean services. Within the ship ser- areas may be substantially reduced. Fur-
vice and maintenance area there can be thermore, pertroleum products, liquified
areas for repair of large vessels or areas gases, and similar cargoes handling facili-
where services are provided for small-craft ties are typically isolated as much as possi-
owners. Specialized areas are usually sepa- ble (subject to special regulations) from
rated from one another, which may have an other port facilities.
Design Principles and Considerations of Port Elements 127

2.9.2 oust and Noise control portal scraper reclaimers, front-end loaders
or bulldozers. Naturally, the enclosed stor-
Stringent environmental regulations now in ages must be equipped with adequate dust
place resulted in continuous progress in re- and fume controls to minimize the risks of
ducing the dust and noise emissions and spontaneous combustion or gasjdust explo-
better wastewater control. sions. Dust collection with air evacation
Dust control in most ports and terminals systems was used in areas such as car
are designed to handle the bulk materials dumpers, main conveyor transfers, and
such as coal, iron ore, construction aggre- shiploader spouts.
gates, petroleum coke, and others; these Dust generation from roads often has
materials are stockpiled in open storages. been of significant concern. This type of
Open stockpiles may be long linear piles, dust may constitute a significant amount of
circular or conical shaped. On the other a total dust emitted from a terminal. The
road dusting is typically reduced by spray-
hand, bulk products, such as fertilizers, bulk
ing of roads with water and frequently by
chemicals, grain, and similar, that must be
reducing speed limits.
kept dry are usually handled in covered
Furthermore, use of tracked mobile
storage systems of miscellaneous designs
equipment for stockpile handling rather
(e.g., A-frame structures, domes, silos, or
than higher-speed rubber-tired loaders is
bins).
generally preferred.
In open storages, the materials are nor-
Noise abatement terminals that are lo-
mally handled by mobile or rail-mounted
cated close to residential areas are often
stackers, bucket wheel reclaimers, or com-
faced with noise control problems. Noise is
bined stacker jreclaimers. Booms of this
one of the most difficult environmental
equipment may be up to 60 m long and it problems with which to cope. In general,
may have a handling capacity in the range the use of electrically powered equipment
of 8000 tonnes per hour for coal and even can reduce noise levels. Railcar dumper
higher for iron ore. house, conveyor drives, and transfers can
If not checked, than handling of dust be totally enclosed to reduced both duct and
generating materials at such high rates noise emission levels. Green fences sur-
could be environmentally unacceptable. The rounding terminals and approach rails
modem terminal dust control techniques andjor conveyors made of trees abate the
generally consist of either application of noise and have also aesthetic benefit; use of
water sprays or use of air evacuation sys- green fences is always encouraged by the
tems where needed. The points of applica- authorities.
tion for water spray may also include a Wastewater Treatments In addition to
surfactant or microfoam agglomerating dust and noise control, the port authorities
agents generally include the railcar dumper, must also comply with many other environ-
main conveyor transfers, perimeter sprays mental regulations that may vary from
around stockpiles, stacker jreclaimer, ship- locality to locality.
loader transfers and loading spout, and sur- Among these regulations, most common
rounding the grab unloader-receiving hop- are these that deal with wastewater dis-
pers. Use of water spray has proved to be charge. These regulations are becoming in-
generally very effective, markedly reducing creasingly stringent. Most modem termi-
the dusting. nals now have the systems to handle storm
The covered storages normally have dedi- and dust control and washdown water. Be-
cated mechanical reclaim systems such as cause the cost of the contaminated water
128 Design Principles and Considerations of Port Elements

treatment and disposal is high, the termi- design are as follows:


nal operators try to achieve the smallest
discharge possible. For this reason they col- • Availability of land, climatic, environmen-
lect, settle, and try to reuse the water as tal, and geotechnical conditions
much as possible. • Train or truck waiting yard layout
• Train or truck loading/unloading rates,
number of cars, and cycle times
• Cargo handling and hauling rates
2.9.3 Berth Requirements • Storage capacity required
• Number of different commodities to be
Waterfront or berthing facility require- handled
• Handling and storage characteristics
ments are based on a berth throughput
• General stockpile layout, and layout and
capacity which is estimated either on the storage capacity for each type of cargo
basis of existing experience or by using • Number of cranes, stackers, reclaimers, or
modern theories and computer-based mod- combined stacker jreclaimer systems or
els. In general, a typical general cargo berth other equipment
may have an annual throughput of 100,000- • Cover storages, if required
150,000 tonnes, whereas a container berth • Shiploadingjunloading rates
may handle 50,000-100,000 containers per • Number and type of berths and ship an-
year. To handle this volume of cargo, high- chorages
capacity loading/unloading equipment for
both on land and onboard as well as effi- Port physical planning, in particular the
cient cargo handling machinery is required. determination of the berthage, cranage, and
In contemporary ports, each general storage area requirements, is not the sub-
cargo or container berth is typically ject of this work and the interested reader
equipped with three to four cranes. Natu- is referred to Agerschou et al. (1983), the
rally, the number of berths required United Nations Conference on Trade and
depends on port throughput. The size Development (UNCTAD) (1973, 1978), and
and type of each individual berth depends Bruun (1989).
on the kind of commodity to be loaded/ As a preliminary calculation, the number
unloaded and type of ships (existing of berths required to handle a design
or expected to be built in the future) to be volume of cargo Cn,) can be determined by
served at the berth. The relevant ship Eq. (2-18) (Tsinker, 1986);
characteristics are her main dimensions:
namely the length, width and loaded draft,
cargo-carrying capacity, type of cargo, cargo (2-18)
handling gear, and shape of the hull.
The typical port planning criteria usually
include the following parameters: where
Qm =port throughput in the busiest month. A
more reliable number of berths is obtained
• Annual throughput by this quantity, as it takes into account
• Rates of seasonal deliveries the seasonal nature of transportation of
• Design handling rates certain commodities.
• Size and distribution of shipping fleet
K 0 = factor of monthly occupancy of the berth.
• Size of railcars and unit trains or trucks
K 0 may be determined on the basis of
and truck fleet
existing traffic and by forecasting port
cargo turnover. Usually the value of K 0
Design variables to be considered for port varies between 0.45 and 0.75.
Design Principles and Considerations of Port Elements 129

Km = factor of monthly irregularity in dock us- rate for loading or unloading operation
age, considering possible meteorological (tjh).
disruption in port operations. Typically, ta =time required for auxiliary operations (h)
the value of Km varies between 0.75 and
0.95.
Pd = average capacity of one berth per day. This The number of auxiliary berths required
can be determined by depends on the characteristics of port oper-
ation (loading and unloading of liquid, bulk,
container, or general cargo) and the general
(2-19) requirements for maintaining auxiliary ves-
sels such as tugs, services boats, and supply
vessels.
where To determine the number of berths re-
G = ship cargo-carrying capacity quired to handle break-bulk general cargo,
t. = standard time of vessel loading or unload- the United Nations Conference on Trade
ing (h); t. = Gjr, where r is the standard and Development Secretariat UNCTAD

AVERAGE NUMBER OP TONS PER GANG HOUR


5 ro q ~ q ~

400

~ 300

""
~
:;
__________ Q

AVERAGE
"."'
!2
NUMBER OF GANGS ~ 100

EMPLOYED PER SIUP PER SHIFr

TONS PER SHIP PER DAY : BERTHED SHIPS WORKED


1400 1%00 8oo 6oo ; 400 200
APPROXIMATE NUMBER OF BERTHS REQUIRED
10

soo NUMBER OF COMMISSION


DAYS PER YEAR

1000
..
"'

"g
=·-- -------------------··
1500
..
~
:z: ,,.
ANNUAL .... .."'
!;;

TONNAGE FORECAST
(100,000)
,..
•soo

,.

Figure 2-37. Break-bulk general cargo terminal planning chart: Berth requirements, 2-10
berths (1 ton = 907 kg). (From UNCTD (1978).)
130 Design Principles and Considerations of Port Elements

(1978) has suggested a number of nomo- are related to the following:


graphic charts. The chart shown in Figure
2-37 can help determine the berth-day re- • Size and type of ship to be accommodated
quirement (the number of days ships are at • Kind and amount of cargo to be handled
berthing points) and the approximate num- • Cargo handling equipment and traffic flow
ber of berths required. • Oceangraphic, hydrographic, and geotech-
nical conditions at the berth site
Example • Degree of protection from environmental
forces
The annual port throughput is forecast to • Berthing and mooring forces
be 600,000 t. (Values of tonnage in this • Kind of utilities required for the dock
example, as well as in Figure 2-37, are operation
given in short tons.) The average number of • Type of operation (e.g., short term or
tons per gang per hour is assumed to be permanent)
12.6/h, and the overall fraction of time in
which the berthed ship will be under load- As noted previously, the type of ship,
ing or unloading operations is 0.6. The aver- specifically her length and draft, have a
age number of gangs employed per ship will decisive effect on the length of the docking
be 2.5, and the number of commission days structure and depth of water required. The
per year is assumed to be 350. As Figure kind and amount of cargo may also have an
2-37 shows, about 1300 berth-days, or ap- effect on the length of the quay structl},re
proximately five berths, are required to and the type of the docking facility. For
achieve port productivity. example, dry bulk carriers, container, or
general cargo ships are typically accommo-
As pointed out earlier, the berths re- dated at quay or pier-type docking facilities,
quirements and overall port performance whereas oil tankers and, in some cases,
can be analyzed by computer-run simula- bulk carriers would normally be loaded/
tion models that consider many factors unloaded while moored at a special berth
included in port operation and manage- that includes a loading/unloading platform
ment. The examples of such models are dis- or pier and breasting and mooring dolphins.
cussed by Guest and Pizzariello (1989), Cargo handling and hauling equipment
Ankundinov et al. (1989), and Wadhwa will effect the general dimensions of the
(1992).
marine facility as well as the strength of its
individual structural components.
Site geotechnical and hydrological condi-
2.9.4 Structures tions, availability of certain construction
materials, and construction conditions (e.g.,
2.9.4.1 General construction in dry or wet conditions) will
Port related marine structures are re- dictate the type of structure that can be
quired to provide the proper conditions for economically built for the docking facility;
safe and efficient operation of the dock. This for example, open piled structure, sheet-pile
will typically include conditions for safe ship bulkheads, gravity-type structures, or
berthingjunberthing, reliable tie-up at the others.
dock while under loading/unloading opera- In general, port related marine struc-
tions, and safe and efficient operation of the tures are built in protected harbors; break-
dock cargo handling equipment. waters and offshore terminals are, perhaps,
Most of the marine structures are nor- the only exceptions to this general rule. The
mally dedicated to a specific purpose which type, layout, and configuration of a marine
Design Principles and Considerations of Port Elements 131

structure varies, of course, from site to site, tation, or effects of waves of low recurrence,
depending on the purpose they were built to are typically ignored.
serve. For example, to reduce the adverse
effect of waves, docking facilities built at 2.9.4.2 Layout
exposed locations are usually constructed in
the form of open piled structures, and at The layout of marine structures depends
locations where the bedrock is present and on the port's locality, and the terminal or
located in a close proximity to the dredged dock specialization. The geometry of the
line, the gravity-type wall may represent layout is typically designed in the form of a
the most economical solution to wharf con- continuous (marginal) structure, in the form
struction. of narrow or wide piers, as individual plat-
Berthing and mooring forces, which are a forms and dolphins, or as a combination of
function of the type and size of ship and all or some of these. For example, a general
degree of her exposure to the local environ- cargo or container berth requires a continu-
mental forces such as waves, wind, and ous land area immediately adjacent to the
current, govern not only the type of struc- ship, and cargo handling equipment (gener-
ture and size of structural elements but ally various types of rail-mounted portal or
also the type and size of the fender system mobile cranes) placed as close as possible to
used to absorb berthing energy, as well as the berthing line. Also, general cargo ships
the number and capacity of mooring acces- of a variety of sizes, having varying amounts
sories required to safely maintain the ship of hatches and different kinds of onboard
while at the dock. gear to handle cargo, can be moored at
Service utilities, such as water, power, various locations throughout the multiberth
air, and so on, required for facility opera- area. Therefore, typical waterfront systems
tion are usually hidden in special ducts or developed to handle general cargo or con-
in special galleries. Sometimes they are tainers must be a continuous structure to
placed at special locations (e.g., just at the provide for ship berthing, mooring, and
berthing line or in the immediate vicinity of cargo handling. Where the available coast
portal crane track); therefore, marine struc- or shoreline is limited, wide piers (width
tures should be designed accordingly to ac- 200-350 m) for general cargo berth(s) are
commodate these elements. used to provide for a continuous waterfront.
The type of dock to be constructed de- When containers are handled by con-
pends on the character of operation (e.g., tainer cranes, the terminal's functional re-
whether it is a permanent or temporary quirements are, in principle, similar to those
installation). These factors govern the kind for general cargo terminals (Fig. 2-38). This
of material used for dock construction, as is not necessarily the case when roll-
well as material allowances such as those on/roll-off (RojRo) ships are used for
for corrosion and requirements for struc- transporting containers. Ro jRo ships are
tural concrete. Allowance for long-term con- often equipped with their own straight or
crete deterioration and wood decay are also slewing type of ramp. For straight ramps,
considered. Hence, for temporary docks, a special transitional structure may be re-
lighter structures are normally used. In the quired for end-on loading/unloading
design of temporary marine structures such (Figs. 2-39 and 2-40). A RojRo ship with
as those used for the construction of differ- a slewing ramp may be loadedjunloaded
ent kinds of projects (e.g., hydro projects, alongside the berth (Fig. 2-41). For slewing
bridges, and others), long-term effects, such ramps, the quay deck must be unobstructed
as potential shoreline erosion, long-term sil- and able to support a heavy concentrated
132 Design Principles and Considerations of Port Elements

Figure 2-38. Container terminal. (Courtesy of Port Autonome de Marseille.)

load. In tidal ports, a slewing ramp may be which service passenger and highway vehi-
inconvenient for full-time operation. In this cle traffic are very similar to Ro jRo opera-
case, to accommodate various tide eleva- tions.
tions, Ro jRo facilities are usually provided Dry bulk cargo facilities can be laid out
with an articulated transfer link span to similar to those required for both general
overcome the problem of a large tidal range cargo and container terminals. However,
(Figs. 2-39 and 2-40). Some practical rec- where the dry bulk storage area is located
ommendations on detailing and dimension- at a significant distance from the loading/
ing of a RojRo ship-to-shore connection and unloading berth(s) and the dry bulk cargo is
ramp systems can be found in publications delivered to or from the storage by a system
by PIANC (1978) and International Stan- of conveyers or is transported by pipes in a
dards Organization (1983). Ferry terminals form of slurry, the cargo may be loaded/
Design Principles and Considerations of Port Elements 133

Figure 2-39. RojRo berth with articulated link span. (Courtesy of Port Autonome de
Nantesj St.-Nazaire.)

unloaded at narrow piers or at special one manifold at midship, only one loading/
berthing facilities. These narrow piers may unloading platform of limited size is re-
be also used at passenger terminals and for quired to accommodate cargo handling
servicing navy vessels. Naturally, the width equipment. To safely accommodate a tanker,
of these piers is dictated by the scale of a system of breasting and mooring dolphins
operation, as well as by the type of is used. Similar to dry bulk terminals, the
loading/unloading equipment used. storage tanks need not be located close to
Liquid bulk (crude oil, liquid natural gas, the berth, as the cargo can be transported
liquid ammonia, etc.) terminals typically do elsewhere by pipeline at low cost.
not require continuous berthing structures From the preceding text, it can be con-
(Fig. 2-42). Because at these terminals cluded that the requirements for marine
cargo handling takes place only through structure layout are determined by the type
134 Design Principles and Considerations of Port Elements

Figure 2-40. Passenger ferry terminal.

and parameters of the design ship, type and typically built for Ro jRo operation. Mar-
location of cargo storages, type of cargo ginal waterfront is typically used in well-
handling and hauling equipment, and the protected harbors, in ports loc.ated along
requirements of safe and convenient ship inland waterways, and quite frequently
accommodation at the berth. In addition, along the harbor side of breakwaters. The
marine structure layout is determined by total length of waterfront required depends
the berthage requirement, space between on the number of berths required to handle
individual berths, mode of ship operation the design volume of cargo.
(e.g., RojRo ship or containership loaded/ A new port, by its very nature, is con-
unloaded by cranes), size and shape of cerned with the future. Therefore, water-
available land (e.g., availability of a natural front layout (berthing factilities) must be
sheltered harbor or the construction of an designed to be adequate to accommodate
artificial one), condition of water ap- the expected number and type of ships and
proaches to the berth (e.g., restricted or to provide for the design cargo throughout.
unrestricted approach), and, finally, the As the probability of need to simultane-
economy of construction. ously accommodate only large ships is low,
In conventional ports, waterfront struc- then the total length of the marginal water-
tures are typically built in the form of con- front required is usually set equal to the
tinuous quays (marginal wharf system), lengths of the required number of average
piers, or a combination of both. The ships to be accommodated simultaneously,
marginal wharves layout is the simplest plus additional space for moorings between
form of waterfront construction. It can be ships and between ends of ships and ends of
constructed straight, curved, or staggered the quay (Fig. 2- 44 and Table 2-6).
(Fig. 2-43). Straightness may depend on Experiments by the UNCTAD secretariat
local site conditions. A staggered layout is (1978) using different typical length distri-
Design Principles and Considerations of Port Elements 135

Figure 2-41. Ro jRo ship with slewing ramp.

butions show a general relationship be- that affect the number of berths, is given in
tween the amount by which the average Agerschou and Korsgaard (1969).
berth length exceeds the average ship Theoretically, a waterfront facility is only
length and the amount of ship waiting time required along the ship hatches. Therefore,
involved (Fig. 2-45). one berth facility can be limited to the
The correction factor indicated in Figure length required for cargo handling only,
2-45 applies to any number of berths in the provided that additional moorings are in-
group. An analytical solution for berth re- stalled to handle the ship mooring bow and
quirements based on a statistical distribu- stern lines (Figs. 2-46 and 2-47b).
tion of vessel arrival rates and required Mooring dolphins are not designed to
service time is found in Agerschou et al. withstand the impact of a ship, as they are
(1983). Also, many computer simulation usually located behind the berthing line and
programs for alongside ports are currently about 30°-40° off the bow and stern, so that
available. One such program for one to nine the length of mooring lines is between 60
berths, which considers many of the factors and 120m.
136 Design Principles and Considerations of Port Elements

Figure 2-42. Crude oil terminal. (From Seibu Polymar Chemical


Co., Ltd. commercial catalog.)

Very large ships may require two addi- and rail track. This enabled long, narrow
tional dolphins, usually located so that one piers to be built inside a harbor to maxi-
group of mooring lines is normal to the pier mize the length of berthage (Fig. 2-47).
and to intermediate dolphins, which hold With the increase in modern port productiv-
the ship against an offshore wind; the moor- ity, finger piers become less efficient than
ing lines which are secured to the end dol- wide piers or marginal wharves, and their
phins provide vessel stability in a longitudi- use is now restricted to locations where
nal direction. Mooring dolphins are pro- there are no cross-currents and where wa-
vided with bollards, quick-release hooks, or ter frontage is of limited length. In this
other means of moorings to handle mooring configuration a maximum number of berths
lines. may be provided within narrow confines.
AB stated earlier, in contemporary ports, Finger piers are especially suited for mumc-
the land requirement for operation is about ipal docks serving passenger traffic (Fig.
200 m 2 per meter of general cargo berth. 2-47a) or for naval bases (Fig. 2-47c). These
Previously, with relatively small ships and piers are usually projected out from the
low handling rates, the figure was smaller shore to provide berths on both sides of the
(often around 50 m 2 ) and included the pier for as many ships as is required.
areas required for the dock apron, sheds, Sometimes, narrow piers placed parallel
Design Principles and Considerations of Port Elements 137

(a> to the shore are linked with it by access


trestles to form aT- or an L-shaped layout
(Fig. 2-48). The latter is usually preferred
where current flow is in one direction only,
so that floating craft can carry our berthing
and unberthing maneuvers safely, using the
land side of the pier. The L-shaped config-
uration also permits the berthing of a
greater number of vessels, thereby increas-
ing pier effectiveness. Using more than one
access bridge generally improves the flow of
(C) traffic and, to some extent, increases pro-
ductivity. The disadvantage is that some or
all of the berths on the land side of the
floating pier cannot be used.
T- and L-shaped layouts are typically
(d) used where the slope of the bottom is very
steep, and, therefore, the presence of very
deep water in the immediate vicinity of the
shoreline makes the construction costs of
long water-projected finger piers pro-
hibitive.
Figure 2-43. Marginal wharves layout: (a) straight,
(b) and (c) bended, and (d) staggered. The number of berths per pier depends

(b)
(8)

« < ( < < ( ( < ( < < < ( < ( • ( • j


c::=:::>. c::=::=:> ( )
I l Ic [2/3L•
IICd,d, lI L min .

(d)

Figure 2-44. Berth typical arrangements.


138 Design Principles and Considerations of Port Elements

1.20

D.9

AVERAGE SHIP LENGTH +BERTHING GAP

Figure 2-45. Berth-length correction factor for break-bulk general cargo


terminal planning. [From UNCTD (1978).]

Table 2-6. Sp~ce between ships and between passenger terminal or heavy lift dock or
end of ships and end of mooring structure provide space for ships waiting their turn to
Length of Average Vessel (m) dock. The piers can be projected into the
> 200 200-150 149-100 < 100 water at 90° to the shoreline or at a certain
angle (usually no more than 60°) (Figs. 2-44
25 20 15 10
10 7 5 3
and 2-49). The former can accommodate
30 25 20 15 more berths than the latter; however, the
20 15 15 10 latter is more convenient and requires less
50 40 30 20 space for both land and sea approaches.
As stated previously, the advantage of a
pier system is the possibility to reduce the
size of shoreline required for port operation
and need of shorter breakwaters. As a rule,
a combination of piers and marginal
wharves represents the most economical
and best solution to new port construction
from an operations viewpoint.
The width between adjacent piers may
be determined as recommended in the
Figure 2-46. Typical one-berth layout of dock struc-
previous section (Fig. 2-30).
ture with mooring dolphins: 1-pier; 2-access trestle;
3-mooring dolphin; 4-gangway; 5-ship. 2.9.4.3 Configuration of
cross section
on the width or, in other words, on its Vertical, partly sloped, or sloped config-
capacity to provide a sufficient space for urations of waterfront structures are com-
berth effective operation. Typically, narrow monly used in port engineering (Fig. 2-50).
piers (i.e., 10-60 m wide) can effectively The vertical configuration is the most
accommodate one to two berths on each commonly used cross section of waterfront
side. At the same time, wide piers (i.e., construction, particularly for general cargo
200-300 m wide) typically accommodate at and container terminals, because it pro-
least three to four berths at each side. The vides for an efficient use of portal and con-
end of the wide pier may be used as a tainer cranes. Cranes are usually placed as
Design Principles and Considerations of Port Elements 139

(a)

(b)

Figure 2-47. Examples of piers constructed to maximize use of land. (a) Inland waterway
municipal passenger docks to serve the local traffic (artistic impression); (b) oil tenninal,
(c) naval base.
140 Design Principles and Considerations of Port Elements

(c)

Figure 2-47. Continued.

tion, is less convenient in terms of ship


, ~: s· c::!;r. berthingjunberthing operations and less
effective where the use of cranes for
loading/unloading operations is required.
It is usually used where extreme water-level
fluctuations occur, such as in inland water-
(a) (b) (c)
ways during spring or fall flood, or where
Figure 2-48. Typical layout of narrow pier: (a) L- the docking facilities are used for loading of
shaped pier; (b) T-head pier; (c) finger pier. 1-Pier dry or liquid bulk. This type of construction
structure: 2-access trestle. may also be efficient for the construction of
passenger terminals on inland waterways
close as practical to the edge of the water- and city embankments. Typically, vertical
front structure that maximizes efficiency of and sloped parts of a partly sloped struc-
crane outreach. On the other hand, cranes ture are separated by a 1-1.5-m-wide berm.
should be located far enough from the berth Usually for operational and safety rea-
edge to ensure its safe performance and sons, breastingjmooring piers are installed
safe and convenient ship berthing and at intervals of about 25-30 m along the
mooring operations. The vertical configura- partly sloped waterfront structure (Fig.
tion is most common for construction in sea 2-51). Mooring devices (i.e., bollards) are
and estuarial ports. It is also very common typically installed at the edges of vertical
for the construction in urban areas where and sloped parts to ensure convenient and
construction space is limited. safe ship mooring at any water level.
A partly sloped configuration, although A sloped configuration is basically used
more economical than a vertical configura- for bank protection and seldom for cargo
Design Principles and Considerations of Port Elements 141

~ -- -- · -

Figure 2-49. Wide pier arrangements: (a) Projected at 90°; (b) projected at 60°. kv k 2 , h 1 and
h 2 are respectively distances of the railway track from the edge of the adjacent marginal wharf
(or slope) and the ship approach (departure) lane from the top edge of a pier.

(a) (b) (c)

Figure 2-50. Basic configurations of waterfront structures: (a) vertical; (b) partly sloped; (c)
sloped.

handling operations. However, in combina- stability. These berms may some time be
tion with other structures such as dolphins, used for miscellaneous traffic as required.
piers, or different kinds of platform, this
type of construction may be extremely effi- 2.9.4.4 structural Types
cient for various bulk loading/unloading
docks, passenger terminals, and especially Port marine structures presently in use
for temporary ports built to serve during are either bottom fixed or floating. As this
construction of different projects adjacent to work is dedicated to bottom-fixed structures
waterway. A sloped configuration is obvi- only, the interested reader is referred to the
ously the simplest type of construction and other work by this author (Tsinker, 1986),
the most economical solution to many which is dedicated primarily to the design
waterfront construction problems. and construction of floating terminals, and
Depending on slope height, one or more to the other relevant literature.
berms can be provided to increase slope The bottom-fixed port marine structure
142 Design Principles and Considerations of Port Elements

Figure 2-51. Waterfront structure of partly sloped configuration with breastingj mooring piers
sitting on the sloped portion of the structure.

can, in general, be classified as follows: soil typically used where the foundation mate-
retaining, piled, and structures on special rial does not permit pile driving or where
foundations. Combinations of all of the heavy ice, waves, or other environmental
above basic types of structures are also used. forces can be dangerous to the piled struc-
There are more than 30 types of waterfront tures. Various kinds of gravity-type struc-
constructions used to date. The most typical ture are used. The gravity-type quay wall
of them are shown in Table 2-7. may be built in the form of mass concrete
All port related marine structures can be walls or walls composed from heavy prefab-
categorized as a soil retaining structures, ricated concrete blocks or elements, as
piled structures, and structures that are floating-in-caissons, cantilever or internally
rested on special foundations. anchored structures, sheet-pile cells, and
Soil retaining structures may be subdi- others. The detailed discussion on gravity-
vided into gravity-type structures, flexible type quay walls is given in Chapter 5.
structures such as sheet-pile bulkheads of Sheet-pile bulkheads are structures
different constructions, or a combination or formed from flexible sheeting restrained by
both. an anchor system and by penetration of
Gravity-type structures are those that de- sheeting below the dredge line. Sheet-pile
velop their resistance to soil pressure and anchorages may be provided in a variety
miscellaneous loads primarily from their of ways, such as by tie backs secured
own weight. This type of construction is to anchor walls of different constructions,
Design Principles and Considerations of Port Elements 143
Table 2-7. Basic types of waterfront constructions

Structural Scheme Definition Comments

A. GRAVITY TYPE STRUCTURES

Cast-in-Place Concrete or
Masonry Wall Material : Concrete, Natural Stone

Prefabricated from Concrete Material : Prefabricated Heavy Concrete


Blocks Blocks

.
' \,• .:
•. Floated-in-Caissons Caisons are of Prefabricated or Monolithic
,, .•,
: ~ '.' Construction.
Material : Reinforced Concrete
·...: .·t
~~~

......
~

Large Diameter Cylinders Cylinders are of Prefabricated or Monolithic


Construction.
Material: Reinforced Concrete

Large Diameter Sheet Pile Cells Steel Sheet Piles


144 Design Principles and Considerations of Port Elements

Table 2-7. (Continued)

Angle Type Wall Built from Prefabricated Elements,


or Prefabricated Sections.
Material: Reinforced Concrete.

Floated in or Material: Timber, Prefabricated Concrete


Ereeled·in·Piace Cribs Elements, Natural Stone.

B.PILED STRUCTURES

Relieving Platform wHh Front Monolithical or Prefabricated Concrete for


Sheet Pile Wall Platform Construelion. Steel or Reinforced
Concrete Piles and Sheet Piles

Relieving Platform with Rear Monolithical or Prefabricated Concrete tor


Sheet Pile Wall Platform Construelion. Steel or Reinforced
Concrete Piles and Sheet Piles.

Piles Supported Platform Monolithical or Prefabricated Concrete lor


Platform Construelion. Steel or Reinforced
Concrete Piles of Regular Construction
or Large Diameter (up to 1.6m)
Design Principles and Considerations of Port Elements 145

Table 2-7. (Continued)

C. SHEET PILE BULKHEADS

Cantilever Bulkhead Steel or Reinforced Concrete Piles

.. Single Anchor Bulkhead Steel or Reinforced Concrete Piles.


Anchors : Regular Steel Tieback wHh
- Deadman, Ground Anchor, Anchor Pile,
Pile System, Others.

__ ..
.............
Mu~i Anchored Bulkhead Same as Single Anchor Bulkhead
146 Design Principles and Considerations of Port Elements

Table 2-7. (Continued)

D. STRUCTURES ON SPECIAL FOUNDATIONS

·r-----_
.. . .~ .
Large Diameter Cylinders wnh
Relieving Platform
Cylinders and Platform are of Prefabricated
or Monolnhic Construction .
Material: Reinforced Concrete.

Platform Supported on Deep Steel or Concrete Caissons, Monomhical or


Caissons Prefabricated Reinforced Concrete Platform.

,/

Platform Supported on Piles Steel or Concrete Screw Piles or Piles with


·-·· ·· ~ with Increased Carrying Capacity Local Widenlngs. Monolithica! or
Prefabricated Reinforced Concrete Platform.
/
/
/
..
... ".. .

Reinforced Earth Wall Prefabricated Concrete Elements


.. ' -... : ... : : .. . ..:. ......
_ ; . ~ and Metal Anchor Strips
~. :·..;-._~_: _ .. :_ ~ .-.. :-~~-...
;".":'.;--/:.· ·:2 ..,·.
:.":/·:.....:- ::~:\~:.::-(::::.: :
,". ·:-: • • _._·.- . . . - ~ . .4 ~- • • •• -.

·::::-::·.:\ . : :.~ -~·-; :,:. · ~


. ·:· ··...,...
: ·-· ·::.-
.. ,; ... .
·-·-.·
I!1-
Design Principles and Considerations of Port Elements 147

by anchor piles, or ground or rock 2.9.4.6 Elevations


anchors. Anchored sheet-pile bulkheads
may have just a single row of anchors, or be The determination of berth (berthing
multianchored. Sheet piles of different structures) surface elevation and the depth
shapes and materials are used for sheet-pile of water or bottom elevation is an impor-
bulkhead construction. The type of sheeting tant phase of the marine structure design
and anchorage typically depends on the process. These elevations, and particularly
height of the structure, the kind of founda- the surface elevation, can be difficult to es-
tion material, and live loading. The details tablish. Normally, port (berth) elevation
on sheet-pile bulkheads design and con- should correspond to a combination of fore-
struction are given in Chapter 6. casted high water level, of a certain proba-
Piled structures are those whose stability bility, and wave action. It is usually a
depends on pile bearing and lateral load- trade-off between the cost of increasing the
carrying capacity. surface elevation and the economic value of
Piles are usually designed to carry verti- the potential losses due to interruption of
cal and lateral loads due to the structure port operation with the possibility of dam-
deadweight and live load, and miscella- ages to the cargo and equipment such as
neous sources of lateral loads such as moor- cranes, which cannot be removed before ex-
ing forces, soil lateral pressure, and others. treme high water occurs. An example of a
Pile cross section and material primarily partly flooded inland port general cargo zone
depends on pile length, foundation mate- is shown in Figure 2-52. In protected
rial, kind of performance (i.e., whether the harbors, a surface elevation equal to
pile is basically designed to carry vertical, about 2.0-2.5 m above the mean sea level is
lateral, or combined load), and pile-driving usually chosen.
techniques (e.g., use of hammer, vibrator, or In inland ports, the surface elevation is
jetting techniques. usually chosen as 1.0 m· above the maxi-
mum design flood level. Where docks, such
2.9.4.5 Structures on as those receiving large and very large
Special Foundations tankers and bulk carriers, are built at un-
protected locations, the top of the berth
These structures are typically repre- structure is taken as 5-7 m above the mean
sented by different kinds of platforms rested sea level or, sometimes, more.
on special foundations such as deep cais- If a sloped or partly sloped configuration
sons, steel or concrete screw piles, piles is used for the waterfront construction, then
with local widenings, or other unusual types wave run-up (h 8 ) must be considered. The
of foundation. This type of construction is preliminary value of the wave run-up can
usually used on weak soils when any con- be determined by. the following empirical
ventional type of construction is cost- equation:
prohibitive.
Besides the above basic types of marine
structure, a combination of some of them (2-20)
are frequently used; for example, piled
platforms with front or rear sheeting,
where
gravity-type structures resting on piles
with or without sheeting, partly submerged kr = roughness factor which depends on type of
gravity-type walls composed of concrete slope protection: for concrete cover, k = 1.0
and for rip-rap cover, kr = 0.77
cylinders of large diameter, or large-
diameter caissons anchored by different Hw = wave height (m)
type of anchorages. a = slope angle
148 Design Principles and Considerations of Port Elements

Figure 2-52. Partly flooded inland port.

For detail design, special analysis must the project has been defined in terms of
be carried out (U.S. Army Corps of Engi- certain engineering objectives parameters,
neers (1984), Shore Protection Manual; van the basic structure proportions and layouts
der Meer and Starn, 1992). are established to satisfy such objectives.
Where applicable, an additional rise in The conceptual design process may be car-
water level due to wind setup effect should ried out in either of two ways: on the basis
be considered. Wind setup is the result of of accumulated experience and data for sim-
the force of the wind blowing over the water ilar structures, or a novel concept. The for-
surface, resulting in raising the level of mer would focus on past practice and inter-
water at that location. polate or extrapolate from similar projects.
The gross underkeel clearance in the dock This method of conceptual design is quick
vicinity is determined in the same way as is
and fairly reliable when adequate data are
recommended for the access channel with
available but only serves as a rough guide if
appropriate assessment of such parameters
a novel concept is to be developed. In the
as squat (Z3 ) and wave response allowance
(Z4 ). See Section 2.3. latter approach, new ideas would be investi-
gated for their technical and economical
feasibility. During conceptual design, the
2.9.4. 7 Design initial proportions of the struct ure should
Depending on project objectives and com- be established.
plexity, the marine structure design process The preliminary design is the next step
may consist of conceptual, preliminary, and of the process. During this stage, more-
final stages. detailed structural analyses are required to
At the conceptual design stage, provided confirm earlier assumptions and to obtain
Design Principles and Considerations of Port Elements 149

more accurate quantities for project cost Unfortunately, sometimes in a headlong


evaluation. The structure is usually roughed pursuit of competitive initial economy, the
out at this stage: its layout, general dimen- longevity of a structure is of a secondary
sions, and, to a significant extent, some consideration to some designers and partic-
details are established. ularly to those with a little practical experi-
During the final design stage, detailed ence. They believe that marine structure
calculations and drawings and the specifi- design consists wholly of stress analysis. As
cations necessary to build the structure are a result, vital considerations of the struc-
produced. ture constructability, endurability, in-
To carry out its function as a part of a spectability, maintainability, and resistance
marine terminal, the structure should pro- to the marine environment, corrosion, tem-
vide an adequate berthing facility to accom- perature cycles, and other detrimental
modate the design ship, to support all kind effects imposed on the structure in a highly
of loads for which it is designed, and to corrosive marine environment are some-
provide for the all required land-based times overlooked or neglected.
facilities. The nonhomogeneity of foundation and
Hence, in determining the general ar- backfill materials still results in uncer-
rangement (layout) of a structure the de- tainty in the structural properties of these
signer should clearly define space and materials and uncertainty to some degree
boundaries for all kind of cargo handling in the mechanism of the soil-structure in-
and hauling equipment, traffic, storage, ad- teraction. The latter suggest that a conven-
tional marine structure still cannot and
ministrative offices, as well as for mooring
probably should not be designed with math-
accessories, fire-fighting equipment, and all
ematical precision. Sometimes, attempts to
kinds of relevant service.
use an elaborate mathematical model with-
In the past, port related marine struc-
out proper input data and proper under-
tures have been traditionally designed with
standing of the mechanism of the soil-
a high degree of redundancy largely be-
structure interaction may result in mislead-
cause· of the relatively rapid deterioration
ing and disastrous design. In general, the
rate of structural materials in the marine precision of any analysis should not exceed
environment but mostly due to lack of the accuracy of the input data. It seems
proper understanding of the mechanism of that some designers do not appreciate the
the structure-ship andjor structure-soil fact that the most of the data used in engi-
interaction. The latter was particularly true neering are, by nature, approximate. They
in designing "flexible" structures such as have developed a false sense of accuracy
sheet-pile bulkheads (Tsinker, 1983). In and quality of a computerized analysis.
modern marine engineering practice, be- Growing reliance on computer-aided
cause of a better understanding of struc- analysis and design without adequate
tural response to different kind of static controls on misuse can lead to a higher
and dynamic loadings, use of better quality incidence of errors and possibly an increase
and better protected structural materials, in the number of failures.
use of sophisticated fender systems to more As stated by Tedesco (1994), " ... many of
efficiently dissipate energy of the docking or the calculations made today are not neces-
moored vessel, the better quality control, sary. Sometimes calculations are produced
and the use of computerized analysis, because the analyst is fascinated with the
greatly enhanced an engineer's ability to program or is taken in by the shear beauty
arrive at a safer and more functional of the (computerized) analysis. At other
structure. times, someone wants to show how many
150 Design Principles and Considerations of Port Elements

pages of calculations have been produced; computerized analysis either graphically or


sometimes this is done to impress a client." numerically to satisfy himself that the pro-
The marine structures designer always grammer made no faulty assumptions or
must be aware that for all of the good sci- omitted or misinterpreted some crucial fac-
ence available to the engineer today, there tors that may dangerously affect structural
are still many problems associated with performance. This is particularly true in
marine structures engineering that can be the case of marine structures engineering,
best solved with an empirical, time-tested which in its many practical aspects still
formulas. remains as much art as science, perhaps
It is appropriate to mention that the pop- more of an art but definitely combining both.
ularity of computers, and particularly of Application of sound judgment that is based
desktop and laptop computers, has resulted on experience is a key to successful project.
in a flood of software. Unfortunately, the Focht (1994) suggested that "judgment must
quality of some software presently on the be applied at least twice-at the beginning
market is questionable. For example, it and at the end of a numerical prediction
would be of great interest, and perhaps a (structural behavior exercise); actually, it is
shock to some, if results of the analysis of necessary throughout the process." An ex-
retaining structures are compared with the cellent commentaries on development of en-
same input data using different programs gineering judgment as a part of a successful
now commercially available. Software users project are found in Ferguson (1992, 1993)
must first understand the program, be and Petroski (1984). In the age of potential
aware of the modeling and algorithmic as- "computer-assisted mistakes," it is very im-
sumptions used, check the program against portant to realize that a computer is just a
other programs, if possible, and also make useful tool, and computer output is only a
some trial solutions without the computer, supplement which cannot replace an engi-
even though in many cases these have to be neer's professional judgment. Hence, engi-
approximate solutions. When the commer- neers must keep current with developments
cially available computer program is com- in practice and academia. In complex de-
plex, the designer has no easy way of dis- sign cases, the use of relevant consultant
covering all the assumptions (made by the services is required to confirm the program's
programmer) that are included in this soft- usefulness and validity.
ware. Consequently, the program user has Furthermore, sometimes output errors
not much choice but to accept the software may have nothing to do with application
and rely on output produced. The structural software. The computer's internal architec-
software are often developed by structural ture may be at fault. The latest advanced
analysts lacking design and construction technology allows the designer to determine
experience, and, therefore, some of the as- fairly quickly structural response for given
sumptions, included in the program, may design conditions. The latter also enables
have an adverse effect on the validity of the him or her to produce a number of interac-
results. Hence, the successful computer- tions in order to arrive at the most economi-
aided design requires a great deal of vigi- cal solution possible. Thus, the design prob-
lance on a part of a designer. As Ferguson lem is transformed into an optimization
(1992) noted, " ... engineers need to be con- problem, which, in general, is a process of
tinually reminded that nearly all engineer- identification of the project variables and
ing failures result from faulty judgments constraints bound by the optimization
rather than faulty calculations." The experi- algorithm.
enced designer will always check, to a suf- To arrive at a safe and functional struc-
ficient depth, some important results of the ture, an appropriate factor of safety (F.S.) is
Design Principles and Considerations of Port Elements 151

usually used. Factor of safety is defined as elaborated provisions for future dock mod-
the ratio of the designed resistance of the ernization to enable it to absorb the load
structure to the applied loads. Naturally, from the new, potentially heavier, cargo
F.S. varies greatly, depending on the impor- handling and hauling equipment and to
tance of the structure, its expected service accommodate larger vessels.
life, the nature of operation (e.g., perma- The marine structures design, similar to
nent or temporary), load combinations the design of any kind of structures, is usu-
(e.g., normal or extreme), and structural ally done according to the available na-
materials (e.g., steel, concrete, or wood). tional standards. If no national standards
It should be noted that safe structure exist, the following documents are recom-
operation as a design concern must take mended for the design of marine structures:
precedence over all other design considera-
tions, and safety factors should never be • U.S. Department of Navy NAVFAC rele-
reduced carelessly. Additionally, all precau- vant design manuals (1980-1986)
tions must be considered with respect to the • British Standard Code of Practice (BSI,
operation of cargo handling and hauling 1984)
• German recommendations on design of
equipment, and handling of dangerous car-
waterfront structures (EAU, 1990)
goes that can be injurious to the safety of • Japanese Standard (PHRI, 1980)
the crew andjor passengers. • Russian Standards SNiP 11-50-74,11-51-74,
In all design cases, and at all design and 11-57-75
stages, the built-in redundancy must be • Miscellaneous PIANC criterias (consult list
sufficient to prevent structural failure. In of references to this chapter)
general, this means that the structure must
be designed in such a way that if one struc- It must be noted that all designers must
tural element fails, other elements can take also comply with numerous local rules and
over the function of the failed element to regulations.
prevent a domino effect, resulting in catas-
trophic failure. The design can be greatly
enhanced and improved by peer review. For 2.9.5 Selection of the
small projects, in-house peer review is usu- Most cost-Effective
ally sufficient; for large projects, outside re- structure for
view may be necessary; and for very large Dock construction
projects, an independent consulting board
may be required. Each project is site~specific. However, in
U.S. Navy Design Manuals DM 2.1 and general, the following basic factors govern
DM 25.6, and the "Waterfront Building the selection of the optimum type structure
Code" proposed by the City of New York for dock construction:
indicate that waterfront structures shall be
designed for a service life of 25 years, basi- • Geological conditions
cally because beyond this period of time, • Depth of water
operating requirements for the facility can • Availability and local costs of certain con-
change drastically to render the structure struction materials and services
obsolete. For some owners, the wisdom of • Environmental constraints
investment to extend the service life for
more than 25 years is therefore question- 2.9.5.1 Geological conditions
able. Nevertheless, the prudent owner will The site subsurface soil conditions, in
protect his investment by having a sound particular the depth to the bedrock or firm
maintenance program in place and well- stratum, usually play a decisive role in the
152 Design Principles and Considerations of Port Elements

process of selection of the optimum type 3. Gathered and evaluated historical perfor-
structure. As a general rule, the presence of mance data on similar structures or prob-
the firm foundation materials that prevent lems in the area, or in similar geologic
pile-driving favor the use of a gravity-type settings.
structure (e.g., floated-in-concrete caissons, 4. Selected an applicable theory for analysis.
steel sheet-pile cells, prefabricated L-shaped
units, blockwork structures, and others). To the latter one the writer should add that
The advantages and limitations of the in a complex field of marine engineering the
gravity-type structures and conditions of marine structures designer must be open
their most effective use for quay construc- minded. Most of geotechnical engineers are
tion are discussed in detail in Chapter 5. familiar with Terzaghi's admonition that if
Where foundation soils are represented the theory does not match with expecta-
by the competent soils that permit sheet- tions based on experience, then new theory
pile-driving to a greater depth than sheet should be developed.
pile bulkheads of miscellaneous designs can
provide the most cost-effective alternative 2.9.5.2 Depth of water
to dock construction. Sometimes, sheet-pile
The tidal regime or seasonal fluctuation
bulkheads are used as a part of relieving
of water levels in inland waterways may
platform structures, which are a combina-
have substantial impact on the selection of
tion of a sheet-piling and a regular pile
the structure. For example, at deep water
construction. Details on sheet-pile bulkhead
locations with a minimal water-level fluc-
design and construction are given in
tuation, sheet-pile bulkheads may not be
Chapter 6. The piled structures such as
economically feasible because of the techni-
piled platforms and pressure relieving
cal difficulties associated with underwater
structures are most suitable for construc-
placement of the anchor system. In this
tion of soft soils with low-bearing capacity
case, piled structures or some types of a
and low passive resistance. Piled struc-
gravity-type structure can be more econom-
tures, however, may be used at shallow
ical.
rock or hard soils locations in which case to
On the other hand, at locations with sub-
provide adequate vertical .and horizontal
stantial water fluctuation and particularly
stability, the piles are anchored to the foun-
where moving ice is present, the open piled
dation via heavily reinforced sockets.
structures can be ineffective. Sheet-pile
Details on construction of piled struc-
bulkheads, relieving platforms, or some
tures and their advantages and disadvan-
types of gravity-type structure will be more
tages are given in Chapter 7.
practical and more cost-effective. Examples
As was noted by Focht (1994), for suc-
of dock structures constructed at different
cessful designs of a geotechnical part of a
water-level conditions are provided else-
project, the designer must have the follow-
where in this book.
ing:
2.9.5.3 Availability of
1. Developed an adequate representation or construction Materials
modeling of the foundation stratigraphy, and Labor
including water level position, and under-
standing of the site and regional geology. These, as well as local construction prac-
2. Developed an appropriate representation tices, are important factors that may affect
of modeling of soil parameters critical to the selection of the dock structure. The
the particular problem, e.g., stability of numerous examples of different types of
the structure, safety factors, and others. marine structure constructed at sites with
Design Principles and Considerations of Port Elements 153

similar geotechnical and environmental be successfully utilized for quay wall con-
conditions are found elsewhere in this book. struction. Examples are found elsewhere in
For example, availability of the dry dock this book.
facility and the construction materials and
labor at low local costs may result in the 2.9.5.5 costs
selection of the floated-in-concrete caissons
for construction of the gravity-type quay In considering the cost-effective alterna-
wall founded on soft foundation soils. As tive for marine structure construction, the
noted earlier, such a foundation condition designer (and the owner) should realize that
would normally favor construction of a piled the initial construction cost is not the sole
structure or sheet-pile bulkhead. determining factor for structure evaluation,
it must be combi:p.ed, as appropriate, with
the cost of maintenance. Normally, steel
2.9.5.4 Environmental conditions and thin-walled concrete structures would
Environmental agents such as waves, require more maintenance than massive
wind, currents, and ice may heavily influ- concrete or granite-block-type structures.
ence the choice of the dock structure (e.g.,
an open piled structure as opposed to a
closed-type construction). For example, at 2.9.6 constructability
the open offshore locations exposed to waves
and current action, the open piled structure During the design stage, the designer
will attract much smaller loads than the should continually be aware that the pro-
structure of a solid-type construction. Be- ject he or she is developing is constructible,
sides, the latter structures will be suscepti- and a constructibility review during design
ble to seafloor scouring by the action of stage can help ensure successful execution
waves and currents. of the project. In general, constructibility
Environmental factors that affect the encompasses a process that has input to
construction schedule require the structure every phase of the project development,
to be completed in part or in full within a from conception to detail design, construc-
certain "weather window." This may re- tion maintenance, and repair; it
quire the use of certain construction tech- basically integrates all these stages in a
niques (e.g. prefabricated components) to single process.
complete the structure or part of it within Marine structures usually constitute a
the required "safe" period of time. The lat- significant portion of the total project cost
ter certainly strongly affects the selection of and, therefore, sound innovative design of
the type of dock structure. marine structures combined with compe-
Again, the presence of ice will require the tent planning and construction manage-
use of structures and materials that can ment can result in a significant savings. As
sustain ice impact loads and ice abrasion pointed out earlier, the structure design is
effects. not merely a stress analysis process, and in
Sometimes, the construction needs to be addition to other important considerations,
carried out in populated areas where local such as longevity, inspectibility, maintain-
conditions such as neighboring buildings or ability, and so on, the structure con-
the presence of sensitive equipment may structibility must be given appropriate
necessitate strict avoidance of noise or shock attention. It begins with a realistic under-
and vibration of the ground which is nor- standing on the part of the designer of how
mally associated with pile-driving. In this separate structural components will be fab-
case, the "quiet" slurry wall techniques can ricated and eventually put together. It is
154 Design Principles and Considerations of Port Elements

not just a contractor's problem as some in- mization of the construction procedures
experienced designers contend, and it that are based on the best utilization of
should be incorporated in the design con- the local resources such a labor force and
cept and eventually, with well-founded pro- available construction facilities.
visions, directives, and restrictions, spell out
in the project specifications. The specifica- Within the context of each construction
tions may also include items of a special stage, suitable procedures have to be devel-
concern that may occur during the life of oped to meet the following criteria:
the structure (e.g., provisions for a potential
increase in depth of water, change in live 1. Compliance with project drawings and
specifications
loads, etc.). By definition, if the contractor
understands not only the scope of work but 2. Assurance of meeting quality require-
ments
the overall purpose of the project he bids
3. Safety considerations
on, he may better accommodate the project
objectives by employing better construction 4. Ability to meet schedule requirements
methods. The latter may include the selec-
tion of construction facilities, the procure- An important aspect of project con-
ment of materials, the fabrication and structibility if providing safe and conven-
assembly of prefabricated structural ient access for personnel and equipment to
components, the selection of construction the areas where they must work.
equipment, the planning construction Last but not least, a survey of construc-
stages, and the organization and super- tion works and quality and safety control
vision of the work. assurance must be in place. For more dis-
Constructibility also comprises the care- cussion on a project constructibility, the
ful planning of construction stages and the reader is referred to Gerwick (1986).
development of basic principles of project
·execution.
For a better result, the designers and 2.10 STRUCTURAL MATERIALS
constructors should work in concert as a
team. This will allow for a better design and Timber, concrete, steel, or a combination of
implementation of principles of construc- these structural materials are used for the
tion such as the following: construction of basic structural members in
waterfront structures. Other materials of
1. Simplification of the structure configura- interest to a port designer include soil,
tion in general and some of its structural stone, rubber, different kinds of plastics,
components in particular. This may in- miscellaneous recycled materials (e.g.,
clude standardization of structural com- rubber tires, hardware, and fastenings), and
ponents and materials. others.
2. Preassembly of the smaller components Timber is mostly used where wood is
into the largest possible at the site. locally available and is therefore expected
3. Concurrently carry out several different to be the least expensive construction mate-
field activities (e.g., excavation, prefabri- rial. It is also used for temporary structures
cation of some structural components, and which are not expected to be in service for a
installation of the other ones). long period of time. Note that in a fresh-
4. Avoidance of some excessively tight toler- water or cold water environment fully sub-
ances, and procedures that may be overly merged wooden piles will not deteriorate for
sensitive to weather conditions. a very long period of time and can even gain
5. Selection of structural systems and opti- some additional strength in some instances.
Design Principles and Considerations of Port Elements 155

Steel is an excellent structural material. of the above-mentioned materials. In-depth


It is, however, highly susceptible to corro- treatment of structural materials for
sion and therefore must be protected from marine engineering can be found in
corrosion. recently published work by Whiteneck and
Reinforced concrete members are usually Hockney (1989).
much heavier than those manufactured
from steel. However, good quality concrete
members if properly protected from physi-
cal damage are almost maintenance-free 2.10.1 structural concrete
and thus can offer economical long-term so-
lutions for marine structures. A combina- The term concrete is often used as a general
tion of steel and concrete members is fre- definition for a variety of materials used for
quently used in marine construction (e.g., a construction of marine structures (e.g., plain
sheet-pile bulkhead with steel sheeting and concrete, reinforced concrete, prestressed
concrete cope wall or a piled platform with concrete, ferrocement, and fiber reinforced
steel piles and concrete deck structure). concrete).
Important considerations in material Plain concrete is an artificial conglomer-
selection for the construction of marine ate of cement and coarse and fine aggre-
structures include the following: gates, including natural sand and gravel, or
crushed stone. Concrete · which has been
1. Structural properties such as specific strengthened to resist the action of tensile
gravity, material strength, resistance to stress by adding of embedded steel bars is
cyclical and impact loading, resistance to called reinforced concrete.
seismic forces, and ductility Prestressed concrete is a concrete that
2. Nonstructural properties such as durabil- has been subjected to a permanent com-
ity in the marine environment, physical pressive stress of the same order of magni-
and chemical interaction with other mate- tude as the tensile stress expected to occur
rials (compatability), and fire resistance
under load. The compressive stress is in-
3. Availability, which may include trans- duced by the action of tensioned high-tensile
portability, mode of transport, and steel wires, strands, or bars anchored to the
handling limitations
concrete. The steel (tendons) may be ten-
4. Maintenance and preservation require-
sioned before placing the concrete, and the
ments, which include annual cost of
maintenance tension is transferred as compression to the
concrete after it has hardened. This tech-
5. Environmental considerations, which in-
clude environmental impacts such as cor- nique is known as prestressing. Alterna-
rosive and pollutant attacks on exposed tively, the concrete may first be cast with
surfaces, effects of sunlight, wave, and the tendons arranged either outside the
currents, severe termperature, ice, concrete or in ducts. The tendons are ten-
marine organisms, periodic wetting and sioned and anchored at their extremities
drying, and human activity, on each when the concrete has attained adequate
material. strength. Finally, the tendons are protected
by grouting. This technique is known as
The purpose of this section is to provide posttensioning. The combination of both
the reader with basic information on princi- prestressing techniques is also possible and
pal structural materials such a concrete, may be useful in prefabrication.
steel, and timber, used in marine engineer- Ferrocement is a composite material usu-
ing. This discussion is intended to highlight ally defined as a concrete consisting of ce-
important aspects of the selection and use ment and fine aggregate, reinforced by
156 Design Principles and Considerations of Port Elements

small-diameter steel wires distributed tailed discussion of concrete deterioration


throughout the body of concrete. M:ore prac- in the marine environment is given in
tically, ferrocement consists of a number of Tsinker (1995).
layers of closely spaced steel meshes into Cracks which occur in concrete are clas-
which the concrete mortar is forced. The sified in two basic categories:
strength of ferrocement is usually consid-
ered to come entirely from the steel, with
(a) Those caused by high stresses such as
the concrete simply operating to keep out thermal gradients while in service, ten-
the water (Turner, 1974). sile stresses, exposure of unprotected
Fiber reinforced concrete is a concrete concrete to freezing and thawing,
containing a random dispersion of small heating-cooling and wetting-drying cy-
fibers which reduce concrete intrinsic crack- cles in service, impact forces, and a vari-
ing and increase it strength. ety of expansive chemical reactions, such
Typically, the unit weights of plain and as those that occur in sulfate attack and
reinforced concretes are about 2300 kgjm 3 in alkali-aggregate attack.
and 2400 kgjm 3 , respectively. The unit (b) Self-imposed or intrinsic stresses. There
weight offerrocement may reach 2900-3000 are three types of intrinsic stresses that
kgjm 3 . cause cracking: plastic or those occurring
Despite evidence of long-term durability in the first few hours; thermal contrac-
of concrete structures in marine environ- tion occurring in one day to three weeks
ments and a generally outstanding perfor- and usually occur during construction of
mance in the ocean, serious concrete deteri- massive concrete structural members;
oration has been reported. The general drying shrinkage, often occurring after
several weeks and in some instances even
cause of deterioration was noted to be
after several months.
cracking, resulting in corrosion of the em-
bedded steel. Corrosion occurs in the pres-
ence of oxygen, moisture, and an elec- In modern engineering practice, various
trolyte. Salt, acting as a catalyst, intensifies design codes and guidelines provide some
the electrolytic properties of concrete, recommendations on limiting crack width
thereby creating a corrosion cell, resulting in concrete structures. For instance, Ameri-
in steel corrosion. The steel corrosion, in can Concrete Institute ACI 224R-80 recom-
turn, leads to the formation of large mends 0.15 mm as the maximum crack
amounts of expansive products (about six to width at the tensile face of a reinforced
seven times the original volume) and concrete structure subjected to wetting and
further cracking of concrete in a direction drying or spraying in seawater. The Federa-
parallel to the reinforcement. This process tion Internationale de la Precontrainte (FIP,
continues and two situations develop: 1985) recommends that crack widths at
points nearest the main reinforcement not
1. The reinforcing bars disintegrate, which exceed 0.004 times the nominal cover (e.g.,
reduces the flexural strength of the con- 0.2 mm for a 50-mm cover, 0.3 mm for 75
crete structure. mm, and 0.4 mm for 100 mm). The Soviet
2. Concrete spalling, delamination, cracks, Code SNiPII-11.14-69 calls for 0.05-0.1 mm
and stains occur. as the maximum crack width for concrete in
tidal zones or in areas exposed to seawater
A comprehensive list of the physical and spray. It should be noted that some authori-
chemical causes of concrete deterioration ties (Beeby, 1980) have found no relation-
can be found in works by Mehta (1980a, ship between the width of transverse cracks
1980b) and Mehta and Gerwick (1982). De- and reinforcement corrosion.
Design Principles and Considerations of Port Elements 157

2.10.1.1 Durability wall or breakwater wall construction, the


blast furnace slag-Portland cement in a
The marine environment is notoriously
ratio of 70 : 30 to the total amount of ce-
aggressive with regard to material durabil-
ment required may be employed. Alkali con-
ity. Chloride-induced corrosion of the rein-
tent should be limited to 0.65% (Na 2 0 +
forcing steel, freeze/thaw attack on satu-
0.65K 2 0) (Gerwick, 1986). Where sulfate
rated concrete, alkali~aggregate reaction,
aggression is anticipated, pozzolanic addi-
sulfate attack on the aggregates, salt scal-
tions to cement have been shown to in-
ing, and abrasion from ice and sediments
crease durability. Inclusion of pozzolanic
are the principal forms of attack. These are
admixtures up to 30% by weight of cement,
assisted and amplified by the constant wet-
replacing a similar amount of Portland ce-
ting and drying in the splash zone and by
thermal strains. For details, consult Tsinker ment, have been reported to produce more
(1995). durable concrete (Tuthill, 1988; Gerwick,
Durability of concrete in marine environ- 1990). Normally, sulfate-resistant cements
ments is dependent on the quality of mate- with a low C 3 A content are used for marine
rials used for the concrete mixture, as well application. However, acidity (pH) of the
as the mixing procedure and the volume cement paste must be maintained at a value
and quality of water used. Top quality con- high enough to prevent corrosion.
crete is the first line of defense against its In North American construction practice,
deterioration. pozzolanos, ASTM Class F (fly ash) or N
Conclusions on materials derived from (natural) with limitations on free carbon,
most reports and publications available to sulphur, and CaO, are used. In special cases
date are as follows: The most suitable where very high strength and impermeabil-
cement for use in construction of durable ity are required, condensed silica fumes are
marine structures seems to be either an used.
ordinary Portland cement, with C 3 A In general, concrete sulfate resistance is
(3Ca0 · A1 2 0 3 ) content less than 8%, or an virtually never a problem in seawater if
ASTM Type-II cement, with a water; cement-rich and impermeable concrete
cement ratio of 0.4-0.45 (by weight). Ce- mixes are employed (Gerwick, 1986). The
ment content should be greater than 350 corrosive action of external chlorides on em-
kgjm 3 • The importance of a low C 3 A con- bedded steel is one of the most severe and
tent was clearly demonstrated by some im- widespread problems of concrete structures
portant studies of cement properties and in a marine environment. Due to the mag-
their effect on concrete durability (Gjory, nitude of the problem, design engineers
1971). Gerwick (1975, 1976) proposed a should rely on corrosion protection methods
lower C 3 A content, giving as an acceptable to protect newly constructed structures
range 3-8% for cement used for construc- against chloride-induced corrosion and
tion of marine structures. other detrimental effects of the marine en-
Mehta (1988) stated that, theoretically, vironment. Corrosion protection measures
for sea concrete structures, the use of ordi- are ,now a standard package provided by
nary Portland cement having a C 3 A content the designer of the modern marine struc-
of 6-12% should be satisfactory, provided ture. This package typically includes the
that the cement is compatible with the ad- appropriate design mix of the concrete and
mixtures used in the concrete mix, which the direct protection of the reinforcing steel.
will also depend on the alkali content of the The former typically assumes that the use
cement. With very thick sections or large of special corrosion protection-type admix-
masses of concrete in a gravity-type quay tures would be added to the concrete mix at
158 Design Principles and Considerations of Port Elements

the batch plant and uniformly mixed chemically. Because the calcium corrosion
through the concrete. inhibitor maintains an active corrosion con-
Microsilica fume and a calcium nitrite trolling system within the concrete mix, the
corrosion inhibitor are two proven corrosion admixture protects the reinforcing bars from
protection products which help to delay the corrosion for the entire life of the concrete
onset of reinforcement corrosion (Berke et structure.
al., 1988). Microsilica fume reduces the per- Some studies have indicated that adding
meability of the concrete, by slowing consid- 2% calcium nitrite to concrete exposed to
erably the ingress of water-borne chlorides. salt used for snow and ice melting can in-
A calcium nitrite corrosion inhibitor crease the concrete service life by 15-50
promotes the stabilization of a natural years (Peel, 1982).
passivating layer on the steel, thereby Permeability is the rate at which water
controlling the corrosion rate. These two or other liquids flow through concrete. This
admixtures may be used individually, or in is an extremely important property of con-
combination for the most severe corrosion crete. Low permeability is indeed the key to
environments. overall good durability of concrete and en-
Microsilica fume, which originated in sures better corrosion protection for the re-
Denmark and Norway, is a product called inforcement and better protection of con-
condensed silica fume with 85% or more crete structures from freeze/thaw damage
content of a silica (Si0 2 ). It has an ex- in cold environments.
tremely fine particle size which allows mi- In most cases of concrete behavior in
crosilica to fill the voids in the cement paste marine environments, it has been found
and between the cement paste and aggre- that concrete durability is a function of its
gate. This results in a far less permeable permeability, which is not so much a func-
microstructure matrix. Usually, a specified tion of the amount of the void volume as it
microsilica quantity is based on the severity is a function of the size, distribution, and,
of the environment. In a typical marine particularly, the continuity of the voids.
environment, microsilica fume may amount Also, a steady increase with time perme-
to 10% by weight of cement. It is more ability rates have been observed for con-
effective with lower water j cement ratio crete in freshwater, in contrast to seawater.
concretes, and when used in combination Decreasing concrete permeability observed
with superplasticizers, its strength in- in seawater may be due to chemical reac-
creases of 20-50% over conventional con- tions between the ions in seawater and the
crete can be achieved with the same hydrated cement that produce crystalliza-
water j cement ratios. tion products or chemical compounds that
The highly alkaline environment of con- precipitate and decrease the pore continuity
crete creates a natural protective, passivat- in concrete (Harvery, 1980).
ing layer of iron oxide film on the steel in The use of cements having a low to mod-
chloride-free concrete. Chlorides which erate tricalcium aluminate (C 3 A) content
eventually migrate through the concrete is beneficial in that it combines with the
when reaching reinforcing steel can break chloride ion to form an insoluble compound
down the passivating layer to initiate the that blocks the pores.
steel corrosion process. Because permeability in concrete occurs
It has been found that calcium nitrite primarily along the interfaces between the
can be used as a corrosion-inhibiting admix- cement matrix and coarse aggregate, selec-
ture; and when mixed directly into plastic tion of the mix with minimum bleed by use
concrete, it aborts the corrosion process of aggregates having surface characteristics
Design Principles and Considerations of Port Elements 159

that promote physical or chemical bond is 2.10.1.3 Aggregates


required. A low waterjcement ratio (0.4-
0.45 by weight) is the principal factor to be It is desirable that aggregates which are
considered in designing a mix of low perme- likely to undergo physical or chemical
ability. changes should be avoided and that natural
The amount of water in the concrete mix sand, gravel, or crushed rock that conform
can be reduced by using water-reducing ad- to ASTM C33, or a similar standard, should
mixtures (superplasticizers). This is partic- be used in the concrete mix. Particu-
ularly useful where reinforcement is very larly, hard and strong aggregates are re-
congested, or when high strength or very quired in tidal and splash zones where
dense concrete is required. resistance to heavy abrasion/ erosion is
Unfortunately, low permeability of con- usually required.
crete is not explicitly specified in North The use of condensed silica fume
American construction practice. As the in- in a concrete mixture improves the bond
dustry becomes more concerned with the considerably between aggregates and
durability of concrete, it becomes apparent the cement paste, and, consequently, the
that permeability must be a specified de- abrasion/erosion resistance of concrete. On
sign property. the other hand, sandstones and quartz ag-
gregates do not form strong chemical bonds
with Portland cement paste. In addition,
2.10.1.2 Freeze/Thaw
coefficients of thermal expansion of sand-
stone and quartz are considerably higher
Concrete used for the construction of ma-
than the cement paste and, therefore, do
rine structures in cold regions has the po-
not provide for good abrasionjerosion resis-
tential for spalling caused by the freeze j
tance.
thaw cycle, by differential expansion, or by
corrosion of the steel bars. To ensure con-
crete durability under freezejthaw condi- 2.10.1.4 water
tions, air entrainment agents are now in-
In general, freshwater should be used for
variably used to provide for the required
quantity and quality of the entrained air concrete mixes used for structural rein-
pore size and spacing. However, these forced concrete. The chloride content of the
agents must be controlled carefully in order water (and the mix) is an important factor
to prevent any detrimental loss of concrete in ensuring protection of reinforcing steel
strength due to overdosage. against corrosion.
For ordinary concrete which has water j Salt water can be used for the prepara-
cement ratio of 0.4 or higher, air entrain- tion of concrete mixes used in unreinforced
ment with a minimum spacing factor of 200 gravity-type concrete structures and some
JLm is usually recommended. Malhotra of the unreinforced concrete elements such
(1986) found that it is difficult to entrain as breakwater armor units. In some cases,
sufficient air with the desired minimum however, salt water can be incompatible
spacing factor when the silica fume content with certain admixtures, and therefore the
is more than 20-30% by weight of cement. reaction must be verified through trial
Mehta (1988), working with particular ma- hatching. The use of salt water for concrete
terials and test conditions, achieved the mixtures in some cases requires the use of
value of the critical spacing factor of 400 retarding agents. Detailed information on
JLm for the concrete with and without silica the use of salt water may be obtained from
fume. various handbooks on concrete technology.
160 Design Principles and Considerations of Port Elements

2.10.1.5 Exothermic Reaction technique is efficient but is rather compli-


cated and expensive to install
The chemical reactions which cause con- • Cooling the mix by injecting a cold, ex-
crete to cure are exothermic. Because con- pandable refrigerant such as liquid nitro-
crete is not a good heat conductor, it is gen (Ibbetson, 1987).
possible to generate high internal tempera-
tures if the rate of heat generation exceeds 2.10.1.6 concrete Preparation-
the rate of dissipation. These high tempera- construction Practices
tures may impair the concrete integrity by
the different mechanisms outlined below. For modern marine structures, the
High temperatures may reduce the final desired properties of concrete are often de-
achieved strength (American Concrete In- manding, and sometimes conflicting, thus
stitute ACI 305R, 1982). Also, a high con- requiring the development of an optimum
crete temperature causes concrete to lose solution. A specification for any structural
its workability and encourages an excessive concrete usually contains a minimum 28-
rate of evaporation from the surface, which day compressive strength requirement that
leads to "plastic" cracking, and because of has historically been the controlling param-
large temperature differentials between the eter by which concrete quality has been
core and outside surface, stresses are set up evaluated. Furthermore, a maximum per-
which the concrete cannot relieve by normal missible water1cement ratio, a minimum
movement so that internal cracks can be cement content, and a minimum consist-
formed. ency for placement and consolidation of
According to FIP (1985), when the mini- fresh concrete are specified. The quality of
mum thickness of concrete is greater than concrete and its durability greatly depend
600 mm and the cement content is more on good construction practices which in-
than 400 kgjm 3 , the application of a ce- volves hatching, mixing, transportation,
ment with a slow release of heat of hydra- placement, consolidation, surface finishing,
tion and other methods of controlling the and curing. The hatching sequence and
temperature rise must be specified. The lat- methods should ensure the homogeneity of
ter may include the following separately or a concrete mixture. For example, a part of
in combination: mixing water may be used for moistening
the dry fine aggregate before adding it to
• The use of mineral materials such as gran- the concrete mixer, or fine materials such
ulated blast furnace slag or pulverized fuel as cement and condensed silica fume can be
ash to replace a portion of the Portland high-speed mixed with plasticizing admix-
cement (Bamfort, 1980; Gerwick, 1990) tures before the aggregates are added.
• Reduction of the rate of concrete place- Properly mixed concrete may be con-
ment veyed to the placing site by a variety of
• Shielding of aggregate piles from the sun, means, including conveyers, trucks, or by
or use of precooling aggregates pumping. If conveyed in trucks, then a con-
• The use of ice to replace part or all of the tinuous agitation is usually required. The
mixing water
conveyer used for concrete transportation
• Shielding of batch plant, cement silos, de-
livery trucks, and so forth from the sun
should be covered to protect the concrete
• Cooling the concrete after pouring by in- from intense sun heat, cold weather, rain,
stalling a cooling pipe system in the con- snow, and so on. When concrete is conveyed
crete as it is being placed. A coolant is by pumping, the forced pressure may in
circulated through the pipes to remove the some instances cause water to be absorbed
generated heat as the concrete sets. This into the aggregates, thus reducing the
Design Principles and Considerations of Port Elements 161

slump. High pressures may also affect en- concrete mixture is done with internal or
trained air. However, experience indicates external vibration as required. For this, the
that pumping, in general, is a reliable concrete should be sufficiently workable, so
method of conveying concrete mixes. In all that compacting equipment will give ade-
cases of concrete transportation and partic- quate consolidation. Consolidation by vibra-
ularly for trucks or conveyers, it is impor- tion is best described as consisting of two
tant to preserve the homogeneity of the stages: the first comprising subsidence or
concrete during conveyance by avoiding slumping of the concrete, and the second,
segregation of its different composite mate- the removal of entrapped air bubbles. The
rials. two stages may, however, occur simultane-
When fresh concrete is placed against ously (Kolek, 1963).
previously placed hardened concrete, then During vibration, the process internal
to obtain a watertight cold joint or construc- friction, which enables the original concrete
tion joint, the surface should be thoroughly mixture to support its honeycombed struc-
prepared by sandblasting or using a high- ture, is reduced drastically. The mixture
pressure water jet. All dirt, foreign materi- becomes unstable and seeks a lower level
als, and laitance should be removed and the and denser conditions. It flows laterally to
coarse aggregate exposed. The indentation the form and around the reinforcing steel.
is considered sufficient when coarse aggre- During vibration, the coarse aggregates
gate is exposed to a depth of 6-10 mm. The are suspended in the mortar, and· continu-
subsequent concreting should start with a ous vibration causes entrapped air bubbles
lift at a regular mix but not containing the to rise to the surface. The mechanism and
coarse aggregate, then followed by the regu- principles involved in the vibration of a
lar mix. The two should be vibrated so that fresh concrete mixture are described in de-
the first lift is well penetrated by the sec- tail in American Concrete Institute ACI
ond. A good bond between previously placed 309.1R (1981). Different types of vibration
and new concrete can also be achieved by equipment are used. These include the
covering hardened concrete with a thin layer flexible-shaft-type (probably the most widely
of epoxy-modified concrete slurry just be- used) vibrator, the electric motor-in-head
fore placing new concrete. In some cases, vibrator, surface vibrators, and different
cold joints may be prevented through the kinds of pneumatic and hydraulic systems.
use of set-retarding admixtures or specially The efficiency of any type of vibrator de-
formulated mortar mixtures. Waterweight pends on its power, frequency, and ampli-
cold joints are particularly important in re- tude. The principal requirement for the
inforced concrete structures. vibrator selector is its effectiveness in con-
Concrete must be properly consolidated crete consolidation and reliability. The ra-
when placed. Even superplasticized "flow- dius of vibrator action depends not only on
ing concrete" should be vibrated, because its characteristics but also on the workabil-
nonvibrated concrete, particularly rein- ity of the mixture and the degree of conges-
forced concrete, has a tendency to entrap tion of the reinforcing steel. According to
3-5% air, thereby reducing the bond Forssbald (1987), with superplasticized con-
strength and prompting settlement cracks crete, the radius of action of an internal
around the reinforcing steel (Forssblad, vibrator is 20-30% greater than with ordi-
1987). nary concrete. In general, this can help to
A standard practice guide for concrete reduce the time of concrete consolidation.
consolidation is provided by American Con- Internal vibration is generally used for or-
crete Institute AC1201R (1977) and ACI dinary construction, provided that the se-
309R (1987). In general, consolidation of a lection is large enough for the vibrator to be
162 Design Principles and Considerations of Port Elements

manipulated; external vibration is usually these admixtures its required, then special
employed when the section is inaccessible measures should be taken to reduce the
internally or is heavily congested with rein- adherence of the concrete to the forms. This
forcing steel. Sometimes it is helpful in a may include special coating of the forms, or
heavily reinforced section to vibrate the ex- the use of steel or steel lined forms. Steel
posed portion of the reinforcing bars. forms should be protected from rusting.
Concrete imperfections may occur Surface finishing plays a significant role
due to undervibration or overvibration. in good concrete performance; a smooth fin-
Undervibration is far more common than ish, that is, a lack of voids andjor other
overvibration. It may result in internal surface defects, helps to improve concrete
honeycombing and excessive entrapped air permeability and can greatly improve its
voids. abrasion resistance. Experience has indi-
Normal density well-proportioned con- cated that the resistance of concrete to
crete with adequate consistency are nor~ abrasion can be greatly improved by chang-
mally not readily susceptible to overvibra- ing the pore structure of the cement matrix
tion. It can, however, occur due to the use of in the surface zone (2-3 mm) by introduc-
grossly oversized equipment or extensive ing some cement to this zone during the
vibration beyond the recommended amount. finishing procedure.
The overvibration may result in mixture Curing is a process of maintaining a
segregation, loss of entrained air in air- proper humidity and temperature condition
entrained concrete and other deficiencies. for a certain period of time after the con-
Because imperfections in concrete are unde- crete mixture placement. Curing is required
sirable, it is preferable to overvibrate the to obtain the desired properties and quality.
concrete mixture rather than underconsoli- The curing process is essentially the pro-
date it (American Concrete Institute, ACI tection of fresh concrete against the loss of
311-4R, 1988). moisture; it is associated with a supply of
Sometimes, in order to ensure a low per- curing water externally. According to Mehta
meability in the concrete mixtures which (1988), when there is doubt about the abil-
bleed on consolidation, a revibration is rec- ity to keep the concrete surface perma-
ommended. The revibration is considered to nently wet for the whole curing period, or
be most effective when performed just prior where there is a danger of thermal cracking
to the initial set of the concrete for mixtures due to too cold a curing water, a heavy-duty
with slumps of 75 mm or more. Revibration membrane curing compound should be used
generally results in the improvement in to seal the concrete surface. In this regard,
concrete compressive strength. It should not Gerwick (1986) warns that with membrane
be used, however, in well-consolidated, curing compounds, heat from the sun or
low-slump concrete because it may be detri- cement hydration may degrade the curing
mental to the bond with reinforcing steel compound; therefore, one or more addi-
(Actowaiji et al., 1984). tional applications may be required during
When placing concrete in hot weather the first day.
(above 30°C) or in cold weather (below 5°C), For the proper maintenance of humidity
certain procedures must be followed. For and temperature, FIP (1985) recommends
details, the reader is referred to the Ameri- that in cold weather the concrete should be
can Concrete Institute ACI 305R (1982) and at least 5°C at the time of placing and
ACI 306R (1988) or similar codes. should be maintained above this tempera-
Furthermore, it is worth noting that poz- ture until it has reached a minimum
zolans and silica fume make the concrete strength of 5 MPa. The temperature may be
adhere to forms. Therefore, if the use of raised by heating the mixing water andjor
Design Principles and Considerations of Port Elements 163

the aggregates. However, the temperature 2.10.1.7 Protective Membrane


of water shall not exceed 60°C by the time
of contract with the cement. Complete concrete structures in the
The exposed surface of fresh concrete marine environment may be protected
should be moisturized with fresh water; additionally by surface waterproofing
seawater should not be used for curing of membranes to minimize the ingress of
reinforced concrete, at least at the initial chloride ions.
stage. There are several traditional and rela-
Some data recommended by leading tively new types of protective membranes
North American and European standards currently used in marine structures engi-
(ACI 201R, 1977; ACI 377R, 1984; FIP, neering. They are hot-applied coal tar, coal
1985) for design and construction of con- tar epoxies, vinyl, polyurethanes and
crete offshore structures are presented in polybutadien rubber coatings, chlorinated
Table 2-8. rubber, and others.

Table 2-8. Concrete for marine structures as recommended by ACI 377R-84 and FIP-85

ACI 377R-84 FIP-1985

A: Material
Cement composition C 3 A content of Portland cement should In the splash zone and atmospheric zone,
not be less than 4% to provide protec- Portland cements with moderate C 3 A
tion for the reinforcement. The maxi- content are recommended. Rapid hard-
mum C 3 A should be limited to 10% to ening cements should only be used for
obtain resistance to sulfate attack. repair. Low heat of hydration cements
are preferred for structures with heavy
dimensions.
Admixtures No chlorides should intentionally be CaC1 2 or admixtures containing more
Chlorides added. Total water-soluble chloride ion than 0.1% chloride by weight of cement
or a concrete mixture from all the should not be used.
component materials should not exceed
0.1% by weight of cement for normal
reinforced concrete and 0.06% for pre-
stressed concrete.
Chemical and air Where freeze thaw durability is required, Air-entraining agents, workability aids,
training the concrete should contain entrained and retarders are often essential to
air as recommended by ACI 201.2R. obtain optimum mix design, but pre-
cautions should be taken to evaluate
the side effects of each admixture type
before use.
Pozzolanic Pozzolans conforming to ASTM C618 (only High quality pozzolanic materials, such
natural pozzolans and fly ash are cov- as special silica fumes, may be added to
ered) may be used provided tests are produce improved strength, durability,
made to ascertain their relative advan- and workability.
tages and disadvantages, especially in
regard to sulfate resistance, workabil-
ity, and corrosion of steel.
(Table 2-8 continues on next page)
164 Design Principles and Considerations of Port Elements

Table 2-8. (Continued)

ACI377R-84 FIP-1985

A: Material
Aggregates Natural sand and gravel, or crushed rock Aggregates likely to undergo physical or
conforming to ASTM C33, and chemical changes in concrete are to be
lightweight aggregate conforming to avoided. Marine aggregates should not
ASTM C330. Marine aggregates may be be used unless the chloride content is at
used, provided they have been washed an acceptable level and unless the ag-
to meet the chloride ions limits. No gregates have a sufficiently low sea-
limits on maximum aggregate size are shell content.
given.
B: Specification
28-day compressive
strength
Minimum: 35 MPa for all zones; 42 MPa where 32 MPa for all zones; 36 MPa where
severe surface degradation likely. abrasion resistance is required.
WaterI cement
ratio
Maximum: 0.45 for the submerged zone, and 0.40 for 0.45 maximum, but 0.40 is preferred.
the splash zone and the atmospheric
zone.
Cement content
Minimum: 355 kgjm 3 320 kgjm 3 and 360 kgjm3 for 40 mm
and 20 mm max. aggregate size respec-
tively; 400 kgjm 3 for the splash zone.
Consistency No requirement No requirement
Permeability
Minimum: No requirement No requirement

Hot-applied coal tar, and coal tar epoxies growing tendency in Japan to use highly
are probably the most common and widely elastic coatings, such as acrylic rubber, in
used protective membranes used for protec- the marine environment.
tion of buried and wetted metals and con- However, before specifYing any of the
crete, either with or without the aid of the above protective membranes, the weather-
cathodic protection systems. Coal tar epox- ing effect of ozone and ultraviolet rays and
ies are known for their longevity and very impermeability to chloride have to be care-
good resistance to damage due to handling fully examined. When specified, membranes
and abrasion. should be placed according to the manufac-
Vinyl coatings have become important turer's instructions against a thoroughly
where salt water or salt spray is a factor prepared concrete surface. The selected ma-
because vinyl is especially effective in alkali terial and method of application must mini-
environments, being better, in fact, than mize the development of "blowouts" and
epoxies. "pinholes" due to water vapor pressure from
Polyurethanes and polybutadien coatings underneath the coating. However, if they
are typically specified where the concrete occur, then a procedure should be consid-
surface needs protection against abrasion, ered to fill them.
chemicals, and general abuse. Oshiro and Finally, it should be noted that in addi-
Tanigawa (1988) reported that there is a tion to the challenges of nature, an environ-
Design Principles and Considerations of Port Elements 165

mentally conscientious engineer has to have reinforcement may be grouped in bundles


a clear understanding of the potential toxic with up to 4 bars stirrups, provided to keep
effect of different kinds of protective mem- reinforcing bars together during placement
branes on the environment. However, it of concrete, should not be less than 10-12
should also be recognized that pressure by mm in diameter. In special cases, stirrups
environmental concerns is sometimes not 16 mm in diameter and more are used.
based on sound technical reason. Therefore, Stirrups are also required to resist some
an engineer always needs to keep an open shear forces acting on structural element.
mind when designing marine structures, They are typically designed to be anchored
particularly when selecting protective mem- within the confined core of slabs on beams.
branes. When the reinforcing steel is congested,
Quality control, which is a series of for- splicing it becomes a problem. To overcome
malized activities and procedures that is this problem, it is placed in a staggered
part of the fabricating or manufacturing pattern. The splices can be made by lap-
operation, should be in place to ensure that ping, through a mechanical means, or by
the physical characteristics of the materials welding using low-hydrogen electrodes. In
and the processes specified are thoroughly congested areas, the location and type of
followed. Appropriate guidelines for con- splices should be specified in detail. If the
crete inspection during the construction coated (by epoxy or other) reinforcement
phase and on completion of construction bar is abraded or damaged by welding, then
works can be found in American Concrete the coating must be repaired in the field,
Institute ACI 311.4R (1988). prior to the placement of concrete. Also, the
It is appropriate to finalize this section grade of steel and coating type should be
by stating the recommendations which were clearly identified by a color code.
made by Munro (1985). He stated that the For prestressed (posttensioned) steel,
costly problems of concrete durability may high-tensile strength bars or single or
be resolved by properly established priori- multistrand seven-wire tendons are
ties. At the top of the list, Munro recom- typically used; the former are generally used
mends picking the waterI cement ratio and for shortlength structural elements, and the
a practical way to control it. Then, in order latter for long-length structural elements.
of increasing importance, would be cement The use of posttensioned tendons has in-
content, curing, cover over reinforcement, creased dramatically in all fields of rein-
and workmanship. forced concrete construction during the past
For more information on the use of con- 25-30 years. The posttensioned tendons as-
crete in the marine environment, the reader sembly is usually composed of steel
is referred to Tsinker (1995). strand(s), with anchors affixed to each end
by toothed wedges, coated with corrosion-
2.10.1.8 Reinforcing Steel resistant lubricant, and encased ~n sheath-
ing. The tendons are placed within
It can be conventional, prestressed, or sheathing, stressed and anchored, and then
posttensioned. The typical port related rein- grouted.
forced concrete marine structure if not pre- This assembly is typically combined with
stressed or posttensioned usually requires a deformed reinforcing bars and encased in
very heavy concentration of conventional concrete to complete the total reinforcing
reinforcing steel because of crack width lim- system. The ducts should be stiff enough to
itations (i.e., 0.1-0.2 mm). prevent local sag. They also must be tied to
To provide adequate space for concrete the conventional reinforcing steel to pre-
placement and consolidation, conventional vent displacement from the design location
166 Design Principles and Considerations of Port Elements

by concrete mix when it is placed or vi- considered for each 20-ton lot. When in-
brated. Ducts should also be protected serted into the ducts, the entrance should
against accidental entry of any forei~ ma- be funneled, smooth, and abrasion-free. The
terials such as miscellaneous debris, con- strand coating, sometimes referred to as
crete aggregates, dust, and so on. Typically, grease, is a high-quality organic protective
during the transportation and installation medium intended to provide lubrication and
stage the ducts should be sealed with to form part of the corrosion protection sys-
plastic caps. tem. Alternatively, a vapor-phase inhibitor
Watertight galvanized sheaths are rec- powder may be dusted onto the strand.
ommended for the construction of post- Anchor castings are made of durable iron
tensioned concrete marine structures. The and typically provide about 71 and 103 cm 2
use of plastics such as polyethylene or of bearing area for use with 12.7- and 15-
polypropylene sheaths should, in general, mm-diameter strands, respectively.
be avoided. Rigid metal ducts should be An anchor casting typically consists of a
used for long tendons, as the frictional char- stiffened rectangular element but can also
acteristics of this type of sheath have proved consist of a cylindrical casting anchor with
to be favorable. The thicker material also is a truncated cone recess. The most common
intended to reduce the danger of damage wedge configuration is half of a 7° truncated
during shipment. The vents in sheathing cone, about 35 mm long for the 12.7-mm-
are usually provided at all high points in diameter strand, machined from 25-mm-
the tendon profile. In some instances, ten- diameter stock of ASTM C12L14 steel, with
dons may be installed into preformed holes, annular ridges or teeth on the inner sur-
which can be made with rubber tubes stiff- face.
ened by inflation with air or water. Holes If the deflected tendons are required by
can also be preformed with rubber hoses deign and without sheathing (prestressed
that are stiffened with steel rods which are construction), then they can be jacked up or
removed before the hoses are pulled out. held down by several means, one of which is
Preformed holes may be used only on short shown in Figure 2-53. In this case, metal
tendons because the friction losses in the anchors provided in the upper surface of
concrete are relatively large. Tendons the prestressing bench are used to anchor
should have smooth curves between desig- the bottom end of the hold-down device and
nated high points and low points and have a center-hole jack is used to jack the ten-
smooth sweeps around blackouts and dons down. A strand chuck anchors the
sleeves. tendons in the deflected position. The strand
The seven-wire strand (12.7-15.2 mm in chuck and the hold-down anchors are ex-
diameter) is made with high-tensile steel. pended with this procedure.
The most commonly used steel has a guar- The jacking procedure should be done
anteed ultimate strength of about 1860 MPa according to a predetermined sequence in
and is the low-relaxation type. The strand order to equalize the effects of friction along
comes on a reel, or in a reelless coil, each the bench; this procedure is similar to pre-
referred to as a pack, containing about 3660 stressing the deflected fenders as described
m of length for the 12.7-mm strand and earlier.
about 2500 m of length for the 15.2-mm Tendons are stressed with a hydraulic
strand. It should be stored in a dry environ- jack equipped with a calibrated pressure
ment. gauge. A calibration curve accompanies each
Stress-strain data are usually not avail- jack to relate gauge pressure to the deliv-
able for each shipment of tendons, and ac- ered force. The jack pushes against the em-
cording to ASTM A416, one test should be bedded anchor casting and pulls the strand
Design Principles and Considerations of Port Elements 167

stressed from each end to minimize friction,


but need not be stressed simultaneously.
Occasionally, the strands slip at the an-
chors. The obvious reason for this is the
failure of the wedges to grip the strand
properly. There are several reasons for such
failure among them: mishaped recess which
can be out of shape, or bottom or top bases
of the truncated cone are to large or too
small; wedges are fouled with dirt, sand
cement paste, or there is an intrusion of
dirt into the recess; wedges are improperly
heat treated, or their teeth are improperly
machined. Very seldom is the strand
broken. Strand breakage may only be the
result of gross abuse during the stressing
operation. It usually occurs at the stressing
end and is typically caused by repeated
stressed, distressing, and restressing to at-
tain the desired elongation. These opera-
tions "chew" the strand severely causing a
notch effect. Therefore, where possible, it
should be avoided, but in no case more than
one repull is recommended. Strand failure
also may occur in the case of gross mis-
alignment of the strand causing it to pass
through the anchor nonconcentrically or
when anchors twist due to nonuniform
bearing failure of the concrete when the
strand is not perpendicular to the concrete
edge. Angled pocket forms should be used to
prevent local failure of concrete.
Figure 2-53. Tendon hold-down device for use with
jacking down at the hold-down points. [From Libby
Mter the tendons are tensioned, they
(1984).] should be grouted as soon as possible. It is
important to ensure that grouting tubes,
hoses, ports, and valves do not leak during
grouting. Leaks may result in a plugging of
to a predetermined gauge pressure that the system and the trapping of water in the
should correlate to within 5-7% of the ex- sheath. The grout is usually introduced at a
pected elongation shown on the shop draw- pressure 0.5-1.0 MPa; vents should be
ing for each tendon (American Concrete closed and the grout forced out through the
Institute ACI 318, 1989). strand end. The grouting mix should be
The elongation is about 200 mm for each selected according to the standard practice
30 m of tendon length and varies with the for minimum bleed. The type of cement used
friction developed along the tendon length. in the grout should be the same type used
Straight tendons, shorter than about 30 m, in basic concrete, because a difference in
are stressed at one end only. Curved the electrolytic properties of the cement in
tendons, or those longer than 30 m, are the grout and basic concrete may result in
168 Design Principles and Considerations of Port Elements

deterioration of the prestressing steel. The concrete cover over reinforcement


Where grout is exposed to significant pres- should be kept to a minimum; too thick a
sure, which may be the case in vertical cover can lead to excessive cracking. It is
ducts or in draped posttensioning systems general practice to establish the minimum
having a significant vertical component, the concrete cover over reinforcement, based on
head of grout may force the water out of the existing codes. However, specifYing concrete
mix into the strand interspace, and then strength and cover requirements without
expel it through the strand end. This allows control of concrete permeability can re-
the grout to settle, sometimes producing a sult in inadequate corrosion protection
substantial void at the tendon end(s). This (Schupack, 1982).
may be prevented by using special admix- In the North Sea, the current practice is
tures which causes the grout to gel as soon to provide a 50-mm concrete cover over con-
as the grouting stops, or by topping up the ventional reinforcement and 70 mm over
grout a few hours after initial grouting. prestressed steel. The current recom-
Shop prefabrication of reinforcing steel is mended practice in the United States
a great benefit to the construction industry (American Concrete Institute ACI 377R,
in general, and to the construction of ma- 1984) requires the following concrete cover
rine structures in particular. Prefabrication for reinforcing bars: 50 mm in the atmos-
that is done under factory conditions en- pheric zone not subjected to salt spray, 65
mm in the splash and atmospheric zones
sures better quality of steel fabrication,
subjected to salt spray, and 50 mm in the
while also shortening construction time, re-
submerged zone. The concrete cover over
quiring less on-site skilled labor and super-
the prestressed or posttensioned tendons
vision, ensuring easy inspection. It is
must be increased by about 25 mm above
extremely efficient where a great deal of
these values.
repetition is required. However, whereas
The Russian Code SNiPII-1.14 calls for a
prefabrication gives all the above advan-
70-mm cover over prestressed tendons, 50
tages, the cost of transportation can be high.
mm over conventional bars, and at least 30
Therefore, in some cases, on-site prefabrica- mm over secondary reinforcement. The code
tion may be a cost-effective alternative. also allows for a 10-mm reduction in con-
crete cover for prefabricated concrete ele-
2.10.1.9 Concrete cover ments. For the tidal zone of offshore produc-
over Reinforcement tion platforms, FIP (1985) recommends a
75-mm cover over conventional reinforce-
The corrosion of reinforcing steel in con- ment and 100 mm over prestressed steel.
crete structures exposed to seawater has Cover over the reinforcement is impor-
often been observed to induce cracks and tant for concrete durability in the marine
promote concrete deterioration, especially environment. However, again too much
in the splash zone. The mechanism of rein- cover increases the width of cracks, whereas
forcement corrosion is discussed in detail in too little cover may lead to easier access by
Tsinker (1995). As mentioned earlier, to chloride and oxygen, and in extreme cases,
prevent such corrosion, care must be exer- to a loss of mechanical bond between con-
cised in the design, mixing, and placing of crete and reinforcing steel.
the concrete. It is also important to prevent Concrete cover as low as 35-40 mm have
cracks in the concrete cover by a proper been successfully used on many prestressed
design of steel reinforcement (American concrete piles, apparently due to the dense
Concrete Institute ACI 377R, 1984; ACI concrete and rich mix employed. Even
224R, 1980). smaller concrete covers, as low as 10 mm,
Design Principles and Considerations of Port Elements 169

have been successfully used on concrete fusion-bond process, abrasive blast-cleaned


ships and pontoons where special care was reinforcing bars are preheated to about
taken to achieve a dense cement paste cover. 230°C, then electrostatic spray guns apply a
charged dry powder onto the preheated
2.10.1.10 Epoxy coating and bars, and then the coated bars are subse-
Galvanization of quently cooled in a water-quench bath. A
Reinforcing steel range of thickness of fusion-bonded epoxy
coating between 130 and 300 JLm has been
A practical reality is that, despite good found to be very effective. For reliable pro-
design and construction practice, corrosion tection, it should not be less than 100 JLm.
of reinforcing steel in concrete marine Pull-out tests for bond strength have indi-
structures can occur. In current construc- cated that the best performance may be
tion practice, reinforcing steel used for con- achieved when the coating thickness does
crete in marine structures is usually pro- not exceed 300 JLm (Read, 1987).
tected in one way or another. Use of fusion-bonded epoxy-coated bars,
The following basic types of corrosion however, should be treated with caution,
protected for reinforcing steel are currently because they may result in increase in crack
available and used in the modern practice width of up to 100%. As stated by Cairns
of marine structures engineering: fusion- (1994), the proposals are made in the United
bonded epoxy coatings and galvanization; Kingdom for revisions to deign rules for
cathodic protection may be used in some the calculation of crack width in British
cases. Standards BS 8110 and BS 8007. It has
Epoxy-coated steel has gained wide- been suggested that the coefficient for de-
spread acceptance and is now used as a sign crack width should be increased by
reliable means of preventing rebar corro- 50% where fusion-bonded epoxy-coated re-
sion and ultimately reducing the chances of inforcing bars are used.
concrete deterioration. Epoxy-coated steel Epoxy-coated bars should be handled
was one of the many innovative concepts with care to prevent the coating from being
initiated by the U.S. Federal Highway Ad- damaged during shop bending, installation,
ministration in the 1960s to fight serious and placing of concrete. Generally, bars are
problems of reinforcing steel corrosion. The bent after being coated. However, despite
initial major field application of epoxy- the good protection provided by epoxies,
coated reinforcing steel was for eliminating there are occasional reports on corrosion
bridge deck deterioration. Later on, it was damage to coated rebars in the marine en-
extended for use in marine and other struc- vironment attributed to debonding of the
tures exposed to hostile environments. In coating (American Concrete Institute, 1988).
current field construction practice in North An acceptable field practice with epoxy-
America, epoxy-coated reinforcement has coated reinforcing steel includes handling
become fairly well standardized. In the with nylon or similar slings and in-place
United States in 1981, the ASTM Standard epoxy touch-up to damaged coating.
Specification for epoxy-coated reinforcing A more recent development is the use of
steel bars was issued. A similar standard epoxy-coated strands for prestressing appli-
for epoxy-coated welded wire fabric was de- cations. A special proprietary epoxy with
veloped in 1988. superior ductility has been developed. Grit
The epoxy coating is typically applied to is added to the epoxy surface to enhance
the reinforcing steel by a fusion-bonding this bond strength.
process. This product first reached commer- It has been found that even a small slip
cial markets in the mid-1970s. In the of coated bar can result in a disruption of
170 Design Principles and Considerations of Port Elements

epoxy coating. This implies that with any Damaged coatings can be repaired by
significant crack size, the performance of simple application of zinc-rich paint. Be-
the epoxy on either side of the crack may be cause of the sacrificial nature of zinc, no
subject to surface damage (ACI, 1988). uncoated reinforcing steel nor any other
Therefore, in order to minimize cracking embedded metal dissimilar to zinc should
while using epoxy-coated reinforcement, the be permitted in the same concrete element
optimum combination of parameters should with galvanized bars, except as part of the
be considered. It seems that with the devel- cathodic protection system. Galvanized bars
opment of an epoxy with superior ductility, should not be electrically coupled to un-
the only serious problem of use of epoxy- coated bars.
coated strands is the potential damage to It is generally accepted that under rela-
the coating which may be scraped off when tively short-term (say less than 35 years)
the strands are pulled through the metallic exposure to a marine environment, galva-
ducts. This may lead to the conclusion that nized and epoxy coatings should be equally
if coated strands are required, then the duct effective in corrosion protection. Therefore,
material should be relatively smooth. the choice between using the two should be
Over the years that epoxy coating has based mainly on cost and availability. Be-
been applied to reinforcement, the cost per cause zinc coating is sacrificed during the
kilogram of a reinforcing bar has decreased short term, epoxy coating should be more
by about 25-30%. It has also become gener- effective if longer-term protection is re-
ally apparent that the use of epoxy-coated quired.
reinforcement has resulted in a significant From economic considerations, coated
increase in the structure's service life. bars are typically installed at the exposed
Galvanization of steel bars gives good side of the marine structure in combination
corrosion protection. However, it is not al- with uncoated reinforcement installed in
ways satisfactory at the splash zone. It corrosion-protected areas.
should be noted that on the acid side, zinc Fiber reinforcement is used typically as a
is not very resistant and significant corro- secondary reinforcement system to mitigate
sion of zinc can start below about pH = 5. problems associated with intrinsic cracking
Additionally, Trendaway et al. (1989) found such as plastic, thermal, and drying shrink-
that galvanized reinforcing steel shows the age cracks. Fiber reinforcement has been
greatest benefit when exposed to concrete commercially used now for more than two
contaminated with low lev~ls of chloride decades. The greatest advantage of fiber-
(less than 1.9%), or which is chloride-free reinforced concrete (FRC) is its toughness
but carbonated. and resistance to both shear and erosion.
In general, galvanized reinforcing bars Fiber reinforcement, sometimes called
form a good bond with concrete. Zinc coat- fibermesh, provides an internal restraining
ings are relatively ductile and, therefore, mechanism which stabilizes intrinsic
galvanized bars can bend normally without stresses, particularly during the first 7 days
cracking. when the concrete is most vulnerable to
On the other hand, Rasheeduzzafar et al. shrinkage cracking. Unlike continuous reg-
(1992) found that the use of galvanized steel ular or stressed reinforcing bars or tendons,
in concrete with high levels of chloride fibers are discontinuous and generally are
merely delays concrete failure by only a randomly distributed throughout the con-
finite period of time. The latter investiga- crete mix.
tors suggested that he best durability per- In general, fibers are not likely to replace
formed is exhibited by the stainless-clad conventional reinforcement. However, in
reinforcing bars. cases such as precast fiber-reinforcement
Design Principles and Considerations of Port Elements 171

panels which are used for permanently in- The concrete used in the mixture with
stalled formwork, rehabilitation, retrofit, added fibers is of the common type, al-
and repair of deteriorated marine struc- though the proportions should be varied to
tures and other relevant applications, obtain good workability and take full ad-
fiber-reinforced concrete may be a viable vantage of the fibers. Each type of fiber
and economical alternative to regular rein- has disadvantages. These, according to
forced concrete. Romakrishnan (1988), may be as indicated
Some full-scale tests have shown that in the following paragraphs.
steel fibers can be effective in supplement- Corrosion of steel fibers in concrete with
ing and even replacing the stirrups in beams a high water1cement ratio may cause dete-
(Williamson, 1978; Craig, 1983; Sharma, rioration. However, in actual applications,
1986). Other full-scale tests have shown that corrosion of carbon steel fibers has been
steel fibers in combination with conven- found to be minimal.
tional reinforcing can increase the moment A major disadvantage of glass fiber is its
capacity of concrete beams (Henager and high vulnerability to deterioration in the
Doherty, 1976; Henager, 1977). alkaline cementations environment because
A great variety of fiber materials in vari- it seriously damages the long-term proper-
ous sizes and shapes have been developed ties of glass-fiber-reinforced concrete. The
for use in FRC. Currently, steel, glass, poly- application of glass fiber is quite complex.
meric, and carbon fibers are commonly used. To date, the major application for glass fiber
The fiber comes in many shapes. Their cross has been the spray-up process in which glass
sections include circular, rectangular, semi- fibers and a cement-rich mortar are sprayed
circular, and irregular or varying cross sec- simultaneously on a surface.
tions. A convenient numerical parameter The major shortcomings of polymeric
used for describing a fiber is the aspect fiber are low modulus of elasticity, poor
ratio, defined as the fiber length divided by bond with cement matrix, combustibility
the diameter (or equivalent diameter in the and low melting point.
case of a noncircular fiber). Typical aspect A major difficulty in the use of FRC is
ratios range from 30 to 150 for length di- the reduction in workability of the concrete
mensions of 6-75 rom. Fibers may be mix caused by the addition of fibers, espe-
straight or bent and come in various cially when large quantities of fibers are
lengths, usually as fabricated bundles. The required. This shortcoming is effectively
mixing action causes the engineered bun- eliminated by the use of superplasticizers.
dles to open and separate into millions of In the past, attempts were made to in-
individual fibers, which are uniformly dis- crease mix workability by the addition of
tributed throughout the mix in a multi- some cement and water. However, this re-
directional pattern. As the concrete starts sulted in an increase in shrinkage strains,
to harden, many microscopic cracks begin thus impeding the widespread use of FRC.
to form. At this stage, these cracks are in- As is the case with regular concrete, FRCs
tersected by the fibers which halt their properties can be improved by the use
growth. Numerous studies have confirmed of silica fume, fly ash, or polymer latex
that a substantial reduction in fiber con- additives.
crete cracking occurs due to drying shrink- Guidance concerning fiber reinforce-
age. Because the failure of fiber-reinforced ments and their use is found in ASTM A820,
concrete is by fiber pull-out, the use of de- in a selected bibliography prepared by Hoff
formed fibers increases the pull-out strength (1976-1982), in Hanna (1977), in proceed-
and, consequently, the mechanical proper- ings of the 1985 U.S.-Sweden joint seminar
ties of the concrete. edited by Shah and Skarendahl (1986), in
172 Design Principles and Considerations of Port Elements

ACI publication SP-105 edited by Shah and 2.10.2.1 concrete Placed


Batson (1987), Ramakrishnan (1988) and in by Bucket
ACI Committee 544 Report (1988). The lat-
ter also contains a comprehensive list of The bucket for placing concrete under-
relevant references. water is an open-top rectangular or cylin-
drical bucket with a drop-bottom or roller
gate opening in the bottom. Generally,
buckets are obtainable in various sizes and
2.10.2 Underwater concreting designs. This type of equipment has been
used for placing underwater concrete in
Underwater placement of concrete, usually bridge piers as deep as 75 m. The bucket
referred to as a tremie process, plays an should be completely filled and the top cov-
important role in the construction, mainte- ered with a canvas flap. It should be low-
nance, and rehabilitation of marine struc- ered gently into the water until it is com-
tures of all kinds. pletely submerged so that eddy currents
Placing underwater concrete is a very will not displace the canvas or wash the
difficult operation. It is used to fill voids in top surface of the concrete as it is being
the seafloor and in existing structures, con- lowered.
struction of different kinds of underwater Care must be taken to prevent discharg-
foundations, and also for placement in cais- ing the bucket before it has reached the
sons, piles, or in between prefabricated con- surface on which the concrete is to be de-
crete or steel members. This process is used posited, thereby avoiding dropping it
to provide added strength to these various through water. Soundings at frequent inter-
structures. Underwater concreting has been vals during construction is desirable to as-
successfully used for many years for the sist in placing so that the top surface of the
construction and repair of major marine concrete is kept level.
structures, such as dry docks, bridges, piers,
and different kinds of port quay. It is also 2.10.2.2 Tremie concrete
used for underwater construction of heavy
unreinforced concrete structures, as well as Placement of tremie concrete is carried
relatively thin reinforced concrete elements out through clean 150-300-mm-diameter
at various depths of water. To date, tremie steel pipes. A single tremie pipe system is
concrete has been successfully employed in typically used where it is required to pour
depths of water up to 260 m (Berner et al., concrete in a narrow, confined structure (i.e.,
1989). pipe, column) and multiple pipes, typically
Concrete can be placed underwater by spaced at 3-5 m, are used when it is neces-
different methods. These include the use of sary to pour concrete over large areas (Fig.
special buckets, sack and container con- 2-54).
crete, tremie, pump, and grout injection The above spacings are usually consid-
concreting. The first three methods repre- ered as optimum, although it is recognized
sent the simplest way of placing concrete that in some cohesive mixes, excellent un-
underwater. These methods are basically derwater concrete has been placed which
used for the repair of existing structures or has flowed as far as 20 m and more. In
as an auxiliary operation such as sealing of general, the distance between adjacent
gaps between forms and the seafloor; the tremie pipes depends on the ability of the
latter is performed before placing under- concrete to flow without excessive segrega-
water structural concrete. tion and washing out of the cement.
Design Principles and Considerations of Port Elements 173

:.:1
.::{

.,
~~~--ll---1-.t---o~====---~~:r------1:
-

Figure 2-54. Placement of underwater concrete over large area:


1-tremie concrete; 2-tremie pipe; 3-hopper; 4-form; 5-sand in
bags to seal form perimeter (altematively concrete in bags, placed from
inside of form).

Segregation can occur when concrete is range from 1:6 to 1:10 (vertical to horizon-
allowed to flow through water while moong tal). A closer space between adjacent tremie
water into the concrete in the process, or pipes will give a more gentle surface. In
due to dynamic effects such as attempting practical terms, the slope of the fresh con-
to vibrate concrete, or attempting to move crete will control the minimum thickness of
the tremie pipe horizontally. Some recently the structure, which should not be less than
developed chemical products, such as con- 100 em.
densed silica fumes, plasticizing admix- The tremie pipe is fabricated from
tures, and methocel (a thixotropic admix- heavy-gauge steel pipe capable of with-
ture), enable fresh tremie concrete to be standing all anticipated handling stresses.
dropped a short distance through water For deep placement, the pipe is fabricated
without segregation or washing cement out. in sections about 2 m long with bolted joints
In a multiple-pipe system, the number of that allow the upper section to be removed
pipes and their diameters depends on the as placement progresses. To prevent any
concreting rate requirements and particu- in-leakage of water and mixing into the
lars of the structure such as thickness, and concrete, joints between sections must have
the presence of reinforcement. In large soft rubber gaskets, and bolts must be prop-
placements, sometimes it is impractical to erly tightened. Before use, the pipe should
space pipes as close as 5 m (center to cen- be thoroughly cleaned and tested for water-
ter) simply because it would be very diffi- tightness.
cult to supply concrete to the number of There are two basic methods of placing
pipes involved. If pipes are spaced more concrete via the tremie pipe. The traditional
than 5 m apart, then the potential slope of practice of placing tremie concrete involves
the surface of fresh concrete must be con- use of a traveling plug (in North America
sidered. This slope is likely to be in the also known as a "pig") which is inserted
174 Design Principles and Considerations of Port Elements

into the pipe to prevent the concrete from cern. In an operation involving a traveling
mixing with water (Fig. 2-55). To control plug, the tremie pipe is placed initially 5-10
the movement of concrete down the pipe, em above the bottom of the foundation.
the plug is suspended from a wire. First, about 0.5 m 3 of a rich sand-cement
Different types of plugs are used. The mixture is poured into the tremie pipe, fol-
most common are steel cylinders, round lowed by the regular concrete mix. The plug
plates, or cones, all provided with neoprene is then slowly lowered down the pipe while
seals which loosely seal the pipe as the plug the pipe is continuously filled with concrete.
moves through it under the weight of con- During this operation, a fast movement of
crete. For relatively shallow pours of lim- the plug down to the bottom may force the
ited depth (not exceeding about 8 m), a steel water to break through the plug's seal and
plug can be replaced by a rubber ball in- flow through the fresh concrete, washing
flated up to 80 kPa, which fits very well into out the cement. Additionally, the rapid flow
the pipe and is readily forced down by the of water out of the pipe forced by the plug
concrete. may cause erosion of the bottom material.
Ideally, the plug will be self-buoyant to Therefore, plug movement must be care-
float to the surface after it leaves the pipe. fully controlled.
However, in the usual case of underwater When the plug reaches the bottom, the
construction, a trapped plug is of no con- concrete will start flowing into the form-

®
I

Figure 2-55. Stages of placement of underwater concrete (traveling plug technique): 1- tremie
pipe; 2- watertightjoint; 3- hopper; 4-traveling plug (pig); 5-wire; 6- rich sandj cement
mixture; 7- tremie concrete; 8-form.
Design Principles and Considerations of Port Elements 175

work, sideways and up. During this process, the hopper by conveyor, controlled bucket,
the tip of the pipe must always be im- or by pump.
mersed into the fresh concrete by at least The second method of pouring tremie
0.7-1.0 m to prevent water inflow into the concrete involves the use of a tremie pipe
pipe and to control the smooth flow of con- with a bottom seal installed at the tip of the
crete as it flows out and up around the pipe. pipe (Fig. 2-56). The seal is usually made
The flow of concrete should be consistent from a steel plate with a soft rubber gasket
with the rate at which it can be delivered tied to the pipe with twine. Naturally, the
into the hopper at the top end of the pipe. sealed tremie pipe must have sufficient
In general, the concrete level in the pipe weight so that it is negatively buoyant when
which is open to the atmosphere is self- empty.
regulating with regard to the external hy- The concreting starts by the installation
drostatic head. To achieve a smooth move- of the sealed pipe on the bottom and the
ment of concrete down the pipe it should be placement of about 0.5 m 3 of the sand-
evenly fed into the hopper. Delivery of con- cement mixture at the pipe's bottom, fol-
crete in large batches which are suddenly lowed by the addition of the regular con-
dumped into the hopper must be avoided. crete mix.
Preferably, the concrete should be fed into When the tremie pipe is filled, it is raised

~
E
/
~

' ~

~ =:;

e ~

E 3

l
I

Figure 2-56. Stages of placement of underwater concrete (bottom seal method). The asterisk
denotes the initial raise of the tremie pipe. 1-tremie pipe; 2- watertight joint; 3- hopper;
4- bottom-fixed seal; 5-form; 6-rich sand/cement mixture; 7-tremie concrete.
176 Design Principles and Considerations of Port Elements

about 150-200 mm, allowing the concrete For the typical underwater concrete in-
to flow out. The following procedure is ex- stallation, the following mix is used.
actly the same as the previously described
method involving a traveling plug. This Aggregates: maximum size in reinforced
technique requires a heavier pipe than in placements should not exceed 20 mm, and in
the traveling plug method to prevent a po- nonreinforced structures should not exceed
tential water break through the plug, thus 50-55% of the total aggregate by weight.
The rest should be complemented by fine
preventing washing out cement. Although
aggregate.
some segregation of concrete during the fall
Cement: relatively rich mixtures, such as 350
down the empty pipe can be expected, there kgjm 3 or more of cementitious material, are
will usually be an adequate remixture of used in tremie concreting. Cement type II,
concrete at the bottom. ASTM or equal, are typically used.
When large areas are to be filled with Pozzolanns: ASTM 616 Type Nor F, approxi-
tremie concrete, the slope of an extensive mately 15% by weight of cementitious mate-
pour will allow laitance (mixture of cement rial, are generally used to improve flow
and water, which under certain conditions characteristics.
may be a very plastic or soft and weak Workability: must be adequate to ensure a
water-laden substance) to flow down and satisfactory flow of concrete. A slump of
collect along the formwork andjor in a far 150-200 mm is generally necessary, particu-
corner. This laitance must be removed be- larly when embedded items such as reinforc-
ing steel obstruct the flow or when relatively
fore another concrete layer is placed. If not
long horizontal flow is required.
properly removed, it may be trapped under
Water I cement ratio: 0.42 or less (0.45 maxi-
good concrete and constitute a weak porous mum) is usually recommended.
chalklike layer within structural concrete.
As stated previously, the concreting
A water-reducing admixture (plasticizer), an
should proceed smoothly and continuously;
air-entraining admixture (to give a total of
the pipe must not become fixed in set con- 5-6% of air), and retarding admixtures (to
crete. In the latter case, fresh concrete increase the setting time as required) are
flowing down the tremie pipe can act as a also used.
hydraulic ram, raising the hardened upper As mentioned earlier, binding additives
layer of concrete. such as microsilica fume can inhibit cement
If the pipe sits too deep into the concrete washout and concrete mix segregation.
or the filling rate of concrete is too slow, the There are also other commercially available
pipe may get plugged. In this case, the flow antiwashout admixtures, which can be very
of concrete may be restarted without re- effective in preventing segregation of con-
moving the pipe from the concrete by the crete exposed directly to water during
use of a steel rod or other mechanical means placement.
to break the plug. The best selection of concrete mix is based
Upon completion of the underwater on test placement. Because the basic tremie
placement of concrete, the pipe should be mix being placed in large blocks will de-
removed slowly from the still fresh con- velop a fairly high heat of hydration, lead-
crete. In order to prevent the outside water ing to possible thermal cracking, a certain
from breaking through into the empty pipe means should be employed to mitigate the
through the fresh concrete, the pipe should problem. Among these is the use of coarse-
be slowly filled with water to balance the grind blast furnace slag in a proportion of
outside hydrostatic pressure. 70-85% of the total cementitious material.
Design Principles and Considerations of Port Elements 177

The reduction of the heat of hydration is and produce a loss of air in the concrete
particularly important for the cement-rich discharged from the pump. Others sug-
high-strength concrete that is required gested that the loss of air content could be
for more than 350 kgjm 3 of cementitious attributed to the possibility of a slug of
material. concrete in the vertical pipe becoming heavy
enough to slide down ahead of the pump
2.10.2.3 Pumped concrete pressure, thus creating a vacuum if air does
not leak past it.
This method of placing concrete has been Formation of a vacuum can draw the
introduced more than 35 years ago. Pump- cement paste away from the aggregate and
ing concrete is a method of conveying con- cause the line to plug. Pumping of concrete
crete by pressure through suitable rigid and the associated pressure surges can in-
pipes or flexible hoses. Pumping may be crease the likelihood of aggregates segre-
used for most concrete construction but is gating from the cement paste, thus increas-
especially suitable where space or access for ing the possibility of plugging both the
construction equipment is limited. Tremie placement line and the pipe.
placement techniques, with minor differ- To avoid the aforementioned problems,
ences, are generally applicable to direct the contractor should test his equipment
pump placement underwater. In direct and techniques to find the best way to de-
pumping, the mechanism causing concrete liver the concrete. This is especially impor-
flow through the pipeline is pump pressure tant when concrete is placed in deep water.
rather than gravity. Therefore, the concrete One solution to reducing the length of a
mix must be proportioned for pumping and vertical pipe is to make a loop in the rubber
for flow after leaving the pipe. Pipes used hose.
for direct pumping are typically smaller Mixes placed by pumping are generally
than those used for tremie concreting. How- the same as those placed by other methods.
ever, the portion of pipe immersed into the However, the maximum size of coarse ag-
concrete must be rigid enough to prevent gregate should be limited to one-third of the
any substantial sideways movement. Some- minimum inside diameter of the concrete
times, however, pump action may cause delivery hose or pipe, and the maximum
some lateral movement of the pipe embed- size of well-rounded aggregates should be
ded into fresh concrete, which may con- limited to 40% of this diameter.
tribute to laitance formation by drawing Typical equipment such as piston, pneu-
fines to the pipe-concrete interface. matic, and squeeze-pressure pumps are em-
Furthermore, in the last few years, there ployed for pumping concrete. These pumps
has been an increasing number of reports of are either trailer or truck mounted and
loss of a significant proportion of the en- range from small units with pump pres-
trained air in pumped concrete (Gaynor, sures of 1.7-2.1 MPa and outputs of 11-23
1992; Hoppe, 1992; Gorsha, 1992; Yingling m 3 jh to large units that exert pressures of
et al., 1992). Most of these cases have in- over 7 MPa on the concrete and have up to
volved large mobile long-boom pumps with 115-m3 jh output potential. The majority of
127-mm-diameter lines. According to some truck-mounted pumps are equipped with
investigators, this has occurred because hydraulically operated placing booms which
when concrete is pumped down a vertical articulate or telescope to place concrete
section of the delivery pipe, the weight of where needed.
concrete can become sufficient to overcome For a detailed discussion on pumped con-
frictional resistance, slide down the pipe, crete, the reader is referred to American
178 Design Principles and Considerations of Port Elements

Concrete Institute, ACI 304.2R (1982) and quired, is usually in the form of a blanket of
Eckardstein (1983). graded granular material, such as gravel or
crushed stones.
2.10.2.4 Grout Injection Wood, concrete, and steel are used for
concreting formwork construction. When a wooden
Under this method, grout pipes are pre- formwork is used, the boards should be of
placed in the forms, then fine-free coarse tongue-and-groove construction and the
aggregate is placed. Grout is then injected. formwork itself must be weighted or
The pipes are withdrawn as grouting anchored in order to overcome uplift forces.
proceeds. The grout forces the water out of Prefabricated concrete formwork which
the forms and fills interstices in the aggre- is incorporated into the final structure
gate. The grout can be mixed and placed in offers a very attractive solution for under-
much the same manner as used in grouting water construction. Concrete formwork
rock formations except that, in general, a panels are strong and stiff enough to with-
somewhat thicker grout is desirable. stand the pressure of fresh concrete with
For best results, a blanket of aggregate minimal deflection. It also provides for a
should be kept 0.5-1.0 m above the maxi- · durable, abrasion-resistant face of the
mum grout level. This is particularly impor- structure. A high degree of bond between
tant where the surface is exposed to moving prefabricated and underwater concrete is
water. It is essential to protect the pre- usually achieved by protruding steel dowels
placed aggregate from contamination by silt from precast members into underwater
or other pollutants between the time of concrete.
placement and grouting. Hence, the time A steel formwork has been extensively
interval should be kept as short as possible. used in the past. Steel is particularly suit-
Grout injection concreting is seldom used able for the construction of cylindrical or
for large placements; however, it is well- other complicated forms, particularly of
suited to small placements in repair works those in tension.
and placement into structural elements The minimum cover provided over rein-
containing many embedded items and forcement for underwater placed concrete
inserts. should not be less than 150 mm.

2.10.2.5 Form work for Placement


of Underwater concrete
The formwork must be robust, simple, 2.10.3 Precast concrete
and tight, especially in flowing water, to
prevent cement from being washed out, pro- Precast concrete is particularly well suited
ducing weak zones, which may effect struc- for use in marine application. It reduces
tural integrity. This is particularly impor- time of the over water construction and
tant at the bottom, where potential gaps ultimately results in overall cost saving.
between side forms and the bottom can be For example, the cost of formwork is usu-
effectively sealed by diver-placed sand bags ally on the order of 10-15% of the total
or concrete-filled bags. construction cost for a typical pier
The formwork must be designed to with- (Gaythwaite, 1990). Furthermore, addi-
stand handling and installation stresses and tional savings are achieved through reduc-
to support the weight and horizontal pres- tion in construction time, labor require-
sure of the fresh concrete with acceptable ments, and downtime due to weather
deflections. Bottom formwork, where re- conditions.
Design Principles and Considerations of Port Elements 179

Prefabricated concrete elements such as curing in order to prevent shrinkage crack-


piles, sheet piles, and miscellaneous con- ing. The compressive strength of the joint
crete armor blocks are used in breakwater concrete at a specified age should be equal
construction, floating-in-caissons, and large to the strength of concrete in adjacent pre-
concrete elements used in quay wall con- cast elements (segments).
struction. These and many other types of Dry-packed joints typically should not
prefabricated structural elements are used exceed 65-70 mm in width. Mortar should
in the construction of virtually all types of be introduced into these joints in batches
marine structures. Regular structural con- not exceeding 5 kg. Each batch should be
crete and prestressed concrete are used for thoroughly tamped and packed before the
precast concrete elements. Precast elements next batch is placed. Mortar should be
are designed for adequate structural rammed into place using a heavy hammer
strength during all construction and service and ram, or a pneumatic tamper. It should
phases, such as fabrication, transportation, be thoroughly mixed and have zero slump.
erection, and service in the completed struc- Maximum aggregate size normally does not
ture. The size and ·weight of precast ele- exceed 5 mm.
ments is govemed by the capacity of avail- Grouted joints should normally not ex-
able conventional and floating equipment ceed approximately 25 mm in width. The
for transportation, and erection or installa- nonshrink grout can be installed either by
tion. gravity or pressure methods. In either case,
The connections of precast elements to provision should be made to contain the
each other should be given special atten- grout from leaking through the forms. For
tion. They should be adequately strong and this, the joint should be tightly sealed. When
durable in the marine environment. Typi- gravity grouting is used, the grout should
cally, joints between precast elements are be consolidated by tamping. The water1
either wide or match-cast. Wide joints may cement ratio of the grout should be kept to
be cast-in-place with concrete, dry packed the minimum possible, but in no case should
with mortar, or grouted. Match-cast joints it exceed 0.45.
are normally bonded with epoxy. For all Match-cast joints: Here, epoxy resins are
types of joints, the joint surfaces must be typically used for bonding precast concrete
clean, free from salt, grease, oil, or other elements. Tensile strength of some epoxies
contaminants. that are commercially available can be equal
Where epoxy bonding is to be used, the to or even superior to the tensile strength of
joint surfaces should be lightly sandblasted the original concrete. Hence, the use of suit-
while maintaining the match-cast fit. able epoxies makes match-cast joints ex-
Cast-in-place joints with overlapping or tremely effective. When match-cast joints
welded reinforcing steel are widely em- are used, the following information concern-
ployed. The typical width of a cast-in-place ing epoxies has to be obtained from the
joint is equal to one or one-half the adjacent manufacturer:
element thickness, but not less than 100
mm. Prior to construction of such joints, the 1. Workability and pot life
adjacent surfaces should be roughened and 2. Compressive strength and modulus of
kept thoroughly wet, or an appropriate elasticity
bonding agent may be applied. The joint 3. Direct shear strength of bonded concrete
concrete should have a very low waterI prisms with notation indicating whether
cement ratio in order to minimize potential failure occurs in the joint or in the
shrinkage and should be well vibrated and concrete
cured. Particular care should be taken with 4. Ability to bond in such a way that the
180 Design Principles and Considerations of Port Elements

tensile strength of the concrete is fully single steel-consuming category of port


restored structures.
5. Creep behavior of the epoxy compound Steel consumption statistics show that
Any special safety and handling require-
hundreds of thousands of tonnes of steel
6.
ments products are used annually in the construc-
tion and modernization of port facilities
alone, not to mention other facilities associ-
All the required properties should be ob- ated with port construction. These struc-
tained at various temperatures expected at tures include breakwaters and different
the construction site during construction kinds of docking facility (bottom fixed and
periods. floating), navigation aids, and other related
The adhesive usually consists of resin structures.
and hardener components. It should be ap- In some cases, even deep water gravity-
plied immediately after mixing, and the type breakwaters have been built in the
joint surfaces should be brought together form oflarge-diameter steel sheet-pile cells,
before the pot life expires. The concrete sur- filled with granular material. In shallow or
faces which are to be bonded must be dry relatively shallow waters, popular designs
and clean. The adhesive should be applied include different kinds of propped or free-
uniformly to both surfaces; for best results standing steel and reinforced concrete
some pressure (minimum 2 kgjm 2 ) should sheet-pile walls.
be applied to bring the bonded surfaces The cell-type construction using straight
together. web-type sheet piles is gaining popularity
In some instances, mating pairs of steel because of its rapid construction and struc-
plates, which are joined by welding, or other tural reliability, improved by the develop-
suitable steel connections can be used. Spe- ment of the prefabricated sheet-pile cell
cial attention is required for welded joints construction.
to prevent cracking or spalling of concrete Quay walls of various kinds such as steel
due to overheat. Additionally, protection sheet-pile cells, various types of steel
against corrosion should be applied to pre- sheet-pile bulkheads, steel sheet-pile
vent staining of adjacent surfaces and to double-wall construction, various kinds of
protect the integrity of welds and contact relieving type and regular platform sup-
surfaces. ported on steel piles, including steel sheet-
pile walls, and others have gained wide
acceptance in marine engineering practice
2.10.4 Structural Steel in because of their relative structural simplic-
ity and economy.
Port Engineering Flexible-type dolphins, which are
columnlike structures designed to with-
2.10.4.1 General stand ship impact and other severe service
Structural steel ductility, in either ten- conditions, are built primarily from steel
sion or compression, its strength and dura- members such as large-diameter heavy-
bility, particularly with effective corrosion walled steel piles joined with steel di-
protection, and its widespread availability aphragms. Where permanent aids for navi-
makes it a viable solution to many prob- gation must be installed at a site having
lems associated with the construction of a considerable water depth, the solution is
great many kinds of port structures. In- often a steel jacket or caisson-type struc-
deed, quays, piers, and other port related ture, fabricated on-shore, then towed to the
marine structures comprise the largest site for installation. Steel-pipe piles are used
Design Principles and Considerations of Port Elements 181

to support heavy portal and container of 914-1524 mm is limited only by the abil-
cranes. ity to transport and handle these pipes.
All steel piling pipes are normally sup-
2.10.4.2 Materials and Welding plied with beveled ends to facilitate splic-
ing. A full-penetration circumferential
In different countries, structural steels welding with backup plate is usually speci-
are generally manufactured to specifica- fied for pipe pile joints. Cast steel mechani-
tions issued by national regulating bodies. cal splices are also used in cases where
In North America, such bodies are the conditions are such that field welding is
American Society for Testing Materials unreliable or uneconomical.
(ASTM) and the Canadian Standards Asso- For permanent installation in marine
ciation (CSA). structures, pipe piles with a wall thickness
In North America, conventional steel pil- less than 10 mm are not recommended.
ing pipes are readily obtained in diameters H-piles are the second largest group of
ranging from 219.1 to 1524 mm with wall piles used in marine engineering. The main
thicknesses ranging from 4.8 to 28.6 mm. advantage of these piles is that they can be
These are available in three basic grades of driven through materials where no other
steel: ASTM A252 Grades 2 and 3 and Stelco types of bearing piles could be used (e.g.,
50 with corresponding minimum yield rip-rap). They can penetrate to bedrock or
strengths of 241, 310, and 345 MPa, respec- other firm strata where piles of other types
tively. In special cases, added strength may would get hung up or be damaged in driv-
be achieved in pipe piles by filling them ing. Furthermore, H-piles can be driven
with concrete. For this purpose, concrete where a close spacing of piles is required in
order to carry heavy superimposed loads,
with a compressive strength of 30 MPa has
with a much smaller resultant displace-
been used in the past and there is a trend
ment or disruption of the soil than is
toward the use of even higher-strength con-
possible with other type of piles.
cretes. In the case of a closed-end pipe pile,
In North America, H-piles are available
a sandjcement mixture is typically placed
in the following grades according to ASTM
first, followed by placement of regular con-
specification: A36, A572 (Grades 50, 55, and
crete. In common practice, for piles up to 60), and A690 (USS Mariner) with corre-
457 mm in diameter, the pipe is considered sponding yield strengths of 250, 345, 380,
to be the chute or spout acting to prevent 415, and 545 MPa. The latter grade of steel
segregation. The workability of this con- (an alloy which includes 0.52% of copper
crete is usually confined to a slump of 50-75 and 0.54% of nickel) can provide up to two
mm. Open-end pipe piles of any diameter to three times the added resistance to corro-
and closed-end piles of 508 mm diameter sion in the seawater splash zone areas.
and greater are filled with concrete in a H-piles can be easily spliced in the field
manner discussed in Section 2.8.2. with the help of commercially available
The available length of steel piling pipe splicers.
depends on the manufacturing process used The tips of pipe piles and H-piles can be
to produce the pipe. In North America, pipes reinforced if required by different kinds of
up to 406 mm in diameter are available up tip points commercially available. In special
to 18 m in one continuous length; pipes cases, extrastrong tip points ("hard bite")
508-914 mm in diameter are available in can be used. More information on steel piles
single lengths of up to 12 m or in double is found in Chapter 7.
(two single welded together) lengths up to Steel sheet piles produced in North
24 m. The lengths of pipes with diameters America are typically manufactured in three
182 Design Principles and Considerations of Port Elements

broadly defined profile types designated as needed, similar to the beam or girder cover
straight web, arch web, and Z-shaped pro- plates. In North America, sheet piling is
file or cross sections (Fig. 2-57). Other sec- manufactured according to ASTM Specifi-
tions such as ARBED, H-type interlocked cations A328 (standard carbon steel with
with an adjacent section by special connect- minimum yield point of 265 MPa), A572
ing bars are available from European steel Grade 50 (high-strength, low-alloy steel
mills (Figs. 2-58 and 2-59). with minimum yield point 345 MPa), and
Straight web sheet piles are designated A690 (high-strength, low-alloy steel "USS
primarily for applications where the tension Mariner" with minimum yield point 345
value of the interlock is the prime consider- MPa). The latter grade has three times
ation, such as in large cellular-type struc- greater corrosion resistance to salt water in
tures filled with granular material. Where the splash zone as compared with other
required, these sheet piles may have grades of piling steel.
heavy-duty interlocks with higher interlock Regular structural steel produced in
tension values. In North America, these North America is covered by ASTM and its
sheet piles are rolled according to ASTM approximate equivalent, Canadian Stan-
A572 Grade 50 with a minimum yield dard CAN3-G40.21. The grades and types of
strength of 345 MPa. In North America, structural steel available in North America
straight web sheet piles are available with according to the above standards are listed
minimum interlock strengths from 2.1 to in Table 2-9. ·
4.9 MN jm. A factor of safety of 2.0 is usu- Note that in CSA G40.21, each grade is
ally applied to the above values. A typical designated by a number, indicating its min-
straight web piling interlock has a swing of imum yield strength in megapascals (the
about 5°-10° between two adjacent sections. minimum yield strengths are reduced for
Arch web sheet piles are adaptable to a heavier thicknesses), and by one or two let-
wide variety of field conditions. Although ters, designating its specific type as follows:
not quite as strong as the Z-type piles, these (For more information on steel sheet piles,
sheet piles possess a much larger bending consult Chapter 6.)
strength than straight web sheet piles.
These sheet piles are normally used in cel- Type G-General Construction Steels. Steels
lular-type construction where interlock ten- of this type meet specified strength require-
sion rather than bending strength is a prime ments. However, the chemical control is such
consideration, and in relatively shallow that ·not all of these steels may be welded
sheet-pile bulkheads and sheet-pile walls satisfactorily under normal field conditions.
included in different kinds of piled soil- They are primarily designed for applications
retaining structures. involving bolted connections, or for welding
under carefully controlled shop conditions.
Naturally, z- and H-shaped sheet piles
Type W-Weldable Steels. Steels of this type
possess the highest bending resistance.
meet specified strength requirements and are
These piles are usually used in sheet-pile suitable for general welded construction
bulkhead construction and typically are not where notch toughness at low temperature is
intended for structures where interlock ten- not a design requirement.
sion is the primary design consideration. Type WT- Weldable Notch Tough Steels.
Sheet piles of this type may be additionally Steels of this type meet specified strength and
reinforced where bending movement re- Charpy V-notch impact requirements and are
quirements exceed the bending resistance suitable for welded construction where notch
provided by the standard shape. In the lat- toughness at low temperature is a design
ter case, plates of required size are shop- requirement.
welded to the flanges in the areas where Type R-Atmospheric Corrosion Resistant
Design Principles and Considerations of Port Elements 183

(a)

(b )

(c)

Figure 2-57. Steel sheet piles: (a) straight web; (b) arch web; (c) Z-shaped.
184 Design Principles and Considerations of Port Elements

"' "' "'


r,.. "'

~ ~ ~ ~ ~ ~

"" "' "'


Figure 2-58. Combinations of ARBED sheet piling: 1-H-sheet pile; 2- arch web
sheet pile; 3-connecting bar.
Steel. Steels of this type meet specified
strength requirements and display an atmo-
spheric corrosion resistance approximately
four times that of plain carbon steels (copper
content not exceeding 0.02%). These steels
may be readily welded up to the maximum
thickness covered by the CSA and ASTM
standards.
Type A-Atmospheric Corrosion-Resistant
Weldable Steel. Steels of this type meet speci-
fied strength requirements and display an
atmospheric corrosion resistance similar to
steel Type R. These steels are suitable for
welded construction where notch toughness
at low temperature is not a design require-
ment. This type is often used on unpainted
structures.
Type AT- Atmospheric Corrosion-Resistant
Weldable Notch Tough Steel. Steels of this
type meet specified strength and Charpy
V-notch impact requirements and display an
atmospheric corrosion resistance similar to
steel Types A and R. These steels are suitable
for welded construction where notch tough-
ness at low temperature is a design require-
ment and are also often used on unpainted
structures.
Type Q-Quenched and Tempered Low-Alloy
Steel Plate. Steels of this type meet specified
Figure-2-59. Port-of-Miami. Construction of steel strength requirements. Although these steels
sheet-pile bulkhead. (Courtesy of ARBED.) may be readily welded, the welding and fabri-
cation techniques are of fundamental impor-
Design Principles and Considerations of Port Elements 185

Table 2-9. Structural steels per CSA standard CAN3-G40.21 with a listing
of approximate equivalent ASTM standards

CAN3-G40.21
ASTM Standard Grade and Type
Plate HSS" Plate and Bars HSS

A283 Gr.D 230G


350G
400G
A36 260W
A572 Gr. 42 A500Gr.B 300W 300W
A572 Gr. 50 A618 Gr. 2 & 3 350W 350W
380W
A572 Gr. 60 400W
A572 Gr. 65 480W
A36KFG<4l 260WT
A572 Gr. 42 KFG 300WT
A572 Gr. 50 KFG 350WT 350WT
A572 Gr. 60 KFG 400WT
A572 Gr. 65 KFG 480WT
A242 Type 1 350R
A588 Gr. C 350A 350A
A242 Type2
400A
480A
A588 Gr.C 350AT 350AT
A242 Type 2
400AT
480AT
A514 700Q
700700 QT

• Hollow structural section.


Source: Stelco Catalog, 5th ed. "Structural steels: section and uses." Stelco, Hamilton,
Ontario

tance and must not adversely affect the prop- strength of 300 MPa is usually defined as
erties of the plate, especially the heat-affected "normal strength steel," and steel with a
zone. greater minimum specified strength is de-
Type QT-Quench and Tempered Low-Alloy fined as "higher strength steel." It should
Notch Tough Steel Plate. Steels of this type be noted that typically the actual yield
meet specified strength and Charpy V-notch strength of steel as furnished exceeds the
impact requirements. They provide good re-
specified minimum values by 5-15%.
sistance to brittle fracture and are suitable
for structures where notch toughness at low When designing steel structures, particu-
temperature is a design requirement. Al- larly for cold regions ( -40°C and less), the
though these steels may be readily welded, designer must select the proper structural
the welding and fabrication techniques are of steel to be used in construction. The use of
fundamental importance and must not ad- improper steel could cause brittle fractures,
versely affect the properties of the plate, espe- even at moderate stress levels, with even-
cially the heat-affected zone. tual structural failure.
The steels presently produced, and the
Steel having a minimum specified means available to the steel mills to im-
186 Design Principles and Considerations of Port Elements

prove notch toughness properties, provide satisfactory, crack-free, sound joint; the
the designer with a wide variety of steel steel is said to be weldable if the required
grades that can meet the severe design con- weld joint can be made without difficulty or
ditions of marine environment and cold without a resulting cost increase.
regions. Some steels are more weldable than oth-
A new class of high-tensile-strength steel ers. Basic structural steels used for marine
with a yield point up to 800 MPa, good construction usually do not require special
fatigue resistance in salt water, adequate precautions or procedures. However, when
toughness to resist impact at very low tem- welding thicker plates, the increased rigid-
peratures ( -50°C to -60°C), and with im- ity and restraint and the drastic quench
proved weldability has been recently devel- effect make the use of the proper procedure
oped (Asai, 1981). This kind of material is vitally important. At the fabrication shop,
typically used in deep water offshore con- the welding procedure, which is usually
structions and for miscellaneous construc- planned well ahead, typically calls for one
tion projects in regions with cold climates. or all of the following:
Steel weldability, and particularly the
weldability of high-strength steel (carbon 1. Joint design and plate preparation
equivalent greater than 0.41), is a decisive
2. Minimal penetration to prevent the weld
criterion for selection of structural steel. In metal from being diluted with the alloy
welded joints, the weld metal and the heat~ elements in the plate
affected zone should have strength and
3. Preheating, controlled interpass tempera-
notch toughness characteristics correspond- ture, and sometimes even controlled heat
ing to those of the base metal; pores and input to retard the cooling rate and
cracks occurring in the welded joints should reduce shrinkage stress
not exceed a critical magnitude related to
the operating conditions.
The main objective of any welding proce-
The mechanical properties of welds are
dure is to join the pieces with an efficient,
determined by the chemical composition of
good quality, crackless weld. A crack in a
the weld and base metals, the formation of
weld is never minor and cannot be toler-
the microstructure in the weld metal and
ated. Cracks in the welds could result from
the heat-affected zone, and the tempera-
many factors such as the following:
ture /time ratio which occurs during weld-
ing.
Low-hydrogen electrodes are usually 1. Joint restraint, which causes high stress
in the weld
specified for the welding of structural steels
with carbon equivalent greater than 0.41. 2. Inadequate bead shape, which could re-
These electrodes are also recommended for sult in high tensile stresses causing a
splicing piles of all kinds in order to prevent longitudinal or internal crack
fracture under impact. 3. High carbon content of the base metal,
Ordinarily, a correctly designed welded which reduces the weld metal through
joint and properly made weld do not require admixture
special procedures to prevent cracks during 4. Rapid cooling rate, which increases hard-
welding or service. The need for special ening and loss of ductility in the heat-
welding procedures increases, however, affected zone
with heavy-plate structural members. Most 5. High-impact loading under low tempera-
steels can be commercially arc-welded with tures
good results. The weldability of a metal 6. Fatigue cracking due to a notch effect
refers to the relative ease of producing a from poor joint geometry
Design Principles and Considerations of Port Elements 187

To avoid cracks due to the above reasons, of a notch causes the stress corrosion crack
the following rules should be followed: to start there. Another type of corrosion
deterioration is corrosion fatigue, which
1. Create a bead with the proper width- is simply the reduction of the metal's
to-depth ratio and a proper surface (e.g., fatigue resistance caused by the presence of
slightly convex). a corrosive medium. Fatigue cracks propa-
2. Keep joint restraint problems to a mini- gate faster in the presence of cyclic tensile
mum. stresses and have a pronounced effect on
3. Avoid using inferior welding-quality steels corrosion fatigue.
that have an excessively high percentage Under some conditions of limited service,
of those elements that adversely affect or where corrosion is insignificant, corro-
weld quality (e.g., sulfur and phosphorus). sion control measures are unnecessary. On
4. Control input carefully by preheating the the other hand, when corrosion potential is
basic metal, heating between weld passes, significant, a corrosion-protection system
or postheating to control cooling rate. This must be used to retard or eliminate deterio-
lowers the shrinkage stresses and retards ration of the structure. Corrosion protection
the cooling rate, helping to prevent exces-
is normally provided to avoid costly mainte-
sive hardening in the heat-affected zone.
nance procedures resulting in loss of service
from prolonged downtime.
2.10.4.3 Protection Against
Most marine structures are expected to
corrosion
perform satisfactorily for 20-50 years, as a
The greatest disadvantage in the use of minimum. Long service-life requirements
steel in seawater application is corrosion. dictate that corrosion protection be pro-
Corrosion losses on metal structures in the vided for permanent structures from the
corrosive marine environment have been outset. While in service, the structure must
costing the American industry hundreds of retain its integrity with an adequate mar-
millions of dollars annually. The protection gin of safety. Safety factors used in marine
of metal structures against corrosion, par- structure design compensate for excessive
ticularly marine structures built of steel, is unpredictable loads, defects in workman-
very important to economic port design and ship, and improper service. Corrosion could
operation. alter adopted safety factors and increase
The factors taken into consideration dur- safety hazards to operating personnel. Con-
ing the planning and design stages of ma- sequently, corrosion protection is necessary
rine structures usually include service life, if these safety factors appear inadequate
existing and expected future environmental over the intended service life. Sulfur, am-
conditions, and the adopted level of inspec- monium sulfate, salt, and raw sugar are
tion and maintenance. examples of corrosive materials that can
Corrosion deterioration limits the service affect a dock structure, cargo handling
life of the structure. When designing any equipment, and their maintenance. To han-
type of marine structure, the corrosion po- dle such bulk materials, appropriate con-
tential-ranging from negligible to severe struction materials and corrosion-protection
-must be evaluated. If not properly con- methods must be selected.
trolled, corrosion deterioration can develop Corrosion-control methods vary from
into stress corrosion, which is the crack- good design and fabrication to use of coat-
ing of a metal under the combined effects of ings, cathodic protection systems, or a com-
tensile stresses and a corrosive environ- bination thereof. However, a knowledge of
ment. A corrosion pit on a metal surface corrosion fundamentals is necessary before
acts as a stress raiser. Corrosion at the tip deciding on the level of protection applied.
188 Design Principles and Considerations of Port Elements

For basic information on corrosion funda- The straight epoxies, those not mixed
mentals, consult Tsinker (1995). with heavier petroleum-based compounds,
There are two basic ways of protecting are normally more expensive than coal tar
structural steel against corrosion: epoxies and usually are used for patch coat-
ings or for applying new coatings to wetted
1. Application of protective coatings and submerged portions of the marine
structure. Epoxies require a near-white,
2. Utilization of cathodic protection sandblasted surface. With proper applica-
tion, epoxies are effective for 10-20 years in
2.1 0.4.3.1 Protective coatings the submerged zone.
Metallic coatings are also used for the
A coating is an economical way to protect corrosion protection of structural steel in
exposed carbon steel marine structures marine environments. Steel metallizing is a
against corrosion. It can also be durable. process of applying a metallic coating to a
The best available coatings, if not mechani- substrata metal surface by spraying molten
cally damaged, generally have a service life metal. For marine applications, suitable
of up to 15 years. sealers are used as top coats, in addition to
In recent years, many coatings have been steel metallizing, because of the relatively
developed and evaluated for service in a high "porosity'' of the metallic coatings. Ad-
marine environment. The coatings com- hesion of the applied metallic coating to the
monly used for submerged and splash zones substrate is purely mechanical and is most
on marine structures include coal tar, coal effective when applied on a roughened,
tar epoxy, epoxy, metallized zinc or alu- white-blasted metal surface.
minum with top coats, and phenolic mastic. Flame-sprayed zinc coatings compare fa-
Due to a good service record and econ- vorably with other high-ranking coatings
omy, coal tar epoxies are the most fre- when top-coated with an organic coating,
quently used coatings for immersion service except in the splash zone. Zinc performs
on steel marine structures. Epoxies have an poorly in the splash zone because of the
outstanding adhesion to metals (National higher water velocities there that erode its
Association of Corrosion Engineers, 1983; protective film. Usually, a zinc coating 0.025
Windler, 1983.) Before applying a coal tar mm thick is specified for each year of pro-
epoxy coating, the metal surface should be tection required (Horvick, 1964).
near-white blasted for good results. This The American Welding Society (1974) has
type of coating forms a hard film with good evaluated the effectiveness of flame-sprayed
resistance to fouling penetration and deteri- aluminum coatings on steel and concluded
oration by cathodic action, thus offering that zinc with a sealer coat performs better
good corrosion protection in all zones of the than sealed aluminum, whereas bare alu-
marine structure. The service life of coal tar minum performs better than bare zinc.
epoxies is a function of the thickness of the However, field metallizing is a relatively
protective film; for submerged and splash slow process, and substantial capital invest-
zones, a minimum dry film thickness of 0.5 ment is required to permit large-scale oper-
mm should be applied. However, some ation. Regardless of the merits of metallized
brands of coal tar epoxies deteriorate coatings, the cost of these systems with top
rapidly and disbond when exposed to ultra- coats ranges from 1.5 to 3 times that of a
violet light. A dull brownish color and pow- good coal tar epoxy coating.
dery disbandment characterize the deterio- According to the American Iron and Steel
ration of the coal tar epoxy; its effectiveness Institute (1981), inorganic and organic zincs
is greatly reduced when this occurs. are most frequently used for protecting the
Design Principles and Considerations of Port Elements 189

atmospheric zone of marine structures. An In general, the total thickness of all pro-
inorganic zinc coating with an intermediate tective coats must exceed 0.2 rom. The ef-
coat of vinyl, red-lead primer, and a top fectiveness of protective coatings relies on
coat of silicone-alkyd copolymer resin is their being maintained intact. The designer
rapidly gaining acceptance in naval instal- must be aware that protective coatings are
lations. This coating shows superior resis- subject to abrasion during transport, han-
tance to the action of destructive environ- dling, driving, or fixing, and from scour
mental agents such as sunlight, heat and action after installation.
cold, wind, rain and snow, and atmospheric
contaminants. The total film thickness for 2.10.4.3.2 cathodic protection
the system is about 0.15-0.2 rom.
Organic zinc-rich coating composites have Cathodic protection is an electrochemical
been developed in recent years and also method of corrosion control. It is an estab-
have had good results in marine atmo- lished means of preventing corrosion, par-
spheric exposures. They generally have bet- ticularly in marine structures. It effectively
ter film-forming characteristics and are eas- protects both bare- and coated-steel marine
ier to top coat than inorganic zinc. The structures that are totally immersed. The
three most common top coat systems are cathodic protection method changes the di-
vinyl, acrylics, and epoxies. rection of flow of electric current by making
The ultimate life of any coating is heavily the structure the cathode of an external
dependent on surface preparation. Blast electrochemical cell. It changes the direc-
cleaning is commonly used in the surface tion of flow of electric current to the steel
preparation of steel structures when a pro- structure and not from it, thus preventing
tective coating is necessary. A surface iron ions from flowing out of the steel, and
blasted to near-white gives the best results from causing its decomposition (galvanic ac-
and is necessary for long-term service of tion). This is accomplished by establishing a
coatings immersed in seawater. However, direct current (de) voltage between the pro-
blast cleaning to near-white is not economi- tected metallic structure and an auxiliary
cal for less corrosive environments such as anode so that the current flows through the
the atmosphere. The removal of oil, grease, water or moist soil to the structure.
rust, and other surface contaminants is con- Marine structures can be protected by
sidered adequate for many protective- two types of cathodic protection systems:
coating systems in atmospheric environ-
ments. 1. A galvanic anode system
As the purpose of coating is to isolate the 2. An impressed-current system
steel structure from its environment, it is
very important that the coating be free of These systems are shown schematically in
voids. At certain locations, the protective Figure 2-60.
coatings can be attacked by marine life that The galvanic anode system consists of a
fastens onto them, perforating the protec- sacrificial anode electrically connected to the
tive coatings and exposing the metal to cor- structure and immersed in an electrolyte
rosive agents, especially in tidal zones. At (seawater). The protected surface acts as a
these locations, a single or double finish cathode (hence, the name cathodic protec-
coat of antifouling or antivegetative paint tion). Galvanic anode cathodic protection
(based on slowly dissolving toxic salts of systems are popular for use on marine
copper or mercury, or organic poisons such structures because once properly installed
as phenol, piridine, acridine) can be used if they require little or no attention for the
economically justified. design service life. Because galvanic anode
190 Design Principles and Considerations of Port Elements

~ r---~~/=w~ote• ••••'~,___----'!'+-~=~

<0
e

--~--

------------
-~__./~- .1-

a-galvanic-anode system b-impressed-current system

1-marine structure 5-insulated wire


2-sacrificial anode (zinc, aluminum, or s~electron flow in external circuit
magnesium) 7-inert anode, or graphite, lead alloy or
3-metallic ions lost into solution other suitable material that will best dis-
4-electric current flowing through elec- charge the immersed current
trolyte a-external source of direct current

Figure 2-60. Type of cathodic system.

systems require no external source of power marketed. Its efficiency is claimed to ap-
and because they are relatively easy to in- proach 95%, and its actual current output is
stall and maintain, they are attractive if about 2820 A-hjkg.
protective current requirements are not Magnesium alloy anodes are less efficient than
many and if low driving voltages are ade- zinc anodes because they waste current on
quate to obtain the required current. self-protection. If used in combination with
Galvanic anodes used in marine cathodic protective coatings in seawater, magnesium
protection technology are as follows: anodes could impress too high a voltage gra-
dient across the coatings. This increases the
probability ·of coating failure due to electro-
Zinc, as an anode for cathodic protection of
osmotic penetration by seawater and of conse-
ship hulls, was introduced more than 150
years ago (Davy, 1824-1825). However, over quent blistering of the coating film by hydro-
the past 15-20 years, the zinc anode composi- gen gas evolution at the metal substrata.
tion has changed from high-purity zinc to Magnesium-clad zinc anodes are a typical zinc
zinc having small additions of aluminum anode with approximately 5.3% of magne-
(0.1-0.3%) and cadmium (0.025-0.06%), with sium cast as a jacket over all faces of the zinc.
allowable increases in iron (0.005% maxi- The basic advantage of this combination over
mum). Once properly installed, zinc anodes a conventional zinc anode is that the magne-
require no further attention. They are very sium, with its higher driving voltage, polar-
efficient and have a design life of 10 years or izes the structure much faster than would
more. zinc, thereby reducing the current demand by
Aluminum alloy anodes, like zinc anodes, can the structure, and this is easily handled
supply current as long as the structure de- by the zinc after the magnesium jacket is
mands it. Various anodes of this type have consumed.
been on the market since 1950, but they did
not perform consistently in long-term marine
service. However, in recent years, a propri- Some typical properties of galvanic an-
etary composition aluminum anode contain- odes in cathodic protection are given in
ing 0.45% mercury and 0.45% zinc has been Table 2-10. The data in Table 2-10 deci-
Design Principles and Considerations of Port Elements 191

Table2-10. Typical properties of galvanic anodes The dissolution rate of iron or steel is
in cathodic protection
about 9 kg/ A-yr, and at high current, an
Aluminum enormous amount of steel is consumed.
Property Zinc Alloy Magnesium Therefore, the bulk of such materials to be
used as anodes must be very large to avoid
Actual current 780 2820 110
output frequent replacement.
(A-hjkg) Graphite anodes are suitable for use in
Actual anode 11.3 3.1 8 both freshwater and seawater and have
consumption been successfully used for cathodic protec-
(kgjA-yr)
Density 6.9 2.8 1.7
tion of marine structures for years. Despite
(gjcma) being insoluble, graphite decomposes in
time due to oxidation, and the graphite
anode must be replaced in 5-15 years, de-
sively favor the use of aluminum alloy pending on its size and location. The theo-
anodes. retical consumption rate for a graphite
The impressed-current system is nor- anode is 1 kg/ A-yr. However, its actual
mally used for marine structures requiring consumption rate is less than 10% of the
large quantities of current for protection. It theoretical rate at low anode current densi-
normally consists of the following: ties of about 10-20 Ajm2 (American Iron
and Steel Institute, 1981).
1. Anodes and associated direct current (de) At normal anode current densities in sea-
positive wiring water, graphite deteriorates approximately
2. A de power supply and means of power 0.35 kg/A-yr to 0.55 kg/A-yr (Singleton,
regulation 1970). Assuming 0.4 kg/A-yr as a conserva-
3. A negative return circuit from the pro- tive value, a 35-kg graphite anode sub-
tected structure to the de power supply. merged in salt water would be completely
consumed, providing 7 A continuously in
Scrap iron and steel, graphite lead alloys, about 12.5 years (35 j0.4 X 7). Practically,
or iron-silicon materials are usually used as however, replacement must be made before
anodes in impressed-current cathodic pro- the anode is greatly reduced in size in order
tection systems. (Basic properties of these to avoid an excessive current density as the
anodes are shown in Table 2-11.) exposed surface of the anode diminishes.

Table 2-11. Property of anodes used for impressed-current cathodic protection

Property or High-Silicon
Characteristic Graphite Iron Alloy Lead Alloy

Recommended anode 2.5-11 11-55 55-270


current density range
(Ajm2 )
Consumption rate 365-1000 135-500 115-245
range (g/ A-yr)
Density (gjcm 2 ) 2.22 7.0 11.3
Application limitations Do not use in freshwater, Lower consumption rates Do not use in freshwater
and do not bury in mud. in freshwater. or polluted water, and
do not bury in mud.
Special operating Do not exceed maximum Do not exceed maximum Use minimum current
requirements current density of 110 current density of 55 density of30 Ajm2 •
A/m 2 • Ajm2
192 Design Principles and Considerations of Port Elements

Graphite anodes can be suspended in the the rate of 1 mil (0.025 mm)jyr (Peabody,
water or mounted in insulated holding 1967). Therefore, if the corrosion rate in the
clamps attached to the structure. Due to submerged zone of bare steel is, say, about
the brittle nature of the material, graphite 5 milsjyr, an average corrosion current of
anodes must be protected against possible c = 21.5 X 5 = 107.5 mAjm2 is flowing
damage by any floating objects or by strong continuously. To stop the corrosion by
current. means of cathodic protection, a current flow
Sometimes, but relatively seldom, iron of 107.5 mAjm 2 (with allowance for
silicon materials (high-silicon alloys), lead nonuniform current distribution) would be
alloys containing 2% silver, and some other required.
materials are used as anodes for marine A coated structure naturally requires less
cathodic protection systems. current flow to be protected against corro-
In applying cathodic protection to the sion. The effectiveness of a sound coating
submerged part of steel marine structures, could be assumed to be 90-95%. Thus, if
rectifiers are widely used to provide direct the corrosion rate in the submerged zone is
current for powering the impressed-current 5 milsjyr, the required current density for
cathodic protection system. A rectifier com- cathodic protection in this zone would be as
monly consists of a step-down transformer follows:
and a rectifier stack (usually selenium, sili-
con, copper oxide, or copper sulfide) to con- Cc = 21.5 X 5 X (0.05 to 0.1)
vert alternating current (ac) to direct cur-
= 5.35 to 10.7 mAjm 2
rent. Both transformer and rectifier stack
are mounted inside a weatherproof, protec-
tive housing. The unit can be air cooled or A margin for coating deterioration must
oil immersed. In the latter case, an explo- again be added to the current requirements
sion-proof housing must be provided. on a coated marine structure. The total re-
The design of a cathodic protection sys- quired current flow is estimated as the total
tem (galvanic anode and impressed current) area to be protected, by means of cathodic
consists of the following: protection, times flow requirements one
square meter (c or Cc).
Because anodes for marine cathodic pro-
1. Determining the protective-current re-
quirements for the structure
tection systems (galvanic and impressed
current) are standardized, their electrolyte
2. Selecting the appropriate type of anode,
resistances are known. Therefore, the total
its size, number, and location as required
for the assumed service life of the marine number of anodes required for cathodic pro-
structure tection can be determined according to
3. Determining rectifier requirements for the
cYI
impressed-current system N=-- (2-21)
4. Designing a reliable wiring system and WUE
specifying a suitable mounting method to
where
mechanically and electrically secure the
anodes to the structure N = number of anodes
c = consumption rate of anode material
The protective current required to protect a (kgjA-yr)
bare-steel marine structure depends on the Y = system of life expectancy (yr)
average annual corrosion rate of the struc- W = mass of each anode (kg)
ture. A current of about 21.5 mAjm2 flow- U = utilization factor of anode material (The
ing continuously corrodes a steel surface at utilization factor is a measure of the pro-
Design Principles and Considerations of Port Elements 193

portion of the anode which can be expected on local conditions. However, these data are
to deliver adequate current at the end of just illustrative and cannot replace the de-
the system's lifetime. It is related to the sign of real current requirements, which is
reduced cross section and length of an al- required for each particular site. Additional
most fully consumed anode. For most
considerations should be given to such pa-
marine structure anodes, this is 0.9)
rameters as the water temperature and the
E = efficiency factor (used with galvanic anodes)
dissolved oxygen content of the water.
I = average total protective current density;
some of the data on protective current den- DESIGN PROCESS OF A CATHODIC PROTEC-
sities are provided in Table 2-12; for de- TION SYSTEM. Before designing a cathodic
tails and practical examples consult Tsinker protection system, certain data should be
(1986) obtained as follows:

The number of anodes required may be in- 1. Customer preferences in system design
creased by considering such factors as 2. The configuration of all structural steel-
nonuniform current distribution between work to ensure adequate distribution of
different structural members or nonuniform anodes with particular reference to
consumption of anodes. highly stressed areas such as anodes
The anodes must deliver adequate cur- 3. Details of all steelwork above and below
rent to polarize the structure and build up seabed level to be protected
cathodic deposits. Therefore, when using 4. Nature of any coating to be applied and
current densities to calculate current re- extent of application
quirements, the following must be consid- 5. Period of time for which protection is
ered: required, together with estimates of
coating performance
1. When protecting uncoated steel marine 6. In the case of existing installations, age
structures, the initial current densities and condition of the structure's steel-
are high and afterward taper off as the work and coating
steel polarizes
7. Availability of electrical power supplies
2. When protecting coated-steel structures, and location of electrically hazardous
densities do not decrease considerably af-
areas
ter polarization but increase with time as
the coating integrity deteriorates 8. Location of structure, weather cycles,
wave heights, water depth, tidal flow,
nature of seabed, and probability of scour
The current densities required for coated
and uncoated steel in moving and stagnant 9. Water resistivity and temperature range
seawater vary over large ranges, depending 10. Details of existing or proposed adjacent
or linked structures and type of cathodic
protection installed or planned
Table 2-12. Protective-current density (mAjm2 )

To Polarize After Polarization The first step in the design is to calculate


the surface area of the structure to be pro-
Bare Steel
320-370 75-105
tected, making due allowance for small
Moving seawater
Stagnant seawater 160-265 40-75 items such as scaffolding clamps, or lad-
Soil zone 40-55 10-15 ders, which are part of the structure and a
Coated Steel further allowance for surface roughness.
Moving seawater 30-55 10-15 The required protection current varies
Stagnant seawater 10-30 5-10 around the world and, indeed, will vary
Soil zone 5-10 1-5 locally in one particular area, depending on
194 Design Principles and Considerations of Port Elements

a variety of conditions. The actual current Once the current requirement for the
density is not normally known, but the ap- structure is known, a choice between im-
proximate range is known for most parts of pressed-current and sacrificial anode sys-
the world where there is marine engineer- tems may be made. The choice is often com-
ing activity (Table 2-13). The actual method plex and is usually based on the lifetime
used to calculate the protection require- economies of the system and past experi-
ments will vary as different classifying au- ence in similar conditions. If a sacrificial
thorities use different basic rules. anode system is chosen, then the number of
The following guidelines for northern required anodes to provide the protection
North Sea applications may be taken as an current for the design time period may be
example: calculated by Eq. (2-21). In an impressed-
current system of cathodic protection, the
1. Lloyds suggests that the system should three essential components are the trans-
be capable of delivering 120 mAjm2 former rectifier or power source, the anodes,
throughout its lifetime, ignoring any con- and the cables joining them. A distributed
sideration of practical utilization of the system based on a required number of an-
anodes or variations in current demand odes mounted over the entire structure can,
throughout the life of the offshore struc- if properly designed, give a protection dis-
ture. tribution as good as that obtainable with
2. Det Norske Veritas presently proposes in sacrificial anodes.
its Recommended Practice B401, Ca- Power sources for impressed-current ca-
thodic Protection Design, that the initial thodic protection are generally custom de-
current density is 180 mAjm 2 , the mean signed for each application. They may be
is 90 mAjm 2 , and the current density based on a transformer rectifier with sepa-
available at the end of life is 120 mAjm 2 • rate controllers for each anode or group of
anodes, or on separate units. Automatic
Although these two philosophies appear control of de voltage and henc.e current out-
different, they are in fact very similar and put may be by thyristor or transistor solid-
systems designed on such bases will only state systems which can be controlled by
vary to a small extent. separate analog circuits or a microproces-
sor-based central controller which senses
structure potential with reference to a net-
Table2-13. Minimum design current densities work of permanent reference electrodes on
(mAjm2 ) for cathodic protection of bare steel the structure.
Cabling to anodes requires adequate sup-
Initial Mean Final
Area Value Value Value port and protection from the marine envi-
ronment, generally by substantial conduits
North Sea (northern) 180 90 120 over those sections exposed to wind, wave,
North Sea (southern) 150 90 100
Arabian Gulf 130 70 90
or tidal action, or where mechanical dam-
India 130 70 90 age may occur. Insulation and sheathing for
Australia 130 70 90 the cables must be selected for high resis-
Brazil 130 70 90 tance to water penetration and also to any
Gulf of Mexico 110 60 80 anodic products which may be generated;
West Africa 130 70 90
Indonesia 110 60 80
chlorine gas is often formed at impressed-
Pipelines 50 40 40 current anodes operating in seawater. Syn-
(burial specified) thetic rubbers, polyethylene, and fluori-
Saline mud 25 20 15 nated hydrocarbons have been used with
(ambient temp.) great success in marine protection systems.
Design Principles and Considerations of Port Elements 195

COMBINED CORROSION CONTROL METHODS. 2.10.4.3.3 Monitoring systems


The corrosion control method that is effec-
To ensure continued effectiveness of the
tive and economical for one zone might be cathodic protection system, whether it is of
unsuitable for another. Therefore, using just the sacrificial anode or the impressed-
one protective system might have draw- current type, it is vital to monitor the per-
backs. Ideally, the structure should be nei- formance of the system. The essential pa-
ther underprotected nor overprotected. The rameter to be measured with respect to a
optimum protection system could be a com- suitable reference electrode is the potential
bination of different protective systems. As of the steel at various locations on the
noted earlier, cathodic systems protect sub- structure. Other useful measurements can
merged steel but not steel exposed to the be provided, for example, in sacrificial an-
atmosphere. On the other hand, many coat- ode systems (the current output of specially
ings are economical and effective in the mounted and monitored anodes) and in
atmospheric and splash zones, but less at- impressed-current systems (the current and
tractive for the submerged zone due to high voltage outputs of rectifiers).
maintenance costs. Using both systems con- Zinc andjor silver/silver-chloride refer-
currently usually offers economic benefits ence electrodes are normally used for poten-
by providing effective and economic protec- tial measurements. In conventional cabled
tion to the structure. However, not all sys- systems the reference electrodes, together
tems are compatible; some coatings suffer with the monitored anodes, are wired to
disbondment when used simultaneously either a relatively simple control panel or to
with an impressed-current system. a panel which can incorporate automatic
If combinations work well, further eco- scanning and data logging.
nomic benefits may be obtained by introduc- Alternatively, the reliability of monitor-
ing the second system when the effective- ing installations can be enhanced by using
ness of the original one has decreased in a acoustic waves to detect corrosion and
particular zone. For example, an epoxy cracking in the reference electrodes and
coating could be applied over the structure anodes.
at the time of construction, and after 10-15 Comprehensive monitoring equipment is
years, a galvanic cathodic protection system commercially available to meet varying re-
could be installed to protect the submerged quirements of particular structures, their
zone. However, the possible permutations environments, and maintenance schedules.
and variations of different protective sys-
tems could be limited by economics.
In some cases, the use of unprotected 2.10.5 Structural Timber
steel with an appropriate allowance for cor-
rosion could be economically justified; for In contemporary marine engineering, tim-
example, if the life of the structure is lim- ber as a structural material is generally
ited by some design constraints. A corrosion used for the construction of temporary ma-
allowance usually in the order of 1.5-2.5 rine structures or at locations where wood
mm is provided, consistent with the antici- is readily available locally. In permanent
pated corrosion rates, coating, and other structures, it usually applies to the struc-
protective systems performance. tural members (e.g., sheet piles, or regular
For detailed information on protection of piles continuously submerged below the
steel marine structures from corrosion, the lowest water level) particularly at locations
reader is referred to National Association of free from destructive biological organisms.
Corrosion Engineers (1983). Probably the best example of timber's
196 Design Principles and Considerations of Port Elements

longevity under "wet" conditions is the more durable hardwood imported from West
than 100-year-old Brooklyn Bridge sup- Africa, has been used for construction of
ported on a timber cribbage foundation, marine structures.
which is in excellent condition where it has Ekki is very dense and heavy (about 1300
always been wet. kgjm 3 ), even-grained, free of knots, and
Destructive biological organisms that extremely strong. It is extremely resistant
may attack timber are usually more active to decay in its natural state-no pressure
in tropical and subtropical locations. The preservative treatment is required (which,
range of these organisms present at the judging from the density of wood, would be
particular location is influenced by local nearly impossible to apply successfully).
conditions, which, in addition to the water Obviously, Ekki would cost more than soft-
temperature, include salinity and degree of wood. However, considering the life of the
industrial water pollution present. The var- structure and the cost of its replacement
ious kinds of destructive insects present in due to decay and deterioration, Ekki could
harbors at different regions are discussed in well be a more economical solution because
Gaythwaite (1981) and Tsinker (1995). of its resistance to abrasion and its natural
Ironically, heavy industrial pollution can durability. In addition, because of Ekki's
be "beneficial" to the timber elements of higher allowable working stresses, a struc-
marine structures. Harbors with such con- tural element made from this material may
ditions rarely experience attack by destruc- be of smaller section, thus requiring a
tive biological organisms, because if biologi- smaller volume of material as compared
cal pollution is present to a marked degree, with softwood. However, there is a substan-
the dissolved oxygen in the water will tial doubt concerning availability of this
be low, thereby killing marine micro material to meet the demand (Western-
organisms. European Institute for Wood Preservation,
On the other hand, if industrial pollution 1987).
is less acute and oxygen values are high Some basic properties of Douglas Fir,
enough to support marine life, the sewage pine, spruce, and Ekki presently used by
will prove to be a major source of food for engineers are shown in Table 2-14. It must
these organisms. The destructive biological be noted that some authorities in the field
organisms will thrive, thus reducing the life of timber piling feel that the values of al-
of submerged structural timber. lowable working stresses presented in Table
Up to the end of the 19th century the 2-14 are too conservative (Graham, 1985).
naturally durable timbers such as black With reference to relatively recent results
locust, cedar chestnut, and white oak were of the full-scale tests conducted at the Uni-
usually used in North America for the con- versity of Colorado on pressure-treated piles
struction of port related marine structures. of dense (minimum six rings per 25 mm)
As the supplies of these timbers decreased Douglas Fir or southern pines, Graham rec-
and, ultimately, naturally durable timbers ommends allowable design compressive
were not available in sufficient, cost- stress value equal to or up to 20 MPa.
effective quantities to meet the demand, the Wood preservation history goes back to
chemically treated less durable timber approximately 2000 B.C. Literature indi-
species such as red oak, maples, Douglas cates that the ancient Egyptians applied
Fir; and the pines have been used instead. "natural oils" to wooden plows to retard the
There are isolated situations in which natu- effects of moisture on wood. The Romans
rally durable imported woods have been also recognized the positive effects of vari-
used in marine engineering. For example, ous "oils" on preservation of woods against
Ekki (or Asobe in French), a naturally attack by different marine organisms.
Design Principles and Considerations of Port Elements 197

Table2-14. Allowable working stresses for timbers• (MNjm 2 )

White Douglas Jack Pine


Properties Ekki Oak Fir and Spruce

Bending 25.5 11.6 11.6 8.1


9.45 9.45 6.6
Tension 23.4 3.31 6.8 4.7
2.24 4.8 3.4
Compression II 22.4 6.74 7.7 5.5
grain 5.55 6.4 4.6
Compression I 8.16 4.92 3.17 1.67
grain 4.92 3.17 1.67
Shear 2.04 0.75 0.57 0.43
0.75 0.57 0.43
Modulus of 17,346 10,100 11,900 8,300
elasticity 11,200 11,900 8,300
Relative hardnessb
Side 1.00 0.42 0.17 0.15
End 1.15 0.44 0.21 0.16

Note: For all species (except Ekki), the upper value is for select structural grade and the lower
value is for No. 1.
•values for Ekki determined by T.N.O. Delft, The Netherlands, and Imperial Institute,
London. All other values (except hardness) from CSA 086-MBO, Table 9a.
b Relative hardness when compared to Ekki having an arbitrary value of 1.00 side. Numbers
calculated using Eastern Forest Products Laboratory Report 21 and a report by the Imperial
Institute, London.

Historically, the development of the Waterborne arsenicals such as ammonia-


wood-treating industry in North America cal copper arsenate, ammoniacal copper zinc
during the late 19th century was based on arsenate, and chromated copper arsenate
the need to protect nondurable wood from are unique as wood preservatives in that
destructive organisms in soil and marine these chemicals react with the wood sub-
environments. In modern construction prac- strate, and subsequently become fixed in
tice, wood is treated by one of three major the wood in an insoluble state.
preservatives: creosote, pentachlorophenol, The U.S. Environmental Protection
and waterborne arsenicals (Graham, 1923; Agency (EPA) in 1985 concluded that expo-
Webb and Gjovik, 1988). All these preserva- sure to the major wood preserving chemi-
tives are well defined in American Wood- cals listed above would not pose significant
Preservers' Association (AWPA) Specifica- risk to public health (Webb and Gjovik,
tions. 1988). The EPA also concluded that the
Creosote is the oldest of the above three benefits of treated wood and the preserva-
products used to treat wood in North tives that are used to treat the wood out-
America. Wood treated with creosote, or weigh any potential risks that may occur.
solutions of creosotejpetroleum, or cre- Numerous studies on this subject cited by
osote/coal tar use either pressure or non- Webb and Gjovik (1988) support this con-
pressure treating methods. These preserva- clusion. AWPA Standards C18-86 and
tives are put in the wood to stay there, C2-85, for pressure-treated materials such
without significant losses into the ambient as lumber, timbers, and so forth used in
environment, for extended period of time. marine construction, call for the use of about
Pentachlorophenol often referred to as 400 kg of creosote or 40 kg of arsenicals to
penta is an oil-type preservative and is con- treat 1 m 3 of wood. The above rates are
sidered to be inactive as a pesticide. applicable for treatment of coastal Douglas
198 Design Principles and Considerations of Port Elements

Fir, southern pines, and western hemlock breakwater is usually termed a mole. Both,
and are used for wood treatment against the offshore and the land-connected break-
attack by teredinids, pholads, sphaeroma waters sometimes are used as docks.
(conditionally) and other biological organ- Breakwaters are used to protect exposed
isms. For more details on wood preserva- harbors otherwise lacking natural protec-
tives, the reader is referred to Lorenz and tion from adverse affects of waves, currents,
Gjovik (1972), Wade et al. (1987), Brooks migrated sediments, and drifting ice and to
(1993) and Baileys (1995). create a calm water area within the harbor
If two or more types of borers are pre- which is required for safe and convenient
sent, as is often the case, then dual treat- ship operation.
ment gives good results. Breakwaters can be shore-connected
Johnson (1987) noted that poor penetra- (moles) and offshore, which have no connec-
tion of southern pines heartwood may con- tion to the shore. The former are basically
tribute to premature failure of bulkheading designed to protect the harbor water area
exposed to teredinid attack. Graham (1983) from all the above-noted environmental
and Chellis (1961) refer to some successful phenomena. On the other hand, offshore
creosoted wood bulkhead construction in the breakwaters are designed primarily to pro-
northeast part of the United States lasted vide protection from wave action to the area
for 30-50 years. of shoreline located on the leeward side of
In North America, pressure-treated tim- the structure. The main purpose of using
ber piles are readily available in length up offshore breakwaters is to dissipate or re-
to 25 m and more. duce the amount of wave energy reaching
In conclusion, it must be pointed out that the harbor in its lee. They may also provide
in recent years, creosote-treated wood in additional protection for harbors, serving as
some states has been recognized as haz- a littoral barrier-sediment trap.
ardous to the environment. For example, in Depending on local site conditions,
California, creosote-treated timber fender shore-connected and offshore breakwaters
piles removed from a site are listed as haz- may be used separately, or in combination
ardous wastes. Their disposal has become with each other (Fig. 2-61). The selection of
very costly, and the Washington State De- the most suitable combination depends on
partment of Fisheries (WDF) in a July 1992 the predominant direction of maximum
permit approval for replacing timber piles waves.
stated that "all piling, lumber, and other The effectiveness of a selected break-
materials treated with preservatives shall water(s) layout may be checked and con-
be sufficiently cured to prevent leaching into firmed by a physical model test. According
the water or bed. WDF discourages the use to their siting in relation to the harbor,
of such materials treated with creosote and breakwaters can be defined as external and
recommends instead the use of steel, internal (Fig. 2-62). The main objective of
concrete, plastic, or other less toxic external or offshore breakwaters is to pro-
alternative." vide protection to the harbor from the ef-
fects of waves or to serve as a littoral bar-
rier. Internal breakwaters are used to pro-
tect some specific harbor areas from the
2.11 BREAKWATERS in-harbor waves generated by wind or pass-
ing ships.
By definition the breakwater is a structure The exposed side of a breakwater is typi-
protecting a shore area, harbor, anchorage, cally designed as sloped, vertical, or com-
or basin from waves. The land-connected bined configuration (Fig. 2-63) and the rear
Design Principles and Considerations of Port Elements 199

(a ) (b ) 5

~ ~
----r--....

(c )
~5

Figure 2-61. Conceptual layout of protected harbor: (a) naturally protected harbor; (b) harbor
protected by offshore breakwater; (c) harbor protected by two shore-connected breakwaters; (d)
harbor protected by shore-connected and offshore breakwaters. 1-Coastalline; 2-harbor;
3-existing island; 4-shore-connected breakwaters; 5-offshore breakwater.

side can be of any stable configuration. ble-mound breakwaters where the design
Structurally, breakwaters can be built as wave is higher than 5- 6 m. In this case,
rubble mounds, composed of floated- they represent a more economical solution
in-caissons, sheet-pile cells, or heavy con- for breakwater construction relating to nat-
crete blocks, piled, floating, pneumatic, hy- ural rocks, even when rocks are available
draulic, or composite structures (Fig. 2- 64). locally. Except in Japan, where there are
To minimize the quantity of rock mate- many artificial units competing in the mar-
rial and to limit overtopping, rubble-mound ket, there are five main types of concrete
breakwaters are often built with a gravity- units being used worldwide, namely, tet-
type concrete superstructure which can be rapods, acropodes, quadripods, tribars, and
constructed in the form of a cast-in-situ dolos (Fig. 2-65). It was been reported that
massive concrete wall, composed from in many practical cases, acropode units pro-
floated-in-caissons, or prefabricated heavy vide an economical solution for the armor of
concrete blocks. The slope of a rubble-mound rubble-mound breakwaters (Vincent, 1989).
breakwater exposed to waves is usually pro- Additionally, ordinary concrete cubes and
tected by heavy concrete blocks of different grooved or modified cubes, sometimes called
configurations or by heavy rocks. Antifer blocks, have also been used exten-
Experience has shown that artificial sively.
blocks are recommended for armoring rub- In some instances, piled breakwaters
200 Design Principles and Considerations of Port Elements

Figure 2-62. Conceptual layout of a large artificial harbor: !-shore-


connected breakwater (mole); 2-offshore breakwater; 3-in-harbor
breakwater; 4-pier; 5-marginal wharf; 6-outfitting pier; 7 -dry
dock; 8-small-craft harbor.

(a) ( b) (c)

Figure 2-63. Breakwater. Typical configurations: (a) sloped; (b) vertical; (c) composite.

were constructed to protect mostly small tions, including the annualized costs of
harbors from the effects of smaller waves. construction and maintenance in port engi-
Depending on wave forces, piled break- neering.
waters were built in the form of a single Floating breakwaters are typically
row of stiff piles (e.g., large-diameter tubu- built in different configurations, using
lar piles, propped sheet-pile walls, or double anchored-in-place concrete or steel pon-
sheet-pile walls) filled with granular mate- toons or as a flexible floating cover over a
rial and completed with a concrete super- certain area of water. In North America,
structure. In some cases, pile-supported the floating cover is typically composed of
curtain walls may present an economical used tires linked together. However, differ-
solution for breakwater construction. ent kinds of inflated covers, or covers com-
Bruun and Smith (1991) suggested gen- posed of different kinds of buoyant materi-
eral principles for optimization of break- als, can also be used.
waters in terms of the cost of port opera- For details on floating breakwater design
Design Principles and Considerations of Port Elements 201

(a)

(b)

(c)

(d)

(e) (f)

Figure 2-64. Types of breakwaters: (a) rubble-mound; (b) gravity; (c) piled; (d) floating; (e)
pneumatic; (0 hydraulic. 1- rock fill; 2-very heavy rocks or concrete blocks; 3- heavy rocks;
4-concrete superstructure; 5- large concrete blocks; 6-concrete caisson; 7- granular fill; 8-
steel sheet pile; 9- large-diameter concrete or steel tubular piles filled with granular mat-
erial or concrete; 10- concrete or steel sheet piles; 11- concrete or sheet piles; 12- rock fill;
13-concrete or steel pontoon; 14- mooring cable; 15-anchor; 16- interconnected used
tires or composition of other floating m aterials; 17 and 18- compressed air supply and distribu-
tion lines; 19- water supply system; 20-water jet.
202 Design Principles and Considerations of Port Elements

OUADRJPOD TETRAPOD

Eleva lion
Elevation

DO LOS
( OOLOSSE, plura l l TRIBAR

Eleval•on
Elevol10n
(a)

(b )

Figure 2-65. (a) The most commonly used armor units [From U.S. Army Corps of
Engineers (1984), Vol. 2]. (b) Acropode used for construction of breakwater at
Haromachi Port, Japan.
Design Principles and Considerations of Port Elements 203

Table 2-15. Pneumatic breakwater, case histories

Place of Use Year Achieved Result

El-Segundo, 1907 Waves 3.6-4.5 m high effectively


California, U.SA. dissipated.
Shipyard. "Million 1908 Heavy waves have been dissipated.
Dollars," U.S.A.
Gross Island, 1908 Tidal waves have been dissipated
Maine, U.SA. during a 15-min period of time.
BasardBay, 1908 Waves have been stopped from
Massachusetts, U.S.A., overtopping ship.
Salvage works on ship
"Yakey''
Sevastopol, former U.S.S.R. 1932 Waves 1.3-1.5 m high have been
1935 reduced to 0.2-0.5 m.
Dover, U.K. 1952 Waves 1.0-1.5 m high have been
1961 reduced to 0.15-0.25 m.
Dover, U.K. 1958 Waves 2.4-3 m high have been
reduced by 40-50%.
Hayana Island 1955 Waves 1.0-1.5 m high have been
(port of Nagasaki) 1956 reduced to 0.6 m.
Japan
Yokohama, Japan 1958 Waves 1.4-2.3 m high have been
reduced by 50-70%.
Kremenchug, Ukraine 1970 Waves 1.2-2.2 m high have been
(former U.S.S.R.) reduced by 50-70%.

and construction, the interested reader is cal installation of pneumatic breakwaters


referred to Oliver (1992), PIANC (1994), are presented in Table 2-15.
and Headland (1995). Hydraulic breakwaters are based on wa-
Pneumatic (bubble) breakwaters have ter jets producing a water current directed
been known since the beginning of the cen- against incoming natural waves, located as
tury. These breakwaters are constructed close to the surface of water as practical.
from perforated pipes placed at a certain Experiments with hydraulic breakwaters
depth below the water surface and inflated show some promising results.
with a compressed air-supply system. Air, It should be noted that experimental and
delivered to the perforated pipe, rushes to practicai installations of pneumatic and hy-
the surface of the water, where it creates draulic breakwaters known to date indicate
zones of circulating water, which, in turn, that the cost of operation of these break-
produces a surface-induced current directed waters in most cases is prohibitive.
against the waves. Interference of the in- Detailed discussion on breakwater de-
duced current with the incoming natural sign and construction is provided in Chap-
waves results in the dissipation of wave ter 10.
energy.
Perhaps because of their limited capac-
ity, pneumatic breakwaters have never been
used for the protection of major harbors. 2.12 IN-HARBOR SLOPE
They have found, however, limited use in PROTECTION
the protection of entrances to dry docks,
navigation locks, and, in some cases, as Slopes in the immediate vicinity of berthing
in-harbor secondary breakwaters. Some re- facilities and slopes adjacent to port land-
sults of successful experiments and practi- based facilities such as storages, roads, and
204 Design Principles and Considerations of Port Elements

so on are usually required to be protected (a)


from erosion by wind or ship generated
waves, current, ice, and ship propeller jets.
Although waves are a major factor to be
considered for slope protection in large sea
harbors, seasonal water-level fluctuations
and current usually govern the design of
slope and bottom protection in inland ports.
In some cases, erosion caused by ship pro-
pellers can also be detrimental to marine
structures and stability of adjacent em-
bankments.
In order to reduce marginal wharf wing-
walls, the slope of the adjacent bank protec-
tion must be as steep as practical. The
steepness of the protected slope largely de-
pends on properties of the soil used for
slope construction. For example, in sandy
soils of average density a slope of 1 : 2 to
1 : 3 (V : H) is generally used. In cohesive
soils, the allowable slope depends to a large
extent on mechanical and physical proper-
ties of the soil and may vary from quite
steep to very gentle. Figure 2-66. Slope protection. Typical configura-
The projected length of the slope can be tions: (a) sloped: (b) broken slope; (c) partly sloped.
reduced substantially by employing a rip- 1-Slope stabilization; 2-rubble-mound; 3-sheet
pile.
rap structure or by constructing slope pro-
tection of the partly sloped configuration
(Fig. 2-66). In most port slope protection
works, a slope can be stabilized by careful 2.13 AIDS TO NAVIGATION
placement of 0.3-1.0-m-thick rip-rap over a
0.3-0.5-m-thick sublayer of granular mate- Navigational aids are provided to give the
rial or geotextile filter cloth. A variety of ship navigator warning of hidden dangers
other slope protection methods are used. such as shoals, rocks, or other obstacles
Among these are rock revetments of miscel- hidden underwater. They are also installed
laneous designs, gabious, fabric-formed con- to mark the width and important turns in
crete revetments and fabric-formed (articu- navigation channels and to guide the ship
lated) concrete mats, flexible precast con- safely at maximum possible speed through
crete block mats, willow mattresses and canals, rivers, or along a coast. The types of
cast-in-place and prefabricated concrete aids required vary from location to location.
slabs; sometimes retaining walls of miscel- They could be either floating or fixed struc-
laneous designs are also used for shore and tures, unlit or equipped with required light-
bank protection. ing, sound-warning, or radar reflectors.
Useful information on a subject matter A combination of floating and fixed navi-
can be found in Span et al. (1980), Fuehrer gation aids is commonly used. Routinely,
et al. (1981), Balanin et al. (1981), Herbich navigation lights are installed on marine
(1992), and PIANC (1992b). structures such as piers, quays, dolphins,
Design Principles and Considerations of Port Elements 205

bridges, and offshore moorings to mark their in areas where there is a limited marine re-
location and outline their limits. quirement, radiobeacons may operate singly.
In the United States and Canada, lights
and colors of navigating aids are governed
by the respective coast guards of these 2.14 MOORING ACCESSORIES
countries and are very similar. The follow-
ing navigating aids are commonly used: Mooring accessories provide the docking fa-
cility with a convenient and reliable means
Lightstation: a fixed structure, such as a of mooring a ship while she is loading or
lighthouse equipped with a light and located unloading at berth. The basic requirements
at prominent sites, that assists the ship in for mooring accessories are as follows:
fixing her position. It may be installed at or
near the shoreline, or on man-made piers in • Simplicity
or near the waterway. • Reliability
Day beacon: an unlit aid primarily used to • Fast operating
assist the ship during the daylight hours • Ability to be operated with least damage
where sight navigation is practical or, when it to mooring lines
is not practical, to operate a light.
Furthermore, mooring accessories must
Fog signal: a sound-producing aid that warns
of danger when usual aids are obscured by
have convenient and safe access and must
weather conditions. It could be of fixed or be placed close enough to the ship to satisfy
floating type. the mooring requirements of all vessels that
Sector light: a kind of aid which consists of a
will use the dock.
single light whose total luminous beam is The principal mooring accessories typi-
divided into sectors of different colors to pro- cally used are as follows:
vide a warning or a leading line for ship
navigation. • Bollards
• Bitts
Range: an aid which consists of two or more
• Quick-release hooks
fixed navigation marks situated at different
• Cleats
elevations to provide a leading line for ship
• Mooring rings
navigation. The range may or may not exhibit
lights.
The most commonly used mooring accessory
Radar beacon: established to assist a radar is a conventional T -head bollard (Fig.
observer in the positive identification of a 2-67 a). It is usually manufactured from cast
charted navigational aid, such as a land posi-
steel or cast iron and, after installation,
tion, a fixed light station, or a buoy. Usually,
these beacons respond to any radar transmis- filled with concrete. A double-head bollard
sion which is within the marine radar band of (Fig. 2-67b) is used to accommodate two
9300-9500 MHz. mooring lines simultaneously and is de-
Radar reflector: used as an enhancement to
signed for a docking facility with two or
the regular fixed or floating aid, or as an more berths. Cast iron, with its natural
independent aid to navigation. resistance to corrosion, is more economical
Radiobeacon: an aid which typically trans-
than cast steel for bollard construction, but
mits signals within the radio-frequency band cast steel provides larger bollard-loading
of 285-325 kHz. These transmissions enable capacity. Quick-release bollards are seldom
ships fitted with direction-finding equipment used (Fig. 2-67 c). This type of bollard con-
to take a bearing or to take several consecu- sists of two main parts-the stem and the
tive bearings which will produce a fix. head. The stem is fixed to the deck of the
Radiobeacons usually operate in groups, and pier by a bolting arrangement, and the head
206 Design Principles and Considerations of Port Elements

(a) (b) four hooks. The rope monitor automatically


indicates when the rope tension exceeds the
limit by producing a warning signal. Then a
shipside can regulate each mooring rope to
prevent its breakage.
For mooring of small boats, mooring rings
or cleats are used.
In inland ports where ships can be
moored to the shore, or to bollards located

m
(d)
on a slope, or to mooring devices placed at
the toe of the slope of a partly sloped con-
(e) figuration quay wall, a concrete block usu-
ally with an embedded concrete-filled steel
I~ pipe is typically used instead of a conven-
tional mooring device (Fig. 2-68). The size
Figure 2-67. Mooring accessories: (a) T-head hoi- of the concrete block and pipe diameter de-
lard; (b) double-head bollard; (c) quick-release bollard;
(d) double bitt; (e) quick-release hook.
pends on the design value of the pull force
transmitted from the ship to the bollard via
mooring line.
Stability of such a type of mooring bol-
is capable of turning through a horizontal lard is determined for the mooring force,
range of 360° and automatically follows the acting at the most unfavorable angle, at the
rise and fall of the mooring line. On release, same time giving consideration to local soil
it tilts forward and sheds the ropes. The conditions, such as the angle of internal
trip mechanism is housed entirely within
friction of the soil and friction coefficient
the bollard; it is lightly loaded and indepen- between the soil and concrete block.
dent of rope tension or direction. The bol-
For better accessibility, mooring devices
lard operation can be controlled manually
are typically located next to stairs on a
as well as by hydraulic, pneumatic, or elec-
sloped portion of the quay wall. Both moor-
tric remote power, which could be on the
ing devices and stairs are normally placed
shore, on the ship, or both.
at a spacing of 20-25 m along the quay
Bitts (Fig. 2-67 d) are used for the same
wall.
purpose as bollards but for smaller ships.
Quick-release hooks (Fig. 2-67e) are usu-
ally used for major docking facilities to re-
duce manpower needed to cast off the vessel 2.15 FENDER SYSTEMS <BY
and when the time for the release mooring
M. SHIONO AND G. TSINKERl
lines is critical (e.g., for oil terminals). These
hooks should be proof-tested to 150% of
their design· load. They are usually de- It is practically impossible to avoid a strik-
signed to allow 180° hook rotation in a hori- ing blow on a berthing structure by a dock-
zontal plane so that a variety of mooring ing ship, and the fenders are typically in-
line arrangements can be handled and loads stalled to prevent direct contact between
acting at an angle of 30° up the horizontal the ship and structure. The principal ftmc-
can be handled. With remote control and a tions of the fender system is to dissipate the
load-monitor device, quick-release hood as- energy of berthing impact and to protect
semblies can be manufactured with up to both ship and structure when the moored
Design Principles and Considerations of Port Elements 207

-~
~
F=
f=
f=
f=
f=
f= ,.
F=
f=
~
f= ~
.L/ y ~
A~
3

Max.W.L.

Min. W.L.

Figure 2-68. Moorings at a quay of a partly sloped config-


uration: 1-quay; 2-concrete block; 3-steel pipe or conven-
tional mooring device; 4-stairs.

ship is exposed to the action of miscella- craft. As noted earlier in this chapter, in the
neous environmental loads (e.g., wave, wind, past 40-50 years the size and shape of
current, and ice). The effective dock fender- ships have changed dramatically. Accord-
ing reduces potential downtime and repair ingly, suitable fendering systems have been
cost for both dock structure and a ship. developed to dissipate berthing energy of
In the past, primitive fenders made from large ships and to reduce the magnitude of
bush or timber have been used to accommo- a striking force on berthing structures (The
date small boats. Later, used tires were Dock and Harbour Authority, 1982a, 1982b).
introduced and utilized. Used tires are still The fendering systems existing as of the
used quite often at berthing facilities, ser- early half of the 1980s are listed in Tables
vicing small commercial ships and pleasure 2-16 and 2-17. Fender systems currently
208 Design Principles and Considerations of Port Elements

Table 2-16. Categories offender types

Mode of Absorption of
Vessel's Kinetic Energy Fender Type

Conversion to potential 1. Gravity


energy by gravity force
Conversion to potential 2. Floating body
energy by buoyancy force
Conversion to potential Compression 3. Spring
energy by elastic 4. Bush
deformation 5. End loaded rubber
6. Pneumatic
7. Foam-filled
Compression/ 8. Side-loaded rubber
bending 9. Buckling
Shear 10. Shear
Compression/ 11. Rubberjsteel sandwich
shear
Torsion 12. Torsion
Bending 13. Flexible pile
Dissipation as heat 14. Hydraulic
energy by friction
Dissipation by plastic 15. Steel corrugated unit
deformation

Source: From PIANC (1984a).

in use for protecting marine structures can the marine structure from ship inipact. In
be classified as follows: most cases, to increase the efficiency of this
type of fender system it is enhanced. with
1. Timber rubber units that are fixed to the dock
2. Solid rubber structure and support heads of the fender
3. Pneumatic piles (Fig. 2-69). The ship impact load is
4. Foam filled transferred to several rubber units and ulti-
5. Mechanical mately to the dock structure through the
6. Other horizontally placed timber system that en-
compasses both piles and the rubber cush-
ings. This system is very flexible and ab-
2.15.1 Timber Fenders sorbs ship impact very well. Its disadvan-
tages, however, are limited capacity and
Timber fenders are the simplest form of vulnerability to concentrated impact loads
fender. They can be wooden logs hung verti- that may break the pile; on the other hand,
cally down to the water level, can be fabri- pile replacement is a difficult and costly
cated and installed as a frame of different procedure.
configurations, or can be in the form of a The stress-grade timber is usually used
wooden strip fastened to the face of the for timber fenders. Pine is the most popular
berth by different means. Worn-out timbers material for this purpose; however, hard-
must be immediately replaced by new ones wood such as oak, green-heart, teak, and
to prevent the ship hull from being dam- Ekki (Azobe) are also used for dock fender-
aged by the fastenings such as bolts or ing. The latter could eliminate many of the
brackets. problems inherent in the use of ordinary
Wooden piles are often used to protect timber for dock fendering (Gower, 1982).
Table 2-17. Fender types and characteristics

Standard Energy Reaction Rated Hull


Sizes Absorption Force Deflection Pressure
Fender Type (m) [tf- m(X10kN + m)] [tf(X10kN)] (%) [tfjm 2 (X 10kNjm2 )]

1. Gravity As Required For Project


2. Buoyancy As Required For Project
3. Spring As Required For Project
4. Bush (typical) D = 1.5 22 40
L = 3.5
5. End-loaded D = 0.25 to 0.6 1 to 17 10 to 75 50 Panel required to
rubber L = 0.37 to 0.9 distribute pressure
6. Pneumatic
(a) Floating D = 0.3to 0.6 0.5 to 840 0.9 to 795 55 to 60 10 to 20
L = 0.3to 12
(b) Fixed H = 0.45 to 2.7 1.3 to 231 13 to 385 55 to 60 29 to 34
D = 0.6to 3.6 (without panel) t:l
a>
Panel may be used to f!l,
distribute pressure '§
(c) Tyres D = 0.6to 3.0 0.25 to 92 7 to 513 21 to 54 70
7. Foam filled D = 0.9to4.5 3 to 546 15 to 574 60 10 (ave.) g.
L = 1.5 to 9.0 12 (peak)
8. Side-loaded rubber ~a>
(a) Cylindrical D = 0.15 to 2.8 0.15 to 500 4.5 to 790 50 30 to 62 rn

(b) D-type 0.15 to 0.5 0.35 to 3.7 12 to 36 50 75 to 100 ~p..


(c) Cubic 0.2 to 0.5 2to 10 65 to 130 50 200 to 300
9. Buckling 0
0
(a) Panel contact H = 0.3 to 2.5 2 to 800 16 to 730 45 to 55 Panel required to ~
-longitudinal L = 1.0to 4.0 distribute pressure s.:a>
(b) Panel contact H = 0.3to 3.0 0.45 to 670 4 to 629 47.5 to 50 Panel required to '"I
II>
-circular distribute pressure p:
0
(c) Direct contact H = 0.15 to 1.3 0.3 to 216 6 to 380 45 to 55 80 to 150 ::srn
-longitudinal L = 1.0 to 4.0
(d) Direct contact H = 0.27 to 2.5 0.45 to 604 4to 629 42.5 to 50 29 to 67 a.
-circular
10. Shear H = 0.25 to 0.6 0.8 to 10 4to 20 165 Panel required to ~
distribute pressure
11. Rubber and H = 0.8 to 1.65 3.5 to 38 17 to 85 35 to 52 Panel required to ~
distribute pressure sa>
steel sandwich
12. Torsion As Required For Project a
rn
13. Flexible pile As Required For Project
14. Hydraulic As Required For Project
15. Steel corrugated unit H = 0.55 7.6 to 30 25 to 135 57 Used with other fender to:!
unit ~
Note: Characteristics are based on perpendicular impacts. Fender performance under angular impacts varies with the type of fender.
Source: From PIANC (1984a).
210 Design Principles and Considerations of Port Elements

fenders should be countersunk. All cuts


made in the field should be painted with
creosote oil. In principle, timber fenders are
used to absorb the energy of relatively small
ships (up to 1000 DWT). The wood itself can
absorb a certain amount of energy through
compression; hence, if the fender is built up
to a substantial thickness, the force of im-
pact can be considerably reduced.
More information on timber fenders de-
sign and construction is given in Tsinker
(1986).

Figure 2-69. Wooden fender piles with timber load 2.15.2 Solid RUbber Fenders
distributing system supported on rubber cushions.
The extensive use of rubber to reduce ship
As described earlier, Ekki is a very dense impact forces on marine structures dates
and heavy (about 1300 kgjm 3 ), even- back to the late 1930s. Nowadays, solid rub-
grained, knot-free, and extremely strong ber is the principal material used for fabri-
kind of wood. Properties of this material are cation of miscellaneous fender systems.
shown in Table 2-14. It is extremely resis- Fenders of this type are able to effectively
tant to decay in its natural state; no pres- absorb the kinetic energy of a berthing ship
sure preservative treatment is required and to reduce ship impact force on the
(which, judging from the density of the berthing structure. It should be noted that
wood, would be nearly impossible to apply "softer" rubber fenders have better energy
successfully). Obviously, Ekki would cost absorption capacity and generate smaller
more than softwood or ordinary hardwood. reaction forces.
However, considering the life of the fender The efficiency of rubber fender systems is
and the cost of its replacement due to decay greatly influenced by the rubber compound
and deterioration, Ekki could well be a more used. Typically, natural rubber, or a blend
economical solution because of its resis- of natural rubber and styrene-butadiene
tance to abrasion and its natural durability. rubber (SBR), is used for fender fabrication.
In addition, because of Ekki's higher allow- Synthetic rubber is also used for fabrication
able working stresses, a structural element of some type offenders. Principal properties
made from, this material may be of smaller of these rubbers, and others used for fabri-
section, thus requiring a smaller volume of cation of marine fenders, are given in Table
material as compared with softwood. 2-18.
To reduce the effect of the adverse ma- Almost each fender-producing country
rine environment, timbers are usually has its own standard applicable to the rub-
creosoted to a retention of 200-250 kgjm 3 ber materials used for fabrication of fend-
of wood, or to refusal in the case of hard- ers. The data produced in Table 2-19 show
wood (oak or similar). However, several the comparison of some rubber properties
other preservatives, such as oil-borne or set up by the Ministry of Transport of Japan
waterborne preservatives, available on the (MOT) and requirements contained in EAU
market, could be used (Channey, 1961). Bolt (1990) (German Standard).
holes should be drilled the same diameter As for the test methods relating to these
as the bolt, and all holes on the face of the standards, both countries have established
Design Principles and Considerations of Port Elements 211

Table 2-18. Properties of rubber

Tensile Elongation Young's


Strength at Break Modulus Specific
(kgjmm2 ) (%) (kgjmm 2 X 10 3 ) Gravity
1.6-1.8 350-450 0.00006-0.0003 1.18

Table 2-19. Recommendations by MOT and EAU (1990) for rubber used for solid rubber fenders

Standard
MOT of Japan EAU(1990)
Item Before Aging After Aging Before Aging After Aging

Tensile strength 16.0 12.8 15.0 12.8


(Njmm 2 )
Elongation 300 280 300 180
(%)
Hardness 72 Original 60-75(±5)
(degree) value +8,
or 76 or less
Tear resistance 80 or more
(Njcm)
Compression set 30 or less
(%)
Abrasion resistance 100
(mm3)

their respective requirements. In general, Performance of rubber greatly depends


the weathering resistance of natural rubber on the ambient temperature. Rubber is con-
or the blend of natural rubber and SBR is sidered as being in the "rubbery state" when
very good, and in common practice the rubber molecules can move easily. An in-
durability of solid rubber fenders produced crease in ambient temperature makes rub-
from these materials is excellent. ber somewhat softer, or, in other words,
Deterioration of rubber caused by oxygen when the temperature increases, the modu-
andjor by heat is more detrimental to rub- lus of the rubber elasticity is reduced.
ber products than that caused by weather- At low ambient temperatures, rubber
ing. This type of deterioration is called molecules have low mobility and the rubber
"aging." tends to crystallize and become very brittle.
Both types of deterioration typically oc- This state of rubber is usually referred to as
cur on the rubber surface. For evaluation of a "glassy state." In this state, rubber be-
rubber aging resistance, a test in an air comes very stiff. The stiffening factor for
heating chamber at 70-100°C is typically most rubber compounds can reach about
conducted. AB shown in Figure 2-70, a rub- 1000 (Fuller et al., 1988). Furthermore, in
ber sheet 1.6 mm thick exposed to such an its "glassy state," rubber cannot function in
aging test deteriorated drastically, but a any engineering application. The tempera-
rubber block 40 mm thick did not suffer ture at which rubber turns into a "glassy
much deterioration. The basic reason for state" differs widely for different rubbers,
this is that the diffusion of oxygen and heat depending mainly on the molecular struc-
into the rubber block is very slow. ture and the chemistry of the material. Typ-
212 Design Principles and Considerations of Port Elements

1.6mn thick
Rubber Sleet

0',, '
~Original
~5
...... £ ~- --·----- ... Aged for 7 days
rn
.>w:
I
I at lDO"C air
0
0 i /tx,;,- · - A - · - ·.:. Aged for 30 days

2S 21 .I j at lDCl"C air
tc It

'L
I I'
10cm X 10cm X 4cm
bOl i! Rubber b1ock

:E.
·u;
Iii
~ 1
I-
Center

0 3 7 0 2(cm)
Days of aging Oi stance fran surface

Figure 2-70. Aging test: effect of specimen's thickness.

Table 2-20. Typical glass-transition temperature


of rubbers used in engineering applications Table 2-21. Temperature effect on performance
of beta-type fender
Glass-Transition
Temperature Reaction
("C) Temperature Load Ratio Deflection
Type
(oC) (r) (%)
Silicon rubber -80 to -125
-75 -30 1.40 41.7
Butadiene rubber (BR)
Natural rubber (NR) -70 -20 1.26 45.0
Butyl rubber (IIR) -65 -10 1.16 47.2
Ethylene propylene -40to -60 0 1.04 50.0
rubber (EPDM) +10 1.02 51.2
Styrene butadiene -60 +20 1.00 52.5
rubber (23% styrene) (SBR) +30 0.99 53.5
Polyurethane -55 +50 0.97 54.2
(polyester type) +70 0.96 55.0
Acrylonitrile butadiene -45 Note: r = R,jRo, where R, = reaction force at ambient
rubber (28% ACN) (NBR) temperature (in Celsius) and Ro = reaction force at 20°C
Polychloroprene (CR) -40 and 52.5% of fender deflection.
Source: Sumitomo Rubber Industries, Ltd., Kobe, Japan.

ical glass-transition temperatures for dif-


Table 2-22. Temperature effect on performance
ferent kind of rubbers are presented in of Pi-type fenders
Table 2-20.
The effect of low ambient temperature on Reaction
performance of some types of fenders is Temperature Load Ratio Deflection
(oC) (r) (%)
given in Tables 2-21 and 2-22 and is also
depicted in Figure 2-71. -30 1.54 40
Seawater has not been found to be dele- -10 1.10 47
terious to natural rubber and most other +10 1.04 51.5
+20 1.00 52.5
types of rubber compounds. In general, a +30-+ 70 1.00 53.5
long maintenance-free life in a seawater
environment can be expected for most rub- Note: See Table 2-21 for definition of r.
ber compounds used for fabrication of rub- Source: Sumitomo Rubber Industries, Ltd., Kobe, Japan.
Design Principles and Considerations of Port Elements 213

2.0 r---. ·- ··-·

--
30°C

t-------- /
J.- I/ -20"C

~
v ~
[7-10"C

11.0 / --------== _.-,:


l?;:1o•c
+20"C
(room tem perature)

g
/~ ~ +30° c. +so- c. + 7 O"C
J / ~
y
f
/
0
- 10

....~'!>.
.......~~
20 30 40 50 Deflection (%)

~
'-.....§
'-..-..---:..:
.....w ......... ~
~
' ......., ........... ~
,,
a: "
' ......... ...................
j
...._ '""'-...""'' +~C. +50.. C,. +7O"C
'' ~

'~ ............... , ...,.,.. +2CrC


1.0

; ' .... ..... ''(room te mperatura)


+10"C
-10°C
.fi
' ' '-2o•c
~-30"C

2.0

Figure 2-71. Temperature effect on performance of Lambda-type fender.


(Courtesy of Sumitomo Rubber Industries, Ltd.)

her fenders. Typically, the useful life of a Rubber used for manufacturing fender
rubber fender is expected to be 15-20 years. systems is commonly specified by its hard-
Shorter life is usually attributed to incor- ness (Table 2-19). In general, softer rubber
rect designs, accidents, or incorrect installa- fenders having the same rated energy-
tion. It should be noted that metal compo- absorbing capacity as stiffer fenders gener-
nents of rubber fender systems, such as ate smaller reaction forces.
anchor bolts, nuts, chains, and frontal The simplest forms of solid-rubber fender
frames typically have a shorter life than systems used in marine engineering are the
rubber components. hollow cylindrical or hollow rectangu-
All kind of rubber used in the marine lar fenders. Hollow cylindrical fenders are
environment absorb some water over a long currently available with up to a 2.8-m out-
service time period. Although this leads to side diameter. Small-diameter cylindrical
an insignificant increase in rubber mass fenders are usually manufactured by an ex-
and volume, it nevertheless does not have a trusion process, whereas, large-diameter
significant effect on the rubber's modulus of fenders are normally built from rubber
elasticity or its strength. sheets by lamination using the vulcaniza-
214 Design Principles and Considerations of Port Elements

tion process. Cylindrical fenders with an The energy absorption of the hollow
outside diameter up to 50 em may be used cylindrical rubber fender is directly propor-
as draped fenders hung over the face of the tional to the reaction force until the fender's
berthing structure (Figs. 2-72a and 2-73). deflection equals approximately 40-50% of
The lowest point in a draped fender is pro- the outside diameter. Beyond this point, the
vided with a drain hole. Larger-diameter reaction force increases much more rapidly
fenders are fastened to the structure (usu- than the energy absorbed. To avoid this
ally in a horizontal position) by means of effect, the fender diameter should be large
bars and chains, or by bars and brackets enough to absorb the required amount of
secured to the face of the dock (Figs. 2-72 energy of a berthing vessel, without deflec-
and 2-73). Typically, the internal diameter tion exceeding 50% of the outside diameter.
of a hollow cylindrical fender is about half Rectangular hollow rubber fenders (Fig.
that of its outside diameter. 2-74) spread the reaction force over a larger

DETAIL 'A'
ITYPJ

(a)

1-rubber fender 3-drain hole


2 - alloy chain 4-holding bar

(b)

Figure 2-72. Cylindrical rubber fender: (a) and (b) draped installation; (c) fenders
fastened by means of bars and chains; (d) fenders fastened by means of brackets.
Design Principles and Considerations of Port Elements 215

Here, a cylinder (or block) of length approx-


imately equal to its outside diameter (or
side) is vulcanized to two steel flanges bolted
to the fender frame and dock structure (Fig.
2.75). The steel components of the fender
are protected against corrosion by a rubber
covering, at least 3 mm thick. The advan-
tage of shear fenders is their simplicity and
lloo ability to function regardless of the direc-
3-coupling link
tion of thrust.
1-cylindrical rubber fender
2-alloy chain 4-U-bolt The number of cycles to full deflection in
the acceptable lifetime of shear fenders is
Figure 2-73. Detail 'N (see Fig. 2-72a). uncertain but is likely in the order of a few
thousand. Fatigue considerations are there-
fore considered secondary to the need to
I ensure an adequate safety factor on ulti-
~
mate bond strength, which leads to the use
I of a design shear strain of 1.0 (100%).
~ Buckling-type fenders include conven-
I tional and improved V-type fenders, buck-
ling columns (such as Pi-type fenders), cell-

,
type fenders, and conical-type fenders such
as Kappa fenders, or similar.
Performance of buckling-type fenders is
rA A
based on a buckling column or cylinder con-
cept to provide for energy absorption. Upon
ship impact, these fenders deflect and
buckle in a given direction at a predeter-
1-rubber fender
2-holding bolt

Figure 2-74. Hollow rectangular rubber fender.

H.W l.

area, thus reducing pressure on the ship


hull. The other advantage of rectangular
fenders is that at equal energy absorption
capacity, they generate similar reaction
force than that produced by cylindrical
fenders. Rectangular fenders can be fas-
tened to the berth face vertically, horizon-
tally, or at an angle. Where practical, cylin- M. W.L.

drical or rectangular rubber fenders could . ".


be placed behind a horizontal or vertical
panel to redistribute ship impact forces over 1-pier
several fender units. 2-fender frame
3-shear fender unit
Solid rubber blocks or hollow cylinders
are sometimes used in shear-type fenders. Figure 2-75. Shear fender.
216 Design Principles and Considerations of Port Elements

mined load. This buckling absorbs the ki- berthing of large ships. Design deflection of
netic energy of the berthing vessel. The this type of fender is up to 52.5%. It com-
greater the available deflection, the more prises of a pair of rubber buffers (columns)
efficiently the fender absorbs the energy of and a protector panel (rubbing board) for
a berthing vessel, thus limiting the force better redistribution of berthing impact
transmitted to the berthing structure. force on the ship hull. The span between
V-type fenders offer a simple solution to buffers is adjustable and depends mainly on
the problem of protecting the berth struc- lateral fender stability requirements, as
ture and ship during docking operation. well as on the required size of the protector
They have more than a 30-years history of panel. The design face pressure on a typical
successful performance and are highly ver- protector panel is 0.2-0.3 MNjm 2 • Pi-type
satile and self-supporting. fenders may be installed either horizontally
An improved standard (Hyper Ace) or vertically without additional support.
V-type fender is shown in Fig. 2-76. Re- However, occasionally, chains are provided
cently, conventional V-type fenders have to support the weight of the protector panel
been substantially enhanced by the devel- and to resist possible high lateral dragging
opment of Lambda (LMD)-type, Beta-type, forces.
and Super M-type fenders (Figs. 2-77, 2-78, A cell-type fender is a cylindrical elas-
and 2-79, respectively). Whereas the design tomeric column having steel mounting
deflection of a conventional V-type fender is plates permanently bonded to the rubber
about 45%, Lambda, Beta, and Super M cylinder (column) through vulcanization
fenders can deflect up to 52.5-55% and are (Fig. 2-82). Under axial load, the cylinder
buckles radially, which controls multidirec-
thus capable of absorbing more energy. Fur-
tional dispersion of energy. Cell-type fend-
thermore, Beta-type and Super M-type
ers are usually fitted with frontal protector
fenders have a wider contact area than
panels.
standard V and Lambda fenders and there-
A Kappa fender (Fig. 2-83) is a kind of
fore allow for lower contact pressure against
buckling cell-type fender. It has a conical
the ship hull. The latter property is particu-
body which provides an enhanced shear re-
larly important because the "softness" of sistance to the fender in all directions. Due
these fenders, along with low contact pres- to the special rubber compound used for
sure, allows the berth to accommodate a fabrication of Kappa-type fenders, they are
variety of ship sizes ranging from small about 15% "softer" and absorb 70% more
craft to medium-size container vessels. energy than standard V-type fenders. The
Two V"type units, placed either top-to-top former enables the Kappa fender to apply
or back-to-hack, can double the energy ab- about 25% smaller contact pressure on the
sorption of a single unit without increasing ship hull.
the reaction force (Fig. 2-80).
In general, all V-type fenders provide a
highly versatile, self-supporting, and simple
solution for protecting the berthing struc- 2.15.3 Pneumatic Fenders
ture. Each unit has a rubber-insulated steel
baseplate which supports the V-type con- Unlike solid rubber fenders, which rely on
figuration of rubber bonded to it. The fender the elastic characteristics of rubber mate-
is fastened by bolts directly to the face of rial or the buckling column phenomenon,
the docking structure. pneumatic fenders use the excellent com-
The Pi-type fender (Fig. 2-81) and simi- pressive elasticity of air. They perform ex-
lar were developed to provide for safe tremely well, especially for mooring opera-
Design Principles and Considerations of Port Elements 217

STANDARD SIZE and PERFORMANCE


Size Reaction load Energy absorption
(Height per meter lenghth per meter length of
inmm) of lender (ton) fender (lon·m)
--
Rubber CV1
Compound CV2 CV3 CV1 CV2 CV3
Y· 150H 13-5 11.3 9.0 0.7 0.6 0.5
V· 200H 1ao 15.0 12.0 1.2 1.0 0.8
y. 250H 22.5 18.8 15.0 1.9 1.6 1.3
y. 300H 27.0 22.5 18.0 2.7 2.3 1.8
y. 400H 36.0 30.0 24.0 4.8 4.0 3.2
y. SOOH 45.0 37.5 30.0 7.5 6.3 5.0
y. BOOH 54.0 45.0 36.0 10.8 9.0 7.2
y. BOOH 72.0 60.0 48.0 19.2 16.0 12.8
Y·1000H 90.0 75.0 60.0 30.0 25.0 20.0
Rated dellectoon. 45%
f-Length(L)--j
Standard lander length: 1, 1.5, 2. 2.5, 3m.
Special length at request.
The fellder of particular compound can be supplied at
request.

~- ..,_..- ...
Example (Size Indication)

-,-
Type
--,-
V - 600H
Height in mm
X
--,-
2500L
-,-
(CV2)
Length in mm Rubber compound
(Performance)
Reaction Load= 45.0 x 2.5 = 112.5Ton
Energy Absorption = 9.0 x 2.5 = 22.5Ton-m

Figure 2-76. V-type (Hyper Ace) fender. (Courtesy ofSumitomo Rubber Industries, Ltd.)
218 Design Principles and Considerations of Port Elements

STANDARD SIZE and PERFORMANCE


Size Reaction load per Energy absorption
(Height meter length of per meter length of
In mm) lender (ton) lender (ton-m)
Rubber
compound
CLO CL1 CL2 CLO CL1 CL2
250H 21 .9 1as 15.6 2.3 2.0 1.6
300H 26.3 22.5 18.8 3.3 4.3 2.4
400H 35.0 30.0 25.0 5.9 5.0 4.2
500H 43.8 37.5 31.3 9.2 7.9 6.6
600H 52.5 45.0 37.5 13.2 11.3 9.5
800H 70.0 60.0 50.0 23.5 20.2 16.8
1000H 87.5 75.0 62.5 36.8 31.5 26.3 1------Lenglh (L)
Rated dellectoon. 52.5%
Standard lender length:1, 1.5, 2, 2.5 & 3m.
Special length at request
The lender of particular compounds can be supplied at
request.

Example (Size Indication)

-,-
Type
- ,- -2500l
LMD - 600H
-,- - ,-
Height in mm
X

Length in mm
(CL2)
Rubber compound
(Performance)
Reaction Load= 37.5 x 2.5 = 93.8Ton
Energy Absorption = 9.44 x 2.5 = 23.6Ton-m

Figure 2-77. Lambda (LMD)-type fender. (Courtesy of Sumitomo Rubber Industries, Ltd.)
Design Principles and Considerations of Port Elements 219

STANDARD SIZE and PERFORMANCE


Si2e Reaction load per Energy absorption
(Height meter length of per meter lengtll of
in mm) lender (ton) lender (ton·ml
Rubber
Compound ceo CB1 CB2 C80 CB1 CB2
B- 300H 24.2 20.7 17.3 3.0 2.6 2.2
B- 400H 32.2 27.6 23.0 5.4 4.6 3.9
B- SOOH 40.3 34.5 28.8 8.5 7.3 6.1 1-Length(l)~
B- 800H 48.3 41.4 34.5 12.2 10.5 8.7
B- BOOH 64.4 55.2 46.0 21.7 18.6 15.5
B-1000H 80.5 69.0 57.5 33.9 29.0 24.2
Rated deflection. 52.5%
Standard lender legnth: 1, 1.5, 2, 2.5 & 3m.
Special length at request.
The fender of particular compound can be supplied a1
request.

Example (Size Indication)

-,-
Type
--,-
8 • 600H X
Height In mm
--,-
2500L
Length in mm
(CB2)
I
Rubber compound
(Performance)
Reaction Load= 34.5 x 2.5 = 86.3Ton
Energy Absorption = 8.7 x 2.5 = 21.8Ton-m

Figure 2- 78. Beta-type fender. (Courtesy of Sumitoino Rubber Industries, Ltd.)


220 Design Principles and Considerations of Port Elements

Pneumatic fenders can be either floating


(Fig. 2-84) or fixed to the face of the berth
(air-block fender) (Fig. 2-85). Usually, the
body of a floating pneumatic fender is pro-
tected against abrasion by an external net.
Floating pneumatic fenders, first manufac-
tured in 1952, were developed as a shock
absorber between vessels and are now
widely used as part of a berthing structure.
The air-block fender is a further develop-
ment of the pneumatic fender. It is avail-
able with and without a protector panel.85
Typically, a pneumatic fender consists of
Figure 2-79. Bridgestone Super M rubber fender. an inner rubber layer that retains the air,
surrounded by cord layers, similar to those
tions during rough weather conditions when in automobile tires. Cord layers are ar-
the ship is subjected to wave and wind ranged at angles that hold the internal air
action. They are particularly suited to piers pressure and distribute the stress evenly.
operated at relatively exposed locations. The outer rubber layer protects the cord

(a) a-top-to-top 1-rubber fender


arrangement 2-steet frame
3-resin plug
b-base-to-base 4-anchor bolt
arrangement 5-wooden blocks
6-shackle
7 -short-link chain
8
t
fr---------+-~L-----~

l
A

Section 8-B c ... Section C-C

(b) tt'~
:H


Figure 2-80. :V-type double fendering system.


Design Principles and Considerations of Port Elements 221

STANDARD SIZE and PERFORMANCE


Si%e Reaction load per Energy ab..,rptlon
(Haight meter length of per mater length of
in mm) buffer (ton) buffer (ton-m)
Rubber CO or C1 or C2 or CO or C1 or C2 or
Compound CPO CP1 CP2 CPO CP1 CP2
Pi- 400H 30.8 26.4 22.0 4.7 4.0 3.4
PI· 600H 46.2 39.6 33.0 10.6 9.1 7.6
Pi- BOOH 61.6 52.8 44.0 18.8 16.1 13.4
HPi.·1000H n.o 66.0 55.0 31.9 27.4 22.8
HPi-12!50H 96.3 82.5 68.8 49.9 42.8 35.6
HPi-1400H 107.8 92.4 n.o 62.6 53.6 44.7
HPi-1700H 130.9 112.2 93.5 92.2 79.1 65.9
HPi·2000H 154.0 132.0 110.0 127.7 109.4 ·91.2
HPi-2250H 173.3 148.5 123.8 161.6 138.5 115.4
HPi-2500H 192.5 165.0 137.5 199.5 171.0 142.5
Rated deflection. P1=50%.HP1=52.5%
Standard rubber buffer length:1, 1.5, 2, 2.5 & 3m.
Speclal length at request.
The rubber compounds of "C" series are applicable to
Pi-type whereas those of "CP" series are for HPi-type.
The fender of other compounds can be supplied at
request.

Example (Size Indication)

--,- ---,--
HPi
Type
- 1250H X
Height in mm
2200L
---,--
Length In mm
--,--
(CP2)
Rubber compound
(Performance)
Reaction Load = 68.8x22 = 151.36Ton
Energy Absorption = 35.6x22= 78.32Ton-m

Figure 2-81. Pi-type fender. (Courtesy of Sumitomo Rubber Industries, Ltd.)


222 Design Principles and Considerations of Port Elements

I
If
...!.. I ...!..

1-pier
2-rubber cylinder
3-steel protector frame covered with
low-friction frontal pads
4-shear chain

Figure 2-82. Cell-type fender.

layers and inner layer from abrasion and 2.15.5 Other Fender Systems
other external forces. Internal air pressure
in a pneumatic fender is usually main- There are other fender systems which can
tained at 50-120 kPa. be considered as alternatives in dock de-
sign. Among these are miscellaneous me-
chanical fenders, the gravity-type fender,
and the Raykin (sandwich-type) shear
2.15.4 Foam-Filled Fenders fender.
Mechanical Fenders. These fenders typi-
Originally, these fenders were developed to cally utilize springs or hydraulic shock ab-
protect offshore platforms and drilling rigs,
sorbers. In the past, attempts have been
but their application was rapidly extended
made to use disk-type or helical-type springs
to docks and ship-to-ship transfer opera-
for fenders. However, they proved to be rel-
tions (Fig. 2-86). These fenders are fabri-
atively complicated in operation and main-
cated from a closed-cell resilient foam,
tenance and were quite frequently damaged
which absorbs a significant quantity of en-
by the docking vessel. Today, they are sel-
ergy when compressed and which is pro-
dom used in modern port engineering.
tected by a thick filament-reinforced outer
Several types of hydraulic shock ab-
elastomer skin. An external net surrounds
the fender body for protection against sorbers have also been used for fendering
abrasion. purposes (Tsinker, 1986). The basic advan-
Foam-filled fenders are relatively soft. tage of hydraulic fenders is that they do not
They have high energy-absorption capabili- generate a force which causes a ship to
ties and, at the same time, a relatively low rebound (which, in fact, is the case with any
reaction force. They can be used either afloat solid rubber, pneumatic, or foam-filled
or fixed by chains to the pier face. Because fender). In hydraulic fender systems, due to
of its foam-filled construction, the fender the constant speed of fender deflection, the
will continue to operate and will not sink maximum reaction force reaches its ulti-
even if the exterior cover is damaged or mate value almost immediately after the
punctured. Unlike the pneumatic fender, ship strikes the fender. This makes hy-
the foam-filled fender does not require draulic fenders relatively stiff compared
internal-pressure adjustments. with rubber, pneumatic, or foam-filled sys-
Design Principles and Considerations of Port Elements 223

STANDARD SIZE and PERFORMANCE


Size
Reaction Load Energy abeorplion
(Height
in mm) per fender (ton) per fender (ton-m)

K·300H 20 2.2
K-400H 35 5.3
K-500H 54 10.4
K-600H 78 18.0
The abow perlorrnMCe Is of fender of CK1 rubber
compound.
Rated deflection: 52.5%
The fender of other rubber compound at request

Example (Size Indication)


K 600H
-~- --~-
Type Height in mm
(Performance)
Reaction Load= 78Ton
Energy Absorption = 18.0Ton-m

Figure 2-83. Kappa fender. (Courtesy ofSumitomo Rubber Industries, Ltd.)


224 Design Principles and Considerations of Port Elements

Figure 2-84. Floating pneumatic fender. (Courtesy of Sumitomo Rubber Industries, Ltd.)

(a ) (b)

Figure 2-85. Fixed pneumatic fenders: (a) regular (ABF); (b) with protector panel (P-ABF)
(from Yokohama Pneumatic Rubber Fenders.)
Design Principles and Considerations of Port Elements 225

1-elastomer skin 3-closed-cell flexible foam


2-nylon·filament reinforcing, integral 4-chain and tire protective net
with skin
Figure 2-86. Foam-filled fender.

terns in which this process is, in fact, (Fig. 2-88). This fender consists of a series
reversed. of connected sandwiches made of steel
Gravity-Type Fenders. A common type of plates cemented to layers of rubber. The
suspended-gravity fender unit is shown in sandwiches are mounted on a steel support
Figure 2-87. When struck by a ship, this which, in turn, is attached to the steel plates
fender is displaced inward and upward on with drilled holes. The system is topped
inclined supports, thereby utilizing the with another steel plate that conveys ·the
weight of the fender to absorb the kinetic ship impact load to the shock-absorbing
energy of impact. sandwiches. This type of shock absorber en-
The Raykin (Sandwich-Type) Fender ables the shear mode of stress to be utilized

1- p<ecast concrete blocks


2- 35-em by 35-em. 120-<:m hardwood
trmbers
3- steel frame
4- steel pon
5- poer

Figure 2-87. Suspended-gravity fender. [From Weis and Blancato


(1959).]
226 Design Principles and Considerations of Port Elements

Postble
compression
j
I 0[

0
u u 0

0
fenders of different designs are illustrated
in Figure 2-89.
~
.!;:!>

~~J 0
0 2.15.7 General Principles
0
0
0
0 in Fender System
·'" .·." Selection and Design
-Width-

Figure 2-88. Rykin (sandwich-type) fender. The major factors considered in selecting a
fender system for a particular situation are
as follows:
to its maximum advantage. The weaknesses
of the sandwich-type fenders are the consid- 1. Energy absorption: The fender must have
erable number of fastening components em- sufficient capacity to absorb the berthing
ployed in each fender unit and the practical energy of a design ship.
difficulties encountered in ensuring their 2. Reaction force: A force exerted on the
long-term protection from corrosion. Any berthing structure during impact in both
perforation of the rubber coating allows wa- perpendicular and transverse directions
ter to reach the steel plate, and corrosion must be minimal.
commences, which, in turn, destroys the 3. Hull pressure: The pressure exerted on a
bond between the plate and rubber. This ship hull by the fender unit must be
destructive process is found to accelerate acceptable.
under the action of a cathodic protection 4. Longevity of the system: In general, under
current when fenders are submerged. normal operating conditions the life ex-
Apart from the particular case of ca- pectancy for solid rubber fenders is 15-20
years. Shorter life may be expected due to
thodic protection, corrosion deterioration
a wrong design, accident, or incorrect
occurs more rapidly if the steel is alterna- installation.
tively submerged and exposed to tidal
movement, particularly in the tropics where
In fender system design, an appropriate
salt concentration occurs on the surface
factor of safety for ship, berthing structure,
from rapid evaporation. For durable perfor-
and fender should be considered. The fac-
mance, Raykin fenders require a thorough
tor(s) of safety to be applied must allow for
maintenance program.
accidental berthing, possible faults in mate-
However, any kind of fender system
rial and workmanship, and effects of im-
should be properly maintained for good and
proper maintenance. The more serious the
durable performance. Fender maintenance
consequences resulting from damage of the
should be conducted on a regular basis
fender, the higher the safety factor that
in accordance with the manufacturer's
should be used.
instructions.
The impact force caused by a docking
ship basically depends on her displacement
tonnage, the angle and speed at which the
2.15.6 Fenders Failure ship approaches the dock, the underkeel
clearance and type of berthing structure
The incorrect design, installation or abuse (for details see Chapter 3, Section 3.5, "Ship
of the fender system may result in fender Impact"). For the purposes of design, unless
physical damage and costly replacements. otherwise specified, it is customary to as-
Some typical damages sustained by rubber sume that the ship is fully loaded or in
Design Principles and Considerations of Port Elements 227

(a)

{b)

Figure 2-89. Typical damages sustained by rubber fender systems: (a)


cylinder type; (b) V-type; (c) buckling column type.
228 Design Principles and Considerations of Port Elements

(c)

Figure 2-89. Continued.

ballast and is approaching the dock at an location. As a guide, the maximum allow-
angle of 10°-20°. Larger angles are usually able hull pressure for several types of ships
used for smaller ships and primarily in in- may be used as are provided in Table 2-23.
land ports. Usually, the assumed amount of The International Commission for Im-
energy to be absorbed by one fender unit is proving the Design of Fender systems
equal to one-half the total kinetic energy of (PIANC, 1984c) has recommended that un-
the ship. In some cases, however, depending der normal operating conditions no plastic
on the shape of a ship's hull and the ap- deformation of the hull should take place
proach angle, she can be in contact with when a ship strikes the fender. As a general
several fender units simultaneously. In this rule of thumb, this commission suggested
case, the design energy should be dis- that the critical pressure (in tjm 2 ) that can
tributed among all fender units involved, in be allowed on a large ship's hull before yield
proportion to each unit's deflection. is reached is about equal (numerically) to
An important factor to be considered in the draft of the fully laden ship, expressed
fender system design is the reaction pres- in meters. This rule of thumb, however,
sure exerted by the fender on the vessel does not apply for liquid gas carriers, Ro j Ro
hull. The average pressure on a ship's hull ships, and ferries that approach the berth
is defined as the reaction force divided by
the contact or footprint area of the fender Table 2-23. Maximum allowable hull pressure
while under compression. The peak pres-
sure is that exerted at the very center of the Allowable
contact area. The maximum allowable hull Hull
Size Pressure
pressure for a given vessel depends on the Type of Ship (DWT) (kPa)
structural design of the vessel and on the
location of the point of loading. As the Tanker 10,000-45,000 200-280
Tanker 65,000 270- 370
strength of the ship's hull is usually not Tanker 100,000 360- 410
known, the fender designer can make his Bulk carrier, All Sizes 120-260
own estimate on the basis of previous expe- including
rience and a thorough evaluation of the LNGj LPG
type of vessels to be used at the particular tankers
Design Principles and Considerations of Port Elements 229

head on. More information on the subject of full fender capacity usually cannot be devel-
hull strength and allowable hull pressures oped if the loading faces are out of parallel.
can be found in Swendsen and Jensen (1970) As was previously stated, angles of 10°-20°
and in American Bureau of Shipping (1982). are frequently specified, and loss in capac-
It is important to realize that fender de- ity can be dangerously high if the design
sign curves supplied by the manufacturer, value is exceeded.
which establish the relationship between The curvature of the vessel hull is also
the fender deflection and its energy absorp- an important parameter in fender system
tion capacity and reaction force, apply to a design. A ship's hull may have a compli-
normal ambient temperature of about 20°C. cated curvature in both the horizontal and
As stated before, cold temperatures in- vertical directions, and the fender, when hit
crease the reaction force and energy absorp- by the vessel, will be compressed in compli-
tion of a rubber fender for a given compres- cated shapes by such curvatures. Generally,
sion, because of stiffening of the rubber. ships such as modern general cargo carriers
Heat decreases the reaction force and en- and tankers have almost straight vertical
ergy absorption, although the effect is less lines in areas where they usually contact
pronounced than the effect of cold. In some the fender system. However, the latest con-
cases, manufacturers do supply information tainer ships do have curvature in a vertical
on the effect of temperature of fender per- direction around the contact area, and this
formance. For example, see technical data must be considered in fender design.
on the effect of cold temperatures on solid Horizontal curvature of the vessel hull
rubber fenders provided in Sumitomo Rub- should be considered in cases such as those
ber Fender catalogs on rubber fenders. shown in Fig. 2-90. Here it is necessary to
In practice, the direction of ship impact determine the spacing of fenders required
is seldom precisely normal to the fender to prevent the ship from hitting the berth
face and is usually accompanied by a drag structure.
force acting parallel to the berth face. The Furthermore, proper fender design re-
drag force is the product of impact thrust quires a thorough evaluation of current,
and the coefficient of friction between ship waves, wind, water depth, space for ship
and fender, which may vary from 0.2 to 0.5,
depending on the contact materials. For ex-
ample, the coefficient of friction for rubber
against metal can be taken as equal to 0.5
and for the plastic cover over the protector
panel against metal as equal to 0.2. Usu-
ally, the value of a friction coefficient is
defined by fender manufacturers. If the
fender's body cannot safely absorb the drag
force, then special fastenings must be pro-
vided. Usually, these are shear chains that
provide flexible support for the fenders in
both the vertical and horizontal directions.
The direction of impact is also important
in that a fender must be wide enough to
keep the pressure on the ship hull at an
acceptable level or to span between the
transverse frames in the ship's hull. The Figure 2-90. Space between adjacent fenders.
230 Design Principles and Considerations of Port Elements

maneuvering, possibility of tug assistance,


and so on.
All these factors have to be taken into
account in selecting the fender type to be
used from the wide range available. On the
basis of all available information, the de-
signer must define the performance re-
quirements for a fender. Economic viability
must also be considered. Early consultation
with fender manufacturers is recom-
mended, as the economics of dock design as
a whole are significantly influenced by the
choice offender. Past experience should also
be considered when selecting a product
designed for long-term use that has high
initial and replacement costs.
To protect both the ship and the struc-
Figure 2-91. Example of incorrect vertical installa-
ture from being damaged during docking
tion of fender. operations, the fenders should be placed
along a berthing structure, and vertically,
in a way to eliminate any possibility of
direct contact between ship and structure.

Figure 2-92. Elevated fender system.


Design Principles and Considerations of Port Elements 231

When the ship's bow curvature is known, The shape of a ship's hull affects not only
the space between adjacent fenders can be horizontal layout of fender units but also
determined by (Fig. 2-90) their position in the vertical plane. A fender
unit incorrectly installed in the vertical
(2.22) plane cannot prevent a ship from direct
2 2 0.5
2l::;; 2[r - (r- h) ]
contact with the berth structure and fea-
where tures such as edge bollards andjor curbs
2l = distanee between adjaeent fenders
(Fig. 2-91). To avoid such contact, the
fender unit should be placed high enough to
r = radius of ship's bow eurvature
ensure safe ship berthingjunberthing pro-
h = depressed height of the fender when design
impaet foree is absorbed by the fender unit cedure. For this purpose, some fender units
have been installed above the apron level
Accordingly, the type of the fender system, (Fig. 2-92). If required, the edge of a berth
and its capacity, are selected on the basis of structure can be protected by special pro-
a design number of units which can practi- tecting systems (Fig. 2-93), or by a custom-
cally be simultaneously engaged in absorb- shaped fender system (Fig. 2- 94).
ing berthing energy. Finally, one must realize that no one
Although, typically the capacity of the fender system design can be described as
fender unit is selected to absorb energy of a the overall best; each has its advantages.
docking ship, it should also be able to effec- However, few materials are as versatile as
tively resist the design wind, waves, and rubber, and current designs do not yet
current forces imposed against a moored exploit the full potential of its energy-
ship. absorbing capacity per unit volume. The

Figure 2-93. Edge protection rubber block: 1-berth structure; 2-ship; 3-rubber
block; 4-conventional fender.
232 Design Principles and Considerations of Port Elements

(a) (b)

Figure 2-94. Edge protection fender. systems. (a) Specially configured fender panel supported
on rubber fender units; (b) double fender units.

next decade must surely bring new, more AGERSCHOU, H., LUNDGREN, H., SORENSEN, T.,
efficient, less expensive rubber fenders. ERUST, T., KORSGAARD, J ., SCHMIDT, L. R., and
CHI, W. K., 1983. Planning and Design of
Ports and Marine Terminals. John Wiley and
Sons, New York.
REFERENCES
ALTOWAIJI, W. A. K., DARWIN, D., and DONAHEY,
ACRES INTERNATIONAL LIMITED, 1989. "Panama R. C., 1984. "Preliminary Study of the Effect
Canal Commission. Relocation of Launch/ of Revibration on Concrete Bond Strength,"
Lindhandler Facilities," Preliminary design, SL Report No. 84- 2, University of Kansas
unpublished report. Center for Research, November.
AGERSCHOU, H. and KORSGAARD, J., 1969. "Sys- AMERICAN BUREAU OF SmPPING, 1982. Rules for
tem Analysis for Port Planning." The Dock Building and Classing Steel Vessels. ABS,
and Harbour Authority, March. New York.
Design Principles and Considerations of Port Elements 233

AMERICAN CONCRETE INSTITUTE, 1988. "ACI AMERICAN WELDING SoCIETY, Committee on Met-
Workshop on Epoxy-Coated Reinforcement." alizing, 197 4. "Corrosion Tests of Metalized
Concrete International, December. Coated Steel," Nineteen-Year Report, AWS
AMERICAN CONCRETE INSTITUTE, ACI 201R, 1977. C2.14-74, AWS, Miami, FL.
Guide to Durable Concrete. American Con- ANKuNDINOV, V. et al., 1989. "Use of Simulators
crete Institute, Detroit. in Harbor and Waterway Development,"
AMERICAN CONCRETE INSTITUTE, ACI 223.1R, ASCE Proceedings Specialty Conference
1983. "Standard Practice for the Use of PORTS'89, Boston, MA.
Shrinkage-Compensating Concrete." Concrete AsAI, S., 1981. "Offshore Structures (Part II),"
International, January. Paper presented at the 15th Annual Confer-
AMERICAN CONCRETE INSTITUTE, ACI 224R, 1980. ence of International Iron and Steel Institute,
Control of Cracking in Concrete Structures. Toronto, Canada, October 12-14.
American Concrete Institute, Detroit. BAILEYS, R. T., 1995. "Timber Structures and the
Marine Environment." ASCE Proceedings
AMERICAN CONCRETE INSTITUTE, ACI 304.2R,
Specialty Conference PORTS'95, Tampa, FL.
1982. Placing Concrete by Pumping Methods.
American Concrete Institute, Detroit. BALANIN, V., BYKOV, L., METELITSINA, G., and
NATALCHISHIN, G., 1981. PIANC Proceedings
AMERICAN CONCRETE INSTITUTE, ACI 305R, 1982.
25th Congress, Edinburgh, Scotland.
Hot Weather Concreting. American Concrete
Institute, Detroit. BAMFORT, P. B., 1980. "In Situ Measurement of
the Effect of Partial Portland Cement Re-
AMERICAN CONCRETE INSTITUTE, ACI 306R, 1988. placement Using Fly Ash or Granulated
"Cold Weather Concreting." ACI Materials Blast-Furnace Slag on the Performance of
Journal, JulyI August. Mass Concrete," Proceedings Institute Civil
AMERICAN CONCRETE INSTITUTE, ACI 309.1R, Engineers, Part 2, London, September.
1981. Behavior of Fresh Concrete During Vi- BARBER, P. and DOUGLAS, J.D., 1985. "The Use
bration. American Concrete Institute, Detroit. of Ship Simulators for the Appraisal and Vali-
AMERICAN CONCRETE INSTITUTE, ACI 309R, 1987. dation of Channel Designs." The Dock and
"Guide for Consolidation of Concrete." ACI Harbour Authority, September.
Materials Journal, September/October. BARRASS, C. B., 1978. Ship Squat. Politech
AMERICAN CONCRETE INSTITUTE, ACI 311.4R, International, Luton, U.K.
1988. "Guide for Concrete Inspection." ACI BEEBY, A. W., 1980. "Concrete in the Oceans."
Materials Journal, JulyI August. CIRIAjUFG Technical Report No.1. Cement
AMERICAN CONCRETE INSTITUTE, ACI 318, 1989. and Concrete Association, W exham Springs.
Building Code Requirements for Reinforced BERKE, N. S., WEIL, T. G., and PFEIFER, D. W.,
Concrete. American Concrete Institute, 1988. Concrete International, December.
Detroit. BERNER, D. E., HAGGERTY, B., and GERWICK,
AMERICAN CONCRETE INSTITUTE, ACI 377R, 1984. B. C., 1989. "260 m Deep Tremie Concrete
Guide for the Design and Construction of Placement for Belled Foundation Rehabilita-
Fixed Offshore Structures. American Con- tion of the North Rankin 'A'," Proceedings
crete Institute, Detroit. Offshore Technology Conference, Houston,
AMERICAN CONCRETE INSTITUTE, ACI 544, 1988. TX, paper 6145.
"Design Considerations on Steel Fiber Rein- BERRIMAN, J. W. and HERBICH, J. B., 1977. "Port
forced Concrete." ACI Structural Journal, Improvement Design Criteria for Texas
September/October. Coast." ASCE Journal of the Waterway, Port,
AMERICAN IRON AND STEEL INSTITUTE, 1981. Coastal and Ocean Division, November.
Handbook of Corrosion Protection for Steel BLAAUW, H. G. and VAN DER KNAAP, 1983. "Pre-
Pile Structures in Marine Environments. diction of Squat of Ships Sailing in Restricted
American Iron and. Steel Institute, Waters," Proceedings of International Port
Washington. Congress, Antwerpen.
234 Design Principles and Considerations of Port Elements

BROOKS, K. M., 1993. "Literature Review and ings Specialty Conference PORTS'95, Tampa,
Assessment of the Potential Environmental FL.
Risks Associated with the Use of Treated DAND, I. W. and FERGUSON, A. M., 1973. "The
Wood Products in Aquatic Environments." Squat of Full Ships in Shallow Water." The
Prepared for Western Wood Preservative In- Naval Architect, October.
stitute, Vancouver, WA. DAND, I. W., 1981. "An Approach to the Design of
BRUUN, P., 1989. Port Engineering, 4th ed. Gulf Navigation Channels." National Maritime
Publishing Co., Houston, TX. Institute Report R104, May.
BRUUN, P. and SMITH 0., 1991. "Economic Opti- DAVY, SIR HUMPHRY, 1824-1825. "On the Corro-
mization of Breakwaters in Port Engineering." sion of Copper Sheeting and Ships by Sea-
The Dock and Harbours, DecemberjJaimary. water and a Method for Protecting the Effect."
BSI, 1984. "British Standard Code of Practice Philosophical Transactions of the Royal
for Maritime Structures," Part I. General Cri- Society, Vols. 114 and 115.
teria, British Standards Institution, BS6349, DELFT HYDRAULICS LABORATORY, 1984. "Tests on
London. Additional Parts: Part 2-"Design of the Controllability of LNG Carriers in Shal-
Quay Walls, Jetties and Dolphins." Part low Waters," Delft Report M1967 (in Dutch).
3-"Design of Dry Docks, Locks, Slipways DICKSON, A F., 1969. "Trends in Vessel Type
and Shipbuilding Berths." Part 4-"Design of and Size-Liquid Cargoes," The Institution of
Fendering and Mooring Systems" (1985). Civil Engineers, Proceedings of Conference
CAIRNS, J., 1994. "Performance of Epoxy-Coated Tanker and Bulk Carrier Terminals.
Reinforcement at the Serviceability Limit DZHUNKOVSKI, N. N., KAsPERSON, A A, SMIRNOV,
State," Proceedings Institute of Civil Engi- G. N., and SIDOROVA, A G., 1964. Ports and
neers. Structures and Buildings, February. Port Structures, Vol. 1 (in Russian). Stroyis-
CAMPBELL, N. P. and ZwAMBORN, J. A, 1984. dat Publishing House, Moscow.
"Richards Bay Harbor-Port Operational EAU, 1990. Recommendations of the Committee
Manual, Mark 1." Permanent International for Waterfront Structures, EAU 1990: Comm.
Association of Navigation Congresses, Bul- for Waterfront Structures, Soc. for Harbor
letin No. 45, Brussels. Eng' g. and The German Soc. for Soil Mech's
CHANNEY, C. A, 1961. Marinas-Recommenda- and Foundation Eng'g., 4th English transla-
tions for Design, Construction and Mainte- tion. Wilhelm Ernst & Sohn, Berlin.
nance. National Association of Engineers and ECKARDSTEIN, K. E., 1983. Pumping Concrete
Boat Manufacturers, Inc. and Concrete Pumps. Publisher SCHWING,
CHELLIS, R. D., 1961. "Pile Foundations, 2nd ed. Heme, Germany.
McGraw-Hill, New York. EDEN, E. W., 1971. "Vessel Controllability in
CoNSTANTINE, T., 1960. "On the Movement of Restricted Waters," ASCE Journal of Water-
Ships in Restricted Waterways." Journal of ways, Harbors and Coastal Engineering, Vol.
Fluid Mechanics, Manchester University, UK. 97, March.
CoRNICK, H. F., 1969. Dock and Harbour Engi- EISIMINGER, S. K., 1953. "Widening and Deepen-
neering. Charles Griffin and Co. Limited, ing the Columbia and Willamette Rivers."
London. The Dock and Harbour Authority No. 34,
CRAIG, R. J., 1983. "Design Procedures for February.
Fibrous Concrete-Shear, Moment and Tor- ERYUZLU, N. E., CAo, Y. L., and D'AGNOLO, F.,
sion," Proceedings of Conference on Struc- 1994. "Under Keel Requirements for Large
tural Concrete Design, New Jersey Institute Vessels in Shallow Waterways," Proceedings
of Technology, May. ·28th Congress PIANC.
CusHING, C. R., 1989. "Vessels ofFuture: A Naval EYRE, J. L., 1989. "The Containerships of 1999."
Architect's Viewpoint." PIANC Bulletin No. PIANC Bulletin No. 66.
66, Brussels, Belgium. FARR, A K., 1986. "Soil Quality Testing: Why?"
DAGGETT, L. L., 1995. "Marine Simulation Capa- American Society of Civil Engineers, Spe-
bilities for Channel Design," ASCE Proceed- cialty Conference PORTS'86, Oakland, CA.
Design Principles and Considerations of Port Elements 235

FIP, 1985. "Recommendations for the Design and CEB-RILEM-IASS Symposium on Behavior
Construction of Concrete Sea Structures." of Concrete Structures in Service, Liege, June.
Federation Internationale de Ia Precon- GERWICK, B. C., 1976. "Materials for Concrete
trainte, London. Ships: Durability and Corrosion Protection
FELD, J., 1968. Construction Failures. John Coating, Inspection, and Repair," Proceed-
Wiley and Sons, New York. ings of the Conference on Concrete Ships
FERGUSON, E. S., 1992. Engineering and the and Floating Structures, Berkeley, CA,
Mind's Eye. MIT Press, Cambridge, MA. September.
FERGUSON, E. S., 1993. "How Engineers Lose GERWICK, B. C., 1986. Construction of Offshore
Touch," American Heritage Invention and Structures. John Wiley and Sons, New York.
Technology, Winter Issue. GERWICK, B. C., 1990. "International Experience
FOCHT, J. A. Jr., 1994. "Lessons Learned from in the Performance of Marine Concrete." Con-
Missed Predictions," ASCE Journal of crete International, May.
Geotechnical Engineering, Vol. 120, No. 10, GJORY, 0. E., 1971. "Long-Time Durability of
August. Concrete in Seawater." ACI Journal, January.
FORSSBLAD, L., 1987. Need for Consolidation of GoLDSTEIN, G., 1983. "Australian Coal Terminal
Superplasticized Concrete Mixes. American Operates Happily Beside Residential Area."
Concrete Institute, Detroit. Civil Engineering, Vol. 53, No. 12.
FuEHRER, M., R5MISH, K., and ENGELKE, G., 1981. GORSHA, R. P., 1992. "Air Loss in Free-Falling
"Criteria for Dimensioning the Bottom and Concrete." Concrete International, August.
Slope Protections and for Applying the New
Methods of Protecting Navigation Canals," GOWER, E., 1982. "Could Ekki be Used to Deck
PIANC Proceedings 25th Congress, Edin- Logging Bridges?" Logging and Sawmilling
burgh, Scotland. Journal, January.
FUEHRER, M. and ROMISH, K., 1982-1983. "The GRAHAM, J. S., 1983. "Design of Pressure-Treated
Effects of Hydrodynamic Forces on Ships Wood Bulkheads," Proceedings ASCE Spe-
Navigating through Canals." A Contribution cialty Conference on the Design, Construc-
to an Estimation of the Maneuverability of tion, Maintenance and Performance of Coastal
Ships in a Restricted ·Channel, PIANC, Bul- Structures, Arlington, Virginia.
letin No. 43. GRAHAM, J. S., 1985. "Allowable Compressive
FuGL-MEYER, H., 1957. The Modern Port. Danish Design Stresses for Pressure-Treated Round
Technical Press, Copenhagen. Timber Foundation Piling," Presented before
FuLLER, K. N. G., GREGORY, M. J., HARRIS, J. A., the AWPA, Reno, NV.
MUHR, A. H., ROBERTS, A. D., and STEVENSON GROTHAUS, W. and RIPLEY, D. M., 1958. "St.
A., 1988. "Engineering Use of Natural Rub- Lawrence Seaway, 1958, 27 ft Canals and
ber," Natural Rubber Science and Technol- Channels." ASCE Journal of the Waterways
ogy, Roberts A. D. (ed.), Oxford University and Harbors Division, Vol. 87, No. 1.
Press, Oxford. GRUY, R. H. and KIELY, W. L., 1977. "The World's
GAYNOR, R. D., 1992. "Current Research at Largest Single Point Mooring Terminals: De-
NRMCA," Concrete Products, April. sign and Construction of the Salm System for
GAYTHWAITE, J. W., 1981. The Marine Environ- 750 000 DWT Tankers," Proceedings of Off-
ment and Structural Design. Van Nostrand shore Technology Conference, Houston, TX,
Reinhold Company, New York. May.
GAYTHWAITE, J. W., 1990. "Design of Marine Fa- GUEST, F. E. and PIZZARIELLO, A. M., 1989. "De-
cilities for the Berthing, Mooring and Repair signing Harbor and Waterway Channels Us-
of Vessels." Van Nostrand Reinhold Co., New ing Marine Simulators." PIANC Bulletin
York. No. 66.
GERWICK, B. C., 1975. "Durability of Structure HAATAINEN, P., 1978. "Experimental Investiga-
Under Water," Proceedings AIPC-PIP- tion on the Squat in Changing Water Depth
236 Design Principles and Considerations of Port Elements

Conditions," Ships Hydrodynamic Labora- eral Highway Administration, Washington,


tory, Report No. 14, Otaniemi, Finland. DC.
HANEMAN, W. M., 1994. "Valuing the Environ- HooFT, J. P., 1974. "The Behavior of a Ship in
ment through Contingent Valuation." Journal Head Waves at Restricted Water Depths." In-
of Economic Perspectives, Vol. 8, No.4. ternational Ship Building Progress, Vol. 21,
HANNS, A. N., 1977. "Steel Fiber Reinforced Con- No. 244.
crete Properties and Resurfacing Applica- HOPPE, J. J., 1992. "Air Loss in Free-Falling
tions," Portland Cement Association Bulletin Concrete." Concrete International, June.
RD049.01P.
HoRVICK, E. W., 1964. "Specifications and Speci-
HARVEY, H. H, 1980. "Permeability of Concrete fying." American Zinc Institute, Inc., Techni-
in Seawater," Proceedings of the Interna- cal Seminar, New York, May.
tional Conference on Performance of Concrete
in Marine Environments, St. Andrews-by-the- HuusKA, 0., 1976. "On the Evaluation of Under-
Sea, Canada, August. keel Clearances in Finnish Waterways," Ship
Hydrodynamics Laboratory Report No. 9,
HAsKELL, A. J., 1989. "Barges and Vessels for Otaniemi, Finland.
the Domestic Ocean Trade." PIANC Bulletin
No. 69. IBBETSON, C., 1987. "Pre-cooling Concrete with
Liquid Nitrogen." Civil Engineering, June.
HAWKEs, P. J., 1985. "HINDWAVE-A Wave
Hindcasting Method," Hydraulics Research INTERNATIONAL STANDARDS ORGANIZATION, 1983.
Report IT 288. "Roll-on/Roll-off Ship to Shore Connection,"
HEADLAND, J. R., 1995a. "Probabilistic Design of ISO 6812-1983, Geneva.
Navigation Channel Depths." ASCE Proceed- JOHNSON, B. R., 1987. Timber Bulkheads,
ings Specialty Conference PORTS'95, Tampa, Geotechnical Special Publication No.7, ASCE,
FL. New York.
HEADLAND, J. R., 1995b. "Floating Breakwaters," JONES, R. M., 1989. "Bulk Ships for the 1990s."
Marine Structures Engineering: Specialized PIANC Bulletin No. 66.
Applications. Chapman & Hall, New York. KHANNA, J. and SORENSEN, T., 1980. "Stand-Off
HEIKKILA, M., 1990. "The Evaluation of Manoeu- Force on Ships Moored in Strong Current,"
vrability through a Narrow Strait using American Society of Civil Engineers Proceed-
a Full-Bridge Shiphandling Simulator. ings of Specialty Conference PORTS'80,
MARSIM and ICSM'90, Tokyo. Norfolk, VA.
HENAGER, C. H., 1977. "Ultimate Strength of KrMoN, P. M., 1982. "Underkeel Clearance
Reinforced Steel Fibrous Concrete Beams," in Ports," Ship Transport Symposium,
Proceedings of Conference on Fiber- Rotterdam.
Reinforced Materials: Design and Engineer-
ing Applications, Institute of Civil Engineers, KNoTT, M. A. (ed.), 1989. Vessel Traffic Controls
London. Systems, ASCE Task Committee on Systems
for Controlling Vessel Traffic. Proceedings of
HENAGER, C. H. and DOHERTY, T. J., 1976. "Anal- Specialty Conference Ports'89. American Soci-
ysis of Reinforced Fibrous Concrete Beams." ety of Civil Engineers, New York.
Proceedings of ASCE, Vol. 12, Paper ST-1.
KOLEK, J., 1963. "Research on the Vibration of
HERBICH, J. B., 1992. "Dredged Navigation
Fresh Concrete," Proceedings of Confer-
Channels," Handbook of Coastal and Ocean
ence on Vibration-Compaction Techniques,
Engineering, Vol. 3, Herbich, J. B. (ed.), Gulf
Budapest.
Publishing House, Houston, TX.
HoFF, G. C., 1976-1982. "Selected Bibliography KRAY, C. J., 1970. "Supership Effect on Water-
on Fiber Reinforced Cement and Concrete," way Depth and Alignments." ASCE Journal
Miscellaneous Paper No. C-76-6 and Supple- of the Waterways, Harbors and Coastal Engi-
ments 1-4, U.S. Army Engineer Waterways neering Division, Vol. 96, No. WW2.
Experiment Station, Vicksburg. Also Chapter KRAY, C. J., 1972. "Closure to Supership Effect
9 Report No. 1 FHWA-RD-77-110, Vol. 2, Fed- on Waterway Depth and Alignments," ASCE
Design Principles and Considerations of Port Elements 237

Journal of Waterways, Harbors and Coastal MOSER, D. A., YOE, C., and HILL, D. J., 1995.
Engineering, Vol. 96, May. "Estimating the Economic Value of Risk Re-
KRAY, C. J., 1973. "Design of Ship Channels and duction from Deep Draft Channel Widening,"
Maneuvering Areas," ASCE Journal of the ASCE Proceedings Specialty Conference
Waterways, Harbors and Coastal Engineer- PORTS'95, Tampa, FL.
ing Division, Vol. 99, No. WWl. MUNRO, E., 1985. "Concrete Durability in a Free
LAFORGE, E. and VANTORRE, M., 1984. "Applica- Market System." Cement Imagination, Vol.
tion on Manoeuvering Simulation Techniques 17, No.3.
for Dimensioning Quay Wall Fenders," NATIONAL AsSOCIATION OF CORROSION ENGINEERS
PIANC Proceedings 28th Congress, Seville, (revised 1983). "Recommended Practice. Cor-
Spain. rosion Control of Steel, Fixed Offshore Plat-
LIBBY, J. R., 1984. Modern Prestressed Concrete. forms Associated with Petroleum Production,"
Van Nostrand Reinhold Co., New York. Houston, TX.
LORENZ, L. F. and GJOVIK, L. R., 1972. Analyzing NAVFAC, 1980-1986. Piers and Wharves. De-
Creosote by Gas Chromatography: Relation- sign Manual 25.1, Department of the Navy,
ship to Creosote Specifications. AWPA 1972 Naval Facilities Engineering Command,
Proceedings, Vol. 68. NAVFAC DM-25.1 (1980). Alexandria, VA.
Additional Parts:
MALHOTRA, V. M., 1986. "Mechanical Properties DM 25.2 "Dockside Utilities for Ship Service"
and Freeze/Thaw Resistance of Condensed DM 25.3 "Cargo Handling Facilities" (1981)
Silica Fume Concrete," American Concrete DM 25.4 "Seawalls, Bulkheads and Quay
Institute ACI SP-91, Vol. 2. Walls" (1981)
McGEE, S. E. and VANN, R. G., 1984. "Determin- DM 25.5 "Ferry Terminals & Small Craft
ing Channel Design Requirements for Norfolk Berthing Facilities" (1981)
Harbor and Channels Deepening Project." The DM 25.6 "General Criteria for Waterfront
Dock and Harbor Authority, July. Construction"
McNOWN, J. S., 1976. "Sinkage and Resistance "Harbors," Design Manual 26.1, Depart-
for Ships in Channels." ASCE Journal of the ment of the Navy, Naval Facilities Engineer-
Waterways, Ports, Coastal and Ocean Divi- ing Command, NAVFAC Facilities Engineer-
sion, Vol. 102, August. ing Command, NAVFAC.
MEHTA, P. K., 1980a. "Durability of Concrete in DM 26.1 Alexandria, VA. (1981). Additional
Marine Environments. A Review." Proceed- Parts:
ings of the International Conference on Per- DM 26.2 "Coastal Protection" (1982)
formance of Concrete in Marine Environ- DM 26.3 "Coastal Sedimentation and Dredg-
ments, St. Andrews, Canada, August. ing"
DM 26.4 "Fixed Moorings" (1986)
MEHTA, P. K., 1980b. "Durability of Concrete in DM 26.5 "Fleet Moorings" (1985)
Marine Environments," American Concrete DM 26.6 "Mooring Design Physical and Em-
Institute, ACI SP-65, Detroit. pirical Data" (1986).
MEHTA, P. K., 1988. "Durability of Concrete Ex- NAVFAC DM-26.1, 1981. Harbors, Design Man-
posed to Marine Environment-A Fresh ual DM-26.1. Department of the Navy, Naval
Look," Proc. 2nd International Conference Facilities Command, Alexandria, VA.
on Concrete in Marine Environment, St.
OLIVER, J. G., 1992. "Floating Breakwaters,"
Andrews-by-the-Sea, Canada.
ASCE Specialty Conference PORTS'92,
MEHTA, P. K. and GERWICK, B. C., 1982. "Crack- Seattle, Washington.
ing-Corrosion Interaction in the Concrete Ex-
OSHIRO, T. and TANIGAWA, S., 1988. "Effect of
posed to Marine Environments." Concrete
Surface Coatings on the Durability of Con-
International, October.
crete Exposed to Marine Environment," Pro-
MINIKIN, R. R., 1961. Winds, Waves and Mar- ceedings of 2nd International Conference
itime Structures, 2nd ed. Charles Griffin and on Concrete in Marine Environment, St.
Co. Ltd., London. Andrews-by-the-Sea, Canada.
238 Design Principles and Considerations of Port Elements

OWEN, M. W. and ALLSOP, N. W. H., 1983. "Hy- PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
draulic Modeling of Rubble-Mound Break- GATION CONGRESSES, 1989. "Economic Meth-
waters," Proceedings of Conference on Break- ods of Channel Maintenance," Supplement to
waters Design and Construction, Institution Bulletin No. 67.
of Civil Engineers, London. PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
PEABODY, A. W., 1967. "Control of Pipeline Cor- GATION CONGRESSES, 1992a. "Capability of
rosion." NACE, December. Ship Manoeuvring Simulation Models for Ap-
PEEL, J., 1982. "Corrosion in Concrete-A proach Channels and Fairways in Harbours."
'Chemical' Solution." Engineering Digest, Vol. Supplement to Bulletin No. 77.
2, No. 10. PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- GATION CONGRESSES, 1992b. "Guidelines for
GATION CONGRESSES, 1978. Report of the In- . the Design and Construction of Flexible
ternational Commission for the Standardiza- Revetments Incorporating Geotextiles in Ma-
tion ofRoll-oniRoll-offShips and their Berths. rine Environment." Supplement to Bulletin
78179.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
GATION CONGRESSES, 1979. Report of Working PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
Group I. Supplement to Bulletin No. 32. GATION CONGRESSES, 1994. "A Practical Guide
for Design and Construction of . Floating
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
Breakwaters."
GATION CONGRESSES, 1980. Report of Working
Group IV "Reception of Large Ships." Supple- PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
ment to Bulletin No. 35. GATION CONGRESSES, 1995. "Approach Chan-
nels. Preliminary Guidelines." Supplement to
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
Bulletin No. 87.
GATION CoNGRESSES, 19821 1983. "Naviga-
tion in Muddy Areas." Supplement to Bulletin PETROSKI, H., 1984. The Engineer is Human. The
No. 73. Role of Failure in Successful Design, St.
Martin's Press, New York, N.Y.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
GATION CONGRESSES, 1984a. Report of the In- PHRI, 1980. Technical Standards for Port and
ternational Commission for Improving the Harbor Facilities in Japan. The Overseas
Design of Fender Systems. Supplement to Coast Area Dev. Inst. of Japan, Bureau of
Bulletin No. 45. Ports and Harbors, Ministry of Transport,
Port and Harbor Research Institute, Tokyo.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
GATION CONGRESSES, 1984b. "Classification of PORT AUTONOME DE MARSEILLE, 1976. "Mesures
Soils and Rocks to be Dredged." Supplement de Squat en vraie Gradeur," Rupport non
to Bulletin No. 47. public.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- QUINN, A. DEF., 1972. Design and Construction
GATION CONGRESSES, 1985a. "Underkeel of Ports and Marine Structures, 2nd ed.
Clearance for Large Ships in Maritime Fair- McGraw-Hill Book Company, New York.
ways with Hard Bottom," Report of a Work- RAMAKRISHNAN, V., 1988. "Materials and Proper-
ing Group of the Permanent Technical Com- ties of Fiber Reinforced Concrete." Civil Engi-
mittee II. Supplement to Bulletin No. 51. neering, April.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- RAsHEEDUZZAFAR et al., 1992. "Performance of
GATION CONGRESSES, 1985b. "Dangerous Goods Corrosion Resisting Steels in Chloride-
in Ports Recommendations for Port Designers Bearing Concrete." Concrete International,
and Port Operators," Report of a Working September-October.
Group of the Permanent Technical Commit- READ, J. A., 1987. "Rebar Corrosion-Coating
tee II. Supplement to Bulletin No. 49. a Cure?'' Civil Engineering, October 1
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI- November.
GATION CONGRESSES, 1987. "Development of REED, W. B., 1980. "Delivering Oil Where It's
Modern Marine Terminals." Supplement to Needed," Proceedings ASCE Specialty Confer-
Bulletin No. 56. ence PORTS'80, Norfolk, VA.
Design Principles and Considerations of Port Elements 239

RIBADEAU-DUMAs, L., 1982. "Squat," Revue Tech- SOVIET STANDARDS


nique des Phares et Balises, No. 53, March. 1-Inland waterway hydraulic structures
RosEMAN, D.P. and JAKOBSEN, B. K., 1986. "Port SNiP II-50-74 (1974)
Design and Ship Characteristics," ASCE Pro- 2-Seaport hydraulic structures SNiP
ceedings Specialty Conference PORT'86, II-51-74 (1974).
Oakland, CA. 3-Wave, ice and ship loadings on hy-
SCHIJF, J. B., 1949. Proceedings Permanent In- draulic structures SNiP II-57-75 (1975)
ternational Association Navigation Con- Stroyisdat Publisher, Moscow.
gresses, 17th Congress, Section 1, Communi- SOHNGEN, B., 1995. "The Effects of Wind on the
cation 2. Lisboa (in French). Design of Canals," PIANC, Bulletin No. 88.
SCHUPACK, M., 1982. "Design of Permanent Sea- SPAN, H. J. TH., et al., 1981. "A Review of Rele-
water Structures to Prevent Deterioration." vant Hydraulic Phenomena and of Recent De-
Concrete International, March. velopments in Research, Design and Con-
SEYMOUR, R. J., 1977. "Estimating Wave Gener- struction of Protective Works," PIANC Pro-
ating on Restricted Fetches." ASCE Journal ceedings 25th Congress, Edinburgh,
of the Waterway, Port, Coastal and Ocean Scotland.
Division, Vol. 103, May. SWENDSEN, I. A and JENSEN, V., 1970. "The
SHAH, S. P. and BATSON, G. B., (eds.), 1987. Form and Dimensions of Fender Front Struc-
"Fiber Reinforced Concrete-Properties and tures." The Dock and Harbour Authority,
Applications," ACI Sp-105, Detriot. June.
SHAH, A P. and SKARENDAHL, A (ed.), 1986. Steel TEDESCO, A, 1994. "Computer Analysis of No
Fiber Concrete. Elsevier Applied Science Pub- Substitute for Experience." Civil Engineering,
lishers, London. February.
SHARMA, A K., 1986. "Shear Strength of Steel
TERMPOL CoDE, 2nd ed., 1983. Canadian Coast
Fiber Reinforced Concrete Beams." American
Guard, Ottowa.
Concrete Institute Journal, Vol. 83, No.4.
SINGLETON, W. T., Jr., 1970. "Performance of THE DOCK AND HARBOUR AUTHORITY, 1982a. A
Various Coating Systems in a Marine Envi- Wide Choice of Fender Designs, March.
ronment." Materials Protection and Perfor- THE DocK AND HARBOUR AUTHORITY, 1982b. Ex-
mance, Vol. 9, No. 11. panding Range of Fender Designs.
SMALLMAN, J. V., 1986. "The Use of Physical and THEVENOT, M. M., 1992. "Evaluation of Width
Computational Models in the Hydraulic De- Criteria for Deep-Draft Channels Using Com-
sign of Coastal Harbours," PIANC Bulletin puter Simulation," ASCE Proceedings Spe-
No. 53. cialty Conference PORTS'92, Seattle, WA
SMIRNOV, G. N., GORUNOV, B. F., KURLOVICH, THIELE, A, 1901. Shiffswiderstand ouf Kanalen-
E. B., LEVACHEV, C. H., and SIDOROVA, A G., Centralblatt der Bauver-Waltung, No. 56.
1979. Ports and Port Structures (in Russian).
THORESEN, C. A, 1988. Port Design. Tapir Pub-
Stroyisdat Publishing Co., Moscow.
lishers, Trondheim, Norway.
SORENSEN, R. M., 1966. "Ship Waves," Technical
Report HEL-12-2, Hydraulics Engineer- TREADAWAY, K. W. J., Cox, R. N., and BROWN,
ing Laboratory, University of California, B. L., 1989. "Durability of Corrosion Resisting
Berkeley. Steels in Concrete," Proceedings of Institute
of Civil Engineers, Part 1, London, April.
SORENSEN, R. M., 1977. "Procedures for Prelimi-
nary Analysis of Tidal Inlet Hydraulics and TRUJENS, C. L., WAGENAAR, W. A, and VANWIJN,
Stability," Coastal Engineering Technical Aid 1980. "The Effect of Contracted Timescales on
No. 77-8. U.S. Army Corps of Engineers, the Learning Ability for Manoeuvring of Large
Coastal Engineering Research Center, Fort Ships," Netherlands Ship Research Centre,
Belvoir, VA TNO, Report 215, December.
SoRos, P., KoMAN, B., and LUBETKIN, M., 1986. TSINKER, G. P., 1983. "Anchored Sheet-Pile
"Innovative Bulk Ports in Latin America." Bulkheads: Design Practice." ASCE Journal
The Dock and Harbour Authority, December. of Geotechnical Engineering, Vol. 109, No.8.
240 Design Principles and Considerations of Port Elements

TSINKER, G. P., 1986. Floating Ports. Design and U.S. NATIONAL RESEARCH COUNCIL (MARINE
Construction Practices. Gulf Publishing Com- BoARD), 1981. "Problems and Opportunities
pany, Houston, TX. in the Design of Entrance to Port and Har-
TSINKER, G. P., 1995. Marine Structures Engi- bors," Panel Report, Wiegel, R. R., Chairman,
neering: Specialized Applications, Chapman National Academy Press, Washington, DC.
& Hall, New York. VAN DE KAA, E. J., 1977. "Discussion of Sinkage
and Resistance for Ships in Channels." ASCE
TucK, E. 0., 1965. "Shallow Water Flows Past
Journal of the Waterways, Ports, Coastal and
Slender Obstacles." Report by David Taylor Ocean Division, Vol. 103, August.
Model Basin, Washington, DC, December.
VAN DE KAA, E. J., 1984. "Safety Criteria for
TuCK, E. 0. and TAYLOR, J. P., 1972. "Shallow Channel Depth Design," Third Meeting of the
Wave Problem in Ship Hydrodynamics." Uni- U.S.-Dutch M.O.U. on Dredging and Related
versity of Adelaide, Australia. Technologies, Charleston, SC.
TuRNER, F. H., 1974. "The Contranstor Concept VANDERMEER, J. W. and STAM, C.-J. M., 1992.
for Cryogenic Carriers." Ocean Industry, "Wave Runup on Smooth and Rock Slopes of
November. Coastal Structures." ASCE Journal of Water-
TuRNER, H. 0. Jr., 1984. "Dimensions for Safe ways, Port, Coastal and Ocean Engineering,
and Efficient Deep-Draft Navigation Chan- Vol. 118, No.5.
nels," Technical Report HL-84-10, US Army VAN WIJK, A. C., 1982. "Wave Induced Ship
Engineer Waterways Experiment Station, Motions in Harbour Entrances, a Field Study,"
Vicksburg, MS. Proceedings I.C.C.E., Cape Town.
TuTHILL, L. H., 1988. "Lasting Concrete in a VICKERMAN, M. J., 1992. "The Influence of
Sulfate Environment." Concrete Interna- Future Vessel Design on Ports," ASCE Pro-
tional, December. ceedings Specialty Conference PORTS'92.
VINCENT, G., 1989. "Breakwater Choices." Civil
UNITED KINGDOM'S NATIONAL MARITIME INSTITUTE Engineering, July.
(NMI), 1981, London.
VLEMMIX, W. et al., 1985. "Reception and Dis-
U.K DEPARTMENT OF TRANSPORT, 1982. Ship Be- charge of Large Coal Carriers," Permanent
havior in Ports and Their Approaches. U.K International Association of Navigation Con-
Department of Transport, London. gresses, Proceedings of 26th Congress, Sec-
UNITED NATIONS CONFERENCE ON TRADE AND DE- tion II, Subject I, Brussels.
VELOPMENT (UNCTAD), 1973. Berth Through- WADE, M. J. ET AL., 1987. Summary Evaluation
put, Systematic Method for Improving Gen- of the Environmental Impact Resulting from
eral Cargo Operators. United Nations, New the Use of Creosoted Pilings in the Historic
York. Restoration of Pier No. 2 at the Charlestown
UNITED NATIONS CONFERENCE ON TRADE AND DE- Navy Yard. Final Report to U.S. Department
VELOPMENT (UNCTAD), 1978. Port Develop- of Interior, National Park Service, Boston,
ment, A Handbook for Planners in Develop- MA.
ing Countries, United Nations, New York. WADWA, L. C., 1992. "Planning Operations of
Bulk Loading Terminals by Simulation."
U.S. ARMY CORPS OF ENGINEERS, 1983. "Hy-
ASCE Journal of Waterway, Port, Coastal and
draulic Design of Deep-Draft Navigation Pro- Ocean Engineering, Vol. 118, No.3.
jects," EM110-2-1613 Headquarters Depart-
WAGENAAR, W. A. and MICKON, J. A., 1968. "The
ment of the Army, Washington, DC.
Effect of Contracted Timescales in Scale
U.S. ARMY CORPS OF ENGINEERS, 1984. "Shore Model Manoeuvering," Institute of Percep-
Protection Manual" Volumes 1 and 2. Depart- tion, TNO, Report, IZF, C3, Soesterberg, The
ment of the Army, Washington, DC. Netherlands.
U.S. NATIONAL RESEARCH COUNCIL, 1992. WEBB, D. A. and· GJOVIK, L. R., 1988. Treated
"Shiphandling Simulation-Application to Wood Products, Their Effect on the Environ-
Waterway Design," Washington, DC., Na- ment. American Wood-Preservers' Associa-
tional Academy Press. tion.
Design Principles and Considerations of Port Elements 241

WEBB, J. C., 1982. "The LOOP Project: Taking WINDLER, F. J., 1983. "The Corrosion Survey
the Port to the Ship." ASCE Civil Engineer- -An Engineer's Guide to Maintenance Paint-
ing, June. ing." Material Performance, Vol. 22, No.5.
WEBB, D. W., 1995. "San Juan Harbor Naviga- WooD, J. S., KHANNA, J., and DEMBINSKI, M.,
tion Channel Design", ASCE Proceedings 1981. "Design of an Offshore Berth for Moor-
Specialty Conference PORTS'95, Tampa, FL. ing Large Ships in a Strong Current," Pro-
WEIS, J. M. and BLANCATO, V., 1959. "Breasting ceedings of Permanent International Associa-
Dolphin for Berthing Supertankers." ASCE tion of Navigation of Congresses, 26th
Journal of the Waterways and Harbours. Congress, Edinburgh.
WESTERN-EUROPEAN INSTITUTE FOR WOOD- WRIGHT, J., CHILD, T. J., REBEL, J. M., and
PRESERVATION, 1987. Minutes of the 37th ROBERTSON, J., 1987. "Port Bonython Jetty,
Meeting of the General Assembly, Portschach, South Australia," Proceedings Institution of
Austria. Civil Engineers, October.
WHITENECK, L. L. and HOCKNEY, L. A., 1989. YARON, S. L. and SHIMON!, J., 1982. "The Hadera
Structural Materials for Harbor and Coastal Offshore Coal Unloading Terminal: A New
Construction. McGraw-Hill Book Company, Approach to Unloading of Bulk in the Open
New York. Sea," Proceedings of Offshore Technology
WICKER, C. F. (ed.), 1965. "Evaluation of Present Conference, Houston, TX, May.
State of Knowledge of Factors Affecting Tidal YINGLING, J., MULLINGS, G. M., and GAINOR,
Hydraulics and Related Phenomena," Report R. D., 1992. "Loss of Air Content in Pumped
No. 3 Committee on Tidal Hydraulics, U.S. Concrete." Concrete International, October.
Army Corps of Engineers. YOUNDALE, J. E. and SHRIVASTAVA, S.M., 1986.
WILLIAMSON, G. R., 1978. "Steel Fibers as Web "Design and Construction of Marine Facilities
Reinforcement in Reinforced Concrete," Pro- for Ridley Island Coal Terminal," Proceedings
ceedings of U.S. Army Science Conference, ASCE Specialty Conference PORTS'86,
Vol. 3. Oakland, CA.
3
Design Loads

3.1 GENERAL with the following:

The purpose of this chapter is to provide • Ship docking and mooring operations, in-
the port designer with guidelines on loads cluding berthing impact, and mooring
and forces to be used in the design of port forces due to winds, waves and current
related marine structures. acting on the vessel
All applied design loads and forces which • Vertical and longitudinal friction forces
should be considered in marine structures between the ship and fenders due to
tidal action, change in ship draft due to
design may be defined in the following basic
loading/unloading, and the ship rubbing
categories: along the structure during berthing ma-
neuvering
• Loads and forces from the water side of • Loads and forces due to the operation of
the structure cargo handling and hauling equipment
• Loads from the land side of the structure such as fixed, rail-mounted, and mobile
• Loads and forces related to dock operation cranes, dry bulk loadingjunloading equip-
• .Load due to the temperature effect ment, conveyers, miscellaneous tire and
crawler track-mounted equipment, trucks
Waterside related loads and forces are and other vehicles, liquid· cargo handling
basically environmental forces such as wind, arms and piping, and others
waves, current, and ice. Land-side related • Uniform distributed loads of a miscella-
neous nature
loads are typically the weight of the struc-
ture, the vertical and horizontal compo-
nents of soil pressure, and the hydrostatic Loads due to temperature effects may
pressure from the groundwater behind the have a substantial impact on the spacing
structure. between expansion joints. In general, these
Loads and forces related to the operation loads may generate considerable horizontal
of the dock are basically those associated loadings on marine structures and be espe-

243
244 Design Loads

cially detrimental when applied to some Safe operation of any structure depends
specific structural components. on the loadings subsequent to completion of
Additionally, loads acting on a marine construction. Because the structure is usu-
structure can be classified as permanent, ally loaded after it is built and not always
temporary, and special. as designed, the selection of design loads is
a problem of statistics and assessment of
Permanent loads include the weight of the probability. In some cases, neither the
structure along with the weights of ware- owner nor the designer is sure what maxi-
houses, sheds, and administrative offices all mum loading the structure may be required
permanently installed on the marine struc- to sustain in the course of its useful life. In
ture, and also the weight of permanently in-
addition, the selected loading is often sub-
stalled fixed cargo handling equipment. In
further discussions, this kind of load will be
jective and largely depends on the designer's
referred to as dead load. (Note that lateral experience. To avoid divergent designs and
soil pressure may also be defined as a perma- to formulate reasonable design loads and
nent load.) load combinations, the designer should fol-
Temporary loads and forces include those due low relevant guidelines, codes, and stand-
to the environment and dock operation. The ards, but also exercise common sense. Rea-
latter includes all kinds of live loads sonable design loads are those that do not
(surcharge, moving equipment), ship berthing result in either prohibitive cost of structure
forces, and mooring forces. construction nor in structure obsolescence
Special loads include accidental loads, earth- shortly after commissioning. The designer
quake forces and other unusual loads. also should exercise great care in formulat-
ing design loads, particularly those gener-
As with the design of conventional struc- ated by mobile cargo handling and hauling
tures, the loads, and their combinations, equipment.
used in marine structure design are distin-
guished as normal and extreme.

Normal load combinations are those that 3_2 ENVIRONMENTAL LOADS


could exist under normal dock operations. The
most severe, realistically expected combina- Wind, waves, current, and ice provide the
tions of dead load and soil pressure, ship principal environmental loads on marine
impact and mooring forces, loads generated
structures. Typically, environmental design
by cargo handling and hauling equipment,
and appropriate surcharge loads are usually
loads are based on storm conditions having
selected as the normal design load combina- an average expected recurrence interval of
tion for a given structure. To proportion the 50 years; some important marine structures
structural elements exposed to normal load may be designed on the basis of an expected
combinations, normal allowable stresses are event recurrence of up to 100 years.
used. Because most of the port related marine
facilities are located in areas sheltered from
The normal loading, in combination with the action of sizable waves and strong cur-
unusual forces such as an earthquake, an rents, the direct action of these environ-
accidental load of miscellaneous nature or mental agents against marine structures
operation under a damaged condition, is are, in general, negligible. Under typical
considered to be an extreme load condition. design conditions, berthing and mooring
In such cases, allowable stresses used in loads and lateral soil thrust are, as a rule,
design would be appropriately increased the dominant horizontal forces acting on
(see "Load Combinations," Section 3.8). a berthing structure. However, even in
Design Loads 245

protected harbors the traveling vessels may Table 3-1. Value offactor C 1
generate substantial waves, the effect of
Length of the Ship
which could be considerable on moored
ships. ::;; 25m 50m lOOm ~200m

However, at exposed locations (i.e., 1 0.8 0.65 0.5


breakwaters, offshore terminals, or some
terminals located in inland waterways) the
effects of wave and current forces must be where
studied closely. Pw = total wind force on the dock structure in
In countries with a cold climate, ice rep- kilonewtons.
resents a major environmental load on dock k = 1.3, shape factor, which considers suction
structures. The ice may impose lateral increase on the leeward side.
thrusts against a structure, produce uplift cl = coefficient that considers the length of the
and pile jacking loads, increase the weight ship; the recommended values of C 1 are
of the structure, and abrade the structural given in Table 3-1.
material. C2 = gust factor; average gust-factor value range
1.25-1.45.

Note: when selecting a design value of C2 the


3.2.1 Wind size of the ship is to be considered. Because
the gust duration may be insufficient to de-
velop a full mooring force or maximum fender
The magnitude and strength of a wind load strain on a large ship (the "inerta effect"),
acting on a dock structure depends on the smaller values of C2 are usually used when
velocity of the prevailing wind, dock orien- dealing with large ships. Fang (1979) has re-
tation, the amount of structure exposed to ported on test results to verify wind and cur-
wind, and the size of a ship moored along- rent loads on tankers 170,000-280,000 DWT.
side the dock structure. In general, the wind It has been found that 20 s of sustained gust
force that acts on a ship is proportional to can fully excite the moored very large crude
the square of the wind velocity. The wind carrier (VLCC), whereas approximately a 1-
min gust duration is required to induce the
effect, which by its nature is dynamic, is
maximum ship response.
usually decomposed into a static action (e.g.,
no variation in wind speed) and a dynamic LAx and LAy
action (e.g., gust) is covered by a special = exposed to wind areas of the ship, cargo

coefficient(s). Two components of wind forces handling and hauling equipment located on
the structure, and the dock structure area
are usually considered: one that acts per-
(in m 2 ) above water line exposed to wind
pendicularly to the dock structure and the respectively in x and y directions. If the
other that acts parallel to it. The maximum dock structure is occupied by ships on both
wind force is usually obtained when the sides, the sheltering effect of the windward
wind is blowing at 90° to the ship's center- ship is taken into account. The wind force
line. on the sheltered vessel is usually assumed
Total wind force, Pw, acting on a ship to be 50% of the value of an unsheltered
alongside a structure may be expressed as vessel had it not been sheltered against the
follows: wind.

Note: The maximum wind load on a dock


Pw = kpwC1C2( LAx sin 2¢ + LAY <f>)
cos 2 structure will be when the ship is in a light
condition (maximum exposed to wind area). If
(3-1)
accurate wind area information for a specific
246 Design Loads

design ship is not available, then approxi- water surface (V10 ), obtained from available
mate values obtained from Figures 3-1 and wind records. If the wind is recorded at an
3-2 may be used for the preliminary design. elevation other than 10 m, the wind speed
cf> = angle of wind direction to ship centerline. at the 10-m level is determined using
Pw = specific wind pressure (in kPa), which
varies with the square of the sustained vw = V10 = ( 1o;z) 1/7 vz (3-3)
wind velocity. The sustained wind velocity
is the average speed over 1 min. The value where
of Pw may be obtained from V 10 = wind speed at an elevation 10 m over
surface of water
Pw = 4.74 X 10- 5 v.,7 (3-2) Z = elevation of recorded wind (in m)
Vz = recorded wind speed at elevation Z.
where Vw =velocity of the wind (in kmjh).
For fetch length, that is, the horizontal ex-
The value of Pw varies from place to place tension of the wave generating area, less
and usually is specified by local building than or equal to 15 km, the wind speed
codes. Wind speed is usually assumed to be recorded over land (VL) should be adjusted
the sustained wind speed 10 m above the to obtain the over-water wind speed (Vw)

300

I
v
I
1/
J
200
I
I
v
""'cc I
~
0

I
.<:
Cl
·;;
,." I
I
"
0 100 J
I
1/
v
v
0
0
r-- -- ./"

2 3 4
Exposed to wind a rea
6 7 8

Figure 3-1. Approximate projected (vessel in ballast condition) to wind area as function of
tankerDWT.
Design Loads 24 7

using For preliminary design Udovichenko and


(3-4)
Yakovlev (1976) recommended the following
equation to determine the ship's exposed to
If the fetch is longer than 15 km, the ratio wind area, A, when she is in a light condi-
R = VwfVL can be obtained from Figure tion:
3-3.
More information on wind speed and A= aL~ (3-5)
wind forces calculations can be found in
NAVFAC DM-26.2 (1982), Gaythwaite where
(1981, 1990), Headland (1995), Bruun A = exposed to wind area of a ship m light
(1989), Sohngen (1995) and PIANC (1995). condition (m 2 ).

50

L
v
/
..,
----.
40

/
v
0
X
I /
v
30

I v
Ill
Cll

~
t::
t::

~
I II"
~ /
v
~
en
·a;

I
20

..
3::
"0
Cll
c

10 l "/ /
/
.,V
0
0 2 3 4 5 6
Ship exposed to wind area (m 2 x 10 3 )

Figure 3-2. Approximate projected to wind area as a function of vessel capacity: 1. general cargo
(light condition); 2. container ship.

Table 3-2. Coefficient a in eq. (3-2)

Length (L 8 )
300m
Type ofVessel lOOm 150m 200m 250m and More
Mixed general
cargo /passenger 0.12 0.12 0.12 0.11 0.11
General cargo 0.11 0.10 0.09 0.08
Tankers 0.11 0.10 0.09 0.08 0.07
248 Design Loads

L 8 =maximum length of a ship (in m). ues of wind load presented in Figure 3-4
a = empirical coefficient which takes into ac- are obtained from miscellaneous sources
count the ship's shape and length. The rec- and should be considered as rather conser-
ommended values of a are given in Table vative. The wind load is transmitted to the
3-2. quay structure via the ship's mooring lines,
Additionally, it should be recognized that the or fender units effectively in contact with a
area of a ship exposed to the wind is not sym- ship.
metrical about the midship line; the latter may In some cases, the berth could be partly
help to develop a moment of rotation about the shielded against the seaward wind by exist-
ship's center of gravity. ing structures, such as warehouses, office
buildings, or by specially constructed wind-
In designing the dock structure for wind protective structures. In the past, such
forces, it is usually assumed that cargo han- structures have been effectively used in
dling equipment such as cranes and loading ports at Le Havre, Marseilles, Dover, and
towers would not operate with a wind other locations (Minikin, 1963).
stronger than 25 kmjh. It is also customary
to consider that a ship would not remain
alongside the dock in light condition in a
3.2.2 currents
severe storm; she would either take on bal-
last or leave for sea. (pw = 0.25 kPa is usu- In the design of marine structures there
ally considered adequate for the dock struc- are three major kinds of currents to be
ture.) In designing the dock, use of unit considered:
wind pressure greater than lkPa should be
justified by special requirements.
1. River currents
For the preliminary design, values of de-
sign wind load (kN jper linear meter of ship) 2. Tidal currents
can be obtained from Figure 3-4. The val- 3. Wind-induced current in harbors

~~~~~~~~~~~ti~~~~~~~F~ORii~~>~3~6~KNO~)TS~
R=0.9
1.0
WJNOSPEEDS ARE REFERENCED
TO 10-METER LEVEL
0.8
0 5 10 15 20 25 30 35 40KNOTS

0 10 45 .._ES PER HOUR

OVERLAND WINOSPEED. VL

Figure 3-3. Ratio (R) of over-water (Vw) to over-land (VL) wind speed as a function
of over-land wind speed. [After Resio and Vincent (1976).]
Design Loads 249

100

90 I
.5!
........
80 I
70 I
1/
><

••cc 80
0

-• J
..!:,.

c
E
•...
50
v
v
40 ~

'A

'D 30
0.
:;; /
fn
20 /
/"
10 ,..._... .... .J

0
0
- 1---
10 20 30
Wind load (kN per lin. m of ship)

Figure 3-4. Approximate wind load per linear meter of ship.

River and tidal currents are usually se- ships and larger values for the dock
lected from available statistical data, structure. These values of care conserva-
whereas the wind-induced currents in still tive and are recommended for the prelim-
water are usually taken as equal to 1% of inary design stages only. The true value
the velocity of sustained wind at 10 m above of c may be obtained from model tests of
the surface of the w~ter. Typically, current the design ship and dock structure.
velocity decreases slowly with depth, and Vc =velocity of the current (in mjs).
current load (drag) on a dock structure de- E A = exposed-to-current areas of the fully laden
pends on the square of the current velocity ship and the structure system (in m 2 ).
and the shape and magnitude of the
exposed-to-current areas of both the
fully laden ship and the submerged part of In a strong current, a dock structure is
the structure. usually placed parallel to the stream. The
The force of the current exerted on the design current force is typically based on a
ship/structure system in both longitudinal current having an expected average recur-
and transversal directions (Pc) (in kN) can rence interval of 5 years.
be obtained from Current force acting on the vessel is usu-
ally transmitted to the dock structure
through the mooring lines and very seldom
Pc = c "'AV
L....
2 (3-6)
c
by the ship bearing against the fender sys-
tem. For more information on the direct
where quasi-static curre:n:t forces imposed against
c = empirical drag coefficient; c = 0.5 to 1.0 ship/dock structure system, the reader is
(kN s 2 jm4 ). Smaller values are used for referred to the British Standard Code of
250 Design Loads

Practice for Maritime Structures, Part 1 velocities of 5 and 7 knots, respectively, and
(1984), NAVFAC (1968, 1982), Palo (1983), dependent on the berth location. The ap-
Gaythwaite (1981, 1990), Bruun (1989), proximate one-third difference in C 8 value
Headland (1995), and PIANC (1995). measured in the above experiments and
Newman (1977) provides an in-depth those measured at lmmingham Oil Termi-
treatment of hydrodynamic current forces nal were mainly attributed to the different
on ships, where, under certain site condi- area densities of piling. Specifically, at Im-
tions, strong current may produce substan- mingham it was about 3% according to
tial standoff forces exerted on the vessel. In Khanna and Sorensen, and in the Wood et
1968, serious problems associated with a al. experiments it was equal to about 1% of
strong tidal current (velocity of about 5.5 the structure's effective area. However, in
knots) was experienced during construction the latter case, a 16% increase in pile den-
and operation of the Immingham Oil Termi- sity did not alter standoff forces signifi-
nal for 100,000-DWT tankers on the Hum- cantly. More details on the effects of pile
ber River in the United Kingdom (Jackson, density and configuration on standoff forces
1973). One problem that appeared was vio- is found in the work of Ball and Wilcock
lent oscillation of unbraced piles across the (1981), and PIANC (1995).
flow. Even when restrained by bracings or a In general, the reduction in the depth/
superstructure, the piles were not immune draft ratio from 1.5 to 1.32 resulted in about
to some vibration and extensive additional a 35% increase in standoff force. The mo-
bracing was subsequently added in order to tion of the moored ship usually started when
ensure safe terminal operation. the flow of water exceeded a threshold ve-
Another problem observed, during a high locity of approximately 2 knots. Under such
current the moored ship was pushed off the conditions, substantial dynamic mooring
berth by considerable standoff force that line loads (two to three times that due to
was built up between the ship and the pier the standoff force) are encountered are
structure. The cause of this force attributed attributed to strong surge-yaw and sway
to the effect of slowing down water passing motions.
through the pier piling, resulting in a pres- A phenomenon similar to that at lm-
sure head by raising the water level. The mingham Oil Terminal can be experienced
increased head of water in the area of the by a ship moored alongside a berth struc-
piles deflected the current outward, with ture located in close proximity to the river-
the result that there was a flow across the bank or in a narrow tidal estuary with
berth line. As a result, a ship moored to a strong currents. In this case, the acceler-
piled berth structure was pushed away from ated flow of water around one side of the
the berth by a combination of the differen- ship may result in differential head (a hy-
tial head and the angular current (Fig. 3-5). draulic gradient across the ship) pushing
The differential head across the moored ship the ship away from the bank.
(tl.h) was found to be equal to tl.h = Therefore, at terminal sites with strong
c.v 2 j2g, where the coefficient c.= 0.42, current and where a potential for standoff
Vis the undisturbed current (flow) velocity, forces exist, the designer should carefully
and g is gravity acceleration. investigate the problem, as no typical de-
Khanna and Sorensen (1980) and Wood sign solution or theory exists in practice at
et al. (1980) reported on model test results this time.
carried out on a 1 : 100 scale model to study In conclusion it should be noted that an
the standoff effect phenomenon. They found essential parameter to be considered in cal-
the value of coefficient c. to range from 0.1 culating the force of current, besides the
to 0.15 and 0.07 to 0.12 for scaled current current velocity, is the underkeel clearance,
Design Loads 251

) I\
Flow

Pressure
• • • •
• • • •
1- _jA
• • • •
• • • •
• • • •
PLAN

!
... h = c.v 2 /2g

- --
-
i

Figure 3-5. Standoff force on moored ship.

i.e., the vertical distance from the bottom of to direct wave forces of significant magni-
the ship to the bed of the basin. According tude.
to PIANC (1995), if the underkeel clearance Goda (1985) suggested that any harbor
is small, the current force can be up to six where a significant wave, H 8 , is less than
times the value of that in deep water. approximately 1.0 m, even under design
storm conditions, can be considered a calm
harbor; he offers the following criteria for
3.2.3 waves harbor tranquility:

As was stated earlier in this chapter, most 1. The harbor should have a broad interior.
marine terminals are built at protected lo- 2. Any portion of the harbor that can be
cations and therefore they are not exposed viewed through the entrance should have
252 Design Loads

a natural beach or wave-absorbing struc- are close to the natural periods of surge,
ture. sway, and yaw for medium and large ships,
3. Small craft should not be viewable from which, combined with usually low damping
the open ocean at any angle. of the ship mooring system, may result in
4. A portion of the waterfront should be wave resonant phenomena (Jensen and Warren,
dissipating in character. 1986; Stammers and Wennink, 1977;
5. Wave reflection from the back side and Viggosson, 1988).
the use of energy-absorbing quay walls Harbor protection from long waves will
such as perforated or slit caissons should require construction of a very long break-
be considered. water, that in most practical cases is uneco-
nomic. Instead, careful design of the berth
In protected harbors, berthing and moor- mooring system and correct positioning of
ing forces typically are much more severe the berth with the harbor is needed to mini-
than those which can be generated by wave mize response of ships to long waves. This
action, and therefore the latter are usually can be best achieved by physical modeling
neglected. However, even in a protected of port (harbor) layout.
harbor moored ship movements can occur, Furthermore, waves created by a passing
as depicted in Figure 3-6, due to the re- vessel under specific conditions may pro-
sponse of the water surface in the harbor to duce standoff forces on a moored ship simi-
excitation by open ocean long-period waves lar to those discussed in the previous sec-
known as seiches. Seiches may cause sub- tion. Naturally, structures located offshore
stantial damage to both ship and mooring should be designed for sea waves of sub-
structures. stantial magnitude. The purpose of this sec-
Long-period waves, or long waves, in the tion is to introduce some basic state-of-the-
open sea are rather small, but they are art concepts for wave loads under the
amplified in coastal regions due to a variety above-noted conditions.
of nonlinear effects. When entering en- In general, the interaction between wave
closed bays or harbors, long waves may be forces and a moored ship is a very complex
further amplified and natural water oscilla- phenomenon, and analytical treatment of
tion may develop. It is very difficult to dissi- the problem typically involves the combined
pate energy of these waves. Their periods effects of wind, wave and current loads,

......,. l

Figure 3-6. Typical ship motions.


Design Loads 253

coupled with ship motion and response of proximity to the bank or another vessel. For
the mooring system. Progress in develop- details, consult Chapter 2. Additional infor-
ment of an analytical solution can be fol- mation on dynamic effects of ship passage
lowed in the works of Wilson (1951, 1958, on moored vessels is found in Wang (1975),
1959), Van Oortmerssen (1976), Van Oort- Jensen et al. (1990), and PIANC (1995).
merssen et al. (1986), Bomse (1980), Vasco As was stated earlier, the evaluation of
Costa (1983), NATO (1965, 1973, 1987), and hydrodynamic loads, associated with long-
PIANC (1995). period waves, and standoff forces on a
In the above works, sophisticated mathe- moored vessel generated by passing ships
matical modeling techniques have been pro" on marine structures is extremely complex.
posed. The proposed methods· formulate a The problem's complexity is aggravated by
mathematical approach for obtaining water the fact that the added mass and damping
surface response, ship motion, and mooring coefficient of an oscillating vessel are dif-
forces due to a given offshore wave spec- ferent for each of the six degrees of freedom
trum for a vessel moored at a particular (Fig. 3-6) and are dependent on the fre-
location in the harbor. It should be noted quency of the ship's oscillation and, to a
that at their present stage of development, great extend, her underkeel clearance. De-
mathematical models of moored ships are tailed treatment of this subject is beyond
best used in preliminary stages of port de- the scope of this text. For a comprehensive
sign, where several alternative solutions are treatment of the subject matter, the reader
usually investigated to achieve the most is directed to the sources cited earlier in
efficient design. Physical scale modeling this section.
may be employed to calibrate (validate) the Offshore marine terminals, such as
mathematical model. Because physical scale transhipment islands or individual cargo
models are expensive and time-consuming loading/unloading berths, are exposed
they are usually employed to model the to substantial wave loads. Perimeter walls
final harbor layout obtained on the basis of of the offshore islands are usually of verti-
a calibrated mathematical model. Principles cal configuration and typically exposed to
of hydraulic modeling are described in most wave forces similar to those acting on
fluid mechanics texts. breakwaters of the same configuration.
As a rule of thumb, standoff forces on a Forces of this type are discussed in detail in
moored ship, generated by passing vessels, Chapter 10.
are proportional to the square of the speed
of the passing ships and the separation 3.2.3.1 wave Force on
distance. Circular Piles and
Additionally, the underkeel clearance, Large-Diameter Cylinders
the vessels' block coefficients, the softness
of mooring lines, and the ratio of length of The majority of individual offshore berth
moored ships (Lms) to length of passing structures, such as piers, breasting and
ships (LP.) affect the maximum standoff mooring dolphins, cargo handling equip-
forces and subsequently, the mooring line ment supporting platforms, and approach
loads. Forces on the moored ship are greater trestles are typically of the open-type con-
when Lps/L88 » 1 and are slightly notice- struction, which, in general, comprises of a
able when Lps/L88 « 1, even for a very concrete deck structure supported on piles
close separation of both vessels. By its very of miscellaneous construction. Circular piles
nature, the effect produced by the passing are used more often than piles of other
ship on a moored vessel is very similar to cross sections because they offer less resis-
that of a vessel transiting a channel in close tance to wave or passing current action.
254 Design Loads

The approach to calculating wave forces ter particles used in the Morison equation
on piles depends on the ratio of the pile are calculated from the conventional wave
diameter (D) to the incident wavelength equations [see for example, the U.S. Army
(L). When the ratio is small (D jL < 0.2) (1984), Shore Protection Manual]. The em-
the wave motion is relatively unaffected by pirical dimensionless coefficients CD and CM
the presence of pile, and the analytical ex- for cylindrical piles are given in Table 3-3.
pression of the wave force acting on a pile For detailed information on the coeffi-
(Fw) is normally calculated using the well- cients CD and CM, the reader is referred to
known Morison equation (Morison et al., Isaacson (1988) and Chakrabarti (1991).
1950). This equation is based on an as- Some preliminary data on CD and eM, ob-
sumption that the pile (cylinder) does not tained from field measurements in the surf
disturb the waves and consists of the sum zone, are reported by T(llrum (1989). Values
of the inertial force (F1 ) and the viscosity of CD and eM for square piles, H-piles, and
drag force (FD): piles of other shapes are found in the British
Standard Code of Practice for Maritime
(3-7) Structures (1984), EAU (1990), and Det
where Norske Veritas (1977).
The assessment of drag and inertia co-
FD 0.5cD(;) Du lui
= (3-8) efficients is made not on the basis of theo-
retical considerations, but principally by
and means of measurements on models and pro-
totypes. They correlate the computed force
(3-9) to the existing force and are greatly depen-
dent on the Reynolds number. They are
where very sensitive to the pile roughness caused
by such effects as fouling, which may sub-
CD = empirical dimensionless drag coeffi- st~ntially increase the pile's projected area.
cient taking into account the resistance
The Reynolds number (Re) is given by
of the pile against the flow pressure
CM = empirical dimensionless inertia coeffi-
cient taking into account the resistance (3-10)
of the pile against the acceleration of
water particles where
g = gravity acceleration (mjs 2 ) v = kinematic viscosity (m 2 js)
'Yw =density of water (kNjm 3 ). For freshwa- um = approximate maximum horizontal water-
ter, 'Yw = 10 kNjm 2 and for seawater particle velocity (mjs) at z = 7Jc (see def-
'Yw = 10.25 kNjm 3 inition sketch, Fig. 3-7)
u = horizontal component of the instanta-
neous orbital velocity of the water par-
ticles at the studied pile location (mjs) Table 3-3. Drag (CD) and inertia (CM) coefficients
for cylindrical piles
dujdt = horizontal component of the instanta-
neous orbital acceleration of the water Cn eM Reynolds Number (Re)
particles at the studied pile location
(mjs 2 ) 2.0 > 2.5 X 10 5
1.2-0.7 1.5-2.0 2 X 10 5-5 X 10 5
D = pile diameter or (at noncircular piles) 0.7 >5 X 10 5
characteristic width of the structural 1.5 <5 X 10 5
member(m). 1.2 < 2.5 X 10 5
Note: Cn = 0.7 and CM = 1.5 are typically used for most
The velocity and acceleration of the wa- design applications.
Design Loads 255

In deep water, and added vectorially. For this, the equa-


tions for Fn and F1 can be given by
7TH
U
m
= _T e2"'zL cos
-1
8
(3-11)
Fn = {fndZ (3-12)
0

where (J is the wave phase angle and um(max) and


will occur at e = 0.
(3-13)
Additionally, the coefficient of inertia eM
takes into account the fact that the added where
mass, subject to the force of inertia, has a
greater magnitude than the one correspond- fv = 0.5cv(;) Du lui (3-14)
ing to the volume of fluid displaced by the
solid.
In order to obtain the total instantaneous
wave force the drag and inertia components
must be integrated over the water depth and z is the depth in terms of vertical

Figure 3-7. Wave force on vertical pile.


256 Design Loads

tant velocity rather than on the velocity


component perpendicular to the pile axis.
With due consideration to these experi-
mental observations, the following proce-
dure has been recommended for calculating
wave forces on batter piles. For a given
location on the pile (X0 , Y0 , Z 0 ) the force
per unit length of pile is taken as the hori-
zontal force per unit length of a fictitious
vertical pile at the same location. For a
detailed procedure and numerical examples
the reader is referred to U.S. Army Corps of
Engineers (1984). It must be noted that
Note: x, Y, and Z a1.aa
ara orthogonal, besides drag and inertial forces, a trans-
verse force due to alternate eddy formation
and shedding generated downstream from
Figure 3-8. Wave force on battered pile. the pile known as the "lift force" (FL), will
also result from the wave-pile interaction.
The lift force acts perpendicularly to both
distance along a coordinate axis with its wave direction and pile axis and is analo-
origin at the bottom (Fig. 3-7). gous to a drag force. It reaches its maxi-
The problem also could be solved by us- mum value at the same moment as the drag
ing nondimensional nomographs and tables force does, but could be four to fives times
provided in the U.S. Army Corps of Engi- greater than the drag force (Rogan, 1978).
neers' Shore Protection Manual, Vol. II For design of rigid structures, NAVFAC
(1984) and in NAVFAC DM-26.2 (1982), or DM-26.2 (1982) recommends adoption of a
by commercially available computer pro- conservative value of CD for determination
grams, [e.g., Pile developed by Bridgeman of lift forces, regardless of the force direc-
(1987)]. tion.
The maximum drag force occurs at the Due to the oscillatory nature of the lift
force in the design of flexible structures, the
wave crest position and the maximum iner-
dynamic response of the structure must be
tial force occurs at about one-quarter wave-
considered (Laird, 1962; Chakrabarti, 1991).
length ahead of the crest. The practical ex-
The magnitude of this lift force can be de-
ample is given in Isaacson et al. (1995).
termined by
For nonvertical (sometimes called batter
or raking) piles, the wave phase angle (}, for
the local coordinates X 0 , Y0 , and Z 0 , is (3-16)
different for each individual section, as de-
picted in (Fig. 3-8). Because wave forces
are perpendicular to the pile axis, it is pos- where CL is the lift force coefficient which
sible to calculate them by Morison's equa- can be taken as approximately equal to
tion using components of velocities and ac- Cn/3 for circular cylinders and steady flow
celeration perpendicular to the pile (U.S. velocities. It generally does not depend on
Army Corps of Engineers, 1984). Experi- pile roughness. For more information on CL
ments by Brusnall and Loftin (1951) sup- values under different conditions, the reader
port this assumption. They found that the is referred to Chang (1964), Sarpkaya
drag force component depends on the resul- (1976), and Sarpkaya et al. (1977).
Design Loads 257

Both linear and nonlinear wave theories The combination of piles is then superim-
are used for the calculation of wave forces. posed over the plot of wave forces, and the
However, use of nonlinear wave theory [e.g., maximum force for the system is found by
the fourth and fifth order of Stoke's wave summing the forces on the individual piles
theory can produce better results (Tsuchiya for small changes in phase angle ( 8) (Fig.
and Yamaguchi, 1974; Isaakson and Qi-Hua 3-9). The phase difference between piles is
Zuo, 1989)]. If nonlinear wave theory is given as
used, the Morison equation is usually solved
by numerical integration. 11 () = (27T' jL) 11 x (3-18)
The moment of wave force about the
mudline is given as follows: where
L = wavelength
(3-17) 11 x = spacing between piles in the direction of
wave advance (i.e., along the x-coordinate
axis).
where z is the depth in terms of vertical
distance along a coordinate axis with its Assuming that the two-dimensional wave is
origin at the mudline level (Fig. 3-7). not changing while moving, the force ap-
The single small-diameter pile structure plied on individual pile and pile groups for
is rare in marine applications. The usual different wave phases may be determined
open-type piled structures are piled piers, as follows (Rogan, 1978):
platforms, or dolphins. It is generally ac-
cepted that unless piles are placed too
close to each other (1.5D or less) the pile
density may be ignored in calculating total
wave force on a pile group (Quinn, 1972; (3-20)
Bushnell, 1977). Wave forces on a combina·
tion of piles in group are calculated by plot- where Fw(n) and Mw(n) correspond to the
ting the forces on a single pile as a function total force and moment acting on the nth
ofwave phase angle (8). pile in a pile group, respectively. The total

Figure 3-9. Wave force on pile group.


258 Design Loads

force (Fw(T)) and moment (Mw(T)) on a at breaking, it is usually given as equal to


structure composed of n piles will then be (gdb) 0·5 , where db is the water depth at
expressed as breaking, which is approximately equal to
the breaker height (Hb). Current velocities
(3-21) of any kind (steady, tidal, or generated
by the sustained wind) should be added
n
vectorially.
Mw(T) = L Mw(n) (3-22) As was stated earlier, the Morison equa-
1 tion is used to obtain wave forces on a
small-diameter pile (D jL < 0.2). This
The maximum wave force on a group of equation simplifies the general problem by
piles generally occurs when the central pile assuming that the pile (cylinder) does not
is located near the wave crest. disturb the incident wave in any way. How-
Wave forces on a pile bracing system are ever, as the cylinder's diameter becomes
determined in a manner similar to that large relative to incident wavelength, the
used for vertical piles; that is, the wave large-diameter cylinder causes reflection
force per unit length, az, of pile (Fig. 3-7) and diffraction of the incident waves, which
is also the drag force per unit length of makes Morison's equation unapplicable, and
horizontal bracing when az =DB, where a diffraction theory must be used instead.
DB is the diameter of the horizontal brac- The diffraction theory assumes that the
ing. Naturally, the wave force on the hori- fluid is homogeneous and inviscous, and
zontal bracing is a function of the bracing that the fluid motion is irrotational. Addi-
position above the seafloor. Because brac- tionally, the wave heights are assumed
ings are typically beams of small diameter, small relative to the wavelength so that
only the drag force is usually considered. terms involving wave steepness to the
The wave load on an angular bracing may second (or higher)-order power may be
be obtained by the integration of loads per neglected. MacCamy and Fuchs (1954)
unit length at different elevations. developed a linear diffraction theory for
Although notable progress has been made calculating wave loads on a vertical circular
in recent years regarding understanding of cylinder extending from above the water
the breaking wave-pile interaction, well- surface to the mudline. This theory was
founded recommendations on a part of the supported by limited experiments con-
breaking wave force on pile are sparse. Apelt ducted by Chakrabarti and Tam (1973) for
and Piorewicz (1987) state that no theory is DjL up to 0.54.
available for predicting wave forces in the According to MacCamy and Fuch (1954),
breaking zone, although a considerable the total force on a large-diameter cylinder
amount of experimental work has been re- (Fw) is given by
ported (Swift, 1989; T¢rum, 1989).
For practical engineering applications in
shallow water, the inertia component of pgHL2
breaking wave forces on a pile is normally Fw = - -2- fa tanh(kd) cos( angle) (3-23)
21T
ignored, and the drag coefficient is usually
taken as equal to 1.75 to 2.5 times greater
than the one used in the deep water (U.S. where
Army Corps of Engineers, 1984; NAVFAC p = 'Ywlg = fluid density
DM-26.2, 1982; Rogan, 1978). 1
Because the wave crest velocity (uc) in
shallow water approaches the wave celerity
Design Loads 259

where introduced to the market a computer pro-


J1 =first derivative of the Bessel function of the gram based on MacCamy and Fuchs' origi-
first kind nal formulation of wave load on large cylin-
Y{ = first derivative of the Bessel function of the ders.
second kind for the angle (wt) - a; here For a discussion on wave loads on square
a= arctan(JJYi) and w = 2TTjT in wave caissons and conical-type structures, the
circular frequency reader is referred to works by Mogridge and
k = 27TjL Jamieson (1976), Jamieson et al. (1985), and
T = wave period Rahman and Chakravartty (1986).
with 'Yw• L, H, and d as previously defined.
3.2.3.2 wave Load on
Mogridge and Jamieson (1976a, b) have continuous Vertical Wall
presented a simplified graphical version of
the linear diffraction theory of MacCamy When the structure width is longer than
and Fuchs to estimate wave forces and mo- the incident wavelength (D jL > 1.0), it can
ments on large cylinders. Garison et al. be treated as a continuous wall exposed to
(1974) and Hogben and Standling (1975) the wave force per unit length along the
have extended the diffraction theory to wall. This case of wave load is treated in
structures of arbitrary geometry. Theoreti- virtually all general textbooks and hand-
cal treatment of the diffraction theory is books on coastal and port engineering [e.g.,
found in Sarpkaya and Isaacson (1981). Ac- U.S. Army Corps of Engineers (1984),
cording to Sarpkaya and Isaacson, the total Gaythwaite (1981, 1990), Bruun (1989),
wave force on a large-diameter cylinder can Herbich (1991/1992), and many others.
be obtained from The extensive discussion on wave loads
and their impact on vertical walls is found
in special issue of Coastal Engineering [Vol.
Fw = ( i)'YwHD 2tanh( 2 ~d )eM cos(wt- 8) 22 (1994)], in Proceedings of International
Workshop on Wave Barriers in Deep
(3-24)
Waters (1994), Osaka, Japan, published by
Port and Harbor Institute, Yokosuka,
where all of the terms are as previously Japan, and in Proceedings of the Annual
defined for small-diameter cylinders. Offshore Technology (OTC) Conferen-
Although considerable efforts have been ces, Houston, Texas, and International Off-
spent on research with wave forces on verti- shore and Polar Engineering (ISOPE) con-
cal cylinders subjected to regular and ran- ferences.
dom waves, studies on the wave-induced In general, where the water depth at the
dynamic pressure distribution around verti- face of the wall is deep enough (i.e., equal to
cal cylinders, especially due to random approximately 1.5-2.0 times the incident
waves, are rather limited. The most recent wave height), the wave will be reflected and
study to date on wave impact pressures on form a standing-wave pattern, also known
large-diameter vertical circular cylinders in as clapoties. In general, it is assumed that a
deep water has been conducted by Sundar standing wave is equal to twice the height
et al. (1989) and Khalfin (1990). It has been of the incident wave.
concluded that the values of the dynamic In shallow water (e.g., H < 1.5d, where
pressure coefficients derived from the pres- H is the height of the incident wave and d
sure spectra are in satisfactory agreement is the depth of the water), the wave may
with the linear diffraction theory of Mac- break against the wall, thus generating
Carny and Fuchs. Bridgeman (1984) has greater forces on it.
260 Design Loads

On the basis of his extensive laboratory level (SWL) in the Minikin method is given
tests of the breaking waves impacting (by by the semiempirical formulation. The total
plunging) directly on vertical, 10°, and 30° dynamic force that is centered about the
backward inclined walls with 1/10 fore- SWL is added to the hydrostatic pressure to
shore slope, Kirkgoz (1992) concluded that obtain the total profile of the pressure dia-
within a range of water depth conditions gram (Fig. 3-10b). Again, complete step-
used in his investigation the most frequent by-step instructions and graphical aids for
location of the maximum impact pressure applying the Minikin method are found in
for all three walls tested remained almost the U.S. Army Corps of Engineering (1984)
at the still-water level. Shore Protection Manual. A complete dis-
Figure 3-10 illustrates typical wave load cussion on wave forces acting on vertical
diagrams attributed to the action of stand- wall breakwaters is given in Chapter 10.
ing and breaking waves. Wave forces gener-
ated by a nonbreaking wave (Fig. 3-10a)
generally are calculated using the methods
of Sainflou and Miche-Rundgren, which ac- 3.3 MOORING LOADS
count for partial reflection, as described in
step-by-step instructions and graphical aids Mooring forces acting on a berth structure
in the U.S. Army Corps of Engineering arise from environmental forces, passing
(1984) Shore Protection Manual. vessels, and forces due to ship berthing ma-
When waves break against the wall, the neuvers. They are transmitted to the struc-
Minikin (1963) and Goda (1985) methods ture via ship mooring lines secured to the
are generally employed. The peak dynamic mooring hardware such as bollards, bitts,
pressure which is assumed at the sea water quick-release hooks, and other mooring
accessories.
As was mentioned in the previous sec-
tions, at most sea port locations wind is a
dominant force to be considered in mooring
system design; at inland ports, located along
riverbands with strong currents, the force
of current may control mooring forces; at
exposed offshore locations, a combination of
all environmental forces (e.g., wind, waves,
and current) must be considered. In the
(a)
latter case, berth operations are usually
= limited to certain acceptable weather
~ conditions.
SWL
H, - Standoff forces due to current andjor
2
d, wave effects, as well as ship surging mo-
~ d. tions, due to effects of long-period waves
must also be considered at certain locations
(as described in previous sections of this
(b) chapter). Naturally, vessel size and type are
key issues in the determination of mooring
Figure 3-10. Wave pressure on a vertical wall: (a) forces on a berth structure and the design
by a standing (reflected) wave and (b) by a breaking
wave. H and Hb are respectively the height of an
of mooring systems.
incident wave and of a breaking wave; h 0 is the setup Typically, the general arrangement for
of the water level. mooring lines for a ship lying alongside a
Design Loads 261

marginal wharf is as shown in Figure 3-11, from surge motions. They run fore and aft,
and a typical arrangement of mooring lines, typically at an angle of 5°-10° to the quay
at tanker berth, is shown in Fig. 3-12. face. Breasting lines are used to provide
lateral resistance to wind and current forces
imposed on the vessel. These lines are
3.3.1 Mooring Lines nearly normal to the quay face.
Arrangement In general, a minimum of four lines are
required to safely secure a ship at a berth;
Ship mooring lines (hawsers) are usually of however, in practice, the actual number of
three types: bow and stern lines, springs, lines in each category basically depends on
and breasting lines. Bow and stern lines vessel size. In the case of mooring a large
are used to resist current and wind forces ship, as many as 24 (and more) lines may
imposed against the vessel and acting par- be employed. As mooring lines are usually
allel to the berth. These lines are attached regarded as shock-absorbing elements, they
to the respective ends of the ship and usu- should not be less than about 30 m in length
ally make an angle of 30°-45° with the face to provide for sufficient elasticity.
of the quay. Larger angles are usually not To be efficient, mooring elements, such
recommended. Springs are also used to ab- as mooring and breasting dolphins, must be
sorb wind generated and current generated properly positioned to handle the expected
forces acting parallel to the quay direction, range in ship sizes. This often means that
but they are also used to prevent the vessel additional mooring dolphins are spotted be-
tween the ones on the extreme ends of the
berth. Sometimes it is also necessary to
have the breasting dolphins double to pro-
vide for sufficient ship support and to han-
dle spring lines. Where practical (e.g., at
berths where berthing is only on one side),
Figure 3-11. Marginal wharf. Typical layout of
the mooring dolphins should be set back
mooring lines: 1- dock structure; 2-ship; 3-bollard;
4- bow line; 5- stern line; 6-breasting line; from the breasting dolphins by as much as
7- spring line. 30-45 m to limit the upward component of

t Sym .

0 .3 Lo (min.)

Figure 3-12. Tanker berth. Typical layout of mooring lines: 1-ship; 2- mooring line;
3-loadingjunloading platform; 4-breasting dolphin:; 5- mooring-dolphin; 6-access
trestle; 7- gangway.
262 Design Loads

the mooring pull (vessel in light condition) lifetime than lines made from natural
and also to reduce navigational hazards. fibres. Compared to natural fibres, the
More information on this subject is given in synthetic fibre lines have a higher load/
Chapter 8). diameter ratio, are relatively light, and are
Bulk cargo terminals frequently employ easier to handle. The elongation of syn-
the same concept for mooring arrangements thetic lines is rather large and their
as is used for tanker terminals (Fig. 3-12). strength characteristics can be affected in
At these terminals, however, a greater case of high temperature (e.g., caused by
number of breasting dolphins may be re- friction).
quired, especially in the cases where the Steel wire ropes are relatively cheap and
ship needs to be shifted into position for have a long lifetime if well maintained. They
access to all hatches by loading/unloading have a low elongation being proportional
equipment. to the mooring line load and a high load/
diameter ratio. The disadvantage of these
lines is that they are stiff and heavy and
3.3.2 Mooring Line Materials therefore are difficult to handle. Also owing
to their low elongation rates, they are vul-
Mooring is provided by mooring lines ex- nerable to a potential breakage under dy-
tended from the ship and secured to moor- namic load conditions. Sometimes steel wire
ing hardware, fastened to the berth struc- rope lines are used in combination with
ture. All mooring lines are typically made synthetic tails which enhance the shock ab-
from the same material. They are usually sorbing characteristics of steel wire rope
made from natural (manila) or synthetic lines. In most cases tails are not longer
(nylon, polyester, terylene, polypropylene, than 10m and are connected to steel wire
etc.) fibre materials, or galvanized steel ropes by means of shackles, protecting the
(wire ropes). For better elasticity, individu- tail eyes with leather or plastic sheathing.
als fibers are woven together into strands, These lines most often are used for mooring
which, in turn, are woven into ropes accord- large tankers.
ing to specific patterns. Fibre ropes are usu- Typical values of breaking strength for
ally of three- to nine-strand construction, different types of hawsers is provided in
and steel wire ropes typically consist of six Table 3-4. The elastic stretch at break
strands, each containing 19, 24, or 37 indi- ranges from approximately 10 to 15% for
vidual wire filaments built around a fibre or manila ropes, to about 50% elongation for
steel core. A detailed discussion on the elas- nylon lines. Consequently, a nominal work-
tic properties of mooring lines is found in ing load is approximately in order of 10 to
Wilson et al. (1958) and Wilson (1967). For 15% of tensile strength for manila ropes,
more information on steel mooring cables and 10 to 20% for lines made from syn-
the reader is referred to NAVFAC (1986), thetic materials where the line stretch can
Tsinker (1986), and Gaywaite (1990). be tolerated. The elongation of steel wire
Natural fibre lines are relatively cheap, mooring ropes at break typically range from
easy to handle (when dry), but have a very 2 to 5%.
moderate resistance against abrasion and
therefore have a relatively short lifetime.
These lines are sensitive to high tempera-
tures (due to sunlight or friction) and chem- 3.3.3 Mooring Forces
icals. They absorb water easily, making the
handling process cumbersome in wet condi- Mooring forces are calculated from the envi-
tions. Synthetic fibre lines also are rela- ronmental design loads, as previously dis-
tively cheap; however, have much longer cussed, and are assume to act on the largest
Design Loads 263

Table3-4. Breaking strength of hawsers (tonnes)

6 X 24 Wire 6 X 37Wire
Diameter Polyester Poly- Rope, Galvanized Rope, Galvanized
(in.) (mm) Manila Nylon (Darcon) propylene Plow Steel Plow Steel

1 25 4 13 10 6 26 29
lft 35 7 23 17 11
1-'-2 38 57 65
2 50 14 48 40 24 100 113
3 75 29 111 79 52
4 100 48 180 136 86
5 125 282

vessel for which the facility is designed. It Table 3-5. Number of simultaneously
is necessary, however, to relate the calcu- working bollards (nB)
lated forces to the breaking strength of the Minimum Distance
mooring lines. It is therefore recommended Length of Ship Between Bollards
that the designer consult with the dock op- (m) (m) nB
erator to obtain information on the design :::; 50 20 2
vessel and on number and size of mooring 150 25 4
lines to be considered. 250 30 6
The critical wind load is transmitted to ~ 300 30 8
the structure when the wind is blowing nor-
mal to the ship's centerline, pushing her
out from the dock. When the wind force is The magnitude of mooring force per bal-
symmetrical, the normal component of lard, and its vertical and longitudinal com-
mooring force per ballard (Ns) is given by ponents, can be obtained by the following
equations (see Fig. 3-13):
(3-25) NB
QB = -.--=-- (3-26)
cos f3
Sill a
where
Vs = QB sin f3 (3-27)
Pw = the total wind force onto the ship (in kN)
expressed by Eq. (3-1) Ts = QB cos a cos f3 (3-28)
ns =the number of simultaneous working hoi-
lards (mooring lines)

The actual number of simultaneous work-


ing bollards (ns) depends on the length of
the ship; unless special mechanisms are
used to equalize mooring forces in hawsers,
n B can be obtained from Table 3-5.
The British Standard Code of Practice
for Maritime Structures, Part 4 (1985) rec-
ommends that when the ship is moored at
six points, one-third of the total force should
be assumed at any one point, and when the
ship is moored at four points, one-half of
the total force should be assumed to act at
any single point. Figure 3-13. Mooring forces.
264 Design Loads

In general, the normal component of the The values given in Table 3-7 are suffi-
mooring force (NB) used in Eqs. (3-26)- ciently accurate for seagoing vessels, and
(3-28) should be increased by at least 20% the values given in Table 3-8 can be used
to allow for nonuniformity of distribution of for the design of inland port mooring
mooring forces between working bollards. If structures.
the actual values of a and f3 are not known, It must be noted that bollards at the
then for the preliminary design they may ends of each berth are usually designed for
be obtained from Table 3-6. double the value of mooring force used for a
As the ship mooring lines are generally regular bollard. At docks where strong cur-
not fully stressed simultaneously and the rents exist the values of QB, obtained from
mooring line forces partially cancel each Tables 3-7 and 3-8, should be increased by
other, in the absence of actual mooring line 10-25%. A smaller value applies to smaller
forces supplied by the port operator, the vessels and a larger value to larger vessels.
values in Table 3-7 and 3-8 should be used Normally, bollards are spaced uniformly
for the preliminary design. along a quay face and symmetrical about

Table3-6. Values of a and {3 in Eqs. (3-36)-(3-28)

{3
Vessel Bollard Laden Vessel in Light
Type Location a Vessel Condition

Seagoing At the quay edge 30° 20° 40°


Second line of 40° 100 20°
bollards
Riverboats
Passenger At the quay edge 0°-45° -30° 30°
vessel
Cargo vessel At the quay edge 0°-30° -30° 30°

Table 3-7. Mooring forces Q8 generated by seagoing vessels

Vessel Vessel
Displacement Line Pull Displacement Line Pull
(tonnes) (kN) (tonnes) (kN)

2,000 100 100,000 1,000


10,000 300 200,000 1,500
20,000 600 > 200,000 2,000
50,000 800

Table 3-8. Typical values of Q8 transmitted by inland waterway ship mooring lines to bollards (kN)

Maximum Displacement (tonnes)


Vessel Type s; 100 500 1,000 2,000 3,000 5,000 10,000 20,000 50,000 100,000 2000,000 ~ 200,000

Passenger and
mixed cargo I
passenger ship 50 100 150 200 250
Cargo ship 30 50 100 130 150 200 250-300 600 800 1,000 1,500 2,000
Design Loads 265

the center of a quay section (between ex- expected to be satisfactory at berths accom-
pansion joints). Other arrangements are modating large vessels. A potential "side
also used, depending on the actual length of effect" of the pretension of mooring lines is
the section and bollard spacing require- the creation of vertical up and down friction
ments. Typically, bollards should be de- forces between the ship and fender system
signed to hold several hawsers simultane- due to tidal variation, or changing draft of
ously, and, therefore, the load per bollard the ship due to loading or unloading. These
must be adjusted accordingly. forces are then transmitted to the dock and
In most seaports, bollards are installed have to be applied to the front of the dock
at ground surface level. In inland ports, structure for calculation of general stability
where the water level may fluctuate signif- and local strength. Naturally, fenders must
icantly, bollards are usually installed on at be strong enough structurally to sustain
least two levels, but quite often they are this load.
installed at various elevations to allow for The mooring system must be designed to
ship convenient and safe operation. Some limit ship movements at berth to prevent
practical examples are found elsewhere in possible damage to the ship, mooring struc-
this book. ture, and loading/unloading equipment.
At exposed locations, a soft mooring sys- Information on ship movement allowances
tem comprising of mooring lines with nylon obtained from miscellaneous sources is pre-
tails in combination with soft fenders are sented in Table 3-9. An extensive research
used (Khanna and Sorensen, 1980; Wood et program was carried out by the Nordic
al., 1980; and PIANC, 1995). Nylon tails countries to establish acceptable ·criteria of
provide an additional elasticity to the moor-
motion of moored ships (Jensen et al., 1990).
ing system, thus reducing the load in the
Motion of moored ships were measured at a
mooring lines due to motions of the moored
number of ports for various ships being
ship. Furthermore, in a multicomponent
loadedjunloaded by different gear, and
mooring arrangement, the length of nylon
ship acceptable motions were recommended
tails can be varied in order to maintain
based on interviews with ship crews and
approximately equal loads among lines un-
port operators. Based on these recommen-
der anticipated ship movements.
Nylon tails are routinely used by oil com- dations PIANC (1995) suggested somewhat
pany fleets, particularly at exposed loca- more liberal motion criteria than those
tions. Soft fenders, although absorbing given in Table 3-9. More information on
equal amounts of the ship's energy, produce acceptable ship motions while under load-
smaller forces on the mooring structure ing/unloading operations is found in Bruun
than stiff fenders. They are provided to be (1989), Bratteland (1974), Slinn (1979),
compatible with the softer mooring lines. Thoresen (1988), Jensen et al. (1990), and
Pulling a ship against a fender system by PIANC (1995).
pretensioning of mooring lines may reduce Safe berth operation to a great extent
her motion and thereby reduce the forces in depends on mooring systems reliability.
mooring lines (Brunn, 1989). Therefore, redundancy should be provided
The magnitude of pretension required in as much as is practical; factors of safety on
the mooring lines is not precisely defined. the mooring system components typically
Normally, a pretension to about 10% of a range from 1.5 to 2.0, depending on the
line's breaking strength will remove the ini- severity of environmental conditions and the
tial sag from a synthetic line. In practical nature and consequence of a failure. It must
terms, a relatively small amount of preten- be noted that standard practice in the wire
sion on the order of about 10 tonnes is rope industry is to use a factor of safety of
266 Design Loads

Table3-9. Movement allowance for ship at berth

Surge Swayb Heave Yaw Pitch Roll


Type of Ship (m) (m) (m) (degree) (degree) (degree)

Tanker 2.5a 2.0 1.0 1.0 1.0 1.0


Ore carrier
(crane operation)d 1.5-3.0 0.5-2.0 0.5 1.0 1.0 1.0
Grain carrier
(at elevator) 1.0 1.0 1.0 1.0 1.0 1.0
Container 0.5 0.3 0.5 1.0 1.0 1.0
RojRo (side ramp)c 0.3-0.5 0.3-0.5 0.3-0.5 1.0 1.0 1.0
RojRo (bow or stern ramp) 0.1 0.1 0.2 1.0 1.0
General cargod 1.5 1.0 1.0 2.0 2.0 1.0
LNG 1.5 1.5 2.0 2.0 2.0
Fishing Vesseld ±0.15-2.0 0.15-1.0 0.4 3.0 3.0 3.0

a Can be larger at a single mooring buoy or single-point mooring systems.


b Movement away from the berth.
c Allowance movements dependent on ramp design.
d Depends on cargo handling equipment.

3-5 on maximum breaking strength when of more than two breasting dolphins, then
specifying the working load. possible misalignment of dolphins (.:1)
Monitoring of the mooring lines forces to andjor other factors (e.g., eccentric load ap-
keep them constant is the latest technologi- plication) could result in an unequal load
cal tool used for improving operations at a distribution between individual dolphins.
berth. This technology is basically used for (Fig. 3-14). To account for this, it is recom-
handling of large ships, particularly at loca- mended that the design load generated by a
tions exposed to waves, wind, and current. ship against individual dolphin be in-
The system triggers alarms when loads ex- creased by a factor of 1.3. Additionally, if
ceed a setup limit, which permits initiation these dolphins are of different construction,
of necessary action, such as early line tend- then the total component of mooring force
ing or use of a tug to hold the ship. normal to berthing line, Ps(p)' must be re-
Finally, if the berth structure comprises distributed among the dolphins according to

-·----~~~ __£]___·_~· _K___~


<3

Figure 3-14. Berthing forces due to misalignment of dolphins: 1-ship; 2-breasting


dolphin; 3-loadingjunloading platform; 4-berthing line.
Design Loads 267

their respective stiffnesses. Therefore, con- handled was increased from 20 kg to a few
sidering the possibility of dolphin misalign- hundred kilograms; the cargo was trans-
ment, the normal component of mooring ferred with the help of powered winches,
load exerted on an individual dolphin, Pd(p)' which enabled lifting of larger and heavier
can be given by units. In the 1950s, palletization and pres-
linging came into common use; the aim was
ps(p)
to standardize units of up to 2000 kg.
Pa<P> = 1.3 I: (EI)a (EI)i (3-29) A pallet is essentially a wooden or metal
base on which the cargo is placed. The cargo
is typically secured with different kinds of
where banding. The pallet is transported by fork-
'[,(EI)a = total stiffness of the system of breast- lift vehicles and may have provision for
ing dolphins in direction normal to the being lifted by sling. Pallets are generally
berthing line loaded to a maximum height of about
(EI)i = stiffness of the particular breasting 2.5-3.0 m and may be stacked up to about
dolphin within the system in direction 5.0m.
normal to the berthing line In the 1960s, the standardization process
took a step further through containeriza-
Essentially, if all dolphins are of the same tion and refinement of other systems for
construction, then products requiring protective containers
(e.g., liquid and bulk cargoes).
Productivity, safety, and the working en-
ps(p))
pd(p) = 1.3 ( -n- (3-30) vironment have been improved with the de-
velopment of the innovative cargo handling
methods. The present trend in industrial
where n is the number of breasting shipping is the use of the roll-onjroll-off
dolphins. (RojRo) systems, which under certain con-
For the case in question, the horizontal ditions may be several times as efficient as
component of a total mooring force Ps(h)' is conventional systems. The chief merits of a
transmitted to dolphins via mooring lines Ro jRo system is greater flexibility toward
and through friction forces between the ship various cargoes, no demand for expensive
and the fender system. port installations, and high utilization of all
port and stevedoring resources. However,
conventionallift-onjlift-off operation is very
3.4 LOADS FROM CARGO often found less costly for high-density
HANDLING AND HAULING cargo.
Naturally, developments which occurred
EQUIPMENT AND UNIFORM in ship construction and cargo handling and
DISTRIBUTED LOADS hauling technology inevitably affected de-
sign and utilization of terminal marine
structures. In modern practice, port related
3.4.1 General considerations marine structures are typically designed to
support relatively heavy loads imposed on
Historically, during the era of sailing ships, them by cargo handling and hauling equip-
vessels were loaded and unloaded manu- ment such as fixed cranes, rail-mounted and
ally. With the introduction of steamships, mobile cranes, vehicular loads, railroad
the average weight of a general cargo unit track loads, miscellaneous material hauling
268 Design Loads

equipment, and miscellaneous uniform dis- The container cranes most commonly used
tributed (surcharge) loads. feature the following characteristics:
In some cases, particularly at dedicated
terminals, the structures are designed for • Minimum height between the lower part of
very specific loads. Depending on structure the spreader and quay level of 30 m.
usage, the formulation of possible design • Minimum outreach, measured from the
loads is practically limitless. During the face of the quay, 35 m, and back reach of
useful life of a terminal the kind of vessels 15m.
calling at the terminal may suddenly • Rail gauge between 16 and 35 m; a rail
change, with a subsequent change in the gauge of 30.5 m is most common.
required quay edge cargo handling equip- • A clearance of at ieast 16 m is provided
ment. Therefore, in the formulation of the between the legs in order to leave room for
dock design loads, unless the dock is de- containers and cargo hatches.
• At least 400 kN of lifting capacity under
signed to serve a highly specialized kind of
the heaviest spreader.
vessel, the designer must exercise a certain
flexibility to strike a balance between a
cost-prohibitive structure and a structure In general, container throughput averages
that becomes repeatedly unusable or even 20 containers per hour per crane. According
obsolete because of an inability to support to PIANC (1987) there is a tendency to
unanticipated loads. At present, various anticipate heavier containers and larger
trends can be observed with regard to the ships, which may result in required con-
selection of quay edge handling equipment. tainer crane lifting capacity under the
For example, if traffic is sufficiently homo- spreader of 500 kN and boom outreach,
geneous, dense, substantial, and stable over measured from the face of the quay, of 40
time, then highly specialized cargo han- m. These cranes will also need a larger
dling equipment is used, whereas in other clearance between the legs to allow for pas-
cases the port authorities prefer multipur- sage of the larger containers. The required
pose, more versatile cargo handling equip- space under the crane is determined by the
ment. horizontal cargo handling operation, result-
The recent changes in quay edge equip- ing in a certain number of traffic lanes.
ment are a function of a dramatic evolution Probably the latest development in ship-
of ships. The ship specialization and their ping technology is the use of specific types
increase in sizes conditioned the need of of RojRo ships with stern, side, andjor
quay edge cargo handling equipment. Also, quarter ramps, the weight of which can be
the loads to be handled in ports are getting quite substantial for some. Because one end
heavier, as the ships' beam increase. Hence, of these ramps is supported on the quay
the lifting capacity of cranes and reaches of structure, the resulting load must be con-
their booms are also modified. sidered in the design of the structure.
The general trend concerning multipur- Ships provided with a ramp can be
pose rail-mounted and mobile cranes ob- moored at any location on the quay; there-
served in modern ports is as follows fore, the whole structure must be designed
(PIANC, 1987): to support the ramp loads. If operation of
Ro jRo vessels with ramps is foreseen, then
• Minimum reach of a crane boom, mea- all edge bollards, and any kind of surface
sured from the berthing line of 35 m rails must be located below the deck surface
• 350 kN of lifting capacity under hook, or level (Fig. 3-15).
280 kN under spreader The following is a guideline for typical
• Lifting velocity of 1-1.5 mjs loads which are to be considered in the
Design Loads 269

3.4.2 Design Load Assumptions

Two concepts are presently employed in the


formulation of design loads acting on ma-
rine structures; one is a "real-life" load as-
sumption which is based on the use of mis-
cellaneous loads which are falling in a line
or concentrated load category (e.g., crane,
outrigger pads, gantry, and road traffic
loads) (Fig. 3-16a); the second concept is
Figure 3-15. Edge bollard installed below grade at based on an equivalent uniform load as-
Ro jRo terminal.
sumption (Fig. 3-16b). The equivalent uni-
form load in many cases is a convenient
design of a modern marine facility. It must simple assumption; it is easily recognized
be understood, however, that each port is and understood and is used in lieu of "real"
different and, therefore, recommended gen- concentrated and line loads. Pallets and
eralized design loads provided herein must containers are typically considered as uni-
be treated with caution. form distributed loads.

(a ) (c )

(b ) (d )

Figure 3-16. Terminal design load assumptions: (a) real loads; (b) equivalent uniform
distributed load; (c) compromise between assumptions (a) and (b); (d) example of loading case
(a)- general view.
270 Design Loads

In general, at cargo berths the equiva- should be designed for both a maximum
lent uniform load is usually assumed as design uniform live load and a maximum
being placed within the apron area (area concentrated live load. The latter usually
between berthing line and transit sheds, or has a very noticeable effect on the deck
the open storage area). It is appropriate to slab, somewhat less pronounced effects on
note here that in current practice 20 m is the supporting beam system, and usually
considered as the minimum acceptable insignificant effects on the foundation. This
apron width for the general cargo berth; is because each successive structural ele-
however, a wider apron (up to 40 m) is ment is affected more by the uniform load
usually more desirable because it provides than by the concentrated loads, the effect of
for more flexibility in cargo handling and which is successively diminished.
hauling operations. In general, regular concentrated loads
In some practical cases, however, use of from permanently installed and mobile
the equivalent uniform load concept may be cargo handling and hauling equipment, such
misleading. As is pointed out by Junius as cranes of different kinds, fork lifts, strad-
(1983), in some instances the gross load dle carriers, tracks, and others, must be
intensity of stacked pallets may only accounted for in the dock structure design.
amount to 20-25 kN jm 2 , but due to the Occasional usage of special heavy lift equip-
specific configuration of the pallet, the real ment such as large mobile cranes and the
local concentrated load may be several times like, must be carefully evaluated. Because
as much. Therefore, depending on the rela- the load from such equipment can substan-
tive configuration of the berth structure, tially increase the cost of the structure, it
serious overstressing may occur locally even can be treated as an extreme load, there-
though the design equivalent uniform load fore, in such cases, the structural members
rating of the dock is not exceeded. In prac- designed for normal loads should be checked
tice, a compromise is recommended in the for larger allowable stresses and smaller
combination of both concepts (Fig. 3-16c). factors of safety. In practice, loads from
Because the quay edge area is usually seldom used heavy lifts are spread out by
the most sensitive to closely situated con- different means (e.g., steel beams, mats and
centrated (point) loads, all kinds of related others).
loads, such as rail-mounted and mobile Gurevich (1969) reviewed available re-
crane outrigger loads, and the like, must be search data on loads produced by 45 portal
considered, provided that the equivalent revolving cranes, used in different combina-
uniform load case is less critical than the tions (total 350), with railway track loads
concentrated load(s). placed under the portals of these cranes. It
It must be noted that in the design of has been concluded that, in most cases, por-
backfilled structures, the design concen- tal revolving cranes, in combination with a
trated load must be considered as dis- railway load, produced an equivalent load
tributed in all directions in relation to the of about 35 kNjm 2 ; some 150-kN lift capac-
berthing line. This may significantly reduce ity cranes occasionally produced an equiva-
the effects of heavy concentrated loads on lent load of about 40 k.Njm 2 • In general,
backfilled structures. Gurevich recommends the use of an equiva-
On the other hand, this must not be lent uniform apron load equal to 40 k.Njm 2
expected in the case of open-type berth for ports operating conventional portal
structures with conventional concrete (or cranes in combination with other loads (e.g.,
other) deck structure supported on piles. railway track and vehicular loads); he also
This type of berth structure, as well as recommends a uniform equivalent load of
approach trestles of similar construction, 20 kNjm 2 to be used in general cargo
Design Loads 271

berth design if conventional mobile andjor the surcharge load of stockpiled bulk mate-
crawler cranes are employed for cargo rial. In the case of bulk material, stockpil-
handling. ing the load diagram as depicted in Figure
Again, it must be stressed that, in all 3-16c should be modified to that as shown
cases, the effect of any kind of heavy con- in Figure 3-17.
centrated load on the local strength of a Typically, palletized general cargos nor-
quay wall, or pile-supported pier structure, mally would produce a uniform load of
must be carefully evaluated. around 20-30 kNjm 2 , and general cargo
shipped in containers may produce a load
up to 45 kN jm 2 • General data on standard
3.4.3 Uniform Distributed containers are found in Table 3-10. In this
cargo Loads and table, the basic characteristics including the
weight of internationally used containers
Miscellaneous Live Loads are given according to ISO (International
Standards Organization) Standard 104.
A discussion on uniform distributed equiva-
lent loads (qc) and conditions ofload appli- However, the actual weight of containers is
cation have been provided in the previous usually 50-75% less than the maximum
section. Additionally, it must be said that values given in Table 3-10 and, in practice,
this kind of load is typically used within the 2-high stacked and 4-high stacked contain-
width of dock apron (b) only. Theoretically, ers produce respectively 20 and 30 kN jm 2
an unlimited surcharge load (q 8 ) may be of equivalent uniform load.
applied beyond the apron because it does
not affect strength and stability of the berth
structure. In practice, however, this load is 2

limited by the very nature of the stored -----~


ar-----.•
cargo (e.g., liquid or bulk material, general I/ I I
II
II
cargo piled on pallets, or in containers, etc.) ~

and by the allowable bearing pressure on


the storage area foundation material. For
example, bulk materials such as sand, stone,
coal, and ore can be stockpiled to produce
surcharge loads up to 100-300 kNjm 2 . The Figure 3-17. Typical load diagram at bulk material
use of retained walls can further increase handling terminal.

Table 3-10. Characteristics of standard container

Maximum Theoretical
Value of Uniformly
Weight of Distributed Load (kNjm 2 )
Dimensions (mm) Max. Container (kN) Number of Layers
Types Length Width Height Weight Aluminum Steel 1 2 3

1A 12,190 2,435 2,435 300 230 420 10.1 20.2 30.3


1B 9,125 2,435 2,435 250 180 325 11.2 22.4 33.6
1C 6,055 2,435 2,435 200 130 230 13.6 27.2 40.8
1D 2,990 2,435 2,435 100 75 135 13.7 27.4 41.1
1E 1,965 2,435 2,435 70 14.6 29.2 43.8
1F 1,460 2,435 2,435 50 14.1 28.2 42.3
272 Design Loads

Thoresen (1988) recommends that aprons to pedestrian use only, are normally de-
and ramps for container traffic should be signed for a minimum uniform live load of 5
designed for a uniform live load of at least kNjm 2 • If piers are open to automobile
40 kNjm 2 • Multipurpose RojRo cargo in traffic, then in addition to the uniform loads
modern port design is usually accounted for given in Table 3-11, the loads recom-
30-50 kNjm 2 uniform load. Some repre- mended by the latest issues of the National
sentative ranges of uniform distributed Bridge Design Codes must be taken into
loads, obtained from miscellaneous sources account.
and recommended for design of marine A uniform live load of 10-15 kNjm 2 on
structures, are given in Table 3-11. the deck area is used for design of oil docks
Loads in transit sheds depend to a great or similar docks, which handle bulk materi-
extent on the height to which palletized als by conveyor or pipeline, and for which
cargo are stacked. Typically, this load may general cargo is of secondary importance.
vary between 20 and 50 kNjm 2 depending Liquid bulk cargos are handled by
on the type of cargo. pipelines which exert local concentrated
If there is any possibility that the cargo loads at pipe supports; bulk cargos trans-
may occasionally be stored in nondedicated ported by conveyers usually exert live loads
storage areas, then it is recommended the transmitted to the supporting structure
same uniform distributed load be applied to through the conveyer's frame. Saturated
the whole terminal area, including the weights should be considered for uncovered
apron. Such an approach will ensure safety bulk materials; NAVFAC (1980) covers dock
of terminal operation and provide for maxi- loads for navy piers and wharves; loads for
mum flexibility in cargo handling and haul- ship repair and maintenance facilities are
ing techniques. discussed in Mazurkiewicz (1995).
If appreciable quantities of cargo are
likely to accumulate on the deck of rela-
tively narrow piers (20-50 m wide), then
the actual uniform load (but not more than 3.4.4 Rubber Tire and
20 kNjm 2 ), in combination with heavy con- crawler Track
centrated loads (if any), should be used in Mounted Equipment
the pier's design. Passenger piers, such as
commuter boat piers and those, accessible Contemporary vehicular hauling equipment
such as mobile cranes, forklift trucks, strad-
dle carriers, heavy trucks, and so on impose
Table 3-11. Uniform distributed loads used heavy concentrated loads on a marine
in modern port design
structure. In fact, these loads may govern
Load the design of the pile-supported dock sys-
tem or apron andjor storage area pave-
Passenger cars and light trucks 5
10
ment.
Cars and trucks of all kinds
General cargo 20 Forklift trucks (Fig. 3-18) are probably
Paper 25-55 the most common cargo handling equip-
Forest products 70 ment used in ports. Normally they have a
Steel products 40-100 lifting capacity in the range 30-100 kN and
Coal, sand 100-200
100-300
up to 420 kN, with a typical wheel load of
Ore
Cement in bags 15-20 118 kN for 200-kN-capacity lift trucks and
Fertilizers 15 215 kN for 420-kN-capacity lift trucks.
Perishable produce and Forklift trucks typically have a turning ra-
grain in bags 15-30 dius in a range of about 4.0-5.0 m. Usually,
Design Loads 273

(a )

7.7

(b )

Figure 3-18. Examples of forklift trucks.


274 Design Loads

general cargo tenninals and specialized straddle carriers used in container tenninal
piers are designed for200-kN-capacity fork- operations (Fig. 3-19) have a lifting capac-
lift trucks. However, heavier trucks must ity of 450 kN with a maximum correspond-
also be considered where applicable. ing load per wheel equal to 170 and 160 kN,
Typically, 3-high and 4-high stacking respectively. Because forklifts, straddle car-

(a)

1 15.30

4 . 65

:-:·.·.·.·. .·.·:·:·.·:·:·:·:·:·:·:·:·:·:·:·:;:~:·:·:·:·:-:
4 .10

A 9.90 A 3 high
- 4 .45 r-

stacking

3 .115

(b )

Figure 3-19. Examples of typical 3- and 4-high stacking straddle carriers.


Design Loads 275

riers, and other rubber tire mounted equip- purposes, the loaded area usually is repre-
ment of different capacities are fitted with a sented as the equivalent rectangular area
number of wheels proportional to their lift- (ab), which is assumed to be uniformly
ing capacities, the usual maximum contact loaded (Fig. 3-20).
pressure remains below 1000 kNjm 2 . The standard truck load typically used in
Normally, terminal facilities are de- North American bridge design practice is
signed to accommodate a standard highway shown in Figure 3-21. The average actual
load. In designing for a vehicular highway dimensions of the standard North American
load, it is typically assumed that a 20-t truck tires (corresponding to approximately
motor truck (traffic load class H20-44 or 36.5-, 54.5-, and 73-kN wheel loads) and the
HS20-44) could, within reason, be placed assumed dimensions of the rectangular
anywhere on the dock's deck, and load dis- wheel-load areas at the deck surface (ap-
tribution assumptions should comply with proximately the actual tire pattem dimen-
bridge construction regulations. However, sions) are given in Table 3-12. In addition
the impact coefficients normally used for to vertical loads, all moving rubber tire
bridge design may be decreased by half, mounted vehicular equipment produces lon-
assuming reduced speed of vehicular traffic gitudinal forces due to traction and braking.
on the deck of the dock and access trestle To account of it, 10% of the vertical wheel
(bridge). load is usually applied longitudinally. Ap-
The vehicular load in pier design basi- plicable design codes such as AASHTO
cally consists of the individual wheel, axle, (1982), CANjCSA S6-M87, or similar speci-
and truck loads, as given by the appropriate fications should be used as a design guide
specification. In the design of a pier deck, for defining the general vehicular loads.
the uniformly distributed live load rarely Mobile cranes are a common cargo han-
govems because of the relatively short dling equipment in ports. They may be rub-
spans of the deck structural members. The ber tire or crawler track mounted. Lifting
critical wheel load on the deck elements capacity of this type of equipment ranges
depends on the wheel contact area. In the from 30 to 100 kN and up to 3000 kN; note
case of the deck wearing surface, the di- that to handle loaded containers, the crane
rectly loaded area of the deck structure is must have at least 400 kN lifting capacity.
somewhat larger than the contact area be- Rubber tire mounted mobile cranes trans-
tween the tires and the deck riding surface. mit their corner loads via outriggers to the
This is due to the effect of the load dis- pier deck structure or to the apron pave-
tributing action of the wearing layer. For a ment of backfilled dock structures. In gen-
concretejasphalt wearing surface, a 45° eral, the wheels of mobile cranes are not
distribution is usually assumed. For design used to transmit loads during hoisting oper-

I a, I

c-Q
~c~
I a •I
Figure 3-20. Wheel load.
276 Design Loads

0
i i
H20-44 36.5 kN t45.5kN
H 15-44 27.5 kN ttO.OkN

T ...
HI0-44 18.5 kN 72.5kN

·-----.-
. I

i i I
HS2G-44 36.5 kN 145.5 kN t45.5kN
HS 15-44 27.5 kN ttO.OkN uo.o kN

.
~
"i

I .. ~
....
.,a

.
o.s!
,.;l!
a

..
·--$·---¥~ 0 ~
1!
I!
a

Figure 3-21. Standard truck load. V is the variable spac-


ing, 4.3-9.2 m inclusive. Spacing to be used is one which
produces maximum stresses. For details, see relevant
bridge design code.

ations. The crane's outriggers are typically This, however, is done at the expense of the
dimensioned in a way that the bearing crane's mobility and, therefore, should be
pressure against the terminal pavement avoided where possible.
amounts at a maximum of about 400 Unless otherwise required, new general
kN jm 2 . This load may be too heavy for berthing piers, or quays should be designed
some older structures. In this case, the load to accommodate a minimum 700-kN-
must be properly distributed to the deck capacity mobile crane (NAVFAC, 1980).
structure by means of mats or steel beams. Using this minimum crane size for design,
The bearing pressure ~ay also be reduced the maximum outrigger corner load may
by enlargement of outrigger bearing area. be on the order of 700 kN. Additionally,
Design Loads 277

Table 3-12. Truck wheel dimensions and loaded area (mm)

Dimensions Used in Design

Nominal Top of
Wheel Wearing Top of
Loads Wheel Dimensions Surface Deck Structure
(kN) T L A s w al bl a b

36.5 178 226 40 X 10 3 290 467 508 254 559 305


54.5 201 274 55 X 10 3 325 526 508 254 559 305
73.0 254 279 71 X 10 3 412 666 610 254 660 305

Table 3-13. Mobile crane maximum loads nature of these loads. PIANC (1987) recom-
Lifting
mends that static wheel loads for mobile
Capacity Wheel Load" Outrigger Load cranes and for forklift trucks be increased
(kN) (kN) (kN) by 20% and 40%, respectively.
Crawler track mounted mobile equip-
150 at 3m 160 180
250 at 5 m 120 300 ment (mostly cranes and bulldozers) are
400 at 28m 110 1300 also used in ports. The load distribution
400 at40 m 150 2400 under the crane's tracks is typically trape-
800 at 14m 140 1680 zoidal (nearly triangular) and the bearing
• Traveling position. pressure under the tracks may peak up to
Source: From PIANC (1987). 700 kNjm 2 • Tread widths are typically
within a range of 85-140 em. Crawler track
cranes may cause horizontal forces due to mounted equipment is usually used on un-
wind, and braking due to movement or paved surfaces of port territory. Otherwise,
during rotation. As a rule of thumb, this timber mats must be used to protect the
force is usually taken as equal to about 10% pavement or concrete deck surface. Mats
of the wheel load and can be applied in any are also recommended for better load distri-
direction. Some representative values for bution where required. In addition to the
mobile crane wheel and corner loads are rated load, an impact factor of about 20% is
found in Table 3-13. Typical wheel and usually applied to account for the dynamic
outrigger arrangements for mobile cranes effect of the crane's performance.
with 250 and 450 kN capacities are de- Generally, specific data on the mobile
picted in Figure 3-22. Other types of cargo handling and hauling equipment are
rubber tire mounted equipment include obtained from the manufacturer's litera-
different types of tractors, with or without ture. However, the manufacturer's general
trailers, front-end loaders, automobiles, brochures can sometime be dangerously
Ro jRo trailers, and others. Some of the misleading in regard to deck load, warns
above vehicles may produce substantial Gaythwaite (1990), because the listed aver-
wheel loads and should be considered in the age tread pressure for crawler-type cranes
structure's design; others, such as automo- may be inadequate for deck structure de-
biles, produce relatively light loads. As it sign. Therefore, if in doubt, the designer
was pointed out earlier, piers designed to should carefully verify any specific equip-
handle Ro jRo and heavy truck traffic must ment characteristics with the manufac-
be able to withstand substantial localized turer.
loads. Regular loads, obtained from the In conclusion, it should be noted that to
equipment manufacturer's literature, must prevent mobile equipment from rolling over
be used with consideration of the dynamic the berth edge and into the water, a safety
278 Design Loads

,. 5,80

"I
_J_
I

0,90~
.I 1~·90
(a)

(b)

15.40

11.00
17.00
-·-
I 4.80

'

(c)

Figure 3-22. Examples of typical wheel and outrigger arrangements for


mobile cranes, 250 kN (a) and 450 kN (b) capacities.
Design Loads 279

curb 20-30 em high is usually installed . small as possible; however, care must be
along the berthing line. This curb is typi- taken to avoid any potential collision be-
cally designed to take a horizontal impact tween the crane and a heeling ship, or the
force of 15-25 kN, depending on the type of ship's flared bow.
mobile equipment used. In inland ports, portal cranes are usually
placed closer to the face of the berth than in
seaports (Fig. 3-23). This is usually done
for the following reasons: riverboats and
3.4.5 Rail-Mounted cargo barges generally do not have flared bow and
and Material Handling therefore cannot hit the crane during the
Equipment high water level while approaching the
dock; they also cannot hit the crane by
This type of equipment includes revolving the stern during departure maneuvers; a
portal and container cranes of miscella- closer crane location is required to improve
neous capacities and construction, railroad ship observation by the crane operator dur-
trains, and a variety of specialized bulk ing loading/unloading operation under a
handling equipment. low water level.
Customary general cargo handling cranes Inland port cranes usually have the front
used in port operation are basically revolv- legs placed about 1.1-1.3 m from the berth's
ing portal level-lifting cranes spanning one, face; this provides a minimum 0.8-m walk-
two, or three railway tracks (6-15 m be- way along the structure's edge. In some
tween legs). Typically, portal crane capacity cases however, and particularly when a
varies in the range of 30-400 kN at a work- crane's front legs rest on sheeting, or at
ing radius of 20-40 m. At seaports, the pier-type structures built on a sloped por-
crane's front rail is typically placed at tion of the quay, the walkway for the crew
1.75-2.25 m from the quay edge. This is and maintenance personnel is provided be-
usually done in order to maintain sufficient hind the front leg of the crane. This type of
room for bollards, to accommodate the out- arrangement, in general, is not desirable
board of a crane leg, to provide working because of its vulnerability to ship impacts.
space for line handlers, and for access to It may also cause the problems associated
the berth. with crane accessibility for maintenance
In general, the distance between front from the water side and difficulties with the
rail and quay edge should be maintained as arrangement of mooring hardware and the

rr=--=---=--=.-1I I
I I

0.1
1/
1/ .e-1.0
\111
1

(a) (b) (c)

. Figure 3-23. Typical positioning of portal cranes in inland ports in relation to the
berthing line.
280 Design Loads

placement and handling of ship mooring picted in Figure 3-24. Again, a dynamic
lines. factor of 1.2 is typically used to account for
The crane boom can slew through 360°. vertical and horizontal inertia forces that
Therefore, the maximum load per leg that occur during lifting or setting down of cargo.
is usually attributed to maximum design A horizontal load of 10% of the wheel load
wind load imposed on the crane can occur in the rail direction, due to crane traction or
at each corner boom position. centrifugal forces, is also considered.
Maximum portal crane wheel loads are The most unfavorable crane load combi-
typically within the range 200-400 kN and nation on a berth structure is created by
are usually obtained from the crane manu- two portal cranes operating at the closest
facturer. Some typical values of portal crane distance (typically 2.5 m) between adjacent
wheel loads used in North America are de- wheels. Naturally, under such conditions

EITHER RAIL

EITHER RAIL

BOOM OVER CORNER


PORTAL CRANE
CAPACITY 23 TONNES AT 27.45 m

EITHER RAIL

EITHER RAIL

BOOM OVER CORNER

PORTAL CRANE
CAPACiTY 32 TONNES AT 27.45m

EITHER RAIL

EITHER RAIL

PORTAL CRANE
CAPACITY 45.5 TONNES />T 35.1 m

Figure 3-24. Standard North American portal cranes wheel load.


Design Loads 281

two neighboring cranes cannot produce 3-25) should be increased by 10-15% due
maximum loads in adjacent legs simultane- to wind action or inertia effects.
ously, otherwise the crane's jibs would in- Maximum wind forces parallel to the
terfere with each other. craneway direction and perpendicular to it
The German code EAU (1990) provides on the order of 200-300 kN are applied to
guidelines on typical portal crane corner typical working cranes, or a load of 600-750
loads. Useful information on rail-mounted kN applied on idle cranes must be consid-
crane loading may also be obtained from ered. Smaller values of wind force are used
NAVFAC (1980). in areas with relatively moderate wind con-
A great variety of rail-mounted container ditions, and larger values in areas with
cranes of various configurations are now severe or special wind conditions. If cranes
used in ports throughout the world. Most with spans of more than 18 m are used,
container cranes have lifting capacity in then maximum corner load, as well as max-
excess of 400 kN. They are typically built as imum wind force exerted on the cranes,
full portals cranes with cantilever beams must be corrected accordingly. For example,
and trollies, whose supports have, as a rule, some container cranes with ali effective
six or eight wheels. The crane gauge is span of 30 m may produce a maximum
typically in the range 15-18 m. An example corner load in the order of 3000-3600 kN
of a typical container crane of 530 kN lifting (EAU, 1990).
capacity (inclusive of spreader) commonly In addition, the rated capacity, impor-
in use is shown in Figure 3-25 (EAU, 1990). tant crane features are its outreach and
These cranes are designed to handle the backreach. These effects on crane loadings
typical 40-ft container as well as 2 X 20-ft must also be considered.
containers (so-called twin-twenty method). Contemporary container ships are usu-
The maximum corner load on a portal sup- ally built with flared bows which may sub-
port of either 3000 kN or 3300 kN (see Fig. stantially overhang the berth line. In order

. (:

(a) (b)

~ ~ ~ ~
1080 1080 1080
170 1370 1370
Otslonce beTween /JUI/ers: 11.50 m

(c) (d)

Figure 3-25. Typical container crane of 530 kN lifting capacity.


282 Design Loads

to prevent bow collision a crane is usually dynamic response of elevated structures un-
place 2 m or more from the berth face. der wind loading must also be carefully
A wide variety of specialized rail- addressed (Phang, 1977). In general, when
mounted equipment is employed in bulk nonstandard equipment is employed, it is
cargo handling terminals. As a rule, this advisable that the designer work closely
equipment is custom built to serve a certain with the equipment manufacturer to prop-
particular project. It includes traveling erly understand the loadings and relevant
gantry cranes (bridge cranes) with miscella- operating criteria.
neous grab devices, traveling bucket eleva-
tors which utilize various conveyor systems,
traveling ship loaders, and others. Their 3.4.6 Fixed-Base Equipment
design parameters are provided by the fab-
ricators and typically include vertical and This may include regular stiff-leg (derrick)
horizontal wheel loads due to material han- cranes (Fig. 3-26), stiff-leg cranes sitting on
dling, and dynamic impact and inertia. Most a ring which can slew the crane through
often, maximum design wheel loads are 360° (ringer crane), loaders, such as that
limited to the design loads on a rail section depicted in fixed in-place revolving cranes,
and, similar to the crane corner loads, can miscellaneous loading towers, liquid bulk
be within 350-450 kN. handling arms (Fig. 3-27), and others.
In most cases, this type of equipment has The weights of the above equipment are
substantial windage area able to attract typically in a range 150-500 kN, with some
significant lateral wind force, so that the up to 15,000 kN, and in special cases, even

Figure 3-26. Stiff-leg crane (derrick).


Design Loads 283

Figure 3-27. Liquid bulk loading arms.

more. The foundation structure for this type with similar port characteristics, such as
of equipment is designed on a basis of load exposure to environmental forces, mode of
data and the operating criteria provided by operation, type of vessels, and others. How-
the manufacturer, along with associated ever, one should realize that the data
forces due to effects of environmental loads obtained from actual measurements of
and berthing and mooring forces as previ- berthing energies cannot be applied auto-
ously discussed in this chapter. matically to new port design. For example,
In addition to normal service loads, an data based on measurements on ships ar-
appropriate impact factor is used. Descrip- riving fully loaded cannot be used for load-
tion of some unique installations of this ing terminals where ships will arrive in
kind can be found in special publications by ballast. In the first case, the current could
ASCE, PIANC, The Dock and Harbour be a dominant factor in ship berthing en-
Authority, and others. ergy, and in the latter, the wind load may
be more dominant than current force. It
should be noted that actual berthing-
energy measurements usually show a
3.5 SHIP IMPACT (BY M. SHIONO significant scatter, which can only be de-
AND G. TSINKER> scribed statistically.
Furthermore, the statistical approach
Ship impact force on berthing structures implies that regardless of which design
can be determined by statistical, empirical, value of berthing energy is chosen there is
or theoretical methods, or by physical always a risk that it can be exceeded. Also,
andjor mathematical modeling. the more berthing operations the facility
The statistical method is based on in- (fender) is .exposed to, the larger this risk
formation obtained from prototype projects becomes.
284 Design Loads

Based on field observation data and risk According to Girgrah, Eq. (3-31) may be
analysis, Dent and Saurin (1969) suggest applied in the design of the end fenders and
the folloWing design values for tanker ter- isolated berthing points where the point of
minals located at exposed-to-Wind and contact is amidship. For interior fenders,
waves locations: the value should be reduce by a factor of
0.5.
1. 2.6 meter-tonnes (18.8 feet-kips) per 1000 It should be mentioned that both the
DWT of design ship at yield stress in the statistical and empirical methods are more
fenders often recommended for determination of im-
2. 1.55 meter-tonnes (11.2 feet-kips) per pact forces generated by large ships.
1000 DWT as a normal maximum related The theoretical (kinetic) method of
to approximately maximum working calculating the ship berthing energy is more
stress in breasting structures and fenders universal than both the statistical and em-
pirical methods and is also applicable to a
For berths located in protected harbors and ship of any size. It tends to account for the
for loading terminals where vessels nor- folloWing factors affecting dissipation of the
mally berth in ballast, five-eighths of these ship's total berthing energy:
values are recommended.
More details on statistical methods for 1. Rotation of the ship
the determination of a ship berthing energy 2. Elastic deformation of the berth structure
is found in works by Balfour et al. (1980), 3. Compression of the fendering system
Piaseckyj (1977, undated in 1982), and in 4. Displacement of water between the ship
PIANC (1984). and berth structure
In the empirical method, the folloWing
parameters are usually taken into account: Thus, in most cases, the fender system must
absorb orily a part of the ship's total kinetic
1. Berthing frequency of the design vessel energy, which is usually referred to as the
2. Exposure conditions "effective berthing energy" of a ship (E).
3. Number and maneuverability of vessels The effective berthing energy may be ob-
4. Berthing method: with or without tug tained from
assistance
5. Berth mode of operation; loading or un-
(3-32)
loading.

Specific values are assigned to the parame- where


ters based on judgment and experience and W1 = displaced weight of the vessel
are related to the cost of repair against both W2 =hydrodynamic or added mass
downtime and the probability of exceeding
Vn = component of the vessel's velocity normal
the ·fendering energy capacity. to the berth structure
Girgrah (1977) has concluded that
g = gravitational acceleration
berthing energy is empirically related to
CE = eccentricity factor
ship displacement (W1 ) and to the inverse
square root of vessel displacement as fol- C8 = ship andjor structure softness factor
lows.
The sum (W1 + W2 ), which is usually re-
ferred to as a ship virtual mass (W), is
wl typically expressed in tonnes. The approxi-
E = 120 + wts (t . m) (3-31)
mate relationship among the ship's dis-
Design Loads 285

placement tonnage and her gross, net, and semiopen) and spring properties of the
deadweight tonnages for a variety of vessels fenders
is given in Chapter 2. 2. Ship characteristics, such as its under-
Hydrodynamic mass, sometimes called water shape and draft, and the way she
"added mass," is the mass of a water en- approaches the berthing facilities (e.g.,
trained by a moving vessel as the result of a sideways, parallel to the pier, or straight
vessel's motion (Valentine, 1967). While in forward at a certain angle)
motion, a ship displaces water and pushes 3. Berthing velocity and underkeel clear-
ance
ahead of her a positive pressure field, ap-
4. Oceanographic and environmental fac-
parent as a raised water level. The dis-
tors, such as water depth and possible
placed water passes back along the hull.
impact of current, wind, and waves.
For a ship moving sideways, much of the
returned flow moves under the keel (Fig.
Some studies indicate that the added mass
3-28a), unless the underkeel clearance is
of many slender structures immersed in
too small, in which case the flow is forced to fluid is comparable to the mass of fluid
return around the stern and bow. displaced by these structures. On this basis,
In principle, the hydrodynamic mass of a and for simplification, some authorities rec-
ship moving in open water is approximately ommend using the value of hydrodynamic
10% of the ship's mass. It increases consid- mass W2 as equal to the weight of water
erably during sideways movement, such as contained in a cylinder that has a diameter
during the berthing process, because the equal to the vessel draft and a length equal
ship's mass is augmented by the large vol- to the vessel length. Therefore,
ume of water involved in the transverse
motion of the vessel toward the berth. This (3-33)
volume varies with the following factors:
where
1. Berthing facility characteristics, such as D = vessel draft
type of structure (open, solid wall, or L = vessel length
p = specific weight of water

According to Eq. (3-33), the added mass of


a fully loaded large tanker, or fully loaded
~
-
N•oal•v• wov•
Pf'OI)OQOtU '--------'

"''o• .,rMI.,
large cargo ship, becomes equal to 0.5-0.6
times the displacement tonnage.
Further recommendations to aid in the
aMJ IIIOW•F"\aftl
calculation of the hydrodynamic or added
(a) mass are available. The majority of these
recommendations describe the virtual mass
s... ....... /"···
..-~--'-'---..--1>...m in terms of its relationship with ship
diplacement tonnage.
Equation (3- 34) presents another means
of determining the virtual mass:
(3- 34)
(b)

Figure 3-28. Water movement and wave propaga- where em is the hydrodynamic mass factor.
tion during ship side-way movement. [After Ball, (1982) For the determination of em, Grim (1955)
and Middendorp, (1981).] proposed Eq. (3-35), which is based on re-
286 Design Loads

sults obtained from model tests: Prior to 1989, the Japanese standard
MOT considered formula (3-39) to deter-
1.8D mine the hydrodynamic mass factor, appar-
em= 1.3 + -B- (3-35)
ently based on a publication by Stelson et
al. (1955):
where
D = draft (in meters)
(3-39)
B = beam of the ship (in meters)

Saurin (1963) published a comparison on where


energy absorption by fender systems using D = draught of ship
theoretical data and data derived from full-
L = overall length of ship
scale observations. The observations were
based on 70 berthings of ships varying from p = specific weight of water at berth
16,000 to 50,000 DWT, with various under- wl = displacement of ship
keel clearances. Despite the scattered re-
sults of these tests, Saurin concluded that a According to formula (3-39) for fully loaded
value of em = 1.3 would represent a rea- large tankers and for fully loaded cargo
sonable average. However, the factual data ships, em approaches 1.5-1.6.
presented by Saurin suggests that to be on Blok and Dekker (1979) remarked that
the conservative side, a value of em = 1.8 expression (3-39) is basically the added
should be used. mass following from potential theory on
Vasco Costa (1964), on the basis of his sway oscillation for a beam-to-draught ratio
experiments, suggested some corrections to of 2 in infinite water depth. They also point
Grim's formula and recommended the fol- out that the value of em is greatly influ-
lowing expression for determination of mass enced by fender characteristics such as
factor em: spring rate and the character of its
load/ deflection relationship.
2D Fenders with different stiffness charac-
em= 1.0 +B- (3-36) teristics by as much as a 1 : 10 ratio were
used by Blok and Dekker in their tests. For
Ruppert (1976) suggested modification to the total impulse versus change of momen-
tum for all fenders they found a value of
formulas proposed by Grim and Vasco
Costa, which lead to somewhat lower em
em = 3.8 (with little scatter). The explana-
tion for the larger hydrodynamic mass fac-
values. According to Ruppert, em can be
tor probably lies in the fact that when a
determined from the following formulation:
light vessel hits a stiff fender, she tends to
1.5D
change her direction of movement very
em =0
. 9 +B-- (3.-37) rapidly. The resulting side movement, how-
ever, meets resistance from ambient water
as well as the momentum of ship move-
On the basis of model tests in five different ment, which prevents the ship from chang-
water depths (HjD = 1.07, 1.14, 1.90, 3.31, ing her forward movement rapidly. This in-
and 3.81), Girandet (1966) proposed the fol- evitably, results in a larger impact load
lowing equation which links em value to compared to that resulting from a collision
depth of water (H): between a ship and a soft fender.
On the basis of field observations and
em= 1.2 + 0.12(-D-) (3-38) model studies, Ueda (1981) concluded that
H-D the second part of Eq. (3-39) should be
Design Loads 287

increased by a factor of 2 and hence by imbalance between the flow displacement


equal to 0.57TDLpjW1 • The latter resulted by the ship and the underkeel flow (Mid-
in em values in the range 1.6-2.3. For cal- dendorp, 1981; Ball, 1982). The difference
culation of em value, Ueda proposed Eq. in water level across the ship produces a
(3-40) in which he introduced the. ship's hydrodynamic force toward the berthing
block coefficient ( eb): structure, which is appended to the ship's
own mass to give the "virtual mass."
(3-40) As it was mentioned earlier, when the
forward-moving ship collides with a stiff
fender at a certain angle, she rapidly
This value of em is currently recommended changes direction, which is resisted by the
by the latest issue of Japanese Standard ship's mass as well as by the resistance of
MOT. It should be noted that for eb = 0.785, ambient water. A laboratory investigation
the outcome of Eq. (3.40) is similar to that of the above mode of ship-dock collision has
obtained from Eq. (3-36) proposed by Vasco demonstrated that the magnitude of hydro-
Costa. dynamic mass depends greatly on the angle
From field measurements conducted from between the dock and approaching vessel;
1966 to 1968 in the port of Le Havre accordingly, a smaller hydrodynamic mass
(France) at oil tanker berths receiving is associated with a small approach angle
tankers ranging from 70,000 to 130,000 (Gorunov, 1980).
DWT despite inherent inaccuracies in the Contrary to an open berthing structure,
observations and calculations and in the when a ship approaches a solidwall struc-
variation of D jH ratios, the value of em ture, a gradual cushioning effect occurs, in-
ranged between 1.4 and 1.7, with an aver- creasing with the decreasing distance from
age close to 1.5 (PIANC, 1984). the structure.
The numerous studies conducted in In general, for a preliminary evaluation
Japan (Motora, 1959; Nagasawa, 1959; of the value of added mass, some authori-
Heyashi, 1962; Fujino, 1968; Mizoguchi and ties recommend considering the following
Nakanayama, 1973; Otani et al., 1974; and factors:
others) have produced various values of em
in a range between 1.9 and 3.6. 1. 1.0 for an open structure of any dimen-
The value of the hydrodynamic mass is sions
largely dependent on the berth structure
2. 0.8 for a semisolid-wall structure longer
geometry. For example, an open piled struc-
than the length of the ship
ture allows water from around the ship to
move relatively unimpeded past the piles. 3. 0.6 for a solid-wall structure longer than
On the other hand, a solid-wall structure the length of the ship
considerably obstructs the flow, particularly
if the underkeel clearance is small. When Similar recommendations are found in
the ship suddenly stops upon impact with PIANC (1984).
the berthing structure, such as an open It should be also noted that the em value
piled structure, the flow displacement rate may be affected by currents (Tyrell,
drops suddenly to almost zero, causing the 1966/1967). For large ships, unless the de-
water level to fall on the ship's side nearest signer has good reason to adopt other val-
to the berth and rise on the other side (Fig. ues, Permanent International Association
3-28b). This phenomenon is a result of the of Navigation Congresses (PIANC, 1984)
momentum of the mass of water under the recommends that the value of hydrody-
keel, which maintains an underkeel flow namic mass (W2 ) range between 1.5W1 (for
when the ship is stopped, thus creating an very large underkeel clearance, say 0.5 X
288 Design Loads

draught) and 1.8W1 (for very small under- much larger berthing energies to be ab-
keel clearance, say 0.1 X draught). sorbed by the fender system. Therefore, the
Depending on site conditions, fender ship's approach velocity data should not be
stiffness, and flexibility of berth structure, relied on exclusively but used as a starting
em values for smaller ships can be much point in fender system design. A discussion
greater than those for large ships. It should with local pilots who are going to use the
be noted, however, that some earlier field berthing facility is also recommended.
measurements of ship berthing velocity The latest technology, such as sonar sys-
show no correlation with vessel size (Dent tems (Fig. 3-29), provides dock and ship
and Saurin, 1969; Svendsen, 1970). operators with a means to keep the berthing
The dock designer must always be aware vessel under better control while docking.
that in all cases of the ship's direct ap- Docking sonar systems assist in safe and
proach to the berthing structure, the value controlled berthing operations by providing
of added mass should be determined by an accurate and highly stable measurement
exercising good judgement and by evaluat- of a vessel's approach speed and distance
ing all berthing conditions previously dis- from a berthing structure. Distance and
cussed. A degree of uncertainty in the speed are measured at the bow and stern
mechanism of ship-dock interaction does independently and are instantly available
present a problem in determining the value to both the pilot on board and the dock
of the added mass, which, apparent in the personnel throughout the berthing opera-
above mathematical equations, probably tion. The data are also printed out for sub-
cannot be solved with mathematical preci- sequent analysis.
sion without further research. The above Docking sonar systems have been used
frequently at oil and LNG terminals which
speculations on a part of the hydrodynamic
typically receive large tankers. However,
mass factor, em, indicate that it has been
because of the relatively high cost of this
and still continues to be a matter of debate
technology it is seldom used in dry bulk and
among practitioners.
general cargo ports. One example of sonar
Despite various methods that have been
system use at a bulk loading facility is
suggested to determine the value of the Ridley coal loading terminal, Price
added mass, most practitioners in the field Ruppert, Canada (Port Development Inter-
still rely heavily on past practice, good national, February 1985).
judgment, and experience. The proper selection of ship speed is very
Approach velocity and angle are very im- important in arriving at a realistic value of
portant components to be considered in the kinetic energy to be used in the design of
evaluation of a ship's berthing energy. As the fender system and subsequently the
seen from Eq. (3-32), the approach velocity berthing structure. As a general recommen-
is an important parameter in determining dation, the following berthing velocities can
energy to be absorbed by the fender system. be assumed for the design of a berthing
The designer must always realize that the structure.
ship's approach velocity is almost entirely
dependent on the size of a ship, mode of 1. For tug-assisted large ships
approach (e.g., assisted by tugs or self- • very favorable conditions, 10 cmjs
propelled), human factors, environmental • in most cases, 15 cmjs
conditions, part'iculars of the site, and the • in very unfavorable conditions with
skill of local pilots. A wrong maneuver of cross-currents and/or much wind, 25
the ship due to human error or a sudden cm/s
change in weather conditions can results in 2. For any ship approaching the dock with
a much larger approach velocity and, hence, no tug assistance, 20-25 cm/s
Design Loads 289

(a)

(b ) (c)

Figure 3-29. Docking sonar system: (a) general view; (b) and (c) display board.

In the first case, the design of the terminal The energy to be absorbed by the dock
is usually based on an approach angle of fender system is usually taken as equal to
10°. In all other cases, an approach angle of one-half of the ship's kinetic energy, based
10°-20° is used based on local approach and on CE = 0.5. The remaining half is assumed
environmental conditions, The larger ap- to be absorbed by the ship and water due to
proach angles are more characteristic for ship's rotation about her contact point with
inland ports, and smaller angles are most the fender. This point is commonly referred
often in seaports receiving large ships. to as a "quarter point" (Fig. 3-30). For other
The likelihood of exceeding these design berthing points, the value of CE can be
berthing velocities is great; hence, caution obtained from Eq. (3-41) or Fig. 3-31.
should be exercised in selecting ship design
approach velocities and subsequently fender
system design. For more information on
(3-41)
berthing velocities, the reader is referred to
PIANC (1984).
290 Design Loads

where When the vessel is brought to rest along-


l = distance from the gravity center of a ship to side of a stiff marginal wharf structure, it is
the point of contact, measured along the face assumed that the impact energy is primar-
of the pier (Fig. 3-30) ily absorbed by the fender system only. Fur-
r = radius of gyration of a vessel's gravity center thermore, it is usually considered that the
on the horizontal plane, commonly, r = potential deflection of any type of continu-
0.25£, where L is the full length of the ship. ous berth structure cannot contribute to ab-
sorption of the energy, as the time it takes
for the structure to react is much longer
than the time over which the impact is
applied. To calculate the impact force re-
sisted by the continuous, or otherwise very
stiff, structure it is necessary to consult
load-deflection curves provided by the
fender manufacturer.
In contrast, flexible breasting structures
(e.g., flexible dolphins and similar struc-
tures) greatly contribute to the absorption
of ship berthing energy (Fig. 3-32). As men-
Figure 3-30. Berthing ship. tioned earlier in this section, rubber tires
are used often as fenders in small-craft har-
I0 bors. With varying degrees of success, they
~0.8
~~ are also used to absorb berthing energy
from larger ships up to 3000-5000 DWT.
•.
~
i or, ~ Load-deflection curves for used tires, or
""-....
Jo.• -...__r---
~
............. other fender systems which include used
tires (e.g., tires filled with a closed-cell foam
01
(Tsinker, 1986), big tires stuffed with
0 IL O.lL II• L O.JL O.•L O.SL
smaller ones, and others), can be obtained
D«tbl"' point'ohhr wnKI (A) experimentally.
Figure 3-31. Eccentricity factor versus vessel As reported by Hendrick (1980), used
berthing point. rubber tires and particularly those of large

.."'
... Pile t Fender
"'
Pile ~ """'
Sea bed
Deflection
(a) (b )

Figure 3-32. Energy absorption by combination of flexible dolphin and fender: (a) flexible
dolphin/fender system; (b) system reaction versus deflection.
Design Loads 291

diameter, can absorb substantial amounts determination of the ship impact load that
of energy, especially when assembled in will act on a marine structure. Practically,
groups of three and more tires. On the basis it is done by consulting the reaction
of rather comprehensive tests that have force I energy absorbtion in fender deflection
been carried out at the Building Research curves, given in the fender manufacturer's
Structures Laboratory of the Canadian catalog. Normally, these curves are ob-
National Research Council, Hendrick has tained from the fender load tests that are
arrived at the following basic conclusions: carried out in accordance with recommen-
dations established by PIANC (1984). Ac-
1. The condition of the tire greatly influ- cording to these recommendations, in order
ences the ultimate load on it; factors to be to obtain the aforementioned curves, the
considered are wear, amount of tread re- fenders are tested at relatively low com-
maining, whether or not the tire has been
pression speeds. This, according to
retreaded, cuts (size and location) and
bruises. Kiedaisch (1995), results in the unneces-
sary redundancy ("reserve of energy absorb-
2. The tire construction and type or tread
design may influence the energy absorb- tion capacity") and therefore in fender
tion capacity and the resulting reaction overdesign. By the time this book is pub-
force. lished at least one fender manufacturer is
using the so-called "velocity correction fac-
Performance of used tires as a fender can tor" to account for a ship actual berthing
be enhanced greatly by filling them with a velocity. This resulted in smaller more eco-
closed-cell resilient foam (Tsinker, 1986). nomical fender units recommended by this
In experimental work, the F = f(dr) manufacturer for dock fendering. Padron
curve is obtained first. Here F is the force (1994) disagrees with this practice. He ar-
applied to the fender and d f is the corre- gues that the velocity correction factor, as
sponding fender deflection (Fig. 3-33). The currently applied, is based on the assump-
area between curve F = f( d f) and the axis tion that the fender unit is compressed at a
dr represents the work (energy) done by the constant velocity throughout the impact cy-
fender. Hence, the curve E = f(d,), which cle, while in practice this velocity varies
is an integral curve in the relationship to from the initial impact velocity to zero as
the curve F = f(d,), can be plotted by inte- the vessel comes to a stop. In the process
grating the appropriate areas. the fender unit deflects at a continuously
The design berthing energy of a ship is decreasing rate. Padron pointed out that
used for selection of a desired energy ab- use of the velocity correction factor does
sorbtion (fender) system and ultimately for have validity with respect to the maximum
reaction force that the fender unit imposes
on the structure because it is important to
the designer to be fully aware of the peak
reaction force for which the structure must
be designed. Padron further suggests that
normally the velocity correction factor
should be applied when considering the
magnitude of the reaction force that the
fender unit can be expected to exert on
the structure, and not applied to fender
Figure 3-33. Typical energy versus deflection dia- energy absorbtion capacity because, as
gram obtained from fender test. presently used, the velocity correction fac-
292 Design Loads

tor tends to overstate the fender energy structure interaction phenomenon. How-
absorbtion capacity. More recent investiga- ever, data obtained from scale modeling
tion carried out by Hayama and Sakuraoka usually suffer from the scale and viscosity
(1996) basically supported the opinion ex- effects and, therefore, require experienced
pressed by Padron. interpretation. In practice, scale modeling,
Certainly, more research is needed to es- and small-scale modeling in particular, can
tablish the effect of ship berthing velocity be interpreted as qualitative only and
on fender performance. This is particularly should not be used quantitatively in the
important for performance of relatively stiff practical design, unless the proper compari-
buckling-type fender units. This also ap- son with prototype conditions is available.
plies to magnitude of the other correction The mathematical modeling approach
factors that should be used to account for to the ship berthing phenomenon is still
effects of fender aging, temperature, repeti- under development, and comparison with
tive compression, etc. All this is on PIANC full-scale berthings are still scarce. Some
agenda for years to come (PIANC, 1995b). information on the subject matter is found
As pointed out earlier, the approach veloc- in Van Oortmerssen (1974), Fontijn (1980),
ity and angle are the most important pa- and Middendorp (1981). Mathematical mod-
rameters that affect the berthing energy eling, which potentially can provide better
and ultimately the ship impact load on a theoretical insight into the mechanisms of
structure. ship-fender-structure interactions with
Unfortunately, in real life the actual ves- consideration of such parameters as added
sel approach velocity, due to variety of rea- mass, underkeel clearance, stiffness offend-
sons, sometimes exceeds the design values. ers, approach angle, and others, undoubt-
This results in damage to fender units, dock edly will be developed to the degree neces-
structure, and sometimes to a ship. This is sary to enable it to be used in practical
why in this writers' opinion the dock fender applications.
system should have a sufficient redundancy Fontijn (1987) has developed the Impulse
and in most practical cases should not be Response Function (IRF) method to de-
designed with mathematical precision. The . scribe the ship-berth interaction phe-
designer must be aware that the cost ofthe nomenon. In the model developed by
dock fender system in most practical appli- Fontijn, the ship is treated as a rigid pris-
cations contributes approximately 3-5% or matic body moving in surge, sway, and yaw
less of the total cost of the dock. Hence, an through still water of uniform depth (i.e., no
insignificant reduction in the cost of the waves or currents). Headland (1992a and b)
fender system will practically not affect the compared both the kinetic energy and IRF
total cost of the project; however, it may methods will full-scale measurements, re-
significantly affect safety of dock operation. ported by Lee (1965, 1966). The comparison
Finally, it should be noted that in a indicated that both methods gave nearly
structure-fender-ship system a portion of identical results. Furthermore, both meth-
the kinetic energy is absorbed by a ship's ods were partially verified by Lee's full-scale
hull elastic deformation. As plastic defor- measurements. A complete verification
mations of a ship's hull are unacceptable could not be found due to inconsistency
and elastic deformations are very small (according to Headland) in Lee's data.
compared with those of fenders, the amount At present, the kinetic energy approach
of energy absorbed by a ship is negligible with many refinements as discussed in this
and is usually ignored (softness coefficient section is still considered as the most prac-
C8 = 1.0). tical method of determining the amount of
Physical scale models are sometimes kinetic energy to be absorbed by a fender
used to determine and study the ship- system. However, again, a sound judgment
Design Loads 293

always must be exercised by the designer not reach its maximum value and the ice
during selection of design parameters to be floe may be stopped or change direction of
used in calculations. motion without exerting a maximum load
PIANC (1984) recommends that the sta- on the structure.
tistical method be restricted to new fender The applicability of either of the afore-
system designs on berthing structures at mentioned conditions depends on the
sites where measurements have been taken ice-structure interaction scenario and the
until sufficient measurements are accumu- magnitude of the driving forces in compari-
lated from a number of locations. PIANC son with the potential local forces between
(1984) also recommends that both kinetic the ice and the structure. Examples of the
energy and statistical approaches be consid- above conditions may be the movement of
ered as complementary. the first-year ice sheet against the struc-
ture, or the interaction between structure
and a large ice feature such as multiyear
3.6 ICE LOADS floe, ice ridge, or similar.
In the former case, the ice sheet is usu-
ally large enough so that the structure can
be fully embedded in the ice, and the ice
3.6.1 General load can be calculated from the local loads
along the length (width) of the structure. In
The ice load depends on port geographical
the latter case, if the driving forces are
locations, harbor and waterway geometry
relatively small in comparison with the po-
and protection, ice control and ice manage-
tential local forces, the maximum ice force
ment in berth areas, and geometry of the
dock structure. may not be reached. This would mean that
The motion of ice against the structure the use of potential ice forces rather than
could lead to the generation of substantial those that practically could be reached could
forces which could be of a global or local lead to the overestimation of the loads.
nature. The global ice loads in some cases The ice loads also depend on the defor-
may control general stability of the struc- mation behavior of ice (e.g., bending, buck-
ture; the local loads may determine the re- ling, .compression, shear), which, in turn,
quired strength of some structural compo- depends on the geometry of the structure
nents. Naturally, the ice generated loads (e.g., vertical or sloping). Vertical structures
depend on ice mechanical properties (e.g., would typically include narrow vertical
crushing and flexural strengths and shear structures such as column (pile) supported
resistance) and the driving forces such as platforms, as well as solid vertical walls.
wind, tides, currents, and waves. If the Sloping structures are those with sloping
driving forces are sufficiently high, the sides, intended to fail ice in bending.
structure may be fully enveloped in the ice Ice sheets, or ice floes, are common ice
feature (i.e., ice floe). In this case, the inter- features to be considered in port marine
face area between the ice feature and the structures engineering. When engaged with
structure is at its maximum and, subse- a structure, they can crash or creep
quently, the ice load exerted on the struc- in the indentation mode, buckle, or fail in
ture reaches its maximum value. On the bending.
other hand, the ice-structure interaction Ice movement against a vertical struc-
may end at an earlier stage if the driving ture is one of the most common ice-
forces are insufficient to overcome forces structure interaction scenarios in marine
exerted by the structure on the ice feature. engineering. On the basis of field observa-
In the latter case, the contact width does tions and test results, Sodhi and Nevel
294 Design Loads

(1980) concluded that for ice sheets collided force (Tc) acting on a floating ice sheet:
with vertical structure, the occurrence of a
crushing or buckling mode of failure de- (3.43)
pends on the aspect ratio D jt, where D is
the width of the structure and t is the
where
thickness of the ice sheet. Buckling was
observed to occur for higher aspect ratios Yw = density of water
(generally greater than 6), whereas a crash- V0 = current velocity under the ice feature
ing failure is characteristic for aspect ratio H 0 = depth of water under the ice undersize
smaller than 6.
For comprehensive treatment, the sub- 2/3
ject of ice physical properties and the modes n = [ nP; n~-5 ] (3-44)
of ice-structure interaction, the reader is
referred to Tsinker (1995).
In Eq. (3-43) and (3-44), n 1 and n 2 are
Manning's roughness coefficients respec-
tively for interface between the water and
3.6.2 Environmental the ice sheet, and the water and the bottom.
Driving Forces They very considerably. For example,
0.005 < n 1 < 0.04 is used, depending on the
Ice motion in rivers and natural harbors conditions of ice formation and accumula-
which is associated with spring breakup tion. More details on the values of n 1 is
begins with local melting and weakening of found in Michel (1971). Extensive discus-
the ice cover. The following massive ice sion on the values of the bottom roughness
movement is typical in many rivers and coefficient n 2 is given in Van Te Chow
large sea harbors. (1959). Note that in Eq. (3-43) and (3-44),
Ice motion in large bodies of water (e.g., SI units are used.
natural harbors or lakes) is caused mainly According to Tryde (1989), the current
by wind friction. The air friction force ex- shear on the underside of the ice sheet can
erted on an ice cover (Tw) can be calculated be expressed as follows:
as follows: (3-45)

(3-42) where
Pw = density of water
where Vwt = velocity 1 m below the underside of the ice
sheet
p = the air density; p = 1.34 kgjm 3 at -10°C.
C = the drag coefficient at the 10-m level. An The maximum value of environmental driv-
average value of C = 0.0022 for a rough ice ing forces is reached when wind- and
cover could be used (Croasdale, 1980). Ac- current-driven forces coincide. In real de-
cording to Tryde (1989), C = 0.006. signs, the environmental external forces on
Vw = the wind speed at the 10-m level. the ice floe depend on many factors, such as
A = the fetch area. ice accumulation, shape and curvature of
the river or navigational channel, geotech-
In rivers and navigational channels, water nical characteristics of the riverbanks, and
friction forces are the principal causes of ice other objects restricting ice movement. More
motion. Michel (1971) developed the follow- information on this very special subject is
ing equation to determine the water friction given in Michel (1971), Eranty and Lee
Design Loads 295

(1981), Fenco Engineers (1984), and in more (F) is expressed as


recent work by Wake et al. (1987).
Fe= pDh (3-46)
where
p = effective ice stress
3.6.3 Ice-crushing Load D = width of indentor
h = ice thickness
By definition, the ice-crushing process is
the complete failure of granularization of Both empirical and theoretical methods are
the solid ice sheet into particles of grain or used to determine value of p for the partic-
crystal dimensions; no cracking, flaking, or ular ice-structure interaction conditions.
any other failure mode occurs during pure In empirical methods, the effective ice
crushing. The immediate clearing of the pressure is usually related to the ice uniax-
broken ice by extrusion along the structure ial compressive strengths by the introduc-
wall or indentor face follows the failure. tion of a number of coefficients to account
When the ice is sufficiently thick, crush- for contact conditions, indentation, and
ing is one of the most common modes of shape of the structure.
failure, particularly for ice interacting with Korzhavin (1962) proposed the empirical
narrow vertical structures. Kry (1980) noted relationship (3-47) for p which, according
that the ice-crushing mode of interaction to Neill (1976), is valid for strain rates of
follows the indentation problem. where a 10- 3 -10- 4 s- 1 and gives values of the
triaxial state of stress develops in the ice effective pressure in the range of 0.9uc <
due to the confinement effect of the ice p < l.6uc.
sheet in the ice-structure interaction area.
As summarized by Blauchet et al. (1989), (3-47)
an observation of indentation tests shows where
the following: I = indentation coefficient which accounts for
the confining (scale) effects on ice compres-
• First, microcracks are formed in the vicin- sive strength.
ity of the indentor. fc = contact coefficient which accounts for nonsi-
• Immediately in front of the indentor, ice multaneous contact between the indentor
then separates into fine grains of dimen- and the ice feature. Essentially, for perfect
sions less than 1% of the indentor width. contact, fc = 1.0. Generally, fc is taken as
• The failed ice is ejected out of the contact 0.4 < fc < 0.7.
zone and accumulates in front or on the m = shape factor which takes into account shape
sides of the indentor. of the indentor. It is usually taken as equal
• The actual tests, measured periods of to 1.0 for flat and 0.9 for circular and semi-
crashing loads decrease from 1 s to 0.06 s circular indentors, and 0.85(sin a) 0 ·5 for in-
with the increase in penetration rate from dentors with wedge angles of 2 a between
0.01 mjs to 0.21 mjs. 60° and 120°.
• The size of the ice pieces decreases with uc = uniaxial unconfined compressive strength
the increase in the penetration rate but of ice.
does not very with the aspect ratio.
• The recurrence period of crushing varies I is a dimensionless number and is the
from 0.02 s to 1 s. ratio of the observed failure pressure to the
measured unconfined compressive strength
On the premise that ice failure due to frac- for the same ice sample (e.g., for a sample
ture does not occur, the load exerted by the the same crystal size, temperature, brine
ice sheet due to crushing mode of failure content or salinity, etc.). There are several
296 Design Loads

Table 3-14. Values of contact coefficient fc sional, where I depends on the aspect ratio
Velocity of Ice Flow
D jh as follows:
Width of Movement (mjs)
Rough contact:
Indentor (m) 0.5 1.0 2.0

3-5 0.7 0.6 0.5 0.35


6-8 0.6 0.5 0.4 I= 1.45 + Djh ~ 2.57 (3-49)

Source: After Korzhavin (1962).


Smooth contact:
different ways to calculate the value of I:
0.37
I= 1.15 + Djh (3-50)
• Assur (1975) recommended that I be de-
termined as equal to I= 1 + 2CI-D/h),
which gives a value of I = 3 at D jh = 0.
• As the indentor size increases, the nature A number of other theoretical solutions have
of the stress field in the ice more closely been proposed. These are summarized in
matches that of plane stress, and the ef- Tsinker (1995).
fective contact crushing stress is reduced It must be pointed out that results of
(Ralston, 1977). relatively recent field experiments reported
• An alternative approach is found in Kry by Danielewicz and Blanchet (1987) and
(1978) and Iyer (1983).
Johnson and Benoit (1987) suggested that
• To some extent, settling of fc to less
than 1 accounts for the brittle nature of
the real ice loads can be significantly lower
the ice and its nonsimultaneous failure. than those measured at smaller scale or
Korzhavin (1962) recommends the values predicted by presently used theoretical
of fc depending on the velocity of ice move- models.
ment as shown in Table 3-14. The ice crushing failure at a wide verti-
• Michel and Toussaint (1977) suggested cal structure (D jh ~ 5) is somewhat dif-
that Korzhavin's formula should be modi- ferent from crushing against narrow verti-
fied to account for strain rate. cal indentor. At wide structure because of
• Croasdale et al. (1977) applied the theory rugged surface of the ice sheet the failure of
of plasticity to the solution of the indenta- the ice sheet edge does not take place si-
tion problem. The latter authors suggest multaneously along its entire length but
that the value of the effective stress can be
only at a few placed along the edge.
determined by
The latter results in dramatic decrease
(3-48) in a global ice pressure upon structure.
Bercha (1986) provides the most recent re-
where I is the indentation factor which view of the state of the art on the effect of
can be determined from upper- and lower- the structure size on ice pressure. A com-
bound solutions. prehensive discussion on mechanism of the
wide structure-ice interaction is given in
Expression (3-48) assumes perfect contact Fenco Engineers (1984).
between the structure and the indentor. As- It should be noted that, at present, the
suming a flat indentor, the problem reduces use of empirical data has become the norm
to the classical Prandtl indentor, for which when it is desired to estimate ice loads on
I= 2.57. For homogeneous and isotropic ice, wide vertical-sided structures. Af3 a rule of
provided that Dis much larger than h, the thumb, Bercha (1986) recommends compu-
lower-bound solution equals 1.0. Between tation of the first-year ice sheet load on a
these limits, the problem is three dimen- wide structure by multiplying ice contact
Design Loads 297

area (Dh) by 1.75 MPa. In the latter case, sheet into blocks typically four to five thick-
D is the width of a structure if it is smaller nesses in diameter (Blanchet et al., 1989).
than the ice sheet; on the other hand, it As an ice sheet approaches a sloping
may be considered as the effective length of structure, at initial contact it begins to
the ice in contact with a structure in case crush at the interface (on the underside of
an ice sheet is smaller than structure. the ice sheet for an upward breaking slope).
Ice-crushing load is basically characteris- The resulting interaction force, acting nor-
tic for the vertical face structure-ice inter- mal to the face of the structure, has a verti-
action. However, this kind of load can also cal and a horizontal component. There will
exist under conditions where ice sheets in- also be a frictional force along the slope.
teract with an inclined structure with an- The vertical and frictional components pro-
gles to the horizontal of greater than 75° duce bending in the ice sheet and the ice
(Michel, 1978). For structures with inclined will fail in bending when these components
surfaces, the vertical component of an ice- are increased to a certain critical level. Once
crushing load should be considered. the ice sheet has failed, the smaller pieces
Marcellus et al. (1987) and Timco (1987) of ice are pushed by the advancing ice sheet
discuss ice sheet interaction with multi- and begin to ride up the face of the struc-
legged structures (e.g., piled piers, offshore ture. This causes a larger interaction force
platforms, etc.). For more information, con- to be generated, because additional force is
sult Tsinker (1995). required to push the broken pieces of ice up
the structure.
On a wide structure, the ice may get
turned back on itself, creating additional
3.6.4 Loads Due to Ice Bending ice on the slope of the structure, which may
Mode of Failure lead to the creation of an ice rubble in front
of the structure. The latter, in turn,
These loads are characteristic for the slop- may inhibit simple bending failure of the
ing or cone-shaped structures which tend to advancing ice sheet against the sloped
induce bending failure in ice sheet. The structure.
resulting ice load is substantially lower as Croasdale (1980) presented a two-
compared to the case where the ice sheet dimensional analysis model of ice interac-
fails in the crushing mode. For example, tion with a sloping structure in which the
PIANC (1984) suggested that when the wall horizontal H and vertical V compc>nents of
is sloped greater than 25° from the vertical, the normal load N are determined as
the resulting ice load is reduced by a factor follows:
up to 4 in comparison with the vertical wall.
Actually, two basic loading conditions can (3-51)
H=Nsina+fNcosa
result in ice bending failure. The first one
may occur due to ice compressive loading in and
combination with ice sheet natural asym-
V = N cos a - fN sin a (3-52)
metry and eccentricity of the load which
results in tensile stresses in ice sheet. The
second process occurs when a vertical load where f is the friction coefficient and a is
is applied to the edge of an ice sheet as a the slope angle with the horizontal.
result of riding up (or down) the side of a When the moment capacity of the ice
sloping (conical) structure. The low tensile sheet is related to corresponding vertical
strength and limited ductility of ice result force required to initiate failure, the hori-
in the formation of cracks, which break the zontal force per unit width of the structure
298 Design Loads

is computed by component is proportional to about h 5 and


the ride-up force is proportional to ice thick-
ness. More information on the numerical
H ( Pwghs ) 0 ·25 (3-53) estimation of ice forces acting on inclined
D = 0.68ur - E - C
structure is given in Lindstrom (1990).
For inclined narrow or conical structures,
where the failure zone typically extends around
the structure, and usually not all broken
D =width of the structure
pieces of ice are riding up the slope. Typi-
ur = ice bending strength
cally, they clear around the structure. For
Pwg = weight density of water
specific information on ice-conical struc-
h = ice thickness ture interaction the interested reader is re-
E = elastic modulus of ice ferred to Edwards and Croasdale (1976),
C = coefficient which relates H and V with Manas'ev et al. (1971), Ralston (1979),
respect to a; that is, H = VC, where C = Frederking (1980), Nevel (1972), Morrison
(sin a+ {cos a)j(cos a- {sin a)
et al. (1988), and Wessels and Kato (1989).
Very little experimental data exists on
Once the ice has failed, the broken pieces downward ice breaking structures. Some
start to ride up the face of the structure and preliminary data were reported by Abdel-
an additional force is experienced by the nour (1981). His experiments indicated re-
structure; the latter is computed from ductions in forces by a factor of about 2,
compared with upward breaking. Results of
model tests on a downward icebreaking cone
carried out in an ice tank are reported by
Lau and Williams (1991). Effects of some
parameters, for example, ice thickness, ap-
where proach speed, and cone waterplane diame-
cl = o.6BC ter, on ice load are discussed.
C2 = C(sin a+ {cos a)+ Most recently, Nixon et al. (1993) pub-
(sin a+ fcos a)jtan a lished results of a total 57 tests carried out
z = the maximum ride-up height on a model of a cable-moored floating plat-
Pig = the weight density of ice form with inverted conical geometry. The
basic findings from this study were that the
As follows from the two-dimensional model forces on the structure increased monotoni-
of the ice forces acting on the sloping struc- cally with ice layer thickness, that reso-
ture, the effect of friction and slope angle nance effects and the process of ice collar
becomes significant above and angle of 45°. formation around the platform complicated
[Some examples are provided by Marcellus the variation of forces with ice speed such
et al. (1987).] The ice strength affects the that a clear trend of platform reaction
icebreaking component but not the ride-up (mooring) forces with ice speed is not read-
component, and in two-dimensional elastic ily apparent and that stiffness of moorings
analysis, the ride-up force in a typical inter- has a marked effect on a platform resisting
action is larger than the breaking force. forces.
Ice thickness is the most significant When considering the appropriate value
parameter affecting ice loads on sloping to use for ice flexural strength in ice loading
structures. As follows from Eq. (3-54), in equations, it is necessary to recognize that
two-dimensional analysis the icebreaking ice flexural strength can vary as a function
Design Loads 299

of the following: greater than the largest tests to date, is


still an issue for speculation.
• Salinity and temperature distribution For a comprehensive review on the state
• Strain rate of the art on ice sheet-conical structure
• Crystal structure interaction, the reader is referred to
• The size of the ice beam (or thickness of Marcellus et al. (1987). Some additional
the ice sheet) information on the subject is found in
Izumiyama et al. (1991).
Whether the same flexural strength should
be used for upward breaking and downward
breaking will depend on the salinity and
temperature profiles through the ice. In this 3.6.5 Forces Due to Ice Sheet
context, it is also necessary to recognize Adfreeze to the
that moment capacity is perhaps more rele- structure
vant than flexural strength, as the position
of the neutral axis within the ice will also The adfreeze forces may arise if an ice sheet
vary with salinity and temperature profiles. has developed a bond with the structure
The issue to size effects in flexural fail- during a period of no ice movement, which
ure is important because in most real situa- is characteristic for the nearshore arctic en-
tions involving ice features they are much vironment and inland waterways where the
larger and have been tested in the measure- ice surrounding a structure can remain sta-
ment of flexural strength. It is not clear, tionary long enough to freeze to the struc-
however, whether a size effect exists in ture. There, the vertical motions of the ice
flexural failure. The ice flexural strength due to tidal action or water-level fluctuation
values discussed in the literature show a can be so small that an adfreeze bond can
range from nearly zero to about 3.5 MPa, be developed between the ice sheet and the
but it is clear that the high values have structure. Once the ice sheet starts to move
been obtained from small-scale tests. This again, sliding motion between ice and a
trend is confirmed by Marcellus et al. (1987), structure first requires breaking the ad-
who indicated that for large beams a flexu- freeze bond. In the process, ice adhering to
ral strength value of 0.7 MPa or less would marine structures may cause substantial
be appropriate. The results also show that horizontal andjor vertical forces on these
multiyear sea ice appears to be weaker than structures. According to Cammaert et al.
low-salinity columnar sea ice. The appropri- (1986), the load required to fail an adfreeze
ate value of flexural strength to use in ice bond on ~ conical structure can be much
load calculations for typical multiyear ice larger than the load associated with bend-
features, with cross sections 10-500 times ing failure.

Table 3-15. Values of factor I in Eq. (3-55)


Failure
Zone Cone Angle
Angle 8 40° 50° 60° 70° goo

75° 1.524 1.672 l.g92 2.240 2_ggg


goo 1.612 1.760 1.9g1 2.329 2.9g7
g5o 1.700 I.g4g 2.069 2.417 3.066
goo 1.7g7 1.936 2.157 2.505 3.152

Source: From Cammaert et a!. (1986).


300 Design Loads

As suggested by Cammaert et al. (1986), With epoxy coated steel, the strength was
the load required to break the bond be- 0.13 MPa.
tween the ice sheet and a conical structure Alliston (1985) reported results of com-
over an angle 2 8 can be obtained from prehensive laboratory tests of low friction
and adfreeze coatings to reduce frictional
CaCsDhTai resistance and adfreeze.
Ha=--.-- (3-55)
Sin a As stated earlier, ice adhering to the ma-
rine structure may cause substantial verti-
where cal uplift or downgrade load. For example,
ca =adfreeze factor to account for incomplete on large linear structures, the uplift load
bonding: ca = 0.3 to 1.0 may range from 15 to 30 kN jm around the
Cs = stress factor to account for nonuniform periphery. Downward loads will result when
stress distributions: Cs = 0. 7 to 1.0 water levels fall. Hanging ice may span
D = structure diameter at the water level distances of 5-10 m between vertical
a = cone angle to the horizontal supports.
I= elliptic integral which varies with a and 0, An upper limit to these loads is provided
the failure zone angle; representative val- by the adhesion bond between the ice and
ues of I in Table 3-15. the structural material, shear strength of
ice, or the bending strength of ice sheet.
Other models to consider the effects of The limit force of two former conditions is
adfreeze forces on ice horizontal forces on calculated from
marine structures have been proposed by
Croasdale (1980) and Gershunov (1985). It
(3-56)
should be noted that all three recommenda-
tions produce substantially different force
values. where
With regard to the adfreeze shear T = ice adfreeze (shear) strength
strength, there have been several studies S = adfreeze (shear) area
performed, usually, on a very small scale
and at high strain rates. The range of val-
ues obtained from these studies is quite
wide. For example, Sackinger and Sackinger 3.6.6 Vertical Loads on Piles or
(1977) measured about 0.5 MPa with sea ice Piers Due to Changes in
and uncoated steel at - 15°C. At higher water Level
temperatures, the adfreeze strength was
somewhat less. As water-level changes through, for exam-
Saeki et al. (1981) measured adhesion ple, tidal action or seasonal fluctuation, the
strengths for steel, concrete, painted steel, ice sheet adhering to a pile or pier can
and corroded steel for sea ice at - 2.0°C. deflect and exert a vertical force on it. The
Typical values for clean steel were in the problem of computing vertical forces on piles
range 0.1-0.3 MPa, for painted steel much is usually solved using the theory of a
lower, and for corroded steel in the range of floating elastic ice plate subjected to static
0,4-0.56 MPa. loads with the following governing differ-
Oksanen (1983) performed adhesion ential equation (Kerr, 1976, 1978, 1986;
strength tests with fresh ice on a variety of Gold, 1984):
materials at temperatures of - 5°C, - 10°C,
and - 15°C. For uncoated steel he meas-
ured 0.46 MPa at all temperatures. (3-57)
Design Loads 301

where intact ice sheets ranging from 100 to 400


Di = flexural rigidity of the ice plate: Di = kN. In practice, these loads can be mini-
Eh 3 /12(1 - v 2 ), where E is Young's modu- mized by reducing ice adhesion values to
lus, h is the ice cover thickness, and v is piles. This can be achieved by using low-
the average Poisson ratio across the ice friction materials and miscellaneous spray-
cover thickness. on coatings, jackets, and wrappings around
V4 = biharmonic. operator piers and piles. Additionally, bubbler sys-
b. = vertical deflection of point at the plate ref- tems can be used to protect the piers and
erence plane piles from ice adhesion.
k = specific weight of water (k = Pwg)
q = vertical load distribution
3.6.7 Ice Load of Thermal
Equation (3-57) may be valid even when E Origin
varies across the thickness of the ice cover,
provided that flexural stiffness of the ice When an ice cover is subjected to a temper-
plate is determined from ature increase, it will expand. This will re-

Di = 1/(1- v 2 ) r-z
zo
0 z 2E(z) dz (3-58)
sult in forces exerted on the surroundings,
which may include miscellaneous marine
structures such as piles piers and gravity-
type vertical walls. The magnitude of these
where z 0 is the coordinate of the ice plate forces will depend on a number of parame-
reference plane which is determined from ters, such as the following:
the condition
• Temperature variation as a function of
time
• Material properties of the ice cover
• Ice thickness, presence of cracks, and other
Solutions for a variety of ice cover problems irregularities and geometry of the ice cover
that are based on Eq. (3-58) were reviewed • Restrictions to expansion along the bound-
by Kerr (1976). aries of the ice cover
Most recently Sodhi (1995) proposed a
theoretical formulation for determining the A number of theories on ice thermal load-
breakthrough load using plastic limit anal- ings have been suggested and several labo-
ysis. This load is obtained by equating the ratory experiments and full-scale observa-
rate of work done by the load to the rate of tions have been conducted to verify these
energy dissipation during compression of theories. A comprehensive review of the
ice caused by radial and circumferential state of the art on thermal ice forces is
wedging of ice during deformation. given in work by Kjeldgaard and Carstens
A comprehensive review of experimental (1980). The latter investigators concluded
studies of uplifting forces exerted by ad- that in spite of the considerable amount of
frozen ice on marine piles is given in work experimental work conducted in laborato-
by Christensen and Zabilansky (1985) and ries, there is a great deal of divergencies
Sodhi (1995). concerning the difficult question of which
In general terms, depending on ice sheet stress-strain relationships should be con-
thickness and ambient temperature, water- sidered appropriate as the basis for the
level fluctuations may produce uplift and method of computation of thermal ice forces.
downdrag forces on dolphins, single piles or The first approximation of thermal ice
a small group of piles, and uplift loads from loads at different locations and climatic
302 Design Loads

zones traditionally done on the basis of 3.6.8 Other Ice-Induced Loads


available empirical values. For example, in
contemporary Canadian practice, empirical These are loads associated with ice accre-
values of ice loads of thermal origin for tion on the structure (icing), ice dynamic
rigid structures such as dams or gravity- loads, ice-induced vibration of the struc-
type marine structures vary from 150 to ture, and so forth, which are beyond scope
220 kN jm, and for the design of relatively of this book. Again, for comprehensive
flexible structures such as sluice gates, val- treatment of ice loads on port related ma-
ues of 70-75 kN/m are commonly used. rine structures, and ice effects that affect
These values are suggested in work by port (harbor) operations, the reader is re-
Michel (1970). Drouin (1970) recommended ferred to Tsinker (1995), a companion work
even higher values of ice thermal pressure to this book.
(up to 300 kNjm) to be used in the design
of gravity-type structures, and PIANC
(1984) suggested that intact ice sheets may 3.7 SEISMIC LOADS
exert thermal thrusts in the range 75-300 <BY W. S. DUNBAR)
kNjm, regardless of ice thickness. Rigid
structures and structures in ice sheets con-
If a port is located in a seismically active
fined by harbor (basin) geometry will expe-
area, consideration should be given to the
rience the larger loadings. More flexible
need for including seismic loading (as an
structures and structures in more open or
extreme load conditions) in the design of
sloping-sided harbors (basins) will have
terminal structures. The first step in seis-
smaller loads. It has to be noted that smaller
mic design should be the establishment of
values of ice pressure exerted on flexible
the seismic design criteria applicable to the
structures must be treated with great cau- particular structure. Such criteria are based
tion because in some practical cases (e.g., on the consequences of failure due to seis-
sluice gates), load relaxation at the middle mic loads. For instance, the destruction of a
of the structure may result in a heavy load warehouse with low occupancy, and there-
concentration at the bearing points. fore, low risk of loss of life would require
In recommendations by Canadian De- less stringent seismic design criteria than
partment of Environment (1971), it is as- say a quay wall which is likely to be highly
sumed that the ice thrust varies linearly occupied and on which different kinds of
with the ice thickness and, therefore, that cargo handling equipment are installed. Ad-
ice loads of 150 and 220 kN jm subse- ditionally, as the loss of a quay wall could
quently correspond to ice thickness of 0.3 disable a terminal for a considerable period,
and 0.6 m, respectively. economic losses may have to be considered.
In cold Siberian regions of Russia, the ice Different design loads, possibly larger
pressure of 300 kPa · is commonly used, than those specified in building codes, may
whereas for somewhat less severe condi- be applied to critical structures whose col-
tions, ice pressures ranging from 150 to 200 lapse would endanger lives andjor entail
kPa is more common (Starosolsky, 1970). significant property damage. Typically, a
In Norway, an ice load of 100 kNjm is two-tiered design criterion is applied to crit-
typically used for an average ice condition, ical structures. The first tier is sometimes
and under especially unfavorable condi- called the maximum probable event (or
tions, the value of the ice load may be as equivalent seismic loading) wherein a criti-
high as 150-200 kN jm (Kjeldgaard and cal structure is designed to remain. opera-
Carstens, 1980). tional in the event the "probable" earth-
Design Loads 303

quake occurs. However, in the event of a

d
maximum possible earthquake, the struc-
ture may not be operational, but total or
catastrophic destruction must not occur.
The applicability of the above types of
design criteria will depend on the details of 0 Epicenter
the port design, requirement of the port
authority, and local building codes. How-
ever, the establishment of design criteria is
an essential step, without which nonuni-
form design criteria and inadequate or
overconservative designs are very likely.
The remainder of this section assumes Figure 3-34. Pictorial definitions of earthquake
that design criteria have been established hypocenter and epicenter.
and deals with the description of seismic
ground motion and its estimation for design The basic and relatively simple ideas of
purposes. plate tectonics provide a good theoretical
framework for understanding the mecha-
nism of plate boundary or interplate earth-
quakes. However, a significant number of
3.7.1 Seismic Ground Motion earthquakes, known as intraplate earth-
quakes, occur within the plates on almost
3.7.1.1 The Earthquake Source every continent. Although the basic mecha-
The basic mechanism for the majority of nism of these earthquakes is the same as
that of the interplate earthquakes (i.e.,
earthquake occurrences is shear rupture
shear rupture along a fault), the mecha-
along a fault in the earth's crust. Such rup-
nism for stress generation within a plate is
turing implies that the shear strength of
not understood. This and the fact that in-
the fault surface has been exceeded. A
traplate earthquakes are relatively rare oc-
major goal of basic research into earth-
currences makes the prediction of seismic
quake mechanisms is to determine both the
hazard due to such events quite difficult.
distribution of strength on a fault and the
Two common terms used in earthquake
behavior of the driving forces which cause it
engineering to define the location of the
to be exceeded.
seismic source are hypocenter or focus and
The origin of the driving forces was un-
epicenter. The hypocenter or focus of an
known until about 1965 when the theory of
earthquake is defined as the point within
plate tectonics begari to be developed. The
the earth where rupture begins. The epicen-
basic idea of this theory is that the outer-
ter is the points on the surface of the earth
most part of the earth is divided into large-
directly above the hypocenter (Fig. 3-34).
scale slabs or plates which move inde-
pendently of each other, presumably as a
result of plastic flow processes deep in the
3.7.1.2 Recordings of Ground
earth underneath the plates. Many of the
Motion
world's earthquakes occur at the boundries Ground motion caused by an earthquake
between the plates where relative plate mo- may be measured by an accelerograph. An
tion results in a buildup of shear stress and accelerograph consists of an accelerometer
ultimately shear failure, an earthquake (the transducer) and a recording mecha-
[see, for example, Bolt (1987)]. nism (e.g., photosensitive paper or magnetic
304 Design Loads

tape). The accelerometer is a motion- 0.10

recording device whose amplification is 0.00

maximum at frequencies greater than 10 -0.10


Hz. The motion recorded by an accelero-
..
-o.20
graph is measured in units of acceleration 0.10

and is called an accelerogram. Accel- §


erograms give a direct measure of the ::3 0.00
G
force applied to a structure; thus, they 'il
!l
play a fundamental role in earthquake -Q.10
0.05
engineering.
Accelerographs may be located on the 0.00
ground (free field) or in structures. At a
typical recording station, three orthogonal -o.05
components (two horizontal and one verti- 0.0 10.0 20.0 30.0

cal) would be recorded. An example is shown Time (sec)

in Figure 3-35. These accelerograms may Figure 3-35. North-south, east-west, and vertical
be integrated numerically to produce veloc- accelerograms of the November 25, 1988 Saguenay,
ity and displacement records, if necessary Quebec M6.0 Earthquake Recorded at St. Andre du
(Hudson, 1979). Lac. (Data provided by the Geological Survey of
Canada.)
3.7.1.3 Seismic wave Types
There are basically three types of seismic caused by a P wave is alternatively back
wave: compressional or P (primary) waves, and forth along the direction of propagation
shear or S waves, and surface waves. P (Fig. 3-36a). This produces zones of com-
waves and S waves are sometimes called pression and dilation along the propagation
body waves. In a given material, P waves path. An S wave causes particle motion in a
are faster than S waves, and surface waves direction perpendicular to its path of propa-
are slower than the two body waves. The gation (Fig. 3-36a). As there are two per-
actual speed of propagation depends on the pendicular directions to the propagation
density and elastic properties of the rocks path, the particle motion of an S wave ex-
and soil through which the waves pass. hibits polarization, and components in two
Generally, the stiffer the material, the orthogonal directions will be dominant, re-
greater the propagation speed. P waves can sulting in SH and SV waves. There are two
propagate through both solids and liquids. types of surface wave. Love waves and
An S wave cannot propagate through a Rayleigh waves. These waves propagate
liquid. The motion of the particles of rock horizontally along the surface of the earth.

*~
·y Love Waves

Direction of Propagation
Direction of Propagation Perpendicular to Page

(a) Body Waves (b) SUrface Waves

Figure 3-36. Particle motion of(a) body waves and (b) surface waves.
Design Loads 305

The particle motion caused by a Love wave half-'space is


if always parallel to the earth's surface; no v.
motion perpendicular to the earth's surface fo = 4H (3-59)
(vertical motion) occurs (Fig. 3-36b). The
particle motion due to a Rayleigh wave is where V. is the shear wave velocity of the
somewhat more complicated, but it is simi- layer. Thus, for the resonance mechanism
lar to that of an ocean wave in that the to be operative, the soil layer must have a
motion forms an ellipse and is confined to a thickness equal to a multiple of one-quarter
vertical plane lying in the same direction as of the spatial wavelength of the waves [see,
the propagation path (Fig. 3-36b). This type for example, Dobry et al. (1976)].
of motion is called elliptical retrograde. Love The effects of both of these mechanisms
waves travel faster than Rayleigh waves. would be significantly modified by damping
Depending on the proximity of the earth- and nonlinear properties. Generally, such
quake, any one of these wave types can be properties tend to decrease amplification in
the main contributor to ground motion at a a soil layer.
site. Near an earthquake source, the motion Prediction of the level of ground
is due to a complex combination of waves motion amplification in a soil layer is not
because the source of seismic energy is simple. At a particular site, all the above
spread over a fault. However, it is generally mechanisms could be operative to a degree
S waves which cause the larger amplitude dependent on the geometry and material
properties of the soil and on the nature of
motion. Surface waves develop at some dis-
the incoming waves. An extreme example of
tance from the source. All of this assumes
ground motion amplification is that which
that no amplification has occurred due to
occurred in the soft clayey lake deposits of
the presence of soils or topographic effects
Mexico City during the 1985 Michoacan
at the site of a port. These effects are dis- earthquake. The surprising aspect of the
cussed in the next section. observed ground motion is that Mexico City
is about 300 km from the earthquake. It
3.7.1.4 Amplification Due to appears that the low-amplitude ground mo-
Local conditions tions in the rock underlying Mexico City
were amplified by resonance in the lake
Several studies (Wiggins, 1964; Murphy . deposits (Anderson et al., 1986; Beck and
et al., 1971; Berrill, 1977) have demon- Hall, 1986; Singh et al., 1988).
strated that, for the same earthquake, Other, more regional, features may affect
ground motion at sites composed of soft the response at some. port sites. For exam-
materials is greater than that at sites com- ple, topographic features such as hills can
posed of stiff materials. There are two phys- cause local amplification of incoming seis-
ical mechanisms responsible for amplifica- mic waves through focusing of rays. Multi-
tion or ground motion at soft sites. The ple reflections of trapped seismic waves in
dominant mechanism is conservation of en- sedimentary basins can also amplify mo-
ergy: Because the wave propagation veloc- tions. Highly fractured andjor weathered
ity decreases toward the surface of a soft rock can lead to unusual dynamic behavior.
site, wave amplitude must increase to con- Effects of this nature are highly complex
serve energy. The other mechanism is reso- and would require wave propagation model-
nance due to the constructive interference ing on a regional scale to estimate resulting
of upgoing and downgoing waves in the soil ground motions. However, such modeling is
layer. The fundamental resonant frequency, beyond the state of practice and is only
{ 0 , of a layer of thickness H overlying a recommended in extreme situations. Exam-
306 Design Loads

ples may be found in Wong and Jennings companies and to engineers or planners
(1975) and Barel and Gabriel (1986). concerned with site selection.
A more quantitative measure of the size
3.7.1.5 Intensity and Magnitude of an earthquake is its magnitude which is
derived from instrumental measurements.
The oldest measure of the size or strength Presently there are several different magni-
of an earthquake is its intensity. The inten- tude scales in use. However, the principle of
sity is a measure of such things as human each is the same, namely that the larger
reaction, ground surface effects, and dam- the earthquake, the larger the amplitude of
age to structures. An intensity scale con- the radiated seismic waves. Thus, each dif-
sists of a series of increasing numbers, typi- ferent magnitude scale is associated with a
cally Roman numerals, each of which is specific wave type. Brief descriptions of the
associated with a description of damage or more common magnitude scales are given
other earthquake effects. The larger the below:
number, the greater the effects or damage.
An example is the Modified Mercalli scale
1. Local magnitude (ML)-This is the origi-
(MMI) developed in 1931 (Richter, 1958). nal Richter magnitude scale and is based
For example, the description of the smallest on the peak amplitude of the seismogram
value of MMI, Roman numeral I, is in the 1-5-Hz range.
2. Body wave magnitude (mb)-This uses
I-not felt except by a very few under espe- the peak amplitude of the P wave at a
cially favorable circumstances frequency of 1 Hz.
3. Surface wave magnitude (M8 )-This uses
whereas the description of the largest value the peak amplitude of Rayleigh surface
of MMI, Roman numeral XII, is waves at a frequency of 0.05 Hz.

XU-damage total. Practically all works of The amplitudes used for ML and mb tend
construction are damaged greatly or de- not to increase with increasing earthquake
stroyed. Waves seen on ground surface. size. This is referred to as saturation of the
Lines of sight and level are distorted. magnitude scale and occurs when the rup-
Objects are thrown into the air. ture dimension of the fault becomes much
larger than the spatial wavelength of the
The selection of an intensity value is seismic waves used to determine the magni-
based on observations made by humans. tude. The result is that ML and mb tend
Consequently, although the descriptions of toward the same anomalously low value for
effects which occur at each intensity level magnitudes greater than about 0.6. To over-
are fairly precise, the selection of a number come this problem, the surface wave magni-
can be subjective or judgmental in some tude, M 8 , is used. Large shallow earth-
cases. quakes produce significant amounts of sur-
Because damage and other earthquake face waves at distances greater than about
effects can depend on local conditions, in- 100 km so that reported M 8 values are
tensities will naturally vary with local con- from relatively distant stations.
ditions. Thus, if intensity measurements are Ideally, if ML> mb, and M 8 were ob-
plotted on a map and contoured, they can tained for the same earthquake, they would
provide valuable information on the distri- be numerically equal. However, this almost
bution of the strong ground motion of past never happens. The reason is that, depend-
earthquakes. In the absence of other infor- ing on their size, earthquakes emit differ-
mation, this may be of use to insurance ent amounts of energy at different frequen-
Design Loads 307

cies resulting in peak amplitudes which are earthquake is analogous to breaking a small
representative of the energy at those fre- dry twig which results in a high-frequency
quencies. snap, whereas a large-magnitude earth-
For very large earthquakes ( > 8.0), satu- quake is analogous to felling a large tree,
ration of the M 5 scale occurs. To circumvent which results in a lower-frequency crack or
this, the seismic moment is used to meas- "thud.") Duration of significant strong
ure the size of the earthquake. The moment ground motion depends mainly on the time
of an earthquake is proportional to the zero taken for fault rapture to occur. The longer
frequency (or long period) amplitude of the the rupture time, the greater the earth-
spectrum of the seismogram. Seismic mo- quake magnitude, so that duration in-
ment is a true physical measure of earth- creases with earthquake magnitude.
quake size and has been used for earth- Figure 3-37a shows the accelerogram due
quakes with local magnitudes as low as 3.0. to a M6.2 earthquake recorded by an ac-
Hanks and Kanamori (1979) developed a celerometer of the SMART array in Taiwan.
moment-magnitude scale based on the A M4.9 aftershock of this earthquake oc-
measured seismic moment. This magnitude curred at about the same location and was
scale is preferred in earthquake engineer- recorded by the same accelerometer; the
ing. However, all other magnitude scales accelerogram of this event is shown in Fig-
are used and care must be taken to ensure ure 3-37b. Note the difference between the
that the same magnitude scales are being frequency content and duration of these two
used to compare earthquake sizes. Similar accelerograms.
care must be exercised when using attenua- Quantitative measures of the duration of
tion relationships for ground motion param- significant shaking are important in the
eters (discussed later), as these relation- analysis of soils. Relationships between
ships are defined for a particular type of magnitude and duration of strong ground
magnitude. Nuttli and Herrmann (1982) motion for the westem United States have
provide an excellent description of the vari- been derived empirically by Trifunac and
ous magnitude scales in use. Brady (1975). Measures of the duration of
strong motion for recorded accelerograms
3.7.1.6 Duration and have been proposed by Bolt (1969) and
Frequency content McCann and Shah (1979).

Seismic waves of all types are attenuated


as they pass through the earth, owing to
the inelastic properties of rock and soil. In 3.7.2 Descriptions of
general, attenuation increases with both Ground Motion
distance and wave frequency so that near
an earthquake source, the wave frequencies In order that seismic design can be per·
would be high (10-20Hz or greater in the formed, it is necessary to have some repre-
very near field), whereas at large distances sentation of the ground motion, typically
from an earthquake, frequencies will tend the ground acceleration. Multiplication of
to be less than 10 Hz. the ground acceleration by mass quantities
Both duration and frequency content gives seismic forces which the structure
vary with earthquake magnitude. Large- must be designed to withstand.
magnitude earthquakes tend to produce There exist several methods of describing
low-frequency ground motion, whereas ground motion for design purposes. Each is
small-magnitude earthquakes produce applicable in some situations, inapplicable
higher frequencies. (A small-magnitude or inappropriate in others. It is important
308 Design Loads

SMART ARRAY rise to greater than normal loads in struc-


Station c-oo M6.2 R=5km NS Component 7/30/86 tures. Usually, but not always, the design
0.3
for gravity loads includes a large enough
factor of safety to accommodate additional
0.2
loads due to vertical ground motion.
"" 0.1

"
0
3.7.2.1 Peak Amplitudes
::: 0.0

~g The most widely used description of


<
-o.1
ground motion is peak amplitude, often
-o.2
peak horizontal acceleration. One philoso-
phy behind the use of a peak acceleration is
-o.3 that if the structure is designed to with-
0.0 5.0 10.0 15.0 20.0
stand the maximum force occurring during
Time (sec)
the ground motion, then it should be ade-
(a)
quately designed for other smaller forces
SMART ARRAY occurring during the ground motion. This is
Station C-oO M4.9 R=5km NS Component 7/30/86 considered a suitably conservative assump-
tion for design purposes.
Unfortunately, peak motions have seri-
ous limitations as meaningful engineering
descriptions of ground motion. It is now
well known that peak motions do not corre-
late well with structural response or dam-
age potential (Stevenson et al., 1984; EPRI,
1988). Peak ground accelerations, for exam-
ple, are contained within high-frequency
( > 25 Hz) ground motion, and as the reso-
0.0 5.0 10.0 15.0 20.0 nant frequency of most civil structures is
Time (sec) less than 25 Mz, such motions cannot in-
(b) duce significant structural response.
The response of a structure also depends
Figure 3-37. (a) North-south accelerogram ofM6.2, on the duration of the ground motion.
July 30, 198 6 event recorded at station C-OO of the
SMART accelerograph array in Taiwan. The accelero-
Short-duration impulselike motions, such as
graph was located at a distance of 5 km from the those due to a small earthquake, do not
earthquake. (b) North-south accelerogram ofM4.9 have sufficient time to induce a significant
aftershock of main M6.2 event also recorded at station structural response. The above applies to
C-OO of the SMART array. Note the long duration and linear behavior of structures. Damage is a
low frequency in the ground motion of the main event
as compared with that of the aftershock. [Data pro-
nonlinear phenomenon which also depends
vided courtesy of the Seismographic Station of the on the amplitude of ground motion as well
University of California at Berkeley and the Institute as its duration and frequency content in a
of Earth Sciences of the Academia Sinica, Taipei.] more complicated manner.

3.7.2.2 Response Spectrum


to understand these descriptions and their
advantages and limitations. The most com- As mentioned earlier, peak ground accel-
mon descriptions are discussed below. erations are often contained within high-
Horizontal components of ground motion frequency ( > 25 Hz) ground motion. Be-
are typically used in design since they give cause the resonant frequency of most civil
Design Loads 309

structures is less than 25 Hz, such motions where u(t) is the time history of the rela-
cannot induce significant structural re- tive displacement of the mass, m, of the
sponse. Therefore, peak ground acceleration oscillator, c is the damping coefficient of the
can only characterize the response of very dashpot, and k is the spring stiffness. The
stiff structures whose resonant frequency is dots indicate differentiation with respect to
high. time. Dividing Eq. (3-60) by m results in
The dynamic characteristics of a compli-
cated structure can be represented in terms ii(t) + 2{3w 0 it.(t) + w~u(t) = -a(t) (3-61)
of its response in fundamental and higher
natural vibration modes or frequencies where w 0 = ..jkjm is the resonant angular
(Chopra, 1981). The response of the first or frequency of the oscillator (in radians Is)
lowest-frequency mode is typically the most and {3 = cj2mw 0 is the damping ratio ex-
significant. The behavior of each mode can pressed as a fraction of critical damping.
be modeled by a linear single degree-of- For most structures, {3 lies in the range
freedom damped oscillator, an example of 0.02 (2% damping) to 0.10 (10% damping).
which is shown in Figure 3-38a. Thus, the The solution to Eq. (3-61) is the relative
maximum value of the response of this os- displacement response time history, u(t).
cillator to the ground motion is the engi- Differentiation gives the relative velocity
neering quantity of interest. A plot of the response history, u(t), and the relative ac-
maximum response of the oscillator versus celeration time history, ii(t). The latter is
the corresponding oscillator frequency or not meaningful in terms of an effect on the
period results in a response spectrum for oscillator. The absolute acceleration, z(t), is
the particular ground motion. Displace- a measure of the force applied to the mass
ment, velocity, and acceleration response and is given by
spectra may be defined. The remainder of
this section is concerned with the computa- z(t) = ii(t) + a(t) = 2{3w 0 it.(t) + w~u(t)
tion and characteristics of response spectra. (3-62)
When driven by the accelerogram, a(t),
the equation of motion of the oscillator
shown in Figure 3-38a is Displacement response spectral ordinates
for a particular value of damping ratio are
mii(t) + cu(t) + ku(t) = -ma(t) (3-60) computed by varying the oscillator fre-
quency and monitoring the computed rela-

b:L,,
tive displacement time history to obtain the
~~m peak value, Siw 0 , {3). Velocity and acceler-
ation response spectra, S u and Sa, respec-
tively, may be similarly defined as

-a(t) Sd(w 0 , {3) =max lu(t, w 0 , {3)1


(a)
Sv(w 0 , {3) =max lit.(t, w 0 , /3)1
Sa(w 0 , {3) =max lz(t, w 0 , {3)1
SDOF Oscillator
a (t) "%·~ u (t)
The procedure for computing S d is illus-
(b) trated in Figure 3-38b. An algorithm for
Figure 3-38. (a) Damped single degree of freedom
numerical computation of the response his-
oscillator. (b) Illustration of the procedure for comput- tories and their maxima is given in Nigam
ing the displacement response spectrum sd. and Jennings (1969).
310 Design Loads

The maximum force in the spring of the damping, PSA is a good approximation to
oscillator is kSd = mw~Sd. The quantity Sa for all frequencies. The approximation
w~Sd has units of acceleration and is known PSV = S v is good in the intermediate fre-
as the pseudo-acceleration (PSA). Pseudo- quency range (approximately 0.5-20 Hz).
velocity (PSV) is similarly defined. Thus, Quantities related by powers of w 0 such as
Sd, PSV and PSA can be plotted on four-way
PSA ~ w3Sd logarithmic paper. For the north-south
component accelerogram shown in Figure
PSA = w 0 Sd 3-35, the four-way logarithmic response
spectrum plot is shown in Figure 3-39a.
For {3 = 0, PSA =Sa. For Low values of Although such plots are useful, in that all
quantities of interest are presented on one
Frequency (Hz)
diagram, four-way log plots can be difficult
10.00 1.00 0.10
to read. The useful quantity for design is
PSA which may be plotted using linear axes
as shown in Figure 3-39b. Given an ordi-
nate on this plot, PSV and Sd may be easily
computed.
The use of the prefix "pseudo" in connec-
tion with velocity and acceleration spectra
is somewhat misleading because there is
nothing false or incorrect about PSV and
PSA; they are merely approximations to Sv
and sa.
Response spectra of accelerograms tend
to certain limits depending primarily on the
0.1
relationship between the frequency of the
.____._..__._..,._....__._~....,..,......._____,___._........,........,

0,01 0.10 1.00 10.00


Period (sec l oscillator and the dominant frequency of
(a) the ground motion. These limits can be de-
duced from Eq. (3-61). As w0 becomes large
O.<l
(high frequency), the first two terms on the
left-hand side of Eq. (3-61) become negligi-
:s0: 0.3 Taft earthquake K7.2 ble and w~(t) = -a(t). The latter approxi-
mation must apply to peak quantities so
..•
0
;I

.
.l! 0.2
<>
that
:1I
...
"..
0

0.1
PSA-+ pga as w 0 -+ large
Equation 4-50
"" ·-·--- ------------------
0.0
0.0 0.5 1.0 1.5 2.0
where pga is the peak ground acceleration
Period (sec)
of the accelerogram a(t). This type of re-
(b)
sponse occurs in stiff structures whose reso-
nant frequency is high relative to the fre-
Figure 3-39. (a) Four-way logarithmic plot of the quencies present in the ground motion so
response spectrum of the north-south accelerogram that only a small relative displacement will
shown in Figure 35. (b) Linear plot of pseudo-
acceleration spectrum for same accelerogram. The produce the induced force. Similarly, as w0
damping ratio of each spectrum is 5%. becomes small, the last two terms on the
Design Loads 311

left-hand side of Eq. (3-61) can be neglected tween a response spectrum and a particular
so that u(t) "" -a(t). Integrating and as- accelerogram. For example, it is possible to
suming peak quantities find short-duration accelerograms, typical
of small earthquakes, that result in approx-
Sd = pgd as Wo ~ small imately the same peak response at resonant
frequencies in the 10-Hz range as that of a
where pgd is the peak ground displacement. long-duration accelerogram typical of a
This type of response would occur in a very larger earthquake.
flexible structure whose resonant frequency
is low relative to the frequencies present in
the ground motion, so that a large relative
displacement would be required to produce 3.7.2.3 Time Histories
the induced force.
Based on these limits, it may be seen
that as the oscillator frequency becomes The most complete description of ground
large relative to the dominant frequency motion at a site would be there orthogonal
of the ground motion, the peak response of components of ground acceleration. These
the oscillator depends on the peak ground could be used as input to a computer or
acceleration. As the oscillator frequency physical model of the structure andjor its
becomes low relative to the dominant foundation. The model could then be used to
frequency of the ground motion, the peak make realistic determinations of response.
response of the oscillator depends on the The use of physical models is expensive, but
peak ground displacement. Intuitively, it is common in research. Computer models of
may be seen that the response at inter- structures and foundations are more com-
mediate periods depends on peak ground mon in practice.
velocity. When time histories are used in design,
The utility of a response spectrum is that it is common to select from a database a
it represents the frequency dependence of number of accelerograms caused by earth-
the response, whereas peak ground acceler- quakes whose characteristics (magnitude,
ation is only representative of the response distance) are similar to those of the design
of a very stiff structure. Despite this, a earthquake. A particularly important char-
response spectrum is still not a complete acteristic is recording site conditions which
representation of ground motion or struc- should be similar to those at the site of the
tural behavior. Two important limitations port.
must be recognized: Selection of time histories for use in de-
sign has been automated by various institu-
1. A response spectrum can only be used in tions and government agencies. For exam-
a linear analysis of structures and cannot ple, the National Geophysical Data Center
directly account for nonlinear structural
(NGDC) in Boulder, Colorado maintains a
behavior.
large database composed of 12,000 (pre-
2. A response spectrum is not representative
of the duration of strong ground motion sently)
accelerograms. A microcomputer-
which, for some structures or materials, based catalog search program is available
may be an important consideration in as- from NGDC which would enable one to gen-
sessing damage potential. erate a list of accelerograms in the database
which have the desired characteristics. The
A corollary of the last item is that there accelerograms can then be ordered from
does not exist a one-to-one relationship be- NGDC or other agencies.
312 Design Loads

3.7.3 Design Ground Motion number of selected accelerograms. The se-


Estimation lection of a design time history (or histories)
would also involve considerations of re-
3.7.3.1 Design Motions quired structural behavior.
The distinction between a ground motion
As can be seen in Figure 3-39, the re- and a design motion must be made at all
sponse spectrum of a particular accelero- times. The former is a seismological de-
gram is an irregular function of frequency scription of ground motion, whereas the lat-
or period. This is due to the highly variable ter is a design requirement. Design motions
nature of ground motion, even ground mo- result when design criteria are applied.
tion due to earthquakes of similar magni- Thus, for example, all response spectra, in-
tude and recorded at similar distances. In cluding smooth spectra, are descriptions of
order to represent the possible variability in ground motion and potential design spectra.
ground motion, a smooth response spec-
trum is used. Such a spectrum is often de-
rived by finding the pth percentile of the 3.7.3.2 scaled Response Spectra
spectra of a number of accelerograms. Ex- The peak oscillator responses, PSA, PSV,
amples are the median spectrum (50th per-
and sd, can be larger in different period
centile) or the median plus one standard
ranges than the corresponding peak ground
deviation (84th percentile) spectrum. The
motions. This is due to the behavior of an
average spectrum may also be computed.
oscillator at low, intermediate, and high pe-
The derivation may be deliberately weighted
or biased in some fashion to reflect known riods described above. Thus, for periods less
seismological data or site conditions. How- than 0.1 s, PSA is large compared to pga,
ever, the result is a smooth function of fre- whereas S d is small. The composite holds
quency or period. at long .periods. At intermediate periods,
Although a response spectrum may be the peak ground velocity, pgv, is amplified
consistent with all known seismological data to give a large PSV.
and site conditions, it may not be appropri- Amplification in different period ranges
ate for design. For a variety of reasons, a allows the delineation of the response spec-
design engineer may require certain kinds trum into three ranges: the low period or
of behavior of a structure. This leads to the acceleration range, the intermediate period
concept of a design spectrum which is a or velocity range, and the long period or
measure of the force the structure must displacement range. In each region, the cor-
resist elastically. Thus, if the structure is responding ground motion is amplified the
capable of absorbing energy (e.g., a ductile most. Based on this, Newmark and Hall
steel frame), the required elastic forces (1969) recommended that constant amplifi-
could be less than those implied by a smooth cation factors, PSAjpga, PSV jpgv, and
response spectrum, because, presumably, Sdjpgd (peak ground displacement), be
the structure could absorb energy in its defined for the acceleration, velocity, and
inelastic range. By contrast, if the structure displacement period ranges, respectively.
is considered critical, it is conservative to Using several earthquake records, median
ignore the possibility of inelastic energy- amplification factors were derived. Multi-
absorbing behavior and to use a design plication of the peak ground motions by
spectrum which may possibly be one of the these amplification factors resulted in
larger percentile smooth response spectra. spectral ordinates which could be plotted on
A similar concept is a design acceleration four-way log paper to produce a smooth
time history which may be a composite of a response spectrum.
Design Loads 313

These initial estimates of the amplifica- and Hall (1982) and in Mohraz (1976) from
tion factors were based on accelerograms which pgv and pgd may be computed. How-
recorded during the 1934 and 1940 El ever, these estimates depend on both earth-
Centro, California earthquakes and the quake magnitude and site conditions in a
1952 Kern County, California earthquake. complicated manner; published values ex-
Following the 1971 San Fernando earth- hibit a large amount of scatter.
quake, more records were available, making The procedure of scaling amplification
it possible to obtain better statistical esti- factors to define a response or design spec-
mates of amplification factors (Blume et al., trum is only valid if the amplification fac-
1972; Newmarket al., 1973). These studies
resulted in 50 percentile (median) and 84
percentile (median + one standard devia- Frequency (Hz)

tion) amplification factors for peak accelera-


tion, velocity, and displacement and for
several damping ratios. The amplification
factors formed the basis for the U.S.
Atomic Energy Commission (now the U.S.
Nuclear Regulatory Commission) Regula-
tory Guide 1.60 (1973) which gave a design
spectral shape for nuclear power plants.
Further studies in the mid-1970s re-
sulted in more refined estimates of amplifi-
cation factors. These are given in Newmark
and Hall (1982) and reproduced in Table
3-16. Figure 3-40 shows an example of the
construction of a response spectrum using
these amplification factors. Period (sec)

The use of these amplification factors re- Figure 3-40. Method of construction of a scaled re-
quires independent estimates of pga, pgv, sponse spectrum. The dashed lines are values of peak
and pgd. Often only estimates of pga are ground motions. The solid lines are spectral bound
available and would be derived using one or computed by scaling the median amplification factors
for 5% damping in Table 3-16 by the peak ground
more of the methods discussed below. motions. At periods less than 0.1 s, the pseudo-
Estimates of the ratios pgajpgv and acceleration spectral ordinates decrease log-linearity
(pga)(pgdjpgv 2 ) are available in Newmark to the value of peak ground acceleration.

Table3-16. Spectral amplification factors

84 Percentile Median
Damping Ratio(%) pga pgv pgd pga pgv pgd

0.5 5.10 3.84 3.04 3.68 2.59 2.01


1 4.38 3.38 2.73 3.21 2.31 1.82
2 3.66 2.92 2.42 2.74 2.03 1.63
3 3.24 2.64 2.24 2.46 1.86 1.52
5 2.71 2.30 2.01 2.12 1.65 1.39
7. 2.36 2.08 1.85 1.89 1.51 1.29
10 1.99 1.84 1.69 1.64 1.37 1.20
20 1.26 1.37 1.38 1.17 1.08 1.01

Source: From Newmark and Hall (1982).


314 Design Loads

tors, which basically define the shape of the This means that the same earthquake could
spectrum, are independent of magnitude, produce different ground motions at differ-
distance, and recording site conditions. As ent sites which are equidistant from the
mentioned previously, the frequency source. The dependence on source and prop-
content of ground motion depends on the agation path imply that earthquake magni-
earthquake magnitude; small-magnitude tude and a measure of source-site distance
earthquakes generate ground motion rich would be useful explanatory variables in an
in high-frequencies, whereas, large-magni- attenuation relationship. A functional de-
tude earthquakes generate ground motion scriptor of site conditions could also be in-
rich in low frequencies. Examples were cluded. A general expression for an attenu-
shown in Figure 3-37. Numerous studies ation relationship is
(McGuire, 1974; Trifunac and Anderson,
1978; Joyner and Boore, 1982) have con- logy = a + { 1(M) +fiR) + f 3 (S) ± E (3-63)
firmed that the dependence of ground mo-
tion frequency on magnitude results in a
where
dependence of spectral shape on magnitude.
There is also evidence of a dependence on y = ground motion parameter
distance from the rupture zone. Spectral M = a measure of magnitude
shape also depends on site conditions in R = a measure of distance
that long-period ground motions tend to ex- S = a set of variables describing the site founda-
hibit larger amplitudes on soft soil sites tion conditions
than on hard sites. Amplification factors for E = a random error term with zero mean and
different site conditions are given in Mohraz standard deviation a.
(1976) and Seed et al. (1976). The possible
bias associated with magnitude and site Each of the functions f 1 , f 2 , and fa contain
condition dependence can be significantly constants to be estimated. Both base 10 and
reduced by scaling amplification factors ap- natural logarithms are used on the left-hand
propriate to the site conditions in question side of the above equation.
by independent estimates of pga, pgv, and The function f 1 is often assumed linear
pgd. in powers of M, whereas f 2 is linear in
powers of R and log R. The function fa is
linear in the variables S, which are often
3.7.3.3 Attenuation Relationships multiplied by 0 or 1 (i.e., 0 denotes one type
Ground motion attenuation relationships of foundation condition, 1 denotes another
are functions relating a ground motion pa- type). Most empirically derived attenuation
rameter, such as peak ground acceleration, relationships use moment magnitude as a
to certain explanatory variables. In general, measure of earthquake size. This provides a
ground motion due to earthquakes depends more consistent magnitude measure for
on three things: earthquakes of all sizes, as discussed above.
The measure of distance R is more prob-
lematic, particularly in the near field. As
1. Source characteristics-magnitude, na-
discussed previously, the source of earth-
ture of faulting
quake energy is rupture along a fault sur-
2. Propagation path-reflection and refrac- face. In the far field (a distance of several
tion, nature of materials through which rupture lengths), this source may be treated
waves propagate as a point source, say at the centroid of the
3. Site conditions-soil or rock site, layer- rupture surface. R is then the hypocentral
ing, heterogeneity depth. However, in the near field, the orien-
Design Loads 315

tation of the fault with respect to the site their validity in other regions is question-
plays an important role in determining the able. However, examples of other such rela-
amplitude and duration of ground motion. tionships and useful references are given in
The distance of closest approach to a. fault Joyner and Boore (1988).
is therefore often used as a measure of Prediction of spectral response using at-
distance. Further discussion of this issue tenuation relationships differs from the
and references are given in Campbell (1985). conventional practice of scaling a normal-
An example of empirically determined ized spectral shape by peak values of dis-
attenuation relationships for both peak placement, velocity, and acceleration, as de-
ground acceleration and response spectral scribed above. Attenuation relationships
ordinates is given in Table 3-17. As only which predict spectral response explicitly
accelerograms from Western North America account for this dependence of spectral
were used to derive these relationships, shape on magnitude discussed previously.

Table 3-17. Spectral attenuation relationships

(logy )a + b(M - 6) + c(M - 6) 2 + d log R + kR + s


y = larger of two horizontal components

8 = { *0 soil site. 2:: 5 m thickness


0 rock site
5 ~ M ~ 7. 7, M = moment magnitude

R = (r 2 + h 2 / 12 , r = distancetoverticalprojectiononearth's
surface of nearest point of rupture
u = standard deviation of log y

T (s) a b c d h k s u
Pseudo-velocity (cmjs) 5%damping

0.10 2.24 0.30 -0.09 -1.0 10.6 -0.0067 -0.06 0.27


0.15 2.46 0.34 -0.10 -1.0 10.3 -0.0063 -0.05 0.27
0.20 2.54 0.37 -0.11 -1.0 9.3 -0.0061 -0.03 0.27
0.30 2.56 0.43 -0.12 -1.0 7.0 -0.0057 0.04 0.27
0.40 2.54 0.49 -0.13 -1.0 5.7 -0.0055 0.09 0.30
0.50 2.53 0.53 -0.14 -1.0 5.2 -0.0053 0.12 0.32
0.75 2.46 0.61 -0.15 -1.0 4.7 -0.0049 0.19 0.35
1.0 2.41 0.66 -0.16 -1.0 4.6 -0.0044 0.24 0.35
1.5 2.32 0.71 -0.17 -1.0 4.6 -0.0034 0.30 0.35
2.0 2.26 0.75 -0.18 -1.0 4.6 -0.0025 0.32 0.35
3.0 2.17 0.78 -0.19 -1.0 4.6 0.0 0.29 0.35
4.0 2.10 0.80 -0.20 -0.98 4.6 0.0 0.24 0.35

Peak Acceleration (g)


0.49 0.23 0.0 -1.0 8.0 -0.0027 0.0 0.28

Peak Velocity (cmjs)


2.17 0.49 0.0 -1.0 4.0 -0.0026 0.17 0.33

Source: After Joyner and Boore (1988).


316 Design Loads

However, the random error, e, associated . For given values of m and y, the dis-
with these attenuation relationships can be tance r is given as the solution to the equa-
large. tion y = f(m, r ), where f(m, r) is the
attenuation relationship for the ground
3.7.3.4 Probabilistic Estimates motion or spectral ordinate y. An earth-
quake of magnitude m within the distance
If an exceedance probability for ground
r from the site will cause a spectral
motions can be specified as a design crite-
ordinate of y or greater. The probability
rion, then probabilistic seismic hazard anal-
P(Y > y I M > m) thus depends on the ge-
ysis methods can be used to estimate the
ometry of the seismic zonation model. For
corresponding ground motions or response
example, if the model restricts earthquake
spectral ordinates.
occurrences to a region whose total area is
Modern seismic hazard analysis is done
A and in which the seismicity is assumed
by variants of the Cornell method (Cornell,
uniform, then
1968). These methods can account for known
or assumed seismological characteristic of
if r not in A
the region surrounding a site. Th methods P(Y>y IM> m) ={~A/A ifrinA
require descriptions of the spatial and tem-
poral occurrence of earthquakes. A seismic (3-65)
zonation, in which zones of uniform seismic-
ity or active faults are identified, is used to where AA is a small area centered at r.
describe the possible spatial occurrences. A A further consideration is the effect of
magnitude-recurrence relationship, de- the error term, e, in an attenuation rela-
rived empirically from historical data, is tionship such as that of Eq. (3-63). In phys-
used to describe temporal occurrences. Us- ical terms, the significance of the error is
ing basic probability concepts, a brief that, in addition to the possible combina-
derivation of the method is given below. tions of magnitude and distance causing the
Let Y > y be the event that a ground event Y > y, the error could also result in
motion or spectral ordinate exceeds a value Y > y. Thus, if the error is independent of
y. The simplest probabilistic hazard model magnitude and distance,
is one where Y > y is caused by the occur-
rence of an earthquake whose magnitude P(Y'>y)= {P(Y>ylv)p(v)dv (3-66)
exceeds m at a distance r from the site. If 0

earthquake magnitudes are assumed to oc-


where Y' = v Y and the probability density
cur in the interval me .:::; m .:::; mx, where
p( v) is usually derived from a lognormal
magnitudes greater than a maximum mag-
distribution. For a given exceedance proba-
nitude, mx, are not considered possible and
bility, this additional integration typically
magnitudes less than me are not of engi-
results in larger ground motions or spectral
neering interest, the probability P(Y > y) is
ordinates than the median values obtained
given by
by ignoring the variability. The difference
increases with the standard deviation of the
P(Y > y) = jmxP(Y > y I M > m)p(m) (3-64) error term in the attenuation relationship.
m, The results of a probabilistic hazard
analysis are sensitive to the assumptions
where p(m) is the probability density func- made concerning attenuation and, particu-
tion of earthquake magnitudes which may larly, to assumptions concerning zonation
be derived from a magnitude-recurrence and recurrence. However, it should be rec-
relationship. ognized that the EPRS is similar to scaled
Design Loads 317

spectra and spectra computed from an at- acceptable provided one can choose values
tenuation relationship in that it represents of magnitude and distance in a rational
the result of a combination of earthquakes. manner. In most parts of the world it is not
One essential difference is that an EPRS difficult to judge a maximum magnitude.
represents the result of all possible earth- However, it is often very difficult or impos-
quakes, whereas scaled spectra and spectra sible to choose a distance in a rational man-
computed from an attenuation relationship ner. By rational, it is meant that the occur-
represent only those earthquakes that were rence of an earthquake of the assumed
used in the estimation of the amplification magnitude at a particular distance is a
factors or the regression coefficients, respec- geophysically1 seismologically possible
tively. Also, the weighting done in combin- event. If the port facilities lie within a seis-
ing the effects of several earthquakes to mic zone of areal extent where the seismic-
produce an EPRS is done according to the ity cannot necessarily be associated with a
probability of exceedance as defined by the causative geologic structure, one is not enti-
recurrence relationship, the zonation geom- tled to assume that an earthquake can oc-
etry, or other factors included in the proba- cur anywhere within the zone. A wholly
bilistic model. deterministic estimate requires that a
Probabilistic hazard analyses have been mechanism for earthquake generation be
performed in a variety of contexts to esti- specified. Past observations in California
mate peak ground motions having a speci- suggest that about 40 km of fault rupture
fied probability of exceedance (Algermissen must occur to produce an M6.0 event. Thus,
and Perkins, 1976; Basham et al., 1985). for example, the assumption of the occur-
The existence of attenuation relationships rence of an earthquake directly beneath the
for spectral ordinates makes possible the site is likely not rational and often results
computation of an equal probability re- in extremely conservative designs. In a de-
sponse spectra (EPRS) [i.e., a spectrum in terministic analysis, everything is assumed
which each ordinate has an equal probabil- to be known or predictable by judgment or
ity of exceedance (see, for example, EERI past experience. The possibility of an earth-
Committee on Seismic Risk, 1989)]. Cur- quake occurring anywhere within a seismic
rently, there is little precedent for the di- zone is a probabilistic notion which must be
rect use of such a spectrum in design. How- excluded from a deterministic analysis.
ever, because an EPRS is consistent in an
unbiased manner with available seismologi- 3.7.3.6 statistical Estimates
cal data, it is useful for comparative pur-
poses, (i.e., to determine the exceedance A statistical approach to ground motion
probability associated with spectra derived estimation begins by defining the magni-
by other means). tude and distance range of the controlling
earthquakes based on the known seismic
3.7.3.5 Deterministic Estimates characteristics of the region surrounding the
site. Recorded accelerograms due to past
Given an assumed magnitude and a dis- earthquakes in these magnitude and dis-
tance to a causative geologic structure, one tance ranges are then used to form a suite
could compute ground motions andjor spec- of representative time histories or to define
tral ordinates using attenuation relation- response spectra at various percentile lev-
ships such as those discussed above. The els, typically median and 84 percentile. The
computed values would be median values. If recorded accelerograms are selected to have
desired, one could also compute 84 per- similar source, propagation path, and
centile values. A deterministic approach is recording site conditions as the controlling
318 Design Loads

earthquake(s). This approach has been base shear force is applied statically to the
adopted by the U.S. Nuclear Regulatory structure.
Commission (USNRC, 1989) as a means of The zone coefficient, Z, is related to the
obtaining a site-specific response spectrum. seismicity of the region surrounding the
The 84 percentile spectrum is used to define structure. It may range from a value of 1 in
the motions due to the "Safe Shutdown regions of low seismicity to a value of 3 in
Earthquake." This would be appropriate for regions of high seismicity (e.g., Uniform
critical port structures. A suitable design Building Code, 1982).
spectrum for a less critical structure might The importance coefficient, I, increases
be the median spectrum. The use of repre- with increasing risk to life and limb posed
sentative accelerograms to define a design by structural failure. For example, in the
spectrum on a statistical basis is feasible, Uniform Building Code (1982), 1 :::;; I :::;; 1.5.
as available databases of accelerograms are The building or structure coefficient, K,
quite large. Usually, suitable accelerograms is related to the type of structure and is
are available to adequately describe the intended to place less stringent demands on
range of possible ground motion due to structures that have performed well in
earthquakes within a small-magnitude earthquakes and raise the margin of safety
range and occurring within a given distance for structures that have performed badly.
range. The principal advantage of the ap- Thus, most building codes specify 0.5 <
proach is that the effects of controlling K < 1 for a double steel frame, but K :::::
earthquakes are directly incorporated by 2.5 for an elevated tank.
only considering accelerograms due to The site factor, S, is a function of the
earthquakes in a restricted magnitude and ratio of the fundamental structural period
distance range. In addition, the estimates to the fundamental period of the founda-
depend on real data and not on a model tion. Formulas for the fundamental build-
such as that implied by an attenuation rela- ing period are usually specified in codes,
tionship or probability model. but, in general, a geotechnical site investi-
gation is required to estimate the funda-
mental period of the site [Eq. (3-59) could
be used for a single-layer site]. In all cases,
3. 7.4 Design Loads s;:::: 1.0.
The base shear coefficient, C, is a func-
3. 7.4.1 Building Code Seismic tion of the fundamental natural period, T,
Design Principles of the structure. In the Uniform Building
Code (1982) the formula for C is
The seismic design provisions of most
building codes, particularly those in North T
America, are based on identical principles. C = -- T~ 0.3s
15/T' (3-68)
The goal in each is to compute the maxi-
mum base shear force, V, as the product of C = 0.12, T < 0.3 s
six factors
The use of the fundamental natural period
V=ZIKSCW (3-67) is justified because, for most structures, the
seismic response is dominated by the re-
sponse in the fundamental mode.
where Z, I, K, and S are dimensionless The weight of the structure, W, is really
coefficients, C is a base shear coefficient, the effective weight in the fundamental
and W is the weight of the structure. The mode, which is typically less than the total
Design Loads 319

3. 7.4.2 Seismic Earth Pressures


As backfill is placed behind the wall and
0.3
as any loads are placed on the surface of the
c
backfill, an active thrust is developed which
.
0
;l
tends to push the wall outward as well as to
"
;;:: 0.2
i overturn the wall around its toe. Sliding
e
<
resistance along the base of the wall and
0.1
passive resistance of any soil lying over the
toe of the wall tend to counteract the out-
0.0
0.0 0.2 0.4 0.6 0.8 1.0
ward horizontal component of the active
Period (sec)
thrust. The weight of the wall and the verti-
cal component of the active thrust resist the
Figure 3-41. Comparison of the seismic coefficient overturning.
computed by Eq. (3-68) and the 2% damped pseudo- In the event of an earthquake, additional
acceleration response spectrum of the 1952 M7.2 Taft
earthquake. inertial forces are applied to the wall. The
most commonly used formulation to com-
pute the total (static plus dynamic) active
thrust on the wall is a modification of the
weight of the structure. However, as the Coulomb lateral earth pressure theory orig-
base shear force depends mainly on the inally presented in Okabe (1924, 1926), and
base shear coefficient, the additional con- Mononobe (1929) and further described by
servatism caused by assuming that W is Mononobe and Matsuo. (1929). It is now
the total weight is not significant. known as the Mononobe-Okabe formula-
The dimensionless of C and its depend- tion. Applications and discussion of the for-
ence on resonant period might lead one to mulation are given in Seed and Whitman
believe that it is the same as a pseudo- (1970), Richards and Elms (1979), Elms
acceleration spectral ordinate, PSA. Al- and Richards (1990), Prakash (1981), and
though it is true that they have comparable Whitman (1990).
meanings, a direct comparison is difficult For a detailed discussion on soil dyri.amic
because allowable stresses and the differ- pressures due to seismic or other dynamic
ences between static and dynamic behavior loads, the reader is referred to Chapter 5.
have been taken into consideration when
developing equations or values for C. A
comparison between the seismic coefficient 3.8 LOAD COMBINATIONS
computed according to Eq. (3-68) and the
2% damped pseudo-acceleration response The structure as a whole and all structural
spectrum of the 1952 Taft earthquake is members should be designed to provide an
shown in Figure 3-41. This is perhaps an acceptable and relatively uniform degree of
extreme comparison because 2% damping safety under different load combinations.
is rather low and the magnitude of the Taft As the dead load, which is the weight of
earthquake was large. However, it does in- the overall structure, including the relevant
dicate that the difference between C speci- weight of the backfill and permanently in-
fied by a building code and the value of the stalled fixed cargo handling equipment and
PSA given as the description of ground mo- miscellaneous structures, is practically con-
tion can be large. Details concerning the stant through the life of a structure, a com-
development of seismic design procedures bination of a dead load with any other load
in building codes are given in Berg (1983). is a basic combination in which appropriate
320 Design Loads

factors of safety are applicable. Note that a EP = earth pressures; both vertical andjor hori-
dead load is the weight in air; therefore, zontal, including unbalanced hydrostatic
where applicable, a buoyant weight should water pressure
be considered. L = the worst possible combination of live
When the dead load plus buoyancy is loads, such as any kind of loads generated
combined with some other loads (see the by cargo handling and hauling equipment,
including impact factors, uniform dis-
following load combinations) because of the
tributed loads, ship impact, and mooring
short duration of some of these loads (e.g., forces
ship impact, seismic load, and others), the
E = the worst possible combination of environ-
simultaneous occurrence of the full design mental loads, such as current, wind,
value of each of the loads involved has a low waves, and ice
probability. For example, seismic load usu- T = contraction or expansion due to tempera-
ally does not need to be considered concur- ture changes, shrinkage or creep in compo-
rently with maximum wind or wave forces nent materials, or a combination of these
or, say, ship impact. Therefore, the use of effects
somewhat smaller values for the factor of S = seismic load
safety or an appropriate increase in permis-
sible stresses in structural elements due to The following percentages of permissible
combinations of the dead load and buoyancy stress are recommended for the above load
effect with other load effects is justified. combinations:
In general, the designer, in consultation
with the dock owner, should scrutinize all • 100% for load combinations in Group 1
load combinations rationally. It should be through 5
noted that critical loading may occur not • 125% for load combinations in Groups 6
necessarily during operation of the struc- and 7
• 135% for load combination in Group 8
ture but rather during construction (trans-
portation, installation, etc.).
Load combinations and subsequent recom-
In designing the dock structure, all of the
mended unit stresses as discussed in this
loads discussed in this chapter plus soil
section are used in the working stress de-
(Chapter 5) and ice loads should be consid-
sign method, which is still basically used in
ered to act in the following basic combina- marine structures design. In the case that
tions, and the combination producing the the designer will select the limit state de-
most unfavorable effect should be selected. sign method, appropriate load factors must
be applied to all design loads.
Group 1 = D
Group 2 = D +B REFERENCES
Group 3 = Group 2 + EP AASHTO, 1992. Standard Specifications for
Group 4 = Group 3 + L Highway Bridges, 13th ed. with annual sup-
Group 5 = Group 3 + E plement, Washington, D.C.
Group 6 = Group 4 + T ABDELNOUR, R., 1981. "Model Tests of Multiyear

Group 7 = Group 5 + T Pressure Ridges Moving onto Conical Struc-


tures." Proceedings IAHR Symposium on Ice,
Group 8 = Groups 3 + 5 Quebec City, Quebec.
AFANAS'EV, V. P., DOLGOPOLOV, Y. V., and
In the above groups of load combinations, SHRAISHTEIN, Z. I., 1971. "Ice Pressure on
Individual Marine Structures." Ice Physics
D = dead load as previously defined and Ice Engineering, Israel Program for Sci-
B =buoyancy entific Translation, 1973, Tel Aviv.
Design Loads 321

ALGERMISSEN, S. T. and PERKINS, D. M., 1976. "A BERRILL, J. B., 1977. "Site Effects During the
Probabilistic Estimate of Maximum Accelera- San Fernando, California Earthquake." Pro-
tion in Rock in Contiguous United States." ceedings of the 6th World Conference on
USGS Open File Report No. 76-416. Earthquake Engineering, India.
ALLISTON, G. R., 1985. "Low Friction and BLANCHET, D., CHURCHER, A., FITZPATRICK, J.,
Adfreeze Coatings." Proceedings Conference and BARDA-BLANCHET, 1989. "An Analysis of
Arctic 1985. Observed Failure Mechanisms for Labora-
ANDERSON, J. G., BODIN, P., BRUNE, J. N., PRINCE, tory, First-Year and Multi-Year Ice." Special
J., SINGH, S. K, QuAAS, T., and ONATE, M., Report 89-5 by IAHR, edited by G. W. Timco,
1986. "Strong Ground Motion from the Hanover, NH.
Michoacan, Mexico Earthquake." Science, BLOK, J. J. and DEKKER, J. N., 1979. "On Hydro-
Vol. 233. dynamic Aspects of Ship Collisions with Rigid
or Non-Rigid Structures." Offshore Technol-
APELT, C. J. and PIOREWICZ, J., 1987. "Labora-
ogy Conference, Houston, TX, Paper OTC
tory Studies of Breaking Wave Forces Acting
3665.
on Vertical Cylinders in Shallow Water."
Coastal Engineering, Vol. 11, No.3. BLUME, J. A., SHARPE, R. L., and DALAL, J. S.,
1972. "Recommendations for Shape of Earth-
AssUR, A., 1975. "Problems on Ice Engineering." quake Response Spectra." AEC Report
Proceedings Third International Symposium WASH-1254.
on Ice Problems, Hanover, NH.
BOLT, B. A., 1969. "Duration of Strong Motion."
BALFOUR, J. S., FEBEN, J. C., and MARTIN, D. L., Proceedings 4th World Conference on Earth-
1980. "Fendering Requirement/Design quake Engineering. Santiago, Chile.
Fender Impact Criteria." Ports '80. ASCE
BOLT, B. A., 1987. Earthquakes, W. H. Freeman
Specialty Conference, Norfolk, VA.
& Co., New York.
BALL, D. J., 1982. "The Concept of Added Mass." BoMZE, H., 1980. "Math Model for Harbor
The Dock and Harbour Authority, Vol. 61. Wharves/Ship Motion/Moorings." Proceed-
BALL, D. J. and WILCOCK, J. A., 1981. "Ship and ing ASCE Speciality Conference PORTS '80.
Jetty Interaction." The Dock and Harbour Norfolk, VA.
Authority, Vol. 62, No. 727. BRATTLELAND, E., 1974. "A Survey on Acceptable
BAREL, P. Y. and GABRIEL, J. C., 1986. "The Ship Movements in Harbours." The Dock and
Seismic Response of Two Dimensional Sedi- Harbour Authority, Vol. 55, No. 647.
mentary Deposits with Large Vertical Veloc- BRIDGEMAN, S. G., 1984. "Wave Force and
ity Gradients." Bulletin of the Seismological Movement on Caisson." Computer Program
Society of America, Vol. 76. 'DIFMOM'.
BASHAM, P. w., WIECHERT, D., ANGLIN, F. M., BRIDGEMAN, S. G., 1987. Computer Software
and BERRY, M. J., 1985. "New Probabilistic "Wave Force and Moment on Pile."
Strong Seismic Ground Motion Maps of BRITISH STANDARD CODE OF PRACTICE FOR MAR-
Canada." Bulletin of the Seismological Society ITIME STRUCTURES, Part 1, General Criteria,
of America, Vol. 75. 1984, BS64349 including 1987 amendment.
BECK, J. L. and HALL, J. F., 1986. "Factors Con- BRITISH STANDARD CODE OF PRACTICE FOR MAR-
tributing to the Catastrophe in Mexico City ITIME STRUCTURES, Part 4, Design of Fender-
During the Earthquake of September 19, ing and Mooring Systems, BS6349, 1985.
1985." Geophysical Research Letters, Vol. 13. BRUUN, P., 1989. Port Engineering, 4th ed., Gulf
BERCHA, F. G., 1986. "Ice-Structure Interaction: Publishing Company, Houston, TX.
Engineering Design and Construction BURSNALL, W. J., and LOFTIN, L. K, 1951. "Ex-
Criteria," Vol. 1 Formal Report AESjSAG perimental Investigation of the Pressure
1-2 : 86-17V1 for Public Works Canada. Distribution About a Yawed Cylinder in the
BERG, G. V., 1983. Seismic Design Codes and Critical Reynolds Number Range." Technical
Procedures, Earthquake Engineering Re- Note 2463, National Advisory Committee for
search Institute, Berkeley, CA. Aeronauties.
322 Design Loads

BusHNELL, M., 1977. "Forces on Cylinder DANIELEWICZ, B. W. and BLANCHET, D., 1987.
Arrays." Offshore Technology Conference, "Measurements of Multi-Year Ice Loads on
Houston, TX, Paper 2903. Hans Island During 1980 and 1981." Interna-
CAMMAERT, A. B., KIMURA, T., NOMA, N., YASHIMA, tional Conference POAC Preprint, Fairbanks,
M., YANO, S., and MATSUSHIMA, Y., 1986. "Ad- AK.
freeze Forces on Offshore Platforms." Pro- DENT, E. G. and SAURIN, B. F., 1969. "Tanker
ceedings International OMAE Symposium, Terminals-Berthing Structures." Sympo-
Tokyo, Vol. 4. sium on Tanker and Bulk Carriers Terminal,
CAMPBELL, K. W., 1985. "Strong Motion Attenua- Institute of Civil Engineers, London.
tion Relationships: A Ten Year Perspective." DET NoRSKE VERITAS, 1977. Rules for the De-
Earthquake Spectra, Vol. 1. sign, Construction and Inspection of Offshore
CAN I CSA S6-M87. Design ofHighway Bridges,
Structures, Nl!lvik, Norway.
Canadian Standard Association. DOBRY, R., OWEIS, 1., and URZUA, A., 1976. "Sim-
plified Procedures for Estimating the Funda-
CANADIAN DEPARTMENT OF ENVIRONMENT, 1971.
mental Period of a Soil Profile." Bulletin of
"Review of Current Ice Technology and Eval-
the Seismological Society of America, Vol. 66.
uation of Research Priorities." Report Series
No. 17, Inland Water Branch. DROUIN, M., 1970. "State of Research on Ice
Thermal Thrust." IAHR Proceedings of Sym-
CHAKRABARTI, S. K., 1991. "Wave Forces on Off- posium, Reykjavik, Iceland.
shore Structures." Handbook of Coastal and
Ocean Engineering, Vol. 2, Herbich, J. B. (ed.), EAU, 1990. Recommendations of the Committee
Gulf Publishing Company, Houston, TX. for Waterfront Structures, Erust & Sohn,
Berlin.
CHAKRABARTI, S. K. and TAM, W. A., 1973. "Gross
EDWARDS, R. Y. and CROASDALE, K. R., 1976.
and Local Wave Loads on Large Vertical
"Model Experiments to Determine Ice Forces
Cylinder." Proceedings Offshore Technology
on Conical Structures." Symposium on Ap-
Conference, Houston, TX, Paper 1818.
plied Glaciology, International Glaciological
CHANG, K. S., 1964. "Transverse Forces on Cylin- Society, Cambridge.
ders Due to Vortex Shedding in Waves." EERI COMMITTEE ON SEISMIC RISK, 1989. "The
SM Thesis, Civil Engineering Department, Basics of Seismic Risk Analysis." Earthquake
Massachusetts Institute of Technology. Spectra, Vol. 5.
CHOPRA, A. K., 1981. Dynamics of Structures-A ELMS, D. G. and RICHARDS, R., 1990. "Seismic
Primer. Earthquake Engineering Research Design of Retaining Walls." ASCE Proceed-
Institute, Berkeley, CA. ings of Specialty Conference on Design and
CHRISTENSEN, T. F. and ZABILANSKY, 1., 1985. Performance of Earth Retaining Structures,
"Review of Experimental Studies of Uplifting Cornell University, Ithaca, NY.
Forces Enerbed by Adfrozen Ice on Marine EPRI, 1988. "A Criterion for Determining Ex-
Piles." Proceedings POAC 1985. ceedance of the Operating Basis Earthquake."
CORNELL, C. A., 1968. "Engineering Seismic Risk Research Report NP-5930, Electric Power Re-
Analysis." Bulletin of the Seismological Soci- search Institute, Palo Alto, CA.
ety of America, Vol. 58. ERANTY, E. and LEE, G. C., 1981. "Introduction
CROASDALE, K. R., 1980. "Ice Forces on Fixed, to Ice Problems in Civil Engineering." Depart-
Rigid Structures." Special Report 80-26, In- ment of Civil Engineering, State University of
ternational Association for Hydraulic Re- New York, Buffalo, NY.
search (U.S. Army Corps of Engineers, Cold FANG, S., 1979. "Field Measurement of VLCC
Regions Research and Engineering Labora- Mooring Loads." Proceedings Offshore Tech-
tory). nology Conference, Houston, TX, Paper 3383.
CROASDALE, K. R., MORGENSTERN, N. R., and FENCO ENGINEERS, 1984. "Arctic Marine Design
NUTTALL, J. B., 1977. "Indentation Tests to and Construction Handbook." Transport
Investigate Ice Pressure on Vertical Struc- Canada Report No. TP 5727E, Vol. 1, Fenco
tures." Journal of Glaciology, Vol. 19, No. 81. Engineers, Montreal.
Design Loads 323

FONTIJN, H. L., 1980. "The Berthing of a Ship to GRIM, 0., 1955. "Das Schiff und des Dalben."
a Jetty." Journal of the Waterway, Port, Schiffund Haven, Vol. 7.
Coastal and Ocean Division, VoL 106, No.2. GUREVICH, V. B., 1969. Marine Structures for
FONTIJN, J. L., 1987. "On the Prediction ofFender Inland Ports, Transport Publisher, Moscow
Forces at Berthing Structures, Parts I and (in Russian).
II." Proceedings NATO Advanced Study Insti- HANKS, T. C. and KANAMORI, H., 1979. "A
tute, Advances in Berthing and Mooring of Moment-Magnitude Scale." Journal of Geo-
Ships and Offshore Structures," Troudheim, physical Research, Vol. 84.
Norway. HAYAMA, Y. and SAKURAOKA, M., 1996. "Effect of
FREDERKING, R., 1980. "Dynamic Ice Forces on Compression Speed on Fender Performance,"
an Inclined Structure." Physics and Mechan- Proceedings of 11th International Harbour
ics of Ice, P. Tryde (ed.), Springer-Verlag, New Congress, Antwerpen, Belgium.
York.
HAYASHI, T., 1962. "Ship Impact During
FuJINo, M., 1968. "Steerability of Ship in Re- Berthing." Proceedings of Annual Presenta-
stricted Water." Shipbuilding Association Re- tions on Coastal Engineering (in Japanese).
port No. 124 (in Japanese).
HEADLAND, J. R., 1992a. "Design of Fender Sys-
GARRISON, C. J., TRUM, A., IVERSON, C. and tems Using a Numerical Berthing Model."
LEIVSETH, S., 1974. "Wave Forces on Large Permanent International Association of Navi-
Volume Structures-A Comparison Between gation Congresses Bulletin No. 77.
Theory and Model Tests." Proceedings of Off-
shore Technology Conference. Houston, TX, HEADLAND, J. R., 1992b. "A Computational
Paper 2137. Berthing Model for the Design of Fender Sys-
tems." ASCE Proceedings Specialty Confer-
GAYTHWAITE, J. W., 1981. The Marine Environ-
ence PORTS '92, Seattle, WA.
ment and Structural Design, Van Nostrand
Reinhold, New York. HEADLAND, J. R., 1995. "Offshore Moorings" in
"Marine Structures Engineering Specialized
GAYTHWAITE, J. W., 1990. Design of Marine Fa-
Applications," by G. Tsinker, Chapman and
cilities for the Berthing, Mooring and Repair
Hall, New York, N.Y.
of Vessels, Van Nostrand Reinhold, New York.
GERSHUNOV, E. M., 1985. "Shear Strength of the HENDRICK, G. J., 1980. "Testing of Used Tires for
Fendering Systems on Marine Structures."
Adfreeze Bond and its Effect on Global Ice
Loads Applied to Mobile Offshore Drilling Public Works Canada, File Q50-8. Unpub-
Units Under Arctic Conditions." Offshore lished Report.
Technical Conference, Houston, TX, Paper Otc HERBICH, J. B. (ed.), 1991/1992. Handbook of
4687. Coastal and Ocean Engineering, Vols. 1, 2,
GIRAUDET, P., 1966. "Recherches experimentales and 3, Gulf Publishing Company, Houston,
sur l'energie d'accostage des navires." TX.
Annales des Ponts et Chausees, No.2. HOGBEN, N. and STANDLING, R. G., 1975. "Experi-
GrnGRAH, M., 1977. "Practical Aspects of Dock ence in Computing Wave Loads on Large
Fender Design." Proceedings of the Twenty- Bodies." Proceedings of Offshore Technology
Fourth Congress, PIANC, SII-4, Leningrad. Conference, Paper 2189.
GoDA, Y., 1985. Random Seas and Design of HUDSON, D. E., 1979. Reading and Interpreting
Maritime Structures, University of Tokyo Strong Motion Accelerograms, Earthquake
Press, Tokyo, Japan. Engineering Research Institute, Berkeley, CA.
GoLD, L. W., 1984. "Ice Pressures and Bearing ISAACSON, M., 1988. "Wave and Current Forces
Capacity." Geotechnical Engineering for Cold on Fixed Offshore Structures." Canadian
Regions, Andersland, 0. B. and D. M. Journal of Civil Engineering, Vol. 15.
Anderson, D. M. (eds.), McGraw-Hill Book ISAACSON, M. and QI-HUA Zuo, 1989. "Nonlinear
Company, New York. Wave Forces on a Circular Cylinder." Cana-
GORUNOV, B. F., 1980. Specializing Terminals in dian Journal of Civil Engineering, Vol. 16.
Sea Ports, Transport Publishing House, ISAACSON, M., ALLYN, N., and ACKERMAN, C.,
Moscow (In Russian). 1995. "Design Wave Loads for a Jetty at
324 Design Loads

Plymouth, Montserrat." Canadian Journal of KERR, A. D., 1978. "Forces an Ice Cover Exerts
Civil Engineering, Vol. 22. on Rows or Clusters of Piles Due to a Change
IYER, S. H., 1983. "Size Effects on Ice and Their of the Water Level." Proc. IAHR Symp. on Ice,
Influence on the Structural Design of Off- Part I, Lulea, Sweden.
shore Structures." Proceedings International KERR, A. D., 1986. "Response of Floating Ice
Conference POAC, Vol. 2, Helsinki. Beams and Plates with Partial Flooding."
IZUMIYAMA, K., KITAGAWA, H., KoYAMA, K., and Proceedings of the 1st International Confer-
UTo, S., 1991. "On the Interaction Between a ence on Ice Technology, Cambridge, MA.
Conical Structure and Ice Sheet." Proceedings KHALFIN, J. S., 1990. Wave Forces on Offshore
of the 11th International Conference on Ports Platforms, Nedra, Moscow (in Russian).
and Ocean Engineering Under Arctic Condi- KHANNA, J. and SORENSEN, T., 1980. "Stand-Off
tions, St. John's, Canada. Force on Ships Moored in Strong Current."
JACKSON, W. H., 1973. "Some Problems Associ- Proceedings ASCE Specialty Conference
ated with Jetties in Fast Flowing Water." PORTS '80, Norfolk, VA.
Proceedings PIANC, 23rd Congress, Section KIEDAISCH, E., 1995. "Fender Systems in the 90's
II, Ottawa. and Beyond." ASCE Proceedings Specialty
JAMIESON, w. w., MANSARD, E. P. D., Conference PIANC '95, Tampa, FL.
and Mogridge, G. R., 1985. "Wave Load on a
K:!RKGOZ, M. S., 1992. "Influence of Water Depth
Conical Gravity Platform." Behaviour of
on the Breaking Wave Impact on Vertical and
Offshore Structures, Elsevier Science Publish-
Sloping Walls." Coastal Engineering, Vol. 18.
ers, Amsterdam.
KJELDGAARD, J. H. and CARSTENS, T., 1980.
JENSEN, 0. J. and WARREN, l. R., 1986. "Model-
"Thermal Ice Forces." In the Special State-of-
ing of Waves in Harbours. Review of Physical
the-Art Report 80-26, ed. T. Carstens. U.S.
and Numerical Methods." The Dock & Har-
Army Corps of Engineers CRREL, Hanover,
bour Authority, October.
N.H.
JENSEN, 0. J., VIGGOSSON, G., THOMSEN, J.,
BJ0RDAL, S., LUNDGREN, J., 1990. "Criteria for KORZHAVIN, K. N., 1962. Action of Ice on Engi-
Ship Movements in Harbours." Proceedings of neering Structures, USSR Academy of Sci-
the International Conference on Coastal En- ence, Siberia. Translated from Russian by the
gineering, Venice, Italy. U.S. Army CRREL in 1971, Hanover, NH.
JOHNSON, R. C. and BENOIT, J. R., 1987. "Iceberg KRY, P. R., 1978. "A Statistical Prediction of
Impact Strength." Proceedings OTC, Houston, Effective Ice Crushing Stresses on Wide
TX. Structures." Proceedings IAHR Symposium on
Ice Problems, Part 1, University of Lulea,
JOYNER, W. B. and BOORE, D. M., 1982. "Predict-
ion of earthquake response spectra." Proceed- Sweden.
ings 51st Annual Convention, Structural En- KRY, P.R., 1980. "Ice Forces on Wide Structures."
gineers Association of California (also U.S. Canadian Geotechnical Journal, Vol. 17,
Geological Survey Open File Report 82-977). No.1.
JoYNER, W. B. and BooRE, D. M., 1988. "Meas- LAIRD, A. D. K., 1962. "Water Forces on Flexible
urement, Characterization and Prediction of Oscillating Cylinders." ASCE Journal of the
Strong Ground Motion." Proceedings of Con- Waterways and Harbor Division, Vol. 88, No.
ference on Earthquake Engineering and Soil WW3.
Dynamics, II. Park City, Utah, ASCE, Wash- LAu, M. and WILLIAMS, F. M., 1991. "Model Ice
ington, DC. Forces on a Downward Breakin Cone." Pro-
JUNIUS, R. W., 1983. "Wharf Decks-Design Load ceedings the 11th International Conference
vs Cost." ASCE Proceedings of Specialty Con- on Ports and Ocean Engineering Under
ference PORTS '83, New Orleans, LA. Arctic Conditions, St. John's, Canada.
KERR, A. D., 1976. "The bearing Capacity of LEE, T. T., 1965. "A Hydraulic-Pneumatic Float-
Floating Ice Plates Subjected to Static or ing Fender." Technical Report R334, U.S.
Quasi-Static Loads." Journal of Glaciology, Naval Civil Engineering Laboratory, Port
Vol. 17. Hueneme, CA.
Design Loads 325

LEE, T. T., 1966. "Hydraulic-Pneumatic Float- MINIKIN, R. R., 1963. Wind, Waves and Maritime
ing Fender." Additional In-Service Tests, First Structures, 2nd ed., Griffin, London.
Series, Technical Report R430, U.S. Naval MIZOGUCHI, M. and NAKANAYAMA, T., 1973.
Civil Engineering Laboratory, Port Hueneme, "Studies on the Berthing Velocities and
CA. Energy of the Ships." Technical Note of Port
LINDSTROM, C. A, 1990. "Numerical Estimation and Harbour Research Institute, No. 170.
of Ice Forces Acting on Inclined Structures MOGRIDGE, G. R. and JAMIESON, W. W., 1976a.
and Ships in Level Ice." Proceedings Annual "Wave Loads on Large Circular Cylinders: A
Offshore Technology Conference, Houston, Design Method." National Research Council.
TX, Paper OTG 6445. Hydraulic Laboratory, Ottawa, Canada.
MAcCAMY, R. C. and FucHs, R. A, 1954. "Wave MOGRIDGE, G. R. and JAMIESON, W. W., 1976b.
Forces on Piles: A Diffraction Theory." U.S. "Design Method of the Estimation of Wave
Army Branch Erosion Board, Technical Memo, Loads on Square Caissons." National Re-
No. 69, December. search Council of Canada Hydraulic Labora-
MARCELLUS, R. W., MORRISON, T. B., ALLYN, N. F. tory, Ottawa, Canada.
B., CROASDALE, K. R., IYER, H. S., and TSENG, MoHRAZ, B., 1976. "A Study of Earthquake Re-
J., 1987. "Ice Forces on Marine Structures." sponse Spectra for Different Geological Condi-
Report AESjSAG 1-2:88-5V2 Prepared for tions." Bulletin of the Seismological Society of
Ministry of Supply and Services, Canada, America, Vol. 66.
1988. MONONOBE, N., 1929. "Earthquake-Proof Con-
MAzuRKIEWICZ, B. K., 1995. "Shiplifts, Marine struction of Masonry Dams." Proceedings,
Railways, Shipways and Dry (Graving) World Engineering Conference, Vol. 9.
Docks." Marine Structures Engineering: MONONOBE, N. and MATSUO, H., 1929. "On De-
Specialized Applications, by Tsinker, G. P., termination of Earth Pressure During Earth-
Chapman & Hall, New York. quakes." Proceedings World Engineering
McCANN, W. M. and SHAH, H. C., 1979. "De- Congress, Tokyo, Vol. 9.
termining Strong Motion Duration of Earth- MORRISON, J. R., O'BRIEN, M. P., JOHNSON, J. w.,
quakes." Bulletin of the Seismological Society and ScHAAF, S. A., 1950. "The Force Exerted
of America, Vol. 69. by Surface Waves on Piles." Petroleum Trans-
McGmRE, R. K., 1974. "Seismic Structural Re- action Journal, Vol. 189.
sponse Risk Analysis, Incorporating Peak Re- MORRISON, T. B., MARCELLUS, R. W., ABLYN, N. F.
sponse Regressions on Earthquake Magni- B., and TSEUG, J., 1988. Ice Forces on Marine
tude and Distance." Research Report R74-51, Structures, Vol. !-Calculations, Ministry of
Department of Civil Engineering, MIT, Supply and Services, Ottawa, Canada.
Cambridge, MA.
MoTORA, S., 1959. "On the Added Mass and
MICHEL, B., 1970. "Off-Shore Mooring Structure Added Moment of Inertia of Hull Movement."
for the Arctic." IAHR Proceeding of Ice Sym- Shipbuilding Association Reports Nos. 105
posium, Reykjavik, Iceland. and 106 (in Japanese).
MICHEL, B., 1971. "Winter Regime in Rivers and MURPHY, J. R., DAVIS, A. H., and WEAVER, N. L.,
Lakes." U.S. Army Cold Regions Research and 1971. "Amplification of Seismic Body Waves
Engineering Laboratory Monograph III, by Low Velocity Surface Layers." Bulletin of
Hanover, NH. the Seismological Society of America, Vol. 61.
MICHEL, B., 1978. "Ice Mechanics." Les Presses NAGASAWA, J., 1959. "Study on Prevention of
de L'University, Laval, Quebec. Damage to Ship Hull During Berthing." Ship-
MICHEL, B. and TOUSSAINT, N., 1977. "Mecha- building Association Report No. 106 (in
nisms and Theory of Indentation of Ice Japanese).
Plates." Journal of Glaciology, Vol. 19, No. 81. NATO, 1965. Analytical Treatment of the Prob-
MIDDENDORP, P., 1981. "Fender Forces from lems in the Berthing on a Mooring of Ships,
Berthing Ships." The Dock and Harbour NATO Advanced Study Institute/ASCE, New
Authority, October. York.
326 Design Loads

NATO, 1973. Analytical Treatment of the Prob- NUTTLI, 0. W. and HERRMANN, R. B., 1982.
lems in the Berthing on a Mooring of Ships, "Earthquake Magnitude Scales." JGT, Vol
NATO Advanced Study Institute, Walling- 108.
ford, U.K 0KABE, S., 1924. "General Theory on Earth Pres-
NATO, 1987. Advances in Berthing and Mooring sure and Seismic Stability of Retaining Wall
of Ships and Offshore Structures, NATO Ad- and Dam." Journal of Japan Society of Civil
vanced Study Institute, Trondhiem, Norway. Engineers, Vol. 10.
NAVFAC, 1968. Harbour and Coastal Facilities,
DM-26, Department of the Navy, Washing- , OKABE, S., 1926. "General Theory on Earth Pres-
ton, DC (Superceded by new series of point sure." Journal of the Japanese Society of Civil
manuals.) Engineers, Vol. 12, No. 1.
NAVFAC, 1980. Piers and Wharves, DM-25.1, OKSANEN, P., 1983. "Adhesion Strength of Ice."
Department of the Navy, Naval Facilities En- Proceedings International Conference POAC,
gineering Command, Alexandria, VA. Vol. 2, Helsinki.
NAVFAC DM-26.2, 1982. Coastal Protection De- OTANI, H., 1974. "A Study of the Berthing Im-
sign Manual 26.2. Department of the Navy, pact of the Large Tanker." Technical Note of
Naval Facilities Engineering Command, Port and Harbour Research Institute No. 176.
Alexandria, VA.
PADROU, D. V. (1994). Personal Communication,
NAVFAC, 1986. "Fixed Moorings, DM-26.7" De-
January 27.
partment of the Navy, Naval Facilities Engi-
neering Command, Alexandria, VA. PALO, P. A., 1983. "Steady Wind and Current
NEILL, C. R., 1976. "Dynamic Ice Forces on Piers Induced Loads on Moored . Vessels." Pro-
and Piles. An Assessment of Design Guide- ceedings Offshore Technology Conference,
lines in the Light of Recent Research." Houston, TX, Paper 4530.
Canadian Journal of Civil Engineering, Vol. PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
3, No.2. GATION CONGRESSES, 1984. "Ice Navigation."
NEVEL, D. E., 1972. "The Ultimate Failure of Supplement to Bulletin No. 46.
Floating Ice Sheets." Proceedings IAHR ICE PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
Symposium, Leningrad.
GATION CONGRESSES, 1984. "Report of the In-
NEWMAN, J. N., 197. Marine Hydrodynamics, ternational Commission for Improving the
M.I.T. Press. Cambridge, MA.
Design of Fender Systems." Supplement to
NEWMARK, N. M. and HALL, W. J., 1969. "Seis- Bulletin No. 45.
mic Design Criteria for Nuclear Reactor Fa-
cilities." Proceedings 4th World Conference on PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
Earthquake Engineering. Santiago, Chile, Vol. GATION CONGRESSES, 1987. "Development of
B.4. Modem Marine Terminals." Supplement to
NEWMARK, N. M. and HALL, W. J., 1982. Earth- Bulletin 56.
quake Spectra and Design, Earthquake Engi- PERMANENT INTERNATIONAL AssOCIATION OF NAVI-
neering Research Institute, Berkeley, CA. GATION CONGRESSES, 1995(a). "Criteria for
NEWMARK, N. M., BLUME, J. A., and KAPUR, Movements of Moored Ships in Harbours. A
K K, 1973. "Seismic Design Criteria for Practical Guide Supplement to Bulletin No.
Nuclear Power Plants." Journal of the 88.
Power Division, ASCE, Vol. 99.
PERMANENT INTERNATIONAL AsSOCIATION OF NAVI-
NIGAM, N. C. and JENNINGS, P. C., 1969. "Calcu- GATION CONGRESSES, 1995(b). Letter by Ad-
lation of Response Spectra from Strong
ministrative Director C. DeCraen to all Mem-
Motion Earthquake Records." Bulletin of the
bers ofPIANC, 29 November.
Seismological Society of America, Vol. 59.
NIXON, W. A., ETrEMA, R., MATSUISHI, M., and PRANG, M. K, 1977. "Dynamic Response of Ele-
JOHNSON, B. C., 1993. "Model Study of Cable- vated Conveyor Structure." ASCE Proceed-
Moored Conical Platform." ASCE Journal of ings of Specialty Conference PORTS '77. Long
Cold Regions Engineering, Vol. 6, No.1. Beach, CA.
Design Loads 327

PIASECKYJ, P. J., 1977. Proceedings of the SARPKAYA, T., 1976. "Vortex Shedding and Re-
Twenty-Fourth PIANC Congress, Leningrad sistance in Harmonic Flow About Smooth
(updated in 1982). Paper SII-4. and Rough Circular Cylinders." Boss '76,
PRAKAsH, S., 1981. Soil Dynamics, McGraw-Hill Trodheim, Norway.
Book Company, New York. SARPKAYA, T. and ISAACSON, M., 1981. Mechanics
PRAKAsH, S. and BASAVANNA, B. M., 1969. "Earth of Wave Forces on Offshore Structures, Van
Pressure Distribution Behind Retaining Wall Nosfrand Reinhold Co., New York.
During Earthquake." Proceedings 4th World SARPKAYA, T., COLLINS, N. J., and EVANS, S. R.,
Conference on Earthquake Engineering, 1977. "Wave Forces on Rough-Walled Cylin-
Santiago, Chile, Vol. A5. ders at High Reynolds Numbers." Offshore
QUINN, A. D., 1972. Design and Construction of Technology Conference, Houston, TX, Paper
Ports and Marine Structures, McGraw-Hill 2901.
Book Company, New York.
SAURIN, B. F., 1963. "Berthing Forces of Large
RAHMAN, M. and CHAKRAVARTI¥, I. C., 1986. "De-
Tankers." Proceedings Sixth World Petroleum
sign Method for Predicting Wave Forces on
Congress, Frankfurt, Section VII, Paper 10.
Rectangular Caissons." ASCE Journal of
Waterway Port and Coastal Engineering, Vol. SEED, H. B. and WHITMAN, R. V., 1970. "Design
112, No.6. of Earth Retaining. Structures for Dynamic
RALSTON, T. D., 1977. "Yield Criteria for Ice Loads." Proceedings ASCE Specialty Confer-
Crushing Failure." Ice Mechanics Workshop, ence on Lateral Stresses in the Ground and
Calgary, Alberta. Design of Earth Retaining Structures, Ithaca,
NY, pp. 103-147.
RALsTON, T. D., 1979. "Ice Force Design Consid-
erations for Conical Offshore Structures." SEED, H. B., UGAS, C., and LYSMER, J., 1976.
Proceedings of the Fifth International Confer- "Site-Dependent Spectra for Earthquake Re-
ence POAC '79, Trondheim, Norway. sistant Design." Bulletin of the Seismological
RESIO, D. T. and VINCENT, C. L., 1976. "Estima- Society of America, Vol. 66.
tion of Winds Over the Great Lakes" (Final SINGH, S. K., MENA, E., and CASTRO, R., 1988.
Report), MPH-76-12, U.S. Army Engineer "Some Aspects of Source Characteristics of
Waterways Experiment Station, Vicksburg, the 19 September 1985 Michoacan Earth-
Mississippi, June. quake and Ground Motion Amplification in
RICHARDS, R. and ELMS, D. G., 1979. "Seismic and near Mexico City from Strong Motion
Behavior of Gravity Retaining Walls." Data." Bulletin of the Seismological Society of
Journal of Geotechnical Engineering, Vol. America, Vol. 78.
105, No.4. SLINN, P. J. B., 1979. "Effect of Ship Movement
RICHTER, C. F., 1958. Elementary Seismology, W. on Container Handling Ropes." The Dock and
H. Freeman, New York. Harbour Authority, August.
RoGAN, A J., 1978. "Wave Forces on Piles." The SODHI, D. S., 1995. "Breakthrough Loads of
Dock and Harbour Authority, Vol. 59, No. 693. Floating Ice Sheets," ASCE Journal of Cold
RUPERT, D., 1976. Zir Bemessung und Konstruk- Regions Engineering, Vol. 9, No.1, March.
tion von Fendern un Dalben, Mitteilungen des
SODHI, D. S. and NEVEL, D. E., 1980. "A Review
Franzius-Institute fur Wasserbau und Kusten
Of Bucking Analysis of Ice Sheets." Working
Ingenieurs wesen aer Technische Universitat,
Group on Ice Forces on Structures. A State-of-
Hannover.
the-Art Report. CRREL Special Report 80-26.
SACKINGER, W. M. and SACKINGER, P. A, 1977.
Hanover, NH.
"Shear Strength of the Adfreeze Bond of Sea
Ice to Structures." Proceedings International SoNGEN, B., 1995. "The Effect of Wind on the
Conference POAC, Vol. 2, St. John's, Canada. Design of Canals." Permanent International
SAEKI, H., ET AL. 1981. "Experimental Study of Association of Navigation Congresses, Bul-
the Compressive Strength of Sea Ice and the letin No. 88.
Ice Forces on a Circular Pile." Coastal Engi- STAMMERS, A J. and WENNINK, C. J., 1977. "In-
neering, Vol. 19, December. vestigations of Vessel Mooring Subject to
328 Design Loads

Wave Action." The Dock and Harbour Author- TSINKER, G. P., 1995. Marine Structures Engi-
ity, Vol. LVII, No. 677. neering: Specialized Applications, Chapman
STAROSOLSKY, 0., 1970. "Ice in Hydraulic & Hall, New York.
Engineering." Norwegian Institute of Tech- TSUCIITYA, Y. and YAMAGUCHI, M., 1974. "Total
nology, Division of Hydraulic Engineering, Wave Force on a Vertical Pile." ASCE Pro-
Trondheim, Norway, Report N70-1. ceedings 14th Coastal Engineering Confer-
ence, Vol. III.
STELSON, T. E., 1995. "Virtual Mass and Acceler-
ation in Fluids." Proceedings of the American TYRELL, B. G., 1966 I 1967. "Mooring Dolphins-
Society of Civil Engineers, Vol. 81, No. 670. Evaluation of Forces and Principles of
Design." The Dock and Harbour Authority,
STEVENSON, J. D., KENNEDY, R. P., and HALL, W.
Vol. 47.
J., 1984. "Nuclear Power Plant Seismic De-
sign-A Review of Selected Topics." Nuclear UDOVICHENKO, V. N. and YAKOVLEV, P. I., 1976.
Engineering and Design, Vol. 79. Marine. Structures for Sea and Inland Ports,
Transport Publishing House, Moscow (in
SUNDAR, V., NEELAMANI, S., and VENDHAM, C. P., Russian).
1989. "Dynamic Pressures on a Large Verti-
cal Cylinder due to Random Waves." Coastal UEDA, S., 1981. "Study in Berthing Impact Force
Engineering, Vol. 13. on Very Large Crude Oil Carriers." Japanese
Port and Harbour Research Institute, Vol. 20,
SVENDSEN, I. A., 1970. "Measurement of Impact No.2.
Energies on Fenders." The Dock and Harbour
Authority, September-October. UNIFORM BUILDING CODE, 1982. International
Conference of Building Officials, Whittier, CA.
SWIFl', R. H., 1989. "Prediction of Breaking Force
U.S. ARMY CORPS OF ENGINEERS, 1984. Shore
on Vertical Cylinders." Coastal Engineering,
Protection Manual. Department of the Army.
Vol. 13.
Washington, DC.
THORESEN, S. A., 1988. Port Design, Tapir Pub-
U.S. ATOMIC ENERGY COMMISSION, 1973. Design
lishers, Troudheim, Norway.
Response Spectral for Seismic Design of
TIMco, G. W., 1987. "Ice Forces on Multi-Legged Nuclear Power Plants, Regulatory Guide
Structures." 3rd State-of-the-Art Report 87-17 1.60, Directorate of Regulatory Standards,
by Working Group on Ice Forces, Sanderson, Washington, DC.
T. J. 0., (ed.), U.S. Army CRREL, Hanover, USNRC, 1989. "Standard Review Plan for the
NH. Review of Safety Analysis Reports for Nu-
T!IIRUM, A., 1989. "Wave Forces on Pile in Sur- clear Power Plants, Section 2.5.2, Vibratory
face Zone." ASCE Journal of Waterway, Port, Ground Motion,. (Revision 2)." NUREG-0800,
Coastal, Ocean Engineering, Vol. 115. No.4. Office of Nuclear Reactor Regulation,
TRIFUNAC, M. D. and ANDERSON, J. G., 1978. Washington, DC.
"Preliminary Empirical Models for Scaling VALENTINE, H. R., 1967. Applied Hydrodynam-
Pseudo Relative Velocity Spectra." Methods ics, Butterworth, London.
for Prediction of Strong Earthquake Ground VAN 00RTMERSSEN, G., 1974. "The Berthings of a
Motion, U.S. Nuclear Regulatory Commission, Large Tanker to a Jetty." Offshore Technol-
Report NUREGjCR-0689, Al-A90. ogy Conference, Houston, TX, Paper OTC
TR!FUNAC, M. D. and BRADY, A. G., 1975. "A 2100.
Study of the Duration of Strong Earthquake VAN OORTMERSSEN, G., 1976. "The Motion of a
Ground Motion." Bulletin of the Seismological Moored Ship in Waves." Ph.D. Thesis, Delft
Society of America, Vol. 65. University, The Netherlands.
TRYDE, P., 1989. "Ice Engineering." Port Engi- VAN OORTMERSSEN, G., 1986. "Computer Simula-
neering, Bruun, P. (ed.), Gulf Publishing tion of Modem Ship Behaviour." Proceedings
Company, Houston, TX. ASCE, Vol. 112, No. 2.
TSINKER, G. P., 1986. Floating Ports, Gulf Pub- VANTE CHOW, 1959. "Open-Channel Hydraulics,"
lishing Company, Houston, TX. McGraw-Hill Ryerson Ltd., New York, N.Y.
Design Loads 329

VASCO COSTA, F., 1964. "The Berthing Ship." The on Earthquake Intensity." Journal of the
Dock and Harbour Authority, VoL 45, Nos. Structural Division, Vol. 90.
523, 524, and 525. WILSON, B. W., 1951. "Ship Response to Ranging
VASCO COSTA, F., 1983. "Moored Ships as Oscil- in Harbour Basins." Transaction of the ASCE.
lating Systems." The Dock and Harbour WILSON, B. W., 1958. "The Energy Problems in
Authority, February. the Mooring of Ship Exposed to Waves." The
VIGGOSSON, G., 1988. "Field Observatory of Ship Princeton University Conference on Berthing
Behaviour at Berth. Advances in Berthing and Cargo Handling in Exposed Locations,
and Mooring of Ships and Offshore Struc- Princeton, NJ.
tures." NATO, AS1 Series E, Applied Sci- WILSON, B. W., 1959. "A Case of Critical Surging
ences, Vol. 146. of a Moored Ship." Proceedings of the ASCE,
WAKE, A., POON, Y-K., and CRISSMAN, R., 1987. Vol. 85, No. WW3.
"Ice Transport by Wind, Wave, and Currents." WILSON, B. W., 1967. "Elastic Characteristics of
ASCE Journal of Gold Regions Engineering. Moorings." ASCE Journal of the Waterways
WANG, S., 1975. "Dynamic Effects of Ship Pas- and Harbours Division, Vol. 93, No. WW4.
sage on Moored Vessels." Proceedings ASCE, WILSON, B. W., LEVINTON, and BAKER, 1958.
Vol. 101, No. WW3. "Berthing and Cargo Handling in Exposed
WESSELS, E. and KATO, K, 1989. "Ice Forces on Locations." Proceedings of Princeton Univer-
Fixed and Floating Conical Structures." sity Conference.
Fourth Special State-of-the-Art Report 89-5. WONG, H. L. and JENNINGS, P. C., 1975. "Effects
U.S. Army Corps of Engineers CRREL, of Canyon Topography on Strong Ground
Hanover, NH. Motion." Bulletin of the Seismological Society
WHITMAN, R. V., 1990. "Seismic Design of Grav- of America, Vol. 65.
ity Retaining Walls." ASCE, Proceedings Spe- Woon, J. S., KHANNA, J., and DEMBINSKI, M.,
cialty Conference on Design and Performance 1980. "Design of an Offshore Berth for Moor-
of Earth Retaining Structures, Cornell Uni- ing Large Ships in a Strong Current." Pro-
versity, Ithaca, NY. ceedings PIANC, 25th Congress Section II,
WIGGINS, J. H., 1964. "Effects of Site Conditions Vol. 1. Edinburgh, UK.
4
Geotechnical Aspects of
Soil- Structure Interaction
Design considerations

4.1 GENERAL investigation techniques, both in the field


and laboratory, an engineer's ability to solve
Marine engineering involves the combina- increasingly complex problems by analyti-
tion of several skills, including an intimate cal and numerical methods has now out-
knowledge of all structural design aspects stripped an ability to understand some rele-
as well as an awareness of soil and rock vant soil properties and parameters, as well
mechanics and foundation engineering. as some peculiarities in soil-structure
Furthermore, a firm understanding of the interaction.
basic principles of soil-structure interac- It should also be realized that the lateral
tion is a must. This is most apparent in soil pressure distribution along a retaining
connection with nonrigid structures such as structure is governed almost exclusively by
sheet-pile bulkheads, or other structures movements or deformations of the struc-
where soil pressure distribution at the in- ture. The magnitude of these deformations
terface is governed by the deformation or (movements) is in:f:l.uenced to a large degree
displacement of such structures. by the type of structure (e.g., gravity wall or
It should be noted that in terms of flexible sheet-pile bulkhead), stiffness of the
soil-structure interaction, modern marine foundation material, and the method and
engineering practice is still far from being sequence of excavation and backfilling.
an exact science, and the designer is ad- Many of the complexities in civil engi-
vised to carefully consider what has already neering are a result of a nearly limitless
been achieved in this field of engineering number of variables associated with soil
and be well informed of the failures that materials of different properties. Further-
have occurred. more, because of potential physical and
The designer should recognize the fact chemical (sometimes biological) changes
that despite exceptional advances in soil brought about by human activities, or

331
332 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

changes in environmental conditions, soil tests indicated substantial increase in pene-


cannot be treated as a material with fixed tration resistance over a period of several
properties. For example, tidal fluctuation or months. The above phenomenon of soil be-
operation of miscellaneous cargo handling havior was, and still is, a big surprise when
and hauling equipment may, with time, sig- discovered. Although there are a few hy-
nificantly change density and other associ- potheses to explain this phenomenon, its
ated properties of freshly deposited granu- precise mechanism is still unexplained.
lar fills behind a quay wall. One explanation offered by Mitchell and
The marine structural designer should Solymar (1984) is that the time-dependent
also be aware of some practical problems strength gain phenomenon involves the for-
associated with the sometimes surprising mation of silicic acid gel films on particle
soil behavior. For example, aging of quick surfaces and the precipitation of silica or
clay or freshly densified or deposited sand other material from solution or suspension
may change the properties of these materi- as a cement at particle contacts.
als quite drastically (Mitchell, 1976, 1986). The above phenomenon, however, is an
Other unusual cases noted by Mitchell are exception rather than the rule. The common
concerned with disintegration in time of an general practice quite convincingly proves
apparently sound time-stabilized soil and that the disturbance of a natural sand de-
the failure of excess pore pressure to dissi- posit without follow-up densification usu-
pate as predicted during typical consolida- ally results in loss of strength. However,
tion of soft clays. dynamic impact of any kind will densifY
Whereas the properties of clays have long
previously disturbed sand, particularly sat-
been recognized as dependent on physico-
urated sand, and increase penetration re-
chemical factors of various types, it comes
sistance several times (Tsinker, 1988). This
as more of a surprise to find that clean
happens because loosely packed saturated
sands may be weakened as a result of den-
sands may easily attain a liquefied state
sification, rather than exhibiting time-
during dynamic impact, accompanied by an
dependent strength gain. Weakening of
sands due to densification was observed by increase in pore pressures. Then the pro-
Durante and Voronkevich (1955) more than cess of particle redeposition takes place,
35 years ago. Loss of strength of granular resulting in densification of sand, accompa-
soil as a result of vibrocompaction has also nied by a gradual decrease in excess pore-
been reported by Dudler et al. (1968). The water pressures and outflow of water to the
latest observation of such a phenomenon surface of the soil (Ivanov, 1967; Tsinker,
has been reported by Solymar et al. (1984), 1972). As noted by Ivanov, after compara-
Solymar (1984), and Mitchell and Solymar tively rapid cessation of the phenomenon of
(1984). In the case in question, the clean liquefaction and compaction of soil over a
riverbed alluvial sand, fine-to-coarse substantial time period, the pressures in
grained with a mean grain size of about 1.0 the soil skeleton decrease slightly, owing to
mm and an average uniformity coefficient of the action of the ascending seepage flow.
2.94 had been densified by vibrocompaction With an increase in the density of sand, the
to the depth of 25 m. Below this level, blast- influence of further shock waves is less pro-
ing was used to densizy the sand to a depth nounced. After a certain amount of dynamic
of 40 m. Cone penetration tests, conducted action, such as vibrocompaction, use of ex-
shortly after blasting, indicated that resis- plosives, or earthquake effects, the phe-
tance of the sand to penetration was less nomenon of densification and the substan-
than before, even though the sand was quite tial increase in excess pore-water pressures
obviously densified. Subsequent penetration is practically stopped.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 333

4.2 SUBSURFACE INVESTIGATION tions. They should be made at close enough


intervals across the site to ascertain the
Usually all design work begins with a sub- actual geotechnical conditions that exist.
surface investigation. The following is a The spacing of the borings will vary with
brief introduction to soil investigation that the soil conditions for each locality. In gen-
is required for design of the marine struc- eral, the boring program should remain
tures. flexible and should be evaluated during the
The proper design of any civil engineer- course of soil investigation, as changes may
ing structure, or marine structure in partic- have to be made to the program to properly
ular, requires adequate information on the define the subsurface soil condition. The
subsurface conditions at the construction soil borings should be to a depth that will
site. This information typically includes but encompass all the soils which can be signif-
is not limited to the following: icantly influenced by the structure loadings.
This depth is typically one to one and one-
half times the size of the loaded area below
1. Areal extent, depth and thickness of each
identifiable soil and rock stratum, within
the point of load application. For piled
a limited depth governed by the size and structures, where the load may be applied
nature of the structure. to the soils at a greater depth, the borings
should be made to the depth of at least 3-5
2. Properties of all soils within the construc-
tion site and surrounding area which may m below expected pile tips.
affect stability of the structure, such as The scope of site investigation usually
density, shear strength, compressibility, depends on project design stage (e.g., con-
degree of cohesiveness, and susceptibility ceptual, preliminary, or detail design) and
to liquefaction. In the case of rock, this is usually accomplished in a phased se-
will also include the state of weathering quence as follows:
or decomposition, spacing and orientation
of joints and bedding planes, and pres-
• Reconnaissance investigation
ence of fault zones.
• Investigation for conceptualjpreliminary
3. Location and chemical content of ground- design
water, and presence and magnitude of • Investigation for detail design
artesian pressures. • Investigation during construction as re-
quired
The required data can be obtained from
existing geotechnical survey reports, by Each phase of soil investigation together
conducting direct field reconnaissance, with the engineering done in that phase
soundings and probings, borings and test emphasizes problems that require further
pits, standard penetration tests, or field investigation in the next phase. The amount
tests such as direct shear tests, plate bear- of soil investigation is typically dependent
ing tests, pile load tests, seepage and water on the nature and size of the project and on
pressure tests, and other relevant tests the degree of complexity of the subsurface
wherein the required engineering proper- conditions. Local building codes usually
ties of in situ soils are measured directly. specify the minimum number of borings
Borings with the collection of samples of required for a given type and size of the
the subsurface materials and laboratory structure.
tests performed on these samples to deter- Sometimes when available subsurface in-
mine their strength and consolidation prop- formation is insufficient for the standard
erties are the most commonly used means design procedure because of tight schedules
for investigating the subsurface soil condi- set for completion of engineering work, the
334 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

design could be carried out based on the formation can result in structural failure or
best assumptions made by the engineer. unacceptable deformation andjor settle-
This course of action, however, must be ment.
approved by the client and the assumptions In conclusion it must be said that the site
made must be further verified in the field. investigation should be as thorough as
The latter may include results of pile tests, practical. Limitation of the scope of site
in situ soil investigation, and so forth. The investigation because of its cost can lead to
field data and observations may indicate false economy; lack of sufficient geotechni-
the necessary changes or modifications to cal information may result in designs which
the structure that was designed based on a are considerably more costly than those
limited soil information. Such an approach based on a comprehensive knowledge of the
is called the "observation method"; it is dis- geotechnical conditions of the site.
cussed further in this chapter. Detailed discussion on subsurface inves-
If the structure in question is based on tigation is beyond the scope of this work but
piles, then the soil investigator must be can be found in almost any standard text
aware of all the potential loads that will be and handbook on foundation engineering
supported by piles and understand the ex- [e.g., Lowe and Zaccheo (1991) and Young
pected load transfer mechanics through the (1991)].
pile-soil system. Loads are generally trans-
ferred from the structure to the subsurface
materials through skin friction along the
pile shaft buried in the soil, pile point, and 4.3 SOIL LIQUEFACTION AND
lateral soil resistance. Depending on struc- EVALUATION OF
ture performance and the existing soil con- LIQUEFACTION POTENTIAL
ditions, either one or a combination of two,
<BY G. TSINKER AND
or all load transfer modes may be used to
transfer the structure loads to the founda- S. DUNBAR)
tion soils. Hence, the soil investigation pro-
gram usually includes information on how The 1964 Niigata and Alaska and, in partic-
the loads will be transferred from the struc- ular, 1995 Kobe earthquakes caused great
ture through the pile system to the subsur- destruction directly attributed to soil lique-
face materials and on the expected conse- faction. The areas exposed to the earth-
quences of the loads and pile installation as quake effects have experienced liquefaction
a result of pile-soil interaction. Further- and permanent ground displacement of 1-2
more, the information obtained from subsoil m, resulting in substantial damages in-
investigation provides the piling contractor flicted to all kinds of marine structures (e.g.,
with information on a part of equipment gravity, piled, sheet-pile bulkheads, etc.).
and methods required for proper installa- The earthquake-induced soil liquefaction
tion of piles. and associated damages gave impetus to
The investigation program should also the intense research on this subject world-
consider examination of the existing struc- wide. In general, the cause of soil liquefac-
tures adjacent to the construction site area tion is the weakening of the interlocking
and of the potential effects of construction mechanism between soil grains by excess
activities (e.g., soil excavation, pile driving, pore pressure induced by ground shaking
etc.) on the integrity of these structures. It during the earthquake. Liquefied soil for a
has to be noted that the program of subsur- short period of time behaves as a heavy
face exploration should be planned with liquid. This causes the heavy structures
care; the lack of sufficient geotechnical in- standing on soil to sink or to drift from
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 335

their original location, and underground During the destruction of the soil's natu-
structures (e.g., storage tanks, pipelines, ral structure, or fabric, say by displacement
and alike) may suddenly acquire buoyancy of the upper layer of soil particles, repre-
and float toward the surface. sented by spheres to occupy the positions as
It should be noted that port facilities shown in Figure 4-1b. This results in a loss
constructed on reclaimed and relatively of contact between spheres. The existence of
loose lands that are comprised of materials spheres in such a position illustrates the
dredged from the local seabed, are espe- liquefied state of a soil. This loss of contact
cially vulnerable to the effects of earth- allows no transfer of contact stress to
quake; densification of loose granular soils neighboring spheres. Thus, the transfer of
that reduces the void spaces between gran- the spheres, weight and any external loads
ular particles also reduces potentials for does not take place and the system of
high pore-water pressure and therefore di- spheres acquires the capacity to spread.
minish potentials for soil liquefaction. One Under the influence of this weight, the
typical manifestation of the excess pore- spheres from the upper layer tend to drop
water pressure is the formation of sand down to fill up the gap between the spheres
boils on the surface. of the lower layer, resulting in a denser soil
The phenomenon of liquefaction of structure (Fig. 4-1c). Because the space be-
water-bearing noncohesive soils may be de- tween the spheres is filled with water, the
fined as a complete or partial loss of the downward movement of spheres of the up-
soil's shear strength or as the transition of per layer does not take place immediately;
a soil to an unstable state, emerging as a this depends on the rate of drainage of
result of the destruction of soil's natural water.
structure and displacement of soil particles Once the pore pressure increases to equal
relative to one another. A qualitative de- the soil pressure, the effective stress (soil
scription of the liquefaction process is quite pressure-pore pressure) equals zero and the
easy to understand. If a saturated cohesion- soil becomes liquefied and loses all capabil-
less soil is subjected to any kind of dynamic
ity to sustain shear. Further ground motion
impact, the particles making up the soil will
results in failure of the soil mass with obvi-
compact. If water cannot drain out of the
ous detrimental consequences for the struc-
soil mass, the resulting reduction in pore
ture. Evaluation of the potential for lique-
volume gives rise to an increase in pore-
faction of a soil mass is an important part
water pressure. This process is schemati-
of marine structure design. Quantitative
cally illustrated in Figure 4-1. Referring to
models of the liquefaction process are avail-
this arbitrary model, the phenomenon, in
able (e.g., Martin et al., 1975) and have
brief, may be explained as outlined below.
been used in detailed time-history analyses
of soil masses to evaluate the liquefaction
potential. However, such models are cur-
rently beyond the state of practice mainly
because they require material properties
that either cannot be reliably obtained or
are not measured in standard field investi-
gations. Consequently, the evaluation of
(a) (b) (c) liquefaction potential relies heavily on past
Figure 4-1. Sand liquefaction. Arbitrary model:
observational data. This approach has been
(a) before liquefaction; (b) liquid state; (c) post- developed and advocated by H. B. Seed
liquefaction state. and his co-workers at the University of
336 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

California at Berkeley and is described in Tokimatsu et al., 1986) used shear wave
detail in Seed and ldriss (1982). velocity (V8 ) to evaluate the liquefaction po-
Seed and Lee (1966) defined the liquefac- tential of soils. For sands, in particular, a
tion phenomenon based on cyclic load tests. good correlation between cyclic strength and
They defined initial soil liquefaction as the shear wave velocity has been demonstrated.
state when the pore-water pressure is equal Shear wave velocity can be determined by a
to the applied confining stress, and the com- variety of methods in the field or in a labo-
plete liquefaction is defined as the complete ratory. It is related to shear modulus at low
loss of soil shearing resistance over a wide strain (Gmax) based on the theory of elastic-
strain amplitude. ity by the following relationship: V8 =
For loose sand, the development of initial (Gmax/p) 0 ·5 , where pis the mass density of
liquefaction and complete liquefaction hap- soil.
pens almost simultaneously. For dense Seed et al. (1983) recommended the fol-
sand, complete liquefaction may never oc- lowing relationship between soil shear mod-
cur following initial liquefaction. A strain ulus and standard penetration blow count
criterion is then used by Seed and Lee to (N): Gmax = 65N, where Gmax is expressed
define liquefaction as the state when the in tons per square foot. For practical de-
cyclic load produces a certain strain ampli- signs, the reduction of the value of Gmax by
tude, typically between 5% and 20%. The 50% in the above equation to account for
latter is termed partial liquefaction.
expected strain levels during earthquakes
Vaid et al. (1985) found that a substan-
has been suggested.
tial decrease in soil resistance to liquefac-
On the basis of their observation of
tion may occur with an increase in confin-
earthquake sites in California, Stocoe and
ing pressure and angularity of sand
Nazarian (1985) found that sand did not
particles. They stated that angular sand
liquefy if its shear velocity V8 > 167.65 mjs
could be susceptible to liquefaction even at
relative densities approaching 100% under and did liquefy when Vs < 137.2 mjs. The
moderate earthquakes if the confining pres- shear wave velocity (or shear modulus) is
sure is high. On the other hand, at low typically required to calculate threshold ac-
confined pressures, angular sand is consid- celeration in the strain approach method
erably more resistant to liquefaction than for prediction of liquefaction proposed by
rounded sand over the entire range of Dobry et al. (1982). However, correlations
relative densities. between shear wave velocity and liquefac-
Assessment of liquefaction potential in tion resistance are not well established at
loose saturated natural or freshly placed present.
granular materials is an important matter Saxena et al. (1988) and Clough et al.
of practical design. As pointed out by some (1989) noted that the cyclic strength (resis-
investigators, cyclic shear strength and dy- tance to liquefaction) of naturally or artifi-
namic modulus of soils both depend on den- cially cemented loose sands, even with a
sity, strain, soil structure (gradation), stress small amount of cement present, increases
history, and so forth (Hardin and Drnevich, significantly over noncemented loose sands.
1972; Townsend, 1977; Saxena and Reddy, It also increases with an increase in rela-
1987; Elgamal et al., 1996). Therefore, use tive density of cemented sand and with an
of the above dynamic soil characteristics to increase in the curing period. As pointed
predict soil liquefaction is meaningful. Some out by Clough et al. (1989), cemented sands
other investigators (Dobry et al., 1982; Seed with unconfined compressive strength ex-
et al., 1983; De Alba et al., 1984; Stocoe and ceeding 100 kN jm2 are practically not
Nazarian, 1985; Liao et al., 1988, and liquefiable.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 337

Seed (1987) suggested that the most pru- drains can effectively prevent soil liquefac-
dent way of minimizing the hazards associ- tion. The Fishermen's Wharf suffered no
ated with liquefaction-induced deforma- damage; there were no indications of soil
tions (deflections) is to design the new liquefaction in the gravel-drain-protected
structure or devise remedial measures in area. In sharp contrast, the adjacent areas,
such a way that high pore-water pressures immediately outside the protected struc-
cannot build up in the potentially liquefi- ture, suffered severe damage from soil
able soil; hence, liquefaction cannot be trig- liquefaction.
gered. Seed also emphasized the impor- The gravel-drain construction technique
tance of paying careful attention to past has been used in Japan since 1978. At its
field performance data, coupled with mean- present development stage it begins with a
ingful in situ testing, in developing solu- casing auger drilling a vertical shaft into
tions to soil liquefaction problems in engi- the ground with as little disruption to the
neering practice. surrounding soil as possible. Then the
As indicated by Sonu et al. (1993), in gravel is poured into the shaft and com-
some instances danger of soil liquefaction pacted by the tapping motion of a tamping
can be precluded by installation of properly rod, while the casing auger is slowly with-
spaced gravel drains (Figures 4-2(a) and drawn by reversing its rotation [Figure
4-2(b)); these drains may help to reduce the 4-2(b)]. The casing auger, 40-50 em in di-
pore-water pressures by providing the es- ameter, may be as long as 25 m; typically,
cape route for the excess pore water. As installation of a gravel drain 20 m long
pointed out by the latter investigators, the takes less than an hour (Sonu et al., 1993).
soil densification could be detrimental for This technology is virtually noiseless, which
effectiveness of the gravel drains, because is an important advantage when construc-
the reduced space of pores would retard the tion is carried out in close proximity to
draining speed of the ground pore water. residential areas. The effects of gravel in
Several marine facilities in Japan protected sand-gravel composites is discussed by
by gravel drains suffered no damage due to Evans and Zhou (1995).
earthquake-induced soil liquefaction. One A typical example of soil liquefaction and
example of this kind is illustrated in Figure its impact on a sheet-pile bulkhead is illus-
4-2(a). The structure shown in this figure trated by the case history, depicted in Fig-
has been constructed at the Fishermen's ure 4-2(c). The bulkhead comprises con-
Wharf in the port of Kushiro, Hokkaido, crete sheet piles, anchor piles, and the cope
Japan. The wharfs bulkhead is made of wall. The central section, 40 m long, is sep-
steel cylinders 9.17 m in diameter and ap- arated from side sections with wing walls
proximately 9.0 m high. These cylinders are by expansion joints. During construction,
filled with dredged sand. Gravel drains 400 the bulkhead was overdredged by about 0.5
mm in diameter and approximately 6.0 m m. The backfill material (medium-to-coarse
long were installed within this cellular grain sand) was brought to the site in
bulkhead at a space of 1.2 m center to barges, then placed by a floating crane be-
center as is shown in Figure 4-2(a). After hind the wall right up to the design eleva-
installation the gravel drains were topped tion with no densification. Densification of
with layers of gravel and paved over. On the backfill material was designed to be
January 15, 1993, the port of Kushiro was conducted by the introduction of a large
struck by a major earthquake (magnitude amount of water on the surface of the fill, in
7.8 on the Richter scale) and the post- which case no significant dynamic forces
earthquake inspection of the port facilities were exerted on the backfill material. How-
decisively demonstrated that the gravel ever, before this was done, a small tugboat
338 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

Gravel dtam
¢ 400mm

C<ushed

Celulat
~ 6,000mm
rode

bulkhead
A-A
(a )

A . cas.ng auger
B. Tampong rod
C. Gravel hOpper

(b)

Figure 4-2. Backfill liquefaction. Case histories. (a) Port of Kushiro


Hokkaido, Japan. Fishennan Wharf; Typical plan and cross section. (After
Sonu et al. 1993). (b) Sequence of installation of a gravel drain (after Sonu
et al. 1993). (c) Damage to the dock structure due to the backfill
liquefaction. ! - concrete sheeting; 2- concrete anchor piles; 3-concrete
capping; 4- design dredge line; 5- actual dredge line; 6- sheet-pile
deflection; 7-back:fill settlement.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 339

(c)

Figure 4-2. Continued

hit the wall. This happened to be enough to tion, lateral earth pressure coefficient,
trigger massive backfill liquefaction, which depth of water table, and effective confin-
for a very short period of time (a few sec- ing pressure
onds) turned the fill material into a heavy 3. Earthquake and other dynamic impacts
liquid (KA > 1). This resulted in a substan- characteristics such as intensity of ground
tial wall displacement, as shown in Figure motion (e.g., peak ground acceleration)
4-2(c). The subsequent sand densification and duration of ground motion
resulted in about 5-10% in fill settlement.
These numbers are in a good agreement Not all of these factors can be directly de-
with data on settlement of saturated sands termined or explicitly included in the evalu-
provided by Tokimatsu and Seed (1986). ation. Those which cannot are implicitly in-
Vaid and Thomas (1995) provide the in- cluded by means of laboratory or in situ
sight into the mechanism of sand liquefac- procedures. The correlation between the re-
tion and its postliquefaction behavior. It sults of these procedures and the effects of
must be stressed that in order to avoid the above factors on liquefaction is the ba-
backfill liquefaction as well as its undercon- sis of the empirical evaluation procedure.
solidation or overconsolidation, the back- The basic steps of the evaluation procedure
filling procedure and method of soil place- are as follows (Seed and Idriss, 1982):
ment and densification must be given proper
attention during the design process. (a) Determine the cyclic shear stresses in-
The following basic factors are to be con- duced by the ground motion at different
depths in the soil mass. This may be
sidered for evaluation of soil liquefaction
done by a computerized ground motion
potential. analysis or by a simplified procedure de-
scribed below. The result is a plot of the
1. Soil properties such as dynamic shear average induced shear stress amplitude
modulus, damping characteristics, unit versus depth.
weight, grain characteristics, relative (b) Determine the cyclic shear stress re-
density, and soil structure quired to cause liquefaction (i.e., the
2. Environmental and construction factors cyclic shear strength) either by means of
such as methods of soil formation, seismic laboratory cyclic loading tests on repre-
history, geologic history, age, cementa- sentative samples conducted at different
340 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

confining pressures or by correlation ofis the depth at which shear stresses are to
shear strength with in situ measure- be computed, and amax is the peak ground
ments. acceleration of the design earthquake. The
(c) Compare the induced shear stresses with factor 0.65 is empirically determined from a
the shear strengths and determine where number of detailed laboratory analyses and
in the deposit the induced stresses ex- converts the maximum shear stress of an
ceed the cyclic shear strength.
irregular time history to a single equivalent
average shear stress amplitude. The factor
The first step incorporates the intensity and
r d is a coefficient whose value is less than 1
duration of ground shaking, whereas the
and accounts for the deformability of the
second incorporates the effects of soil type
soil column which results in a reduction of
and environmental factors affecting the
shear stress. The value of r d depends on
deposit.
depth in the manner shown in Figure 4-3.
Different procedures may be used to per-
form the ground response analysis. A com- Often the induced stresses are presented
mon computer model is SHAKE (Schnabel in the form of the ratio -ravl u~, where u~ is
and Lysmer, 1972) which basically uses the effective overburden pressure at the
shear beam theory to compute shear stress depth h. This is known as the cyclic shear
time histories in a layered soil deposit due stress ratio, and the origin of its use lies in
to horizontal ground motion at the base of the fact that laboratory strength data on
the deposit. Applications of this model are soil samples are often expressed in terms of
described in Chugh and VonThun (1985). such a ratio.
However, considerable expertise is required The Standard Penetration Test (SPT)
to perform such analyses. is the most commonly used in situ test for
The laboratory cyclic simple shear stress the determination of the cyclic shear
test is a desirable method of determining strength of soils. In North American engi-
cyclic shear strength. Although the test pro-
vides the best representation of induced
stresses on a soil sample during an earth-
quake, its accuracy is limited by the level of
disturbance of the soil sample. The effect of
sample disturbance has been investigated
by Mulilis et al. (1977) and Ladd (1977) and 10
has been found to be significant. Owing to 20 AVERAGE VALUES
the expense of obtaining "undisturbed"
samples and of conducting the laboratory 30
tests, in situ measurements are often used· RANGE FOR OIFFEREHT
to determine cyclic shear strength. SOIL PROFILES
A simplified procedure for computing in-
duced shear stresses was originally devel-
oped by Seed and Idriss (1971). The basis of
the procedure is the equation

where -rav is the equivalent average shear


stress amplitude (of a shear stress time Figure 4-3. Range of values of r d for different soil
history), y is the unit weight of the soil, h profiles. [From Seed and ldriss (1971).]
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 341

neering practice, the test measures the Because the liquefaction of a soil mass
number of blows, N, of a 63.6-kg (140-lb) depends on the duration of shaking (vibra-
hammer falling freely through a height of tion) which, in the case of earthquake loads,
76.2 em (30 in.) required to drive a standard depends on earthquake magnitude, curves
sampling tube (51 mm O.D., 38.1 mm I.D.) relating cyclic shear strength to SPT blow
305 mm into the ground. Standard test con- counts can be expected to depend on earth-
ditions and drilling procedures have been quake magnitude. Extensive studies by Seed
suggested by Kovacs et al. (1977). and his co-workers [outlined in Seed and
The SPT resistance, N, reflects both the Idriss (1982)] have resulted in the curves
influence of soil properties and the effective shown in Figure 4-4. These curves repre-
confining pressure, a 0. In cohesionless soils, sent the lower bound of cyclic shear strength
measured values of N tend to increase with required for liquefaction to occur. From the
increasing depth due to the increasing con- curves it may be seen that for a given cyclic
fining pressure. Such large values may not shear strength, the penetration resistance
be representative of the relative densities required for no liquefaction increases with
that actually exist. For this reason, a factor increasing earthquake magnitude.
eN, depending on overburden pressure The empirical method described in this
(depth), is applied to N to obtain a cor- section is based on :field performance of soil
rected blow count, N 1 : deposits and differs considerably from
methods based on testing and analyses. Al-
though an empirically based approach is
appealing, it must not be used without
Peck et al. (1974) give the following rela- judgment and due care concerning its appli-
tionship between eN and the effective over-
burden pressure at the depth of measure-
ment:
~1 1 ,.I I I
::SI I ~I ~I ~I
(4-2) J.,' 11..,j .:.~' .:....,1 o/,~1
11 I I ~I II II
:r:l :r:, :r:l :r:l :r:,'
I I I I
I I 1
where a 0 is measured in tonsjft 2 • The I
I
I
I

equation is valid for ao ~ 0.25 tonsjft 2 and


I

a wide range of relative densities (40-80%).


Graphs of eN versus ao are given in Seed
"'
and Idriss (1982) and give results similar to ...~
the above equation. "'....
Field data may be used to establish a ....>..J
....
relationship between the cyclic shear stress
ratio required to cause liquefaction and CURVES APPLICABLE FOR
standard penetration resistance. If such re- CONDITIONS WHERE o' < 1
TON PER SQ FT v
lationships were available, then the lique-
faction potential of a soil mass could be
0 o~----~,o------~2~0------~30~----~40'
evaluated by obtaining the values of N 1
HOOIFIED PENETRATION RESISTANCE, H1 blows/ft
and determining the corresponding values
of Tav =I= ao· These values are then compared Figure 4-4. Chart for the evaluation of liquefaction
with the cyclic shear stress ratio induced by potential for 'sands for earthquakes of different magni-
the design earthquake. tudes. [From Seed and ldriss (1982).]
342 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

cability. The advice of competent geotechni- ity to precisely control properties of the fill
cal engineers should be obtained. materials, and difficulties of predicting an
It is commonly believed that cohesive exact mode of structure movement, or de-
soils, particularly clays, are not susceptible flection (deformation) during construction
to liquefaction. However, as pointed out by and while in operation.
Seed and Idriss (1982), some types of clay Structural movement is very important,
may liquefY or at least suffer some loss of because it affects greatly the character of
strength due to earthquake ground motion. soilj structure interaction. This is probably
Thorough field investigations and testing of why in geotechnical engineering, and its
all types of soil deposits are necessary be- application to marine engineering in partic-
fore reaching conclusions on liquefaction ular, there is a tendency to be conservative
potential. especially in selection of the design soil
The SPT test has been used for many characteristics, prediction of structural
years to determine resistance or in situ movement and deflection, use of specific de-
strength. However, it is difficult to perform sign methods and values of factor of safety.
the test at depths greater than 100 ft and
through large depths of water. The Cone
Penetration Test (CPT), wherein a 1.4-in.-
diameter cone is pushed into the soil and 4.4.1 Modern Trends
the resistance to penetration is measured,
is a more rapid test and provides a continu- In the past 35-40 years there has been
ous record of penetration resistance. Suffi- [started by Casagrande (1964)] an increas-
cient CPT data are available to provide a ing amount of research into statistics and
basis for the empirical method described in risk analysis aimed at improving the cur-
this section. rent design practice in geotechnical engi-
neering. The milestones of this process are
outlined by Jamiolkowski (1988). Statistics
4.4 BASIC DESIGN AND provides procedures for obtaining informa-
tion from given quantitative measure-
CONSTRUCTION ments, which permits analysis of how the
CONSIDERATIONS aforementioned uncertainties of soil and
other parameters involved in soil-structure
As stated earlier in this chapter, geotechni- interaction may affect the design of the ma-
cal engineering in general, and particularly rine structure.
in the analysis of problems of soilj structure Risk analysis is a set of decision-making
interaction, is very far from being an exact procedures dealing with difficult design cir-
science. This is in a great part attributed to cumstances, where many components inter-
the fact, that " ... sources of success or fail- act such that there is more than one mode
ure (of foundations) are hidden deep in the of failure. In practice, for a great number of
ground ... " (Terzaghi, 1951). Despite great practical problems, it is common to use sim-
developments in geotechnical engineering plified and idealized soil models.
that have occurred in the past 50 years Some of the basic risk analysis proce-
turning this kind of science into a mature dures have been summarized and evaluated
branch of modern civil engineering, it still by Whitman (1984). It should be noted that
is to a great extent an art rather than although these procedures enable an engi-
science. This is primarily due to the limit- neer to better understand the relationship
less variation in properties of natural soils between uncertainty and safety, they are
at particular construction sites, our inabil- not commonly used in engineering practice.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 343

As stressed by Jamiolkowski (1988), statis- an economic solution on the one hand, and
tical and risk analyses are not to be consid- ensures environmental protection and pub-
ered as a substitute for conventional deter- lic health and safety on the other. This
ministic design methods, but rather they method is a departure from the traditional
represent a systematic approach to making design process because it allows decisions to
decisions. The designer must also look be- be made in the future, both during con-
yond just formal stress analysis and must struction when uncertainties become under-
always be concerned with environmental stood and during facility operation. The
and public health and safety aspects. Tradi- latter is particularly important where long-
tionally, decisions associated with facility term changes in soil-structure interaction
design are made during the design stage. are expected. More specifically, the observa-
The decision-making process usually not tional monitored decision method includes
only takes into consideration the best avail- an in situ check of assumed geotechnical
able information, but also allows for poten- parameters, as well as short- and long-term
tial adverse effects from industrial pollution monitoring of the soil-structure interaction
and uncertainties in the behavior of founda- process. This shifts the focus of the designer
tion soil. Sometimes potential changes in from just design to design-construction op-
facility operation requirements are also eration. The method also considers contin-
considered. gency plans if the working hypothesis
Due to a worsening legal climate for the requires some corrections or if unexpected
practice of engineering, designers are often deformations during facility construction
not willing to accept potential risks associ- andjor operation occur. The observational
ated with more economical or innovative monitoring decision method may stimulate
design. Conservatism, which is causing economical, innovative design, because from
economic problems and limits innovative the outset it considers improvements to the
approaches to design and construction prac- original design in case anything goes wrong.
tices, is very often justified by the uncer-
tainties in soil-structure interaction and by
insufficient knowledge of a project's
geotechnical conditions. 4.4.2 Bottom-Fixed structures
In 1948, Terzaghi pointed out that in
geotechnical (foundation) engineering, the The primary purpose of a bottom-fixed ma-
basic design should be treated as no more rine structure analysis is to predict its re-
than a crude working hypothesis, and the sponse to the anticipated design loads. The
project's success " ... depends primarily on a design process typically involves analysis of
clear perception of the uncertainties in- the structure's sliding, overturning and
volved in the fundamental assumptions and overall stability, analysis of the bearing
on intelligently planned and conscientiously stresses andjor stresses within the founda-
executed observations during construction." tion itself, and determination of deforma-
Peck (1969) gave structure to Terzaghi's tions within the soil-structure system.
observational approach, and D'Appolonia These analyses are divided by Poulos (1987)
(1990) extended the observational approach and Poulos and Hull (1989) into three cate-
to the method he called "monitored deci- gories: essentially empirical methods, meth-
sions." ods based on simplified theory and usually
The observational (monitored decisions) amenable to hand calculations, and meth-
method provides the designer with flexibil- ods based on more advanced numerical pro-
ity in the decision-making process because cedures and usually performed using a com-
it allows for innovations and, therefore, for puter with site-specific data as input.
344 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

Essentially, the use of more sophisticated specifically for the determination of the
design methods requires the use of more required soil parameters.
detailed soil parameters in order to gauge The most sophisticated analyses used in
the sensitivity of the soil-structure interac- geotechnical engineering are typically per-
tion to parameter variability. formed with the help of computers. Com-
Of course, the use of a sophisticated ana- puters help an engineer to conduct a com-
lytical procedure with crude soil parame- plex analysis much faster and more effi-
ters cannot be expected to provide any bet- ciently than it can be done by other means,
ter design than if a simpler analytical such as the pocket calculator. It should be
procedure were used with the same data. noted that the digital computer has now
Design methods utilizing empirical data been used in geotechnical engineering for
usually do not rely on the fundamental more than 30 years. The first widely used
principles of soil mechanics. computer application in geotechnical prac-
Methods falling into the second category tice was in the slope stability computation,
are based on simplified soil mechanics prin- where the analytical process becomes dubi-
ciples. Most sophisticated methods which ous when the "slip circle" becomes noncircu-
fall in the third category rely on the funda- lar. The use of a computer enables an
mental principles of soil mechanics. They engineer to examine "slip circles" of any
are based on theory (both linear and nonlin- configuration quickly. Computers have led
ear, elastic or elastoplastic) using site- to the development of essentially new tech-
specific analysis. niques such as finite element analysis,
As suggested by Poulos and Hull (1989), which were not used before the advent of
the following factors are to be considered in this electronic machinery. The latter en-
choosing a method of analysis and design ables many alternatives to be examined
for a structure: quickly. However, again, one must always
realize that soils are not linear in their
behavior and the properties which soils ex-
• Project importance hibit in the field may significantly differ
• Available budget from test values obtained in laboratory.
• Available geotechnical data Therefore, the use of inadequate soil data
• The complexity of geotechnical and load
or the introduction of an inadequate
conditions
• Design stage (e.g., conceptual, preliminary
soiljstructure model in finite element anal-
or final design) ysis may sometimes lead to a catastrophic
structural failure (Pilecki, 1987).
Nowadays, computers are widely used in
It is obvious that for small projects and geotechnical engineering. However, 'at pre-
simple geotechnical conditions, the simplest sent, as noted by Smith (1988), they are
design methods can be used. Sometimes the largely used in an automation mode rather
attempt to obtain overly detailed geotechni- than in a design-driving mode, and many
cal information may prove to be more costly engineers are still not confident of the abil-
than the results obtained by a conservative ity of computer analyses to accurately pre-
approach to structure analysis and design. dict soil-structure interaction. Physical
On the other hand, a sophisticated analysis models, and centrifuge modeling in particu-
would be appropriate for the final design lar, observations made on prototypes, and
stage in a major project utilizing relevant collection of field data are still the basic
geotechnical information obtained from an reliable means of resolution of some compli-
extensive site exploration program and from cated problems in soil-structure interaction
laboratory and/ or in situ tests carried out (Scott, 1987). This situation may change in
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 345

the future with the balance rapidly shifting fore based on a number of specific consider-
toward mathematical modeling. However, it ations related to a particular project. Focht
is difficult to expect that physical modeling (1994) grouped these considerations into
in geotechnical engineering, and particu- four categories:
larly large-scale modeling, will be replaced
completely by computational modeling. • Quality of soil parameters data (volume of
data, adequacy of test technique, scatter of
data, quality of samples)
• Appropriateness of analytical technique
4.4.3 Safety considerations (new, long established, or multiple proce-
dures, applicability to geologic environ-
mental compatibility with soil strength
The safety of geotechnical construction and
data)
miscellaneous soil-structure systems per-
• Type of loading (sustained, intermittent,
formance should always be given proper cycling, dynamic, seismic, single event,
consideration as part of the overall design mode of failure, redundant structure) and
process. Safe design is usually assumed to confidence in magnitude of load
be a proper proportioning of the structural • Consequence of failure (loss of life, struc-
elements and their interactions with the ture collapse, major damage to the struc-
foundation soil or fill material that ensures ture, aesthetic damage)
the structure's performance without unac-
ceptable deformations during its design life. All of these considerations determine the
Dangerous or unusual operations such as confidence level the designer wishes to
trenching, demolition, or lifting heavy com- achieve while predicting the soil-structure
ponents have to be strictly controlled, or interaction consequences. Some of the con-
completely prohibited if expected to be siderations are amenable to statistical and
detrimental to the safe construction and/ or probabilistic analyses, but others can only
operation of the structure. Safety require- be approached on a judgmental basis be-
ments are usually satisfied by employing a cause statistical data are not available. Not
suitable factor of safety. By definition, the included at all in any of these factors is the
factor of safety is the ratio of maximum uncertainty of construction quality.
available resistance of a structure to the An alternative design approach to the
resistance mobilized under the applied conventional working stress design method
loads. Traditionally, the working stress de- is one commonly referred to as the limit
sign method is used for design of marine states design method. This method is based
structures. Meyerhof (1970, 1984) examined on miscellaneous factors such as load fac-
the conventional working stress design fac- tors, resistance factors, load and resistance
tors of safety in relation to the probability modification factors, and other factors which
of failure used in geotechnical engineering. are used instead of the conventional factor
He suggested that the total factors of safety of safety. Baikie (1985) has reviewed the
for earth-retaining structures and founda- above two methods and concluded that, for
tions respectively are 1.5-2.0 and 2-3, with soil-retaining structures, both methods are
corresponding probabilities of failure of 10 -a in good agreement, provided that in the
and 10- 4 • latter method the load and resistance modi-
Focht and O'Neill (1985) and Focht (1994) fication factors are assumed to be unity.
presented discussions on choices of and ap- At present, the limit states design is
proaches to selection of factors of safety. rapidly taking over structural design of steel
They stated that the choice of a factor of and concrete structures. It is also gradually
safety should be project-specific and there- entering geotechnical foundation design,
346 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

particularly for piles. Hence, in the case at the base of the wall and is subject to
where the superstructure design is based some displacement due to deformations of
on a limit state design approach while foun- the foundation soil.
dation design is based on the conventional In all of the aforementioned cases of
working stress method, factors of safety soil-structure interaction, the distribution
used for both parts of the entire structure of soil pressures behind the wall is differ-
must be compatible. ent. Typically, gravity-type walls built on
As Focht (1994) pointed out, for success- stiff foundation materials are designed to
ful outcome of the design the designer must resist soil pressures under the "at rest" con-
have clear understanding of a site and re- dition, whereas the same walls built on rel-
gional geology and an appropriate modeling atively soft foundations are analyzed as-
of soil parameters, critical to the particular suming fully active soil conditions behind
problem [e.g., stability of the structure, the wall. The design of a "flexible" structure
soiljstructure interaction, safety factor(s), in one way or another should take into
and others]. The designer also must select account the potential soil arching effect be-
an applicable theory to conduct the analy- hind a wall.
sis. He or she must, however, utilize the Traditionally, the analysis and design of
judgment by questioning the applicability of flexible retaining structures are carried out
the selected theory to predict behavior of assuming limiting pressures, active behind
the structure. the wall and passive in front, to determine
To make the right decision, the marine the required depth of the wall. Further-
structures designer must have good and more, some empirical rules are applied to
broad understanding and appreciation of all determine wall movements, bending move-
ments, and shear forces. These usually take
engineering aspects involved with his or
into account the stiffness of the wall and
her. design (e.g., civil, geotechnical, struc-
location and type of anchorages.
tural, etc.). The designer must also not lose
In recent years many computer programs
sight of the fact that, in general, the techni-
(many based on finite element methods)
cal solutions that are included in the ma-
have been developed to analyze retaining
rine structures must be compatible with
structures. In many cases these programs
associated structures and their con- are quite complex in their formulation and
structibility. In a complex field of marine modeling technique, and thus susceptible to
engineering, the marine structures designer errors and misuse. Indiscriminate use of a
must be open-minded. Most geotechiiical computer program's output sometimes is
engineers are familiar with Terzaghi's ad- cause for concern, and a sheet-pile bulk-
monition that if the theory does not match head failure in Texas (Pilecki, 1987) is a
with expectations based on experience, then very characteristic illustration of computer
new theory should be developed. program misuse.
As it will be seen later, the conventional Pappin et al. (1986) developed a com-
approach adopted for the analysis and de- puter program which claims to be suffi-
sign of retaining structures depends on ciently simple and inexpensive for numeri-
whether the structure behaves as essen- cal analysis of flexible retaining walls. In
tially rigid (e.g., a gravity-retaining struc- this program, soil stiffnesses are modeled
ture built on a very stiff foundation) or a by using precalculated flexibility matrices
relatively flexible structure (e.g., a sheet-pile obtained from finite element computations
bulkhead, or a gravity-type structure built for elastic soil blocks. The earth pressure
on a "soft" foundation). In the latter case, limits allow for the known effect of soil
the wall provides restraint to the retained arching in addition to the active and pas-
soil by mobilizing gravity and friction forces sive pressure limits. For more information
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 347

on this subject, the reader is referred to tom contours are not steep, then soft sedi-
Chapter 6. ments may be displaced by progressively
placing a granular fill starting immediately
behind the structure (Fig. 4-6). Underwater
placed backfill material is normally settled
4.4.4 construction Procedure in a loose state and, as stated earlier, is
susceptible to liquefaction if exposed to any
Construction procedures (construction se- kind of dynamic effect (e.g., waves load,
quence) have a pronounced impact on re- earthquakes, ship impact, and miscella-
taining wall performance in general and on
flexible walls in particular. There are two
basic modes of sheet-pile bulkhead con-
struction: backfilled and dredged construc-
tion (Fig. 4-5). In the former method, the
bulkhead construction proceeds in the fol-
lowing steps: dredging of the natural bot-
tom to the design level; sheet-pile driving;
construction of anchor system; backfilling.
The latter proceeds in the following se-
quence: sheet-pile driving; construction of
anchor system; backfilling (if necessary);
dredging of the material bottom in front of
the structure to the design level.
In all cases, a free-draining granular ma-
terial is typically used for backfilling. Weak
soils such as organic silts, soft sediments,
and the like are usually removed from ar-
Figure 4-6. Fill placement behind a sheet-pile bulk-
eas immediately adjacent to a retaining head: ! - sheeting; 2-anchor system; 3-foundation
structure. This is done to avoid large lateral soil; 4- loose soil strata (e.g., organic silt, soft sedi-
stresses and deformations. If existing bot- ments); 5-granular fill material.

(a)

(b )

Figure 4- 5. Construction sequences: (a) "backfilled" bulkhead; (b) "dredged" bulkhead.


348 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

neous dynamic loads transmitted by cargo rials (sedimentary, igneous, and metamor-
handling and hauling equipment and phic) will affect the type and nature of the
others) and, therefore, should be densified. soils formed. A soil can be described by its
Vibrocompaction is a commonly used grain-size classification, appearance, and
method for backfill densification. This structure, and compactness or hardness.
method involves soil vibration by differ- There are several soil classification sys-
ent means (Dobson, 1986; James, 1973; tems, but the one most widely used in North
Anderson, 1974). The vibration is usually American engineering practice is the Uni-
delivered to the sand via a probe of required fied Soil Classification System (USCS). It is
length that is joined with a vibrator of re- defined in ASTM Standard D2487 and MIL-
quired capacity. The vibration causes a den- STD-619A. The general soil characteristics
sification of the loose sand as the probe is are discussed in the following paragraphs.
placed and extracted. The probe is kept Soil is by nature nonhomogeneous and
within the densified material until the anisotropic, displaying variable physical
desired density is obtained. and mechanical properties, even when sam-
There are some other soil compaction ples are obtained from the same deposit in
techniques, such as vibro:flotation, impact close proximity. Soils commonly found in a
compaction, blasting, preloading, and oth- marine environment are typically charac-
ers, which may be employed not only for terized as cohesionless (nonplastic) and co-
backfill densification but also for site im- hesive (plastic). Sand and gravel represent
provement or "ground modification" at the the first group of soils, and clay is charac-
construction site. Useful discussions on this terized as a plastic material. A third soil
subject matter are found in Ivanov (1967), category commonly found in a marine envi-
Solimar and Reed (1986), Brown (1977), ronment is silt, which is relatively nonplas-
Harder et al. (1984), Whiteneck and tic. Most soil deposits in a marine environ-
Hockney (1989), Hilf (1991), Broms (1991), ment consist of a sand-clay-silt mixture,
Chow et al. (1992), and others. such as sandy clay, silty sand, and so on,
The compaction of fills above the water where the second term is the predominant
level may be carried out by any method. material and the first indicates the filler.
Usually, the required density, moisture lim-
its, and lift thickness are specified and the
contractor is allowed some selection in com-
paction methods. 4.5.1 Gravel and Sand

Gravels and sands are mineral particles


known as coarse-grained soils. Coarse-
4.5 SOILS AND BEDROCK grained materials are such that 50% or
more of the materials by weight are re-
In a marine environment, soils are often tained on the No. 200 sieve. The USCS
referred to as sediments. The word soil cov- defines a soil as gravel when its size ranges
ers a large assortment of materials of vari- between 76.2 mm and the No.4 sieve. Mate-
ous origins; for engineering purposes, the rials larger than 76.2 mm are designated as
materials are generally classified as gravel, cobbles. A soil is defined as sand when its
sand, silt, clay, and organic material. Most grain size is between 4.8 and 0.075 mm (No.
soils are composed of mixtures containing 4 and No. 200 sieves, respectively). The
two or more of those materials. Different uses developed a further classification:
geological processes (such as alluvial, resid- The sand is coarse when its grain size varies
ual, glacial, and loessial) and parent mate- between 4.8 and 2.0 mm (No.4 and No. 10
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 349

sieves, respectively), medium when be- sands may be referred to as dirty gravels or
tween 2.00 and 0.4 rom (No. 10 and No. 40 sands. As noted previously, the particle
sieves, respectively), and fine when between shape has an influence on the density and
0.4 and 0.075 rom (No. 40 and No. 200 the stability of the coarse-grained soils.
sieves, respectively). Gravels and sands, both dry and satu-
Sands are cohesionless materials, but rated, generally provide good bearing resis-
they present an apparent cohesion when tance and low compressibility for either
damp or moist because of the surface ten- static or dynamic (cyclic) loads, provided
sion effects of pore fluids. The effects disap- they are in a relatively dense state. Satu-
pear when the sand is dry or completely rated sands in a loose state as a natural
saturated. Gravel and sand are recognized deposit or as fill may be susceptible to liq-
either visually and manually or more for- uefaction. Cyclic shear testing data which
mally by the following: indicates soil-water-pore pressure response
relative to stress level and number of load
• Effective grain size (D 10 ): grain size such cycles may be used in marine structures
that 10% by weight of the materials are design (Rocker, 1985).
finer Gravels that are of alluvial deposits are
• Uniformity coefficient, Cu = D 60 /D 10 usually mixed with sands to a greater or
• Coefficient of curvature, Cc = D~0 /D 10 D60 lesser degree. Dense gravelly and sandy
soils have a high resistance to the driving of
Because most soils are .composed of more concrete or wooden piles; steel piles or other
than one constituent, the uses makes the piles driven with the help of a water jet
following distinctions for sands and gravels: are typically used if deep penetrations are
necessary.
Well-graded gravel (GW) or sand (SW):
All particle sizes are represented within the
constituent limits; Cu is greater than 4 or 6,
respectively; Cc is between 1 and 3; and the
4.5.2 Silt and Clay
fraction smaller than the No. 200 sieve size
does not exceed 5%. Silts and clays are mineral particles known
Poorly graded gravel (GP) or sand (SP): as fine-grained soils. Fine-grained soils are
Some particle sizes are missing or are in ex- such that 50% or more of the material by
cess within the constituent limits; gradation weight passes the No. 200 sieve. They are
requirements for GW or SW are not met; and distinguished either visually and manually
the fraction smaller than the No. 200 sieve or by means of the Atterberg limits. The
size does not exceed 5%. uses does not make any size distinction
Silty gravel (GM) or sand (SM): More than between silt and clay because the engineer-
12% by weight is finer than thl;l No. 200 sieve, ing properties of fine-grained soils are more
and the fines have little or no plasticity. closely related to plasticity characteristics
Clayey gravel (GC) or sand (SC): More than to grain size. The USCS distinguishes
than 12% by weight is finer than the No. 200 the following:
sieve, and the fines are plastic.
Silt, clay, and organic silt and clay having
When the fraction smaller than the No. 200 liquid limits less than 50%
sieve size is greater than 5% and less than Silt, clay, and organic silt and clay having
12%, a dual symbol should be used. liquid limits greater than 50%
Well-graded and poorly graded gravels
and sands are further referred to as clean Silts and clays usually ·have a low perme-
gravels or sands. Silty or clayey gravels and ability (10- 7 -10- 9 cmjs) with silty soils
350 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

being somewhat more permeable than appropriate soil improvement program must
clayey ones. Organic materials tend to lower be instituted.
the strength characteristics of the soil, lower Clay is distinguished by its fine particle
the maximum density, increase the time for size and cohesive strength, which is in-
consolidation, and increase the optimum versely related to its water content. For
water content. that reason, clay's performance as a founda-
Silt is a fine-grained soil (mineral parti- tion material is strongly influenced by its
cles ranging in size from 0.05-0.075 mm stress history. In situ overconsolidated
maximum to 0.005-0.006 mm) of low plas- clays, clays which have been loaded to
ticity which may exhibit an apparent cohe- higher stresses than the present load, may
sion that is due to capillary forces. It is perform quite well in foundations. Normally
called organic silt if it contains appreciable consolidated or underconsolidated clays
quantities of organic materials, and inor- typical of estuaries will generally experi-
ganic silt if no organic materials are pre- ence large settlements when loaded. The
sent. Silts have relatively poor strength minerals included in the clay composition
characteristics, except when they are dry or influence the properties of the soil. For ex-
in the form of siltstones. Confined, rela- ample, montmorillonite is a highly active
tively dense silts may perform satisfactorily mineral, and soils containing such a min-
as foundation soil, but they must be evalu- eral will present high swelling and shrink-
ated on a case-by-case basis. Most coastal age characteristics. Two other commonly oc-
curring minerals are illite (less active than
silts are found in combination with some
montmorillonite and commonly found in
clay, which increases cohesion and im-
marine clays) and kaolinite (the least active
proves foundation characteristics.
mineral).
A uniform inorganic silt deposit can ex-
hibit appreciable strength values under
static loads even though natural water con-
4.5.3 Bedrock
tents are near or above the liquid limit
(Nacci, 1969). However, they may lose much Depending on their formation, rocks are
of their strength when disturbed by dy- classified as igneous, sedimentary, or meta-
namic or cyclic loads, but regain a great morphic. Igneous rocks have solidified from
deal of it with time. For example, silt de- a molten or partly molten siliceous solution
posits usually exhibit nearly complete loss (magma); sedimentary rocks are naturally
of strength when disturbed during pile- consolidated or unconsolidated transported
driving; however subsequent pile load tests materials. Metamorphic rocks are the result
indicate a substantial regain of lost of subjection of igneous or sedimentary
strength. rocks to elevated temperatures andjor
Organic silts are typically soft, highly pressures. Igneous rocks comprise about
compressible, and exhibit low shear 80% and metamorphic rocks about 15% of
strengths (Pierce and Calabretta, 1978). Or- the terrestrial and suboceanic earth crust,
ganic materials come from the decay of veg- leaving about 5% as sedimentary rocks.
etable matter. They are recognized by their However, about 75% of the subsurface of
odor, which is intensified by heating, and by continental platforms and a considerably
their dark color (although some dark soils higher proportion of the oceanic floors carry
may be inorganic). The presence of organic a veneer of sediments (Winterkorn and
soils is almost universally unacceptable for Fang, 1975).
construction of marine structures. Soil of According to Gay (1957), rocks can be
this kind must be either removed or an classified based mainly on composition and
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 351

texture as granite and related rocks, basalt which is the metamorphosed crystalline
and related rocks, carbonate rocks includ- equivalent of either type.
ing limestone and marble, sandstone, and In the marine structures construction in-
miscellaneous rocks (including chert, shale, dustry, the term limestone is applied to
slate, tufaceous volcanic rocks, corals, and many types of rock that contain a high
others). percentage of calcium carbonate, although
Granite and Related Rocks. The term large proportions of other substances also
granite is commonly applied to medium- to may be present. They also commonly con-
coarse-grained igneous rocks that consist tain clay, silt, and sand grains. A high per-
mainly of feldspar and quartz, and ordinar- centage of clay commonly weakens carbon-
ily contain subordinate proportions of ferro- ate rock, and a high content of sand grains
magnesian minerals. Mica may also be or silica may harden it. Marble is similar to
present. In small quantities, mica is not limestone chemically, but it has been sub-
particularly harmful, but in larger quanti- jected to a metamorphic process which has
ties it sets up planes of structural weakness made it more crystalline in structure and
and provides a starting point for disintegra- harder.
tion. Sandstone is a sedimentary rock which is
Although granites vary widely in texture composed of particles mainly in the size
and appearance, most are dense and have a range of about 0.25-6.4 mm in diameter.
porosity of less than 1%. Most unweathered Although some sandstones consist almost
granitic rocks are hard, strong, tough, and wholly of quartz grains, most sandstones
resistant to abrasion, impact, and chemical are feldspathic; some contain a high propor-
attack. These properties make granitic rocks tion of ferromagnesian minerals. The
well suited as a foundation for gravity-type strength and durability of sandstones are
marine structures and for use as riprap and mainly determined by the type of material
quarrystone armor. that cements the grains together. The
Basalt and Related Rocks. The term porosity of sandstone is typically high,
basalt is applied to any of the dense, fine- ranging from 5% to 25%.
grained, dark gray or black volcanic rocks. Chert is a sedimentary rock composed
The term ordinarily includes rock types that almost entirely of silica, in the form of opal,
geologists classify as dacite, andesite, chalcedony, or microgranular quartz. It
basalt, trachyte, or latite. Basaltic rock has commonly occurs in thin-bedded deposits.
solidified by the cooling of lava either as Conglomerate is a sedimentary rock con-
flows on the earth's surface or as shallow taining abundant fragments of pebble size
intrusive bodies beneath the surface. or larger in a matrix of sand and finer-
Basalts are among the heaviest of rocks grained materials. Conglomerates show
with an average specific gravity of 2.9-3.2, various degrees of induration which depend
but in certain areas they may contain many largely on the nature and amount of ce-
small cavities (vesicles). Baf!altic rocks are menting material-clay, calcium carbonate,
characteristically hard, tough, and durable, iron oxides, or silica-in the matrix.
so they are well suited for use as a reliable Shale is a very fine-grained thinly bed-
foundation material and as aggregate, ded sedimentary rock composed mostly of
riprap, and quarrystone armor units. clay- and silt-size particles. Pre-Mesozoic
Carbonate rocks are broadly divided by shales are commonly well indurated, if not
geologists into limestone, which consists al- metamorphosed.
most entirely of calcite (CaC0 3 ), dolomite, Slate is a thinly foliated metamorphic
which consists mainly of the mineral rock composed essentially of muscovite
dolomite (CaC0 3 • MgC0 3 ), and marble, (sericite), quartz, and graphite, all in grains
352 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

of microscopic or submicroscopic size. Slate basically depends on its composition and


is formed by compaction and partial recrys- the way in which it was formed,
tallization of shale and is commonly dark physical-chemical reaction, and stress his-
colored and moderately hard. Because slate tory. A soil may be composed of one size
has been subjected to intense pressure dur- fraction of narrow range, such as the case
ing formation, it has a low porosity and, with sands, or be composed of any number
consequently, a high strength. Its modulus of size fractions in continuous or gap grad-
of rupture is relatively high, and it is also ing. The size composition of a soil is called
resistant to weathering and to mechanical its texture. Almost all soils with which the
abrasion. marine engineer has to deal with are satu-
Tuff The term tuff includes pyroclastic rated; that is to say, the voids in soil skele-
volcanic types, most of which would be ton are completely filled with water. Other
classed as rhyolite or dacite tuffs or tufa- soils are only partly saturated and the voids
ceous sediments. Most tufaceous rocks are are filled partly with water and partly with
only moderately hard, although on exposure air (gas). Under bearing stresses of the
to air they commonly harden appreciably. magnitude with which the marine engineer
Bedrock characteristics usually of con- usually deals, water is considered as incom-
cern to the marine structures designer are pressible, so that volume change cannot
the mechanical properties such as Young's take place in a saturated soil unless there is
modulus, shear strength, compressive a flow of water. This is perhaps the most
strength, chemical composition, degree of important single consideration in soil me-
weathering, and the amount of fracturing. chanics. The relationship among various soil
These become important in the design of phases is illustrated in Figure 4-7.
any type of structure founded on bedrock, There are three important volume rela-
design of rock anchors, and also when rock tionships: porosity (n), void ratio (e) and
materials are used for harbor and coastal degree of saturation (S). Porosity is the
construction. Details on properties of mis- ratio of the void volume to the total volume
cellaneous rocks, effects of joint and frac- (n = VvfV), and the void ratio is the ratio
ture orientation, stability of rock masses of the void volume to the solid volume (e =
and requirement for rocks as construction VvfV.). Porosity is usually given in percent.
materials are found in the U.S. Army Engi- Both n and e indicate the relative portion
neer Waterways Experiment Station's of the void volume in a soil. This void vol-
Coastal Engineering Research Center, Re- ume is filled with water or air (gas). The
port No. 3, February 1983, in Whiteneck porosity and the void ratio interrelate as
and Hockney (1989) and in standard hand-
books on foundation engineering, or in texts
on rock mechanics.

4.6 PROPERTIES AND


ITf
~f
lL
CHARACTERISTICS OF SOILS

In addition to mineral and biogenic parti-


cles, soil is also composed of air (gas) and Volumes Weights

water. The solid particles form a matrix, (a ) (b )

called the soil skeleton, filled with air (gas) Figure 4-7. Relationship among soil phases: (a) ele-
and water. The soil matrix configuration ment of natural soil; (b) element separated into phases.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 353

follows: The density of granular soils is usually


characterized in terms of relative density
e n (Dr), defined as
n=-- and e = - -
1+e 1-n
emax- e 'Yd(max)( 'Yd - 'Yd(min))
The degree of saturation (S = Vw!V) indi- D =----
cates the percentage of the void volume r emax- emin 'Yd( 'Yd(max) - 'Yd(min))
filled with water. Thus, S = 0 indicates a
dry soil, S = 100% indicates a saturated where
soil, and a value between 0 and 100% indi- emin = void ratio of soil in densest condition
cates a partly saturated soil. emax = void ratio of soil in loosest condition
The most useful relationship between e = in-place void ratio
phase weights is water content (w ), which
'Yd(max) = dry unit weight of soil in densest condi-
is the weight of water divided by the weight tion
of the solid in the soil element (w = WwfW5 ).
'Yd(min) = dry unit weight of soil in loosest condi-
The following are expressions for specific tion
gravity (G) and various unit weights ( 'Y)
'Yd = in-place dry unit weight
(i.e., the weight of a given volume):

Specific gravity (G) Approximate values of emax' Emin' 'Yd(max)'


and 'Yd(min) for some granular soils are given
Mass: Gm = 'Yti'Yo in Table 4-1.
Water: Gw = 'Yw/'Yo Sand would generally be considered as
Solids: G = 'Ysi'Yo loose if Dr < 0.3 (cone penetration resis-
Note: 'Yo =unit weight of water at 4°C "" 'Yw tance below 7.5 MNjm 2 ), medium dense if
and Gw = Se Dr = 0.3 to 0.6 (cone resistance 7.5 to 15
Unit weight ( y) MNjm 2 ), and dense if Dr> 0.6 (cone resis-
Total: 'Yt = wjV = [(G + Se)j tance above 15 MNjm 2 ).
(1 + e)]yw = [(1 + w)j(1 + e)]Gyw Atterberg limits and related indices are
Solids: useful characteristics of assemblages of soil
Water: 'Yw = WwfVw particles. Atterberg limits are based on the
Dry: 'Yd = W.!V = [G/(1 + e)]yw = concept that a fine-grained soil can exist in
Gywf(1 + wGjS) = '}'1/(1 + w) any of four states, depending on its water
Submerged (buoyant): 'Yb = 'Yt - 'Yw = content; solid when dry, and upon the addi-
{[G - 1 - e(1 - S)]/(1 + e)}yw tion of water, proceeds through the
Submerged (saturated): 'Yb = 'Yt - 'Yw = semisolid, plastic, and finally liquid states,
[(G - 1)/(1 + e)]yw as shown in Figure 4-8. The detailed proce-

Table 4-1. Maximum and minimum void ratio and dry unit weight
for granular soil

Void Ratio Dry Unit Weight (gjcm 3 )


emax emin "Yd(min) "Yd(max)

Standard Ottawa sand 0.8 0.5 1.48 1.77


Clean uniform sand 1.0 0.4 1.33 1.89
Fine-to-coarse sand 0.95 0.2 1.36 2.22
Silty sand and gravel 0.85 0.14 1.43 2.34
Uniform inorganic silt 1.1 0.4 1.28 1.89

Source: After Hough (1954).


354 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

dure for determining Atterberg limits is strength, compressibility, and permeability.


found in Lambe (1951), Bowles (1970), U.S. Geotechnical problems typically encoun-
Army Corps of Engineers (1970) and others. tered in the design of marine structures
The plasticity index (IP) is determined as which involve these characteristics are slope
IP = w 1 - wP. A useful way to characterize stability, bearing capacity, settlement, and
the plasticity (liquidity) of a soil is with erosion. The presence of pollutants derived
water-plasticity ratio (B), or liquidity in- from industrial wastes, such as toxic heavy
dex (LL) metals (mercury, cadmium, lead, and ar-
senic), chlorinated organic chemicals (DDT
and PCBs), and pathogens (bacteria,
viruses, and parasites) should also be con-
sidered in the evaluation of the use of any
where water content is at the plastic limit, soil in marine structures of any kind. Con-
the percentage of weight of solids and water taminated soils should not be used in
content at the liquid limit and the percent- general practice.
age of weight of solids, wP and w 1, respec-
tively are shown in Figure 4-8 and wn is
the soil natural water content. 4.6.1 Shear strength
The plasticity index, which indicates the
range of water content over which the soil The shear strength of soil systems is usu-
remains plastic, and the liquidity index, ally expressed by the Coulomb equation
which indicates the nearness of a natural
soil to the liquid limit, are very useful engi- S=c+utan<f> (4-3)
neering characteristics of soil. One must
realize, however, that all of the limits and where
indices are determined on soil samples that
have been thoroughly worked into a uni- c =cohesion
form soil-water mixture and therefore give <P = angle of internal friction
no indication of particle fabric or residual u = normal stress on the shear plane
bonds between particles which may exist in
natural soil. These natural characteristics The main objective of shear tests is to
are destroyed in preparing the specimen. determine the shearing strength at failure
The most significant engineering proper- at different normal stresses from which the
ties of soil to be considered are shear cohesion (c) and friction angle (<f>) may be
computed.
The three types of laboratory tests com-
~
Fluid soil-water monly performed to determine soil strength
mixture liquid State
are designated as follows:
- - - - - - - - liquid Umit, WI
c
.!! Plastic State Unconsolidated-undrained triaxial test, com-
8 monly known as the UU test
l.
r:::
- - - - - - - - Plastic Umit, Wp
Semisolid State
Consolidated-undrained triaxial test, com-
·;;; monly known as the CU test
j Solid State
Shrinkage Umit, w,
Consolidated-drained triaxial test, commonly
known as the CD test
Cll
Dry soil
These conditions are used to evaluate the
Figure 4-8. Attenberg limits and related indices. effect of pore-water pressure.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 355

In the UU test, no drainage is allowed


during application of confining pressure (u3 )
"'• •cr1 +lltr
or normal load in direct shear. In the case
of a structure constructed rapidly over a
soft clay deposit, the UU analysis should be
performed. The unconfined compression test
is considered as a special case of the uncon-
solidated-undrained test with confined
pressure equal to zero.
In the CU test, drainage is allowed dur-
ing application of the confining or normal
load. The sample is consolidated with re-
spect to the applied pressure as observed
via drainage (or vertical deformation in the
direct shear test). No drainage is allowed
during the shear test. In the case of rapid
construction of an embankment or other
structure on a natural slope, the CU test
should be performed.
-IT

The difference between the CD test and


the CU test is that drainage takes place
during the test and the test is slow enough
so that pore pressures do not build up. The
CD test is conducted in the case of a struc- Figure 4-9. Mohr's failure envelope.
ture constructed very slowly over a soft clay
deposit.
From the results of the tests, the stress- Mohr's envelope, several triaxial tests
strain characteristics under various loading should be performed on specimens of the
conditions and the conditions of failure for same soil using various confining (cell)
the soil are established. The strength of a pressures (u3 ); see Figure 4-9. From Figure
soil is usually defined in terms of the stress 4-9 the stresses (T) and ( u) can be ob-
tained either graphically or by means of
developed at the peak of the stress-strain
formulas (4-2) and (4-5):
curve and is presented in the form of Mohr
circles and a Mohr failure envelope (Fig.
4-9). The strength is then expressed in U l - Ug •
(4-4)
r= 2 sm2a
terms of cohesion (c) and the angle of inter-
nal friction ( 4> ).
Ul + Ug U l - Ug
Shear stresses in the specimen are cre- u = 2 + 2 cos2a (4-5)
ated by applying additional vertical stress.
This additional vertical stress (Au) is called
deviator stress. The deviator stress is where
steadily increased until failure of the speci-
'T = shear stress
men occurs. Drainage of water from the
specimen is measured by burette. Details of a-1 =major principal stress (a-1 = u 3 + Au)
triaxial test and pore-water pressure mea- u 3 =minor principal stress (confined pressure)
surements are described by Bishop and a = angle between normal stress and major
Henkel (1962) and Bowles (1970). To obtain principal stress
356 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

Point A in Figure 4-9 have the coordinates results in the water is called hydrostatic
excess pressure. If the water drains from
cr - cr
x= and y = 1 3 (4-6) the soil pores, the hydrostatic excess pres-
2 sure and its gradient gradually decrease
and the load increment is shifted to the soil
The line or curve connecting these points is structure. The transfer of load is then ac-
called a stress path. The stress path is just companied by a change in volume ofthe soil
like a Mohr circle construction as it repre- mass equal to the volume of water drained.
sents states of stress. This process is known as soil consolidation.
Shear tests may also be conducted in the Soil consolidation leads to increased density
field (Winterkorn and Fang, 1975; Chaney,
and a corresponding increase in shear
1991; Young, 1991).
strength and decrease in permeability. It
The effective stress value (u '), which is
also accounts for the settlement of struc-
to be considered in shear stress evaluation,
tures founded on compressible soils. Its un-
is equal to the total stress ( u) minus the
derstanding, together with pertinent labo-
pore-water pressure (u); that is,
ratory tests, permits the calculation of total
cr'=cr-u (4-7)
settlement.
If the present effective overburden pres-
As was noted earlier, in marine structures, sure is the maximum pressure to which the
soil is usually submerged for its entire soil has ever been subjected at any time in
depth. However, in some areas, such as a its history, the deposit is referred to as
tidal zone, the soil may not be submerged normally consolidated. A soil deposit that
for some time due to water-level fluctua- has been fully consolidated under a pres-
tion. If the soil is submerged for its entire sure larger than that of the present over-
depth, the effective stress at a particular burden is called overconsolidated. If soil de-
location within the deposit remains un- posits are not fully consolidated under the
changed by fluctuations in the water level. present overburden pressure, as is the case
However, if the water level decreases to a for recent fills, they are called underconsoli-
point such that the entire depth of soil is no dated. '
longer submerged, the effective stress in Consolidation is a time-dependent pro-
the soil is increased (Carchedi and Morgan, cess that is directly related to the soil per-
1990). Furthermore, as pointed out by meability, thickness of the compressible
Pappin et al. (1992) the resilient stress- layer, and drainage conditions of the con-
strain behavior of granular materials, e.g., fining soils. Because many clays have very
sand, gravel, etc., saturated with water is low permeability, their consolidation under
identical to that of dry material provided static stress may take place over decades.
that full drainage is allowed. Overconsolidated clays can sustain loads up
to the maximum past pressure without sig-
nificant settlement. Some marine clays ex-
4.6.2 compressibility hibit overconsolidated behavior even though
<consolidation> they have not been subjected to past load-
ing. This phenomenon, known as apparent
When a load is applied to a saturated soil overconsolidation, is caused by physico-
mass, the load is usually carried initially by chemical bonding that takes place at the
the water in the pores because the water is particle contact. Apparent overconsolidation
relatively incompressible when compared is found generally in clays which have sedi-
with the soil structure. The pressure which mented slowly, allowing sufficient time for
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 357

particle bonding to occur before the soil A theory relating pressure, time, and vol-
structure is placed under significant stress. ume change was proposed by Terzaghi
Rapidly sedimented clays, which are usu- (1943) and has become known as the
ally found in river deltas, typically show Terzaghi theory of consolidation. One of the
underconsolidated behaviors. These clays major assumptions in the Terzaghi theory
tend to have high water content, and low is that volume change and the outflow of
shear strength, and are highly susceptible pore water occur in one direction only. For
to stability failure (Rocker, 1985). this reason, it is sometimes referred to as
The simplest consolidation test is the the one-dimensional consolidation theory.
one-dimensional, laterally confined com- Its theoretical derivation may be found in
pression test (often referred to as the standard textbooks (Taylor, 1948; Leonards,
odometer test). In this test, the soil sample 1962; Wu, 1966).
is placed within a restraining ring and
loaded with special types of plates on either
the top or the bottom, or both. The change 4.6.3 Permeability
in sample height is measured by a deflec-
tion gauge and is used to calculate the Permeability is an important soil property
change in void ratio (e) at different normal that indicates the relative ease with which
pressures (P). If the soil is saturated, the a fluid will flow through the soil. As men-
sample is placed between two porous disks tioned before, it has a profound impact on
that permit the water to drain away during the rate of soil consolidation. The coefficient
compression. That, in turn, leads to infor- of permeability of a soil is defined as the
mation which permits plotting of the so- average percolation velocity divided by the
called e log P relation. From e log P plots hydraulic gradient in the soil at that partic-
for sands, silts, clays, or mixtures of them, ular point. It is seen then that the coeffi-
factors are quantified for consolidation and cient of permeability has units of velocity.
settlement estimates. It should be empha- Permeability depends on the characteristics
sized that in such tests the lateral expan- of both the pore fluid and the soil.
sion is restrained. In field situations, con- Several field and laboratory test proce-
straint is only approximated by the loading dures are available to determine soil per-
of relatively thin layers of compressible soil meability (Bowles, 1970; Lambe, 1951; U.S.
through load distribution over a large area. Army Corps of Engineers, 1970). It is al-
A detailed test procedure is found in Bowles ways important to realize that because nat-
(1970). ural soil deposits are typically nonhomoge-
Predictions of static settlement on silts neous and anisotropic in nature they cannot
and clays are usually made on the basis of be adequately represented by a small sam-
consolidation or odometer tests on undis- ple. That is why for projects where massive
turbed samples. The rate of settlement and dewatering is required, a large-scale pump-
the time for essential completion of primary ing or field surcharging test should be
consolidation can be predicted on the basis conducted.
of the test. Typically, silts are less com- In general, a qualitative approximation
pressible than clays. of the permeability of the materials can be
Compressibility of sands is usually very made on the basis of grain size. For exam-
low (1.5-2.0% by volume) and is governed ple, clean gravels will have permeabilities
mainly by their relative density (Tsinker, ranging from 10 to 100 cmjs. Clean
1972). It is nearly elastic in nature and medium-to-coarse sands will have perme-
takes place almost immediately upon load- abilities ranging from 10- 2 to 1 cmjs. Very
ing. fine sands will have permeabilities ranging
358 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

from 10- 5 to 10- 4 cmjs. Organic and inor- active, or passive. Pressure at rest is as-
ganic silts, mixtures of sand, silt and clay, sumed where no deformations or displace-
glacial till, and some stratified clay deposits ments of a retaining structure occur, and it
will have permeabilities ranging from 10- 6 is usually applied to the heavy gravity-type
to 10- 5 cmjs. Clays, which are practically quay walls built on a bedrock foundation.
impervious, will have permeabilities rang- Active or passive soil pressures exist where
ing from 10- 9 to 10- 7 cmjs. For more the relative displacement between soil and
information on soil properties related to structure (the structure's elements) may
design of marine structures, the reader is cause soil either to expand (active state) or
referred to EAU (1990). General detailed to contract (passive state). It usually ap-
information on soils and soil properties plies to any type of soil-retaining structures
is found in any standard text or handbook built on soft foundation material. Any
on soil mechanics and foundation engineer- structural tilt produces lateral strain in the
ing. soil and alters the soil's horizontal stress
conditions. Depending on the magnitude
and direction ofthis tilt, the final horizontal
stress can lie anywhere between two limit-
4.7 LATERAL SOIL PRESSURE ing' conditions: active and passive.
The effect of wall tilt on lateral pressure
By definition, lateral soil pressure is the of sand is shown in Figure 4-10. The mag-
pressure exerted by soil against an engi- nitude of wall tilt (Y jH) required to reach
neering structure or, in a broader sense, it soil failure (active or passive) conditions in
is the pressure acting on a surface sur- various soil types is indicated in Table 4-2.
rounded by an earth mass. The magnitude The ratio of lateral (u1) to vertical(o) unit
of the soil pressure depends on the physical pressure is termed as the coefficient of lat-
properties of soil, the geometry of the con- eral pressure (K)
tact surface, and the character of the
soil-structure interaction such as displace- (4-8)
ment (deformation) of the retaining wall
under soil pressure. The vertical unit pressure is usually taken
Since Darwin's (1883) observations relat- as being equal to the weight of overburden
ing lateral pressure of sand with movement ( yh), where r is the soil density and h is
of the retaining wall, hundreds of subse- the depth of overburden. Accordingly, the
quent investigations have been carried out coefficient of active pressure K A = lTz(A/ lTv,
to confirm Darwin's model tests. Terzaghi and the coefficient of passive pressure K P =
(1934) observed that for submerged com- lTz(P)I uv, where lTz(A) and lTz(P) represent
pacted sand the coefficient of lateral earth soil lateral pressure at active and passive
pressure decreased from about 0.6 at no states, respectively.
yield, to about 0.13 at 0.0005H forward Couplet (1726-1728) and Coulomb (1776)
wall movement (here H is the height of the were the first to propose a theoretical solu-
wall). However, after less than 2 days of tion for the determination of soil pressure
rest this coefficient increased to 0.27. Simi- exerted on retaining structures. Since then,
lar, or equivalent, results have been ob- hundreds of papers and books have been
tained in numerous subsequent large-scale published on the subject, and a number of
and full-scale tests that have been carried increasingly complex theories and proce-
out in Europe and North America. dures for determination of lateral soil pres-
In relationship to wall movement, lateral sure have been proposed. However,
soil pressure is usually defined as at rest, Coulomb's theory with some corrections and
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 359

:.c 10.0 ...-.,--,--r--r---r-....,r-....,r--r--r-"T"-T""'


8.0
"'
=
"'
s.o
6.0
'I v,l vI
--- ----
:;; 4.0
....
<(
v
:1.0
.
,,
__ , , LOOSf
.. 2.0
... . ~

"'> K
0 J,O :~ :::·:!::·::\=,:•:'',0 o I : :
p

... 0.8
....
z
<(

0
0.6
0.5
0.<4
ACTIVE STATE PASSIVE STATE

...
N 0.3

~ 0.2

.
0

~ 0.1 ~~~~~~~~~--~~~~~_.--~--~

WALL ROTATION, :f..


H

Figure 4-10. Effect of wall movement on earth pressure in sand.

Table 4-2. Magnitude of wall tilt to reach active


or passive soil pressures
4.7.1 Active Earth Pressure
Wall Tilt(Y/H)• Coulomb developed his theory at a time
Soil Type
and Condition Active Passive
when trigonometric functions were not yet
in use. Accordingly, he expressed all rele-
Dense cohesionless 0.001 0.02 vant soil pressure values as ratios: Later
Loose cohesionless 0.004 0.06 on, several authorities in the field of soil
Stiff cohesive 0.01 0.02
Soft cohesive 0.02 0.04 mechanics offered their own interpretation
of Coulomb's theory. Kezdi (1975) trans-
•Y = horizontal displacement; H = height of wall. formed Coulomb's original equation to cor-
respond to present terminology and devel-
oped formula (4-9) for the determination of
interpretations is still used in most practi-
the coefficient of active soil pressure (Fig.
cal applications.
4-11):
According to Coulomb, when soil failure
occurs behind a wall, a wedge of soil behind 2

the wall moves downward and its weight


presses simultaneously against the wall's
sin( J3- cp)
(4-9)
interior and against the earth mass along [sin( J3 + 8)] 0 ·5 +
the sliding surface. sin(¢+ 8) sin(¢- a) ) 0 '5
(
In Coulomb's earth pressure theory, it is sin( J3- a)
assumed that the sliding surface is straight.
In reality, however, it is very close to where
straight in the upper reaches of the sliding J3 angle between back of wall and horizontal
=
surface and curved in the lower portion due ( J3 = 90° equals vertical)
to frictional forces acting on the wall's cp = angle of internal friction
interior. 8 = angle of friction between the wall and the
360 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

Table 4-3. Effect of angularity and grading on


angle of internal friction ( 4>) of granular soils

Angularity and
Grading Loose Dense
Rounded, uniform 30° 37°
Rounded, well graded 34° 40°
Angular, uniform 35° 43°
Angular, well graded 39° 45°
H

the KA value obtained from formula (4-9):


(4-10)

In practical design, horizontal EA(h) and


Figure 4-11. Earth active pressure.
vertical EA(u) components of earth pressure
EA are usually used. The horizontal compo-
nent of earth pressure can be given by
soil, commonly referred to as wall friction
a = slope angle above horizontal behind the wall EA(h) = EA sin( {3 + 8) = 0 .5-yH 2KA(h) (4-11)

. The coefficient KA<h> = KA sin( f3 + o) for


For granular soil, the maximum slope angle some practical cases is given in Table 4-4.
(a) usually does not exceed the angle of Correspondingly, the vertical component is
internal friction ( 4J) (so-called steady-state obtained by
or residual soil strength), which depends on
soil density and gradation. For example, EA{u ) = EA(h) tan( {3- 90° + 8) (4-12)
evenly graded soil has a smaller void ratio,
which results in both larger density and Terzaghi (1934) has demonstrated that the
angle of internal friction. This trend is prediction of active soil pressure obtained
by Eqs. (4-10) and (4- 11) is in good agree-
shown in Table 4- 3, which is confirmed by
ment with actual measurements. According
tests conducted by Holtz and Gibbs (1956).
to Brown (1948), the friction between the
The general accuracy of results obtained
wall and the soil can be neglected in most
by Eq. (4-9) for o = 0 to o = 0.54J are in a
practical cases so that the active pressure
good agreement with a more elaborate
for any kind of soil can be obtained by Eq.
method developed by Sokolovski (1965). (4- 13) (Fig. 4-12):
Fang and Ishibashi (1986) confirmed that
active soil pressure on a rigid wall rotating EA = 0.5-yH 2 KA- 2cH tan(45o - t/J/2) (4- 13)
about the base, although not necessarily
linear, is in close agreement with values where
predicted by Coulomb. They also indicated
that the magnitude of the active lateral
earth pressure exerted on the wall rotating For granular soils, the coefficient K A can
about the top is higher by about 17% than also be expressed by Eq. (4-14) derived by
the value estimated by the Coulomb theory. Rankine (1857):
For the simple wall case shown in Figure
1 - sin tjJ
4-11 the earth pressure on the wall (EA) KA = ---- (4- 14)
can be calculated from formula (4-10) with 1 +sin tjJ
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 361

Table4-4. Coefficients of the horizontal component of earth active pressure (values of 1000Kah)

a= 0° ao = </> ao = </>

cos f3 4> 8=0 4>/2 ~4> 4> 8=0 t4> ~4> 4> Any Value

0.2 15° 523 480 469 449 589 552 542 524 835
20° 417 378 367 348 482 446 435 418 763
25° 330 295 286 270 388 354 345 329 675
30° 257 229 221 207 306 278 270 256 587
35° 198 175 169 158 237 214 208 195 496
40° 148 132 126 118 178 160 155 146 406
45° 108 96 93 86 129 116 102 95 320
0.1 w 556 510 499 475 627 587 576 556 883
20° 454 409 397 376 526 485 473 453 822
25° 368 327 316 296 434 396 384 365 747
30° 295 260 250 233 353 319 309 291 666
35° 234 205 196 181 282 253 245 228 580
40° 182 159 152 140 220 197 190 176 492
45° 139 121 116 106 168 149 140 133 405
0 15° 588 538 524 500 665 621 609 587 933
20° 490 440 426 401 569 523 510 486 883
25° 406 359 345 322 482 436 423 400 824
30° 333 291 279 257 402 360 334 326 750
35° 271 235 224 205 330 293 283 262 672
40° 218 187 183 161 267 235 226 207 587
45° 172 148 145 125 210 185 177 160 500
-0.1 15° 619 564 549 521 701 654 640 615 983
20° 525 469 453 424 612 561 545 518 948
25° 443 389 373 345 529 477 461 434 900
30° 372 321 306 280 452 402 387 359 839
35° 309 264 251 226 381 335 318 294 768
40° 254 216 204 180 316 275 263 237 689
45° 207 174 164 143 257 223 212 188 605
-0.2 15° 648 588 571 541 737 684 669 642 1036
20° 559 495 477 444 654 596 579 548 1016
25° 479 416 398 365 576 516 498 465 982
30° 409 349 332 299 502 442 424 390 933
35° 347 292 275 244 432 376 360 323 872
40° 292 243 229 197 367 316 300 265 800
45° 243 200 186 157 307 262 247 213 720
Source: From Kezdi (1975).

The Soviet Standard SNIP 11-U.l0-65 face at the back of the wall, the value of
(1966) stipulates that the following condi- 8 can be increased from 0.5¢ up to ¢.
tions should be considered when choosing (c) In all cases of saturated medium to
the value of 8: fine-sand backfills and for those granular
soils exposed to dynamic loads, 8 = 0.
(a) In all regular concrete wall construction,
the value of 8 can be taken as 0.5¢. The British Code of Practice (1951) for walls
(b) In the case of a substantially rough sur- subjected to dynamic loads, such as vibra-
362 Geotechnical Aspects of Soil-Structure Interaction Design Consideration s

2 ••• (4~·-<t>/2)

Figure 4-12. Active pressure of cohesive soil.


A

wall with gen-


tions transmitted from operating or passing Figure 4-13. Earth pressure against
tly sloped interior.
equipment, specifies that the angle of wall
friction ( 8) should be taken as zero.
practical cases the for-
Because in most practical cases of ma- show that for most
plane of failure (as-
rine structures constructio n the backfill is mation of a secondary
be expected when
made of cohesionles s saturated granular suming 8 = 0.5¢)
may
materials, the soil active pressure (EA) is (} ~ 65°.
typically expressed by Eq. (4-10) in which According to Dubrova, in the case of re-
KA = tan 2 (45o - ¢/2). It follows from Eq. taining
walls with a gentle back slope, the
will be determined as
(4-9) that the inclination of the back of the active soil pressure
wall and the slope of the backfill behind the follows:
wall significantl y affect the magnitude of (4-16)
lateral soil pressure.
When the back slope of the wall is too In the latter case, the soil pressure is as-
gentle ( {3 > 135° - ¢/2), the lateral soil sumed as applied at the conditional failure
thrust should be considered to act on a plane AB, and in the calculation of the KA
conditional secondary plane of failure AB value by Eq. (4-9), 8 can be taken as equal
as depicted in Figure 4-13. According to to 0.5¢ to ¢ in the case of dense granular
Dubrova (194 7) the wall back slope (for hor- fill material. However, it must be reduced
izontal surface of the backfill) may be con- to zero in the case of saturated medium-to-
sidered gentle if the angle fine sand.
A similar approach is extended to the
e = f3- 90° design of regular cantilever retaining walls
(Fig. 4-14). Very often the lateral soil pres-
:::::.arctan [
1 - KA
+ ( (1 - KA) 2 _
2 .
KA)o.sl sure on a cantilever retaining wall is deter-
2tan 8 - 4tan 8 mined at the 00 1 plane (Fig. 4-14a). How-
(4-15)
ever, it is more consistent to consider a
failure wedge behind the wall, as indicated
in Figures 4-14b and 4-14c, and determine
where
soil pressure as acting on plane 00 3 in the
KA = tan 2 (45°- ¢/2) case of a regular cantilever wall (Fig.
4-14b), or on the broken plane 00 3 0 4 in
The calculation s of (} by formula (4-15) the case of a short foundation slab (Fig.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 363

0 Ot 0
.... !;:,:e.. , . ~-.::•".u~:;;.:";.' f 'J·~~..-..Yiij·-~:; ·

\ I I
\\ ~1 I f, /

•• I
4$··-/2

---
_.___ _ _ .. . l l/
___,f
0 0

(a) (b) (c)

Figure 4-14. Active earth pressure on cantilever wall.

4-14c). In the latter case, two components lateral soil pressure is a combination of
of earth pressure should be considered; i.e. three superimposed components that are:
EA 1 that is acting directly on a wall, and pressure at plane OB, based on an assump-
EA 2 is acting on the soil failure plane 00 3 • tion of horizontal backfill surface at the top
In calculating both EA 1 and EA 2 the value of a wall (a= 0); pressure at plane OC,
of 8 may be used as previously discussed. also based on an assumption of horizontal
More information on this subject is found in backfill surface extended to point C; and
Barghouthi (1990). pressure at plane OA, based on an assump-
In most practical cases, in granular fill tion of a slope, extended continuously from
the angle of the failure plane (e) behind a point A and up at angle a. The subsequent
wall can be taken with sufficient accuracy design soil pressure diagram is outlined as
as equal to (45° - cfJ/2). OBabc.
An example of active soil pressure on a A uniform surcharge load (q) may be
wall with a broken profile of backfill is de- conveniently expressed in terms of a layer
picted in Figure 4-15. In this case, the of soil having the same density ( -y) as an

Figure 4-15. Active earth pressure on wall with broken profile of backfill.
364 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

K- - - - - - - - - ing the method recommended in the


h.= qj"( 1\ Canadian Foundation Engineering Manual
+---- ---~~~~~~~~~~~~ q (1985) (Fig. 4-18). More information on this
type of loading is found in Gerber (1929),
Sprangler (1938), and Dubrova (1963).
In many practical cases of marine wall
H
construction, riprap (</J = 40° to 45°) placed
behind the wall is used to reduce the lateral
soil pressure. In this case, the failure plane
within the riprap structure is typically as-
sumed as being projected from the base of a
wall (point 0) at 28°-30° to vertical, to the
intersection with the top of the riprap (point
0 1 ). Then the soil pressure below point 0
exerted on the wall is considered as at-
tributed to pressure from a continuous layer
or riprap. Hence, in an example depicted in
Figure 4-16. Effect of unifonn surcharge surface Figure 4-19 at 8 = 0 the design ordinates
load on lateral earth pressure. of a soil pressure diagram will be deter-
mined as follows:
underlying soil and a weight equal to q.
The height of this assumed soil layer will
then be h s = q I y (Fig. 4-16). al = (q + Y1h1)KA(l)
An example of soil pressure due to the
effect of a continuous uniform strip load is a2 = (q + Y1h1 + Y2hz)KA(2)
depicted in Figure 4-17. A uniform dis-
tributed load over a limited surface area a3 = (q + Y1h1 + Y2h2)KA(3 )
should be distributed in all directions.
a4 = (q + 'Y1h1 + Y z hz + Y3h3)KA(3)
Where the surcharge behind a wall consists
of a point or line load, whose intensity is
small compared to total backfill load (e.g., Note that in the case of backfill made from
total load on a wall from surcharge is less sand, KA(l ) = KA(2); since KA(2) > KA(a)' then
than 30% of the active load), the additional a2 > aa.
soil pressure may be calculated by employ- Lateral soil pressures can also be re-
duced by means other than riprap. Among
them is construction of the wall with a
negative slope of the back side ( (3 0 < 90°)
which is fully or in part leaning against the
backfill (Fig. 4-20a), or use of miscella-
neous pressure-relieving structures (Fig.
4-20b). It should be noted that pressure-
relieving platforms of different designs are
not only a very efficient method of soil pres-
sure relief but they also help to maintain
YHK, I uniformly distributed bearing pressure at
the base of the gravity wall. For more infor-
Figure 4-17. Effect of continuous strip of unifonn mation on this subject, the reader is re-
surcharge load on earth lateral pressure. ferred to Tsinker (1995).
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 365

0. 2

.
'"'
II
0.4

0
;
..... 0.6
ph (~) h

I
> hp m P

0.1 0.60H 0.2,0.78,0.59H


0.3 0.60H 0.4 0.78 0.59H
0.8
0.5 0.56H 0.6 0.45 O.HH
0.7 0.48H

1 .0
0 .2 0. 4 0.6 c. 8 I .0 o. 5 I .0 1.5 2.0

VALUE OF crh (&J VALUE OF crh ( _QHP2)

LINE LOAD QL POINT LOAD Qp

FORm S 0.4:

CT (H)_ 0.20n
h QL- (0.16+ 0 2)2

PH=0.55QL

4-r~'--Ph· FORm > 0.4:

FORm l! 0.4:

0.28n 2
(a;; l = io:-167:2,
H2
.,.h 3
RESULTANT Ph= 0 · 64 QL
FORm > 0.4:
(m2 + I)
H2 1.77m 2 n 2
PRESSURES FROM LINE LOAD QL crh (o,:) = (m2 + 0 2)3
(BOUSSINESQ EQUATION MODIFIED BY EXPERIMENT)

cr~ = crh cos 2 (1 .1 9)

SECTION o- o
PRESSURES FROM POINT LOAD Qp
(BOUHINESQ EQUATION MODIFIED
BY EXPERIMENT)

Figure 4-18. Horizontal pressure on a wall due to point and line load surcharge.

Piles driven behind a soil-retaining An example of this reduction in soil lat-


structure have a profound impact on lateral eral pressure on a sheet-pile wall that is
soil pressure distribution; piles not only re- part of a piled relieving platform structure
inforce the fill behind the wall but they also is shown in Figure 4-21. In this figure three
shield a portion of the total soil lateral pres- possible alternatives of pile arrangement
sure against the wall. are depicted as follows: vertical piles ar-
366 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

ing support while the rest of the soil re-


mains stationary (Bosscher and Gray, 1986).
This transfer of pressure from a yielding
mass of soil onto adjacent stationary parts
(piles) is commonly called the "arching ef-
fect," and the soil is said to arch over the
yielding part of the support. In this case the
soil arches over piles as it attempts to move
through fixed piles which are firmly embed-
Figure 4-19. Lateral earth pressure diagram in ded in a nonyielding base and the deck of
multilayered backfill. the structure. Thus, the reduced lateral soil
pressure acts on a sheeting: when the space
between piles is too wide, arching will not
rangement, where lateral support to the
develop. Tsinker (1972) observed formation
structure is provided by an independent an-
chor system, RA (Fig. 4-21a); batter piles of stable arches by a loose medium sand
(Fig. 4-21b); and a combination of a vertical while the space between piles with a cross
and batter pile system (Fig. 4-21c). section of 35 X 35 em varied from 2.5 to
In all the aforementioned cases the lat- 3.0m.
eral soil pressure exerted on sheet piling is The relative density and associated angle
reduced due to the presence of piles. The of internal friction of the sand are very
piles penetrate the soil failure wedge be- important parameters in arch formation.
hind the sheeting and absorb part of the Wang and Yen (197 4) and Ito and Matsui
soil pressure which is a result of soil arch- (1977, 1978) developed a theory which pre-
ing over these piles. dicts an optimal relative width between
In general terms, the soil arching is a supports at which the soil arching can be
phenomenon which can be described as fol- the most effective.
lows: If only a localized area of the support The shielding plane behind piles (AF,
for a mass of soil yields, the soil adjoining A'F', and AEF') can be determined in the
the yielding zone displaces with the yield- manner depicted in Figure 4-21. The shape

(a) (b)

Figure 4-20. Lateral earth pressure for different shapes of wall interior. (a) Back of wall with
negative slope; (b) wall with pressure-relieving structure.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 367

J
I'
1/
1/
___ _]II

(a) (b) (c)

Figure 4-21. Active earth pressure on sheeting with piles behind it. (a) Sheet-pile wall shielded
with row of vertical piles; (b) the same with batter piles; (c) combination of (a) and (b).

of a shielding surface depends on the soil's determined as acting on a vertical plane


properties as well as on the pile arrange- projected from the intersection of sheet pil-
ment. In fact, the presence of a shielding ing with the deck of a relieving platform.
surface creates a soil pressure condition The reliability of the previously de-
similar to those existing in a silo. scribed method to determine the soil pres-
The maximum active pressure at point B sure exerted upon a pile-shielded sheet-pile
(a 1), where h 1 = d cot(45° - </J/2), is deter- wall has been proved by investigations and
mined to be equal yh 1 KA- In the case of by actual design and construction practices
vertical piles, this pressure only is assumed (Smirnov et al., 1979).
as constant down below point B, and the A considerable number of large-scale
soil pressure diagram upon the sheet-pile models tests and field experiments on large
wall has a shape (abcf) as depicted in walls have been conducted in recent years.
Figure 4-2la. These experiments help engineers to better
In the case of batter piles (Fig. 4-21b), understand the conservatism built into con-
the soil pressure at point D is taken as ventional earth pressure theories such as
equal to zero and the maximum active pres- the Coulomb and Rankine theories and the
sure (a 2 ), at point B', is determined as for need to determine some factors not consid-
a 1. The final soil pressure diagram has a ered in these theories. These factors basi-
broken shape (abcf) as depicted in Figure cally include miscellaneous wall move-
4-21b. The soil pressure diagram for both ments; time-dependent factors such as soil
vertical and batter piles (abecf) is depicted consolidation that may occur due to expo-
in Figure 4-21c. For more information on sure of the wall to different kind of static
soil arching and soil lateral pressure distri- and dynamic loads, groundwater fluctua-
bution, consult Chapters 6 and 7. tions, etc.; soil creep; and chemical changes
In some instances a sheet-pile wall is in soil. The soil pressure may also be af-
designed to be slightly inclined toward the fected by residual built-in stresses due to
fill. The soil pressure in this case is usually berth operations (e.g., repeated placement
368 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

and removal of miscellaneous uniform work was concerned with at rest and active
andjor concentrated loads), both static and pressures, and only few were associated
dynamic. For more useful information on with passive pressure.
active soil pressures the reader is referred Fang et al. (1994) conducted laboratory
to Canadian Foundation Engineering Man- studies on passive earth pressures with
ual (1985) and Clough and Duncan (1991). various movements of rigid wall against dry
Last but not least it must be pointed out sand. Based on the experimental data, the
that active soil pressures against a retain- latter investigators concluded that for a wall
ing wall can be substantially reduced by under transnational movement, the passive
reinforcing the backfill material. This can pressure distribution is essentially hydro-
be done by use of galvanized steel, stainless static at each stage of wall motion. The
steel and aluminum alloys strips, and plas-
passive state is reached at different depths
tic in mesh, grids, and strips, installed at a
of soil mass almost simultaneously, and the
certain pattern during placing the backfill
passive earth pressure coefficient, K P,
material behind the wall. Most recently,
is found to be in good agreement with
soil reinforcement and its effects on soil
Terzaghi's prediction based on classical
pressure have been discussed by Laba and
wedge theory. However, when the wall was
Kennedy (1986), Leshchinsky and Boedeker
(1986), Kampton et al. (1985), Juran and rotated about the point located above its top
Christopher (1989), Saran et al. (1992) and and below its base, the passive pressure
others. More information on this subject is distribution was far from linear. In the for-
provided in Chapter 5. mer case the measured total passive thrust
was apparently lower than that calculated
from classical soil pressure theories
(Coulomb, Rankine), and its point of appli-
4.7.2 Effects of cation was found at about 0.18H above the
Wall Movement wall base. In the latter case, high stresses
were measured near the midheight of the
The basic weakness of the conventional the- wall, and the point of application of the
ory of lateral soil pressures is that it does resultant force was found at approximately
not provide a solution for soil redistribution
0.55H above the wall base. In both cases, H
along the back of a wall due to wall move-
is the wall height. Examples of soil lateral
ments.
pressure distribution along the back of a
A great deal of laboratory and full-scale
wall resulting from a variety of wall move-
tests conducted in the past 50-60 years
(e.g., Terzaghi, 1932; Schofield, 1961; ments are depicted in Figure 4-22. In the
Matteofti, 1970; Bros, 1972; Sherif and case (a) when the wall rotates about its
Mackey, 1977; Fang and Ishibashi, 1986; base (point B) and away from the fill, the
Rowe and Peaker, 1965; Mackey and Kirk, hydrostatic soil pressure distribution simi-
1967; James and Bransby, 1970; Fang et lar to that predicted by the Coulomb theory
al., 1994; and many others) indicate that in may be expected.
a great many cases the soil pressure, Dubrova (1963, 1968) proposed a method
although very close in magnitude to that to determine the active lateral soil pres-
determined by the Coulomb or Rankine the- sures which, in general, is based on classic
ories, is, in fact, redistributed along the soil lateral pressure theory and takes into
contact surface area depending on wall de- account miscellaneous wall movements.
formations (movements). It should be Dubrova's method is illustrated in the fol-
pointed out that most of the experimental lowing examples.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 369

A A I

-~
'\
-~
~ ..
\

H
\
\
~·· --2

\ 1
I
\ ~
. ...:.
B B
(a ) (b ) (c)

A' A A

(d ) () (f)

Figure 4-22. Typical wall movements and r esulting lateral earth pressure diagrams:
!-according to Coulomb's KA condition; 2- actual earth pressure diagram; 3-earth
pressure according to Coulomb's K P conditions.

case 1 At the o= 0 condition, Eq. (4-17) is trans-


formed into
The wall rotates about point A and away
from the backfill (Fig. 4-22b). The lateral
soil pressure (stress) at any depth behind
the wall (ay) assuming the o*- 0 condition
is calculated by y 11/JI Y2 + q I<PI y ( 2 )
- 2H tan u 1 + tan u

a = 1' cos 2
u0 [y _ I<PI y 2
( m + sin u 0 )] (4-18)
Y n 2 cos 8 H n cos u 0
(4-17)
where u = 45° - 14>1 y j2H.

where
Example 1 (for case 1}
u0 = 11/JI y j H
m = (1 + tan 8j tan r/1) 0 ·5 Determine the soil pressure exerted upon a
n = 1 + msin u 0 retaining wall rotated about point A away
370 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

from the fill under the following conditions:

Height of wall H= 6.0m


Soil density y = 20kNjm3
I
Angle of internal friction 4> = 30° I
Friction angle between I 3

the wall and fill 0 = %4> = 20° I


Angle between back of the
wall and horizontal f3 = 90°
0
II) I
Backfill slope angle a=O I
I
I
According to Eq. (4-17),
I
I
lcf>ly
30y
u 0 = - - = - - = 5y (in degrees)
~----~J~~--~-LL-~~~
B' o 30 40
H 6
kN/ m 2
tan 0 ) 0 ·5
m= ( 1 + - - Figure 4-23. Example 1: Earth pressure diagrams.
tan 4> 1 and 2-Earth active pressure diagrams determined
by Coulomb's theory correspondingly for 8 = 0 and
0.364 ) 0 "5 8 = t <P conditions. 3 and 4- Earth pressure diagrams
= (1 + -- = 1.28
0.577 according to Dubrova (1963) (8 = 0 and 8 = j,P).
n = 1 + m sin u 0 = 1 + 1.28sin(5y)
dition is calculated by
lcf>l y (30 X 0.0175)y 2
H 6 (m + sin u 0 ) )
= 0.0875y 2 (in r adians) n cos u 0
(4- 19)
Results of soil pressure (stress) calculations
for different depths (y) are presented in
Table 4-5 and in Figure 4-23. Similarly, where u 0 = lcf> l y j H -lc/>1, and m and n are
soil pressure (stress) values can be obtained as in Eq. (4-17).
by Eq. (4-.,18) for the 8 = 0 condition. For Under the 8 = 0 and q = 0 condition,
results, see Table 4- 5.
a = y t an 2 ( 45° - - uo)( y - 2
-14>1 -Y -1- )
case 2 Y 2 H cos u 0

The wall rotates about point B toward the (4-20)


fill (Fig. 4-22c). The lateral soil pressure
(stress) aY under the 8 =I= 0 and q = 0 con- where u 0 = lcf>l y j H -lcf>l.

Table 4-5. Case 1- soil pressure calculations

y (m )

0 1 2 3 4 5 6

8 = 20° 0 15.2 21.8 22.8 20.9 17.0 12.2


8 =0 0 15.1 23.4 25.8 24.7 21.2 16.2
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 371

Example 2 (for case 21 tan 10o


(1 + ---
)o.s
m = = 1.22
tan20°
Determine the soil pressure (stress) exerted
upon a retaining wall rotated about point B n = 1 + 1.22sin u 0
toward the fill (Fig. 4-24) under the follow-
ing conditions: 'Y 20
- - = - - =20.3
cos 8 0.985
Height of wall H= 10m 14>1 20 X 0.0175
Soil density y = 20 kNjm3 10 = 0.035 (radians)
H
Angle of internal friction cf> = 20°
Friction angle between
wall and fill 8= 0.5cf> = 10° Soil lateral pressure (stress) under the 8 = 0
Angle between back of the condition is determined with the help of Eq.
wall and horizontal f3 =goo (4-20). Hence,
Fill slope angle a=O
lcf>ly
u0 = H - 14>1 = 2y - 20
0
(degrees)

According to Eq. (4-19)


lcf>l y 2 (20 X 0.0175)y 2
H 10
20y
u = - - 20° = 2y - 20° (degrees) = 0.035y 2 (radius)
0 10

The results of relevant calculations are pre-


sented in Table 4-6 and in Figure 4-24.

case 3
0
The wall rotates about point A toward the
g fill (Fig. 4-22£). The lateral soil pressure
(stress) ay at the 8 =I= 0 and q = 0 condition
is given by

8
20 40 110 eo 14 0 y cos 2 u 0 [ 14>1 y 2 ( m +sin u 0 )]
0
a = y+ - -
kN/ m 2 Y n 2 cos8 H ncosu 0
Figure 4-24. Example 2- Earth pressure diagrams. (4-21)
1 and 2- Earth active and passive pressures diagrams
determined by Coulomb's theory for the 8 = 0 condi-
tion; 3 and 4- soil pressure diagram according to where u 0 = -14>1 yjH, and m and n are
Dubrova (1963) (8 = 0.5cp and 8 = 0). the same as in the previous cases.

Table 4-6. Case 2- soil pressure calculations

Y (m)
0 2 4 6 8 10

8 = 10° 0 74.1 112.4 125.0 123.8 116.3


8=0 0 65.5 103.6 124.6 131.9 130.0
372 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

Similarly, under the B = 0 and q = 0 sures, as is shown in Cases 2 and 3, are


condition the soil lateral pressure (a) is valid for relatively small wall movements
given by compatible with those required for the de-
velopment of active soil pressure.
a = 'Y tan2{45o - -Uo)( y + -
14>1Y
- -1- )
2 As is shown earlier, the substantial wall
Y 2 H cos u 0 movement is required for full development
(4-22) of the passive pressure exerted on a retain-
ing wall. This, however, may affect the soil
where u 0 is the same as in Eq. (4-21). characteristics and character of soil failure
plane as assumed by Dubrova. For this rea-
Example 3 (for case 31 son, Dubrova's method (Cases 2 and 3) may
Determine the soil pressure (stress) exerted be invalid or will need some adjustments if
on a retaining wall rotated about point A substantial wall movement is anticipated.
toward the soil under conditions similar to
Case 1 (i.e., soil and wall characteristics). case 4
Results of the relevant calculations are
given in Table 4-7 and presented in Figure The wall is displaced away from the fill
4-25. parallel to its original position (Fig. 4-22d).
The results listed in Table 4-7 and de- In this case, the soil pressure is considered
picted in Figure 4-25 indicate that the soil as an average between pressures obtained
passive pressure is not hydrostatic and for the wall rotated about points A (Case 1)
changes with depth. It also may increase and B, away from the fill. It will be an
substantially under B 0. * average value between soil pressure ob-
It should be noted that the solutions pro- tained by the Coulomb theory (the wall ro-
posed by Dubrova for determining soil pres- tates away from the fill about point B) and

Table 4-7. Case 3-soil pressure calculations

y (m)
0 0.5 1.5 3.5 5.5

ar 15 = 20° 0 12.7 53.6 266.9 1136.6


(kNjm2) 15=0 0 11.4 44.2 171.9 461.3

..

.. " .. •• •• n• ,.

Figure 4-25. Example 3-Earth pressure diagrams. 1 and 2-Earth passive pressure
diagrams determined by Coulomb's theory for 15 = 0 and 15 = fcp conditions; 3 and 4-soil
pressures diagrams according to Dubrova (1963) (15 = 0 and 15 = t c/J).
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 373

pressure as described in Case 1 (the wall the wall structure displaced parallel and
rotates away from the fill about point A). toward the fill. In this case, it will be aver-
A typical example (Example 4) of the age soil pressures obtained by methods de-
above approach to calculating the soil pres- scribed in Cases 3 and 4.
sures exerted on the wall (H = 8.0 m, cp =
30°, y = 18.0 kNjm3 , and 8 = 0) is pre- cases
sented in Table 4-8 and depicted in Figure
4-26. Here, the deflection of the wall may be
A similar approach can be taken in order described in terms of a case in which the
to calculate the soil pressures exerted on upper half of the wall is affected by defor-
mations characteristic for Case 1 (Fig.
4-22b) and the bottom half is affected by
deformations characteristic for case de-
picted in Figure 4-22a. Accordingly, soil
pressure distribution for both halves may
be determined by relating equation (4-17)
or (4-18) for the upper part of the wall, and
by Coulomb's theory for the bottom part.
Example 5 illustrates the above ap-
proach to determine the soil pressure redis-
tribution at a wall 8.0 m high, where y =
50 18.0 kNj m 3 , cp = 30°, and 8 = 0. Calcula-
tion results are presented in Table 4-9 and
depicted in Figure 4-27.
Figure 4-26. Example 4-Earth pressure diagrams. A great deal of laboratory and full-scale
1-Active earth pressure determined by Coulomb's
theory (o = 0); 2-soil pressure diagram (AB) accord-
observations conducted to date indicated
ing to Dubrova (1963) (o = 0), Case 1; 3-average soil that under Case 5 conditions the total soil
pressure (o = 0). pressures exerted on retaining walls is

Table4-8. Case 4- soil pressure calculation

Y (m)
0 2 4 6 8

Coulomb 0 11.9 23.8 35.6 47.5


theory
ay Case 1 0 24.0 31.0 28.0 18.6
(kNjm2 )
<s = o) Average 0 17.95 27.4 31.8 33.5

Table4-9. Case 5- soil pressure calculations

Y (m)

0 1 2 3 4 5 6 7 8

ay 0 12.0 15.4 14.0 9.3 30.0 36.0 42.0 48.0


(kNjm2 )

Notes:
1. Pressure for y = 0 through 4 m obtained by using Eq. (4- 20), and pressures for y = 5 through 8 m are obtained by
Coulomb's theory ay = gHKA, where for ll = 0, KA = tan 2 (45° - </>/ 2).
374 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

about 5-8% smaller than that determined Finally it is important to note that the
by Coulomb's theory. To account for it, total value of the soil pressure exerted on
Dubrova suggested the approach as de- retaining structures obtained by the
pictedin Figure 4-27. Dubrova method (with the exception of Case
In the latter case, values of the soil pres- 5) is equal to that obtained by Coulomb's
sure for the upper part of the soil pressure theory. In each respective case, the total
diagram (h 1 ) are determined as recom- soil pressure value (E) can be obtained by
mended in Case 1 (curve Ab). For the bot- integrating the soil pressure diagram:
tom part of the wall (h 2 ), the soil pressure
from point c, which is located at %h 2 from E = .LayC.:ly) (4-23)
the base of the wall and down (cD portion),
where a Y is the soil pressure ordinate at
is distributed according to the classic the-
the respective depth y. Accordingly, the
ory. Then points b and c are joined by a
center of gravity of the soil pressure dia-
gentle curve; the final shape of the soil
gram is determined by
pressure diagram will be AbeD.
Soil pressures for all potential combina-
tions of wall movements may be determined (4- 24)
as average values obtained for the related
wall movements [e.g., parallel displacement Effects of any kind of live loads (uniform or
(Case 4) combined with outward deflection concentrated) placed on fill can be deter-
(Case 5), and others]. mined by one of the earlier described meth-
ods and added to soil pressure diagram.
For more useful information on effects of
wall movements .on redistribution of soil
pressures consult Canadian Foundation
Engineering Manual (1985) and Clough and
Duncan (1991).

4_7.3 Effects of
Time-Dependent
Changes in Soil
I
I As was mentioned earlier, soil strength and
I
I lateral pressures may be significantly in-
\ c fluenced by miscellaneous time-dependent
factors, such as consolidation, chemical

-=~\B~~~--~~~~
0 10 20 30 40
0
50
changes, creep, and built-in stresses due to
berth operations.
All of the above factors independently or
kN / m 2
in combination may sometimes result in a
Figure 4-27. Example 5-Earth lateral pressure. substantial increase in soil lateral pres-
1- Active earth pressure determine by Coulomb's the- sures on a retention structure. Nearly all
ory (8 = 0); 2- soil pressure diagram (AB) according investigators involved with postconstruc-
to Dubrova (1963) ( 8 = 0), Case 1; 3- transitional por-
tion of soil pressure diagram (be). Note: Total area of tion soil pressure measurements report
diagram BAbcD is by about 5- 8% smaller than dia- higher than active pressure values. How-
gram BAD. ever, in some field measurements, lateral
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 375

soil pressure has been found smaller than restraint provided by the forward shifted
normal active pressures exerted on retain- material. An example of this phenomenon is
ing structures (Gould, 1970; Kany, 1972; shown in Figure 4-28.
Tsinker, 1972, 1983; Matsuo et al., 1978). In this example, the soil-retaining struc-
For a comprehensive list of relevant refer- ture is loaded with a uniform distributed
ences, see Bjerrum et al. (1972), Tsinker surface load q(X1 , X 2 ). According to Budin
(1983), and Duncan et al. (1990). and Demina (1979), when this load is re-
It should be noted that, in general, nearly moved, the return movement of the wall's
all results of earth pressure measurements top~ - ~~usually does not exceed the value
reported to date (particularly those taken of (0.1-0.2)~, where ~ is the original de-
from field investigations) are scattered and flection of the wall corresponding to
sometimes inconclusive. This is basically at- q(X1 , X 2 ), and ~' is the return movement
tributed to two factors: first is that almost after the surcharge load is removed. Accord-
all investigators used different instruments ingly, a soil pressure ordinate, a(q ), is re-
to take soil pressure measurements, and duced to a value a(q) - a'(q ), in which a(q)
second, most of the field measurements rep- and a'(q) respectively represent the ordi-
resented just a single case of soil-structure nates of the original and the residual soil
interaction. In general, a single measure- pressures created by the surcharge load
ment of soil pressure has doubtful value q(X1 , X 2 ). Note that ordinate a in Figure
because it is never clear whether a single 4-28 represents an ordinate of the original
measured value represents the average be- soil pressure with no surcharge load on the
havior or an aberration. surface.
Carder and Krawczyk (1975), Weiler and Inherent in structural residual stresses
Kulhawy (1978, 1979), and Dunnicliff(1988) and deformations (deflections) are bending
offer a good discussion concerning the dif- moments that can be expressed through the
ficulties involved in earth pressure mea- factor~= M'(y)jM(y), where M'(y) is the
surements and techniques available to
achieve credible results.
The fundamental reasons for time-
dependent increases in soil pressure are not

X.
well understood. It seems, however, clear
that the "at rest" pressure condition repre-
sents the long-term situation for "nonyield-
ing'' walls and, in some cases, for flexible
sheet-pile bulkheads.
Budin and Demina (1979) offered the fol-
lowing explanation of the above phe-
nomenon. During its functional life, a re-
taining structure is repeatedly exposed to
the effects of miscellaneous surface loads.
These loads develop residual stresses in the
retaining structure and adjacent backfill
material. These stresses have associated de-
formations (deflections) in both the struc-
ture and soil. For example, when a live load
is temporarily removed, the deformed (de-
flected) shape of the structure cannot re- Figure 4-28. Effect of repeated loading on a retain-
turn back to its original shape because of ing structure. [After Budin and Demina (1979).]
376 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

residual bending moment created by sur- summer, the wall was pushed away from
charge load, q(Xv X 2 ), and M(y) is the the fill, and the earth pressure decreased.
bending moment produced by the total soil In winter, the slab contracted and the pres-
plus surcharge loading. The factor g de- sure increased. Although, in general, the
pends on the soil's relative density, D, and temperature effect is unlikely to cause wall
stiffness of the structure. According to failure, it must be recognized and evaluated
Budin and Demina, for sheet-pile bulk- particularly during field earth pressure
heads, the value of g varies from 0.85 to 0.9 measurements.
for D;::: 0.75 and 0.75-0.85 and 0.65-0.75
for 0.75 > D > 0.5 and D:::;; 0.5, respec-
tively. 4.7.5 Effects of
Another peculiarity associated with the
above phenomenon is that stresses at the Backfill Freezing
wall will start increasing again only after a
new surface surcharge load will exceed the In most cases, marine structures are back-
value of q*(X1 , X 2 ) that corresponds to the filled with good quality granular material
residual stresses that already exist in the which usually is not susceptible to freezing.
structural components. Observation of this However, in some rare cases, when backfill
phenomenon may sometimes be the key to is made from clay or even silty sands the
proper understanding and interpretation of water within voids may freeze, resulting in
the results of some field investigations. soil expansion. This effect may sometimes
produce substantial lateral pressures ex-
erted on a retaining structure.
Sandegren et al. (1972) described a case
4.7.4 Effect of Ambient in which freezing caused failure of some of
Temperature on the tie rods anchoring a sheet-pile bulkhead
Earth Pressures retaining a clay backfill. Rehnman and
Broms (1972) reported results of large-
A number of investigators (Granger, 1965; model tests on a cantilever wall backfilled
Symons and Wilson, 1972; Broms and with uncompacted silty sand. In these tests,
Ingelson, 1972; Coyle and Bartoskewitz, the earth pressures increased due to freez-
1976; Brandl, 1987) have noted that daily ing by 38 to 48 kPa at some levels behind
and seasonal cyclic variations in meas- the wall. However, freeze-induced pressure
ured earth pressures were caused by the decreased during the thaw period to levels
temperature-induced wall movements. slightly lower than the pressures before
In most of the cases discussed, the earth freezing.
pressure increased in the summer when the In North America the depth of earth
face of the wall was relatively warm, then freezing may be obtained from charts devel-
expanded relative to the back, and, hence, oped by the U.S. Corps of Engineers (1961).
deflected toward the fill. The earth pressure
decreased in winter when the wall con-
tracted and deflected away from the fill. 4.7.6 Passive Earth Pressure
The measurements made by Symons and
Wilson (1972) are an exception to the rule If a retaining structure is pressed against
because in their case the wall at the top an earth mass such as in the case of a wall
was connected to a pavement slab extend- rotating about its base toward the fill, or in
ing across the top of the backfill. As the the case of forward movement of a sub-
temperature of the slab increased in the merged part of a sheet-pile bulkhead, the
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 377

/ I
/ I
/ . I
/ . I
/ . I
. I
A I
H / '-l I
I
)
2

p, =45"-<~~ 2 I
......._ _ __.,.,. / /
Figure 4-29. Earth passive pressure. !-Straight (classic) plane of failure; 2-curved
plane of failure.

passive pressure of soil or rather the soil's Experiments indicate that the curved
passive resistance can be developed. The portion of a sliding surface may be approxi-
conditions required for the development of a mated by a circle or a logarithmic spiral. AB
passive resistance were discussed earlier in suggested by Kezdi (1975), if the vertical
this section. Passive earth pressure EP is wall movement is small, the wall friction
the force which is required to displace an (S) and its effect on the shape of the sliding
earth mass along a sliding surface located surface will also be small, and the sliding
in front of a wall. surface itself will be close to planar. If the
Classic Rankine theory assumes that the movement along the wall interior is such
earth sliding surface is more or less planar. that significant frictional forces are mobi-
However, the real geometry of the sliding lized, then a curved sliding surface will
surface is curved. In the latter case, the develop.
upper portion of the sliding surface is rela- The classic theory of earth pressure as-
tively planar, but the lower part is defi- suming a planar sliding surface (Fig. 4-30)
nitely curved as indicated in Figure 4- 29. defines the passive earth pressures (EP) as
EP = 0.5yH 2 KP + 2cH tan(45° + c/J/2) (4-25)

where the coefficient of earth lateral pas-


sive pressure K P is given by
H 2

sin( J3 + <f>)
Kp =
[sin( J3 + 8)) 0 '5
_ [ sin(8 + cp) sin(cp +a) ]o.s
sin( J3- a)
(4-26)
Figure 4-30. Cohesive soil passive pressure.
378 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

The method of determination of passive For granular soils, Rankine (1857) ex-
earth pressures developed by Sokolowski pressed the coefficient of passive earth pres-
(1965) considers the curved geometry of the sure, KP, as follows:
failure sliding surface. According to
Sokolowski, the coefficient of soil passive 1+sin¢
K = --:------:-----:- (4-29)
lateral pressure (stress) is calculated by P 1-sin¢

1 + sin 4> cos e It should be noted that Sokolowski's solu-


K = e~tana (4-27)
P 1-sin¢ tion presents a less conservative but more
realistic approach to K P calculation.
where Lee (1961) suggested that wall friction
( 8) may be conservatively estimated by
sin
e = 8 + arcsin ( - . -
8) multiplying the value of K P obtained by Eq.
sm 4> (4-29) by a factor of 1.5 for soils in which ljJ
exceeds 20°.
Values of K P derived from Sokolowski Example calculations of the passive soil
theory are given in Table 4-10. pressure for different configurations of soil
In the case when a = 0 and 8 = 0, for- profiles in the passive zone are depicted in
mula (4-28) can be used to determine KP: Figure 4-31. Effects of wall movements on
passive pressure distribution are discussed
(4-28) in works by Rowe and Peaker (1965),

(a ) {b )

(c ) (d )

Figure 4-31. Earth passive pressure. (a) Effect ofunifonn surcharge surface load; (b)- (d)
- effect of different backfill profiles in the passive zone.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 379

Table4-10. Coefficients of earth passive pressure (KP)

f3- goo
.p 8 -30 -20 -10 0 10 20 30
100 0 1,20 1,25 1,33 1,42 1, 51 1,62 1,76
50 1,27 1,36 1,46 1,56 1,68 1,81 1,97
10° 1,34 1,44 1,54 1,66 1, 79 1,94 2,21
15° 0 1,32 1,40 1,50 1,69 1,93 2,08 2,34
50 1,44 1,56 1,68 1,87 2,15 2,39 2,71
100 1,60 1,63 1,86 2,03 2,36 2,69 3,10
15° 1,68 1,84 2,03 2,20 2,53 2,91 3,48
20° 0 1,4S 1,60 1,80 2,04 2,33 2,79 3,18
50 1,56 1,76 2,00 2,28 2,64 3,13 3,36
100 1, 71 1,9S 2,22 2,55 2,98 3,52 4,22
lS 0 1,84 2,09 2,40 2,78 3,26 3,87 4,67
20° 2,00 2,26 2,61 3,04 3,58 4,28 5,20
2S 0 0 1,58 1, 75 2,15 2,46 3,00 3,68 4,30
so 1, 75 1,98 2,38 2,88 3,41 4,25 5,13
100 1,93 2,26 2,64 3,25 3,84 4,88 5,95
15° 2,13 2;5o 2,94 3,62 4,33 5,54 6,82
20° 2,28 2,72 3,21 3,98 4,89 6,28 7,71
25° 2,49 2,98 3,55 4,35 5,52 7,13 8,67
30° 0 1, 72 2,02 2,43 3,00 3,70 4, 71 6,10
50 1,91 2,30 2,81 3,47 4,37 S,63 7,49
100 2,15 2,60 3,23 4,00 5,12 6,68 8,87
15° 2,40 2,96 3,66 4,62 5,91 7,85 10,46
20° 2,62 3,28 4,10 5,16 6, 70 8,88 11,97
2S 0 2,87 3,64 4,58 5,78 7,56 10,10 13,69
30° 3,22 4,04 5,10 6,S5 8,S5 11,44 15,61
35° 0 1,90 2,28 2,90 3,69 4,70 6,46 9,00
so 2,21 2,68 3,43 4,37 5,83 7,93 11,80
10° 2,50 3,13 4,02 5,13 7,11 9,53 14,88
15° 2,82 3,58 4,64 5,9S 8,48 11,22 18,12
20° 3,17 4,07 5,34 6,82 9,96 13,28 21,02
25° 3,53 4,64 6,14 7,86 11,57 15,71 23,61
30° 3,93 5,26 7,04 8,96 13,34 18,50 26,60
40° 0 2,14 2,66 3,42 4,60 6,25 8,86 13,00
50 2,50 3,16 4,20 5,72 8,03 11,50 17,56
100 2,74 3,76 5,02 6,91 9,8S 14,30 22,50
15° 3,23 4,35 S,90 8,22 11,89 17,39 27,42
20° 3,66 5,00 6,87 9,68 14,03 20,75 32,80
25° 4,19 5,74 7,99 11,35 16,64 24,95 39,30
30° 4,75 6,58 9,24 13,25 19,50 29,41 46,20

Source: From Klein and Cherkasov (1985).


380 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

Mackey and Kirk (1967), Narain et al. The thickness of riprap to be taken as a
(1969), James and Bransby (1970), and design surcharge load (h.) is assumed to be
Fang et al. (1994). the distance between lines (a-a) and (b-b),
In some practical cases, such as piled which are fixed by points A 1 and A 2 . Then
platforms with retaining structures placed the calculation of passive soil pressure is
behind the platform or the modernization of conducted as previously discussed. In the
existing berthing facilities, the passive latter case the lateral earth passive pres-
pressure on retaining structures may be sure diagram is a combination of superim-
affected by the presence of other structures posed components of passive pressure dia-
(e.g., piles). The presence of these struc- grams related to horizontal (KP) and sloped
·tures may substantially increase soil pas- (Kp(a)) surfaces.
sive resistance. Because a theoretical ap- For more useful information on soil pas-
proach to the problem has not yet been sive pressures consult Canadian Founda-
developed, an approximate method called "a tion Engineering Manual (1985).
fictional dredge line method" is often used,
as shown in Figure 4-32. Under this method
a sloped (or other) surface in front of the
wall (BD) is replaced by an assumed dredge 4.7.7 Earth Pressure at Rest
line (a-a), which intersects with the surface
of a slope in front of the wall at point A 1 . As stated previously, the pressure "at rest"
Then the passive soil pressure diagram is applies to relatively unyielding rigid walls
calculated as previously discussed. If there where no deformation or displacement of
is riprap present on the top of the soil slope soil mass occurs. The magnitude of the co-
(line BD), then it may be considered as a efficient of lateral pressure at rest (K0 ) de-
surcharge load in calculations of a passive pends on the amount of frictional resistance
soil pressure. mobilized at contact points between soil

I
I
'Y2,<1>2 "f!
; 11
I I
--b
I I
a
I I H

(a) (b) (c)

Figure 4-32. Earth passive pressure on sheeting placed behind piles. (a) Piled platform with
rear sheet-pile wall; (b) passive pressure without riprap; (c) passive pressure with riprap.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 381

particles (Lambe and Whitman, 1969). Ex- rial, compaction-induced pressures are un-
perimental values of K 0 are best repre- likely to exceed 20-30 kPa and the depth to
sented by an expression proposed by J aky which compaction pressures are significant
(1944) as will not exceed 3-4m.
Several empirical and analytical meth-
K 0 = 1- sin f/J (4-30) ods have been proposed to evaluate residual
lateral earth pressures induced by soil com-
For a sand fill placed behind an "unyield- paction (Rowe, 1954; Sowers et al. 1957;
ing'' quay wall, K 0 will typically have a Broms, 1974; Seed and Duncan, 1985;
value between 0.4 and 0.5 (Tschebotarioff, Duncan and Seed, 1986; Peck and Mesri,
1973). For more information on value of K 0 1987; Clayton and Symons, 1992). Duncan
an interested reader is referred to Lambe et al. (1991) developed charts for estimation
and Whitman (1969). of the compaction-induced pressure. These
charts are based on hysteretic theory devel-
oped by Duncan and Seed (1986). Common
4.7.8 compaction-Induced to all of these methods is the idea that
Pressure compaction represents a form of overconsol-
idation wherein stresses resulting from a
Compaction of soil in layers adjacent to un- temporary or transient loading are retained
yielding retaining wall induces lateral earth to some extent following removal of the peak
pressures in the fill that can be significantly load.
higher than normal "at rest" values as indi- It should be noted that a standard granu-
cated in the previous section. This has been lar soil densification procedure cannot
found to be attributed basically to overcon- change lateral earth pressure substantially.
solidation of the backfill material. As pointed out by Polshin and Sinelschikov
Hence, in the case of compaction-induced (1953), the change in soil density automati-
earth-pressure under the K 0 condition, the cally leads to an increase in the angle of
resulting structural stresses and deforma- internal friction. For example, a change in
tions should be of concern in the design and the sand porosity coefficient from 0.8 to 0.6
analysis of gravity-type quay walls, and would lead to an increase in an angle of
particularly those constructed on bedrock internal friction by up to 22%. Accordingly,
foundations. the effect of increased density on lateral
A large number of laboratory and full- pressure is neutralized by a decrease in the
scale experiments on compactation-induced coefficient of lateral pressure. Ultimately,
earth pressures have been carried out dur- the increase in lateral pressure would be in
ing the past 50-55 years. A comprehensive the order of 7 to 8%, which is actually
list of references on these studies is found within the accuracy of determination of soil
in Duncan and Seed (1986) and Clayton parameters. The latter case is illustrated by
and Symons (1992). The basic general ob- data provided in Table 4-11. Similar re-
servation of these studies is that the depth sults have been reported by Kostilyova
to which compaction increases lateral earth (1956).
pressures is a function of the dimensions
and vertical thrust of the compaction roller,
varying from 2-3 m for small vibratory 4.8 FRICTION FORCES ON WALLS
rollers to as much as 15 m for heavy com-
paction equipment. A large number of field observations show
Clayton and Symons (1992) suggested that all kinds of retaining walls (freestand-
that in the case of granular backfill mate- ing and stiff) are exposed to downward-
382 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

Table4-11. Effect of sand densification on soil loads such as vehicular traffic, cargo han-
lateral pressures
dling equipment, or blasts and explosions
Porosity Coefficient are usually not considered. This is because
Soil Parameters 0.5 0.6 dynamic effects caused by slow-moving
vehicles and trains or by cargo handling
Angle of internal 40 35 equipment are relatively small and can
friction <!> (degree)
readily be taken into account in the design
Density g (kNjm 3 ) 17.4 15.3 by the application of an appropriate equiva-
Coefficient of lateral lent surcharge load.
pressures (.S = 0):
On the other hand, blasts or explosions
Active, KA 0.22 0.27
Passive, KP 4.6 3.69 are very unusual kinds of loads to be con-
At rest, K 0 0.36 0.43 sidered in general-purpose port design. Ef-
Intensity of soil lateral fects produced by these loads are only con-
pressures (u = gK): sidered in design of special purpose ports or
Active 3.83 4.13 terminals (e.g., military bases) or terminals
Passive 80.04 56.46
designated to handle explosives or highly
At rest 6.27 6.58
explosive materials stored in close proxim-
Source: From Polshin and Sinelschikov (1953). ity to the waterfront structures.
Similar to static lateral pressure, soil dy-
namic pressures are not well defined and
directed shear forces (Terzaghi, 1934; understood. In many cases the degree of
Gould, 1970; Fukuoka, 1980; Brandl, 1987). dynamic earth pressure which· a retaining
Typically, these shear forces are generated structure is able to withstand depends to a
by downward movements in the backfill as considerable extent on the margin of safety
it settles relative to the wall (Clough and provided for static load conditions. In most
Duncan, 1971). Only 2-5 mm of relative case histories described in the literature the
shear movement between the backfill and waterfront structure damage or failures due
the wall is needed to mobilize the full shear to dynamic loads were attributed to factors
strength on the interface. The shear force such as increase in soil lateral pressure
can increase in magnitude with time. This behind the structure and complete or par-
has been observed in tests performed on an tial loss of shear strength of the saturated
instrumented retaining wall (Duncan et al., backfill andjor foundation material (Amano
1990) and in test results reported by
et al., 1956 (a), 19-6 (b); Hayashi et al.,
Fukuoka (1980). Although very often
1966· Pitilakis and Montsakis, 1989; Ho et
downward-directed shear forces are ignored
al., i990). A summary of some quay wall
by designers, they may have a very impor-
failures is found in Seed and Whitman
tant stabilizing effect on retaining walls. (1970).
The earliest methods for determining the
dynamic lateral pressure on retaining
4.9 DYNAMIC SOIL PRESSURES structure was that developed by Mononobe
(1929) and Okabe (1926), and further de-
<BY G. TSINKER AND
scribed by Mononobe and Matsuo (1929). It
S. DUNBAR) is now known as the Mononobe-Okabe for-
mulation. Applications and discussion of the
Dynamic loads on marine earth-retaining formulation are given in Seed and Whitman
structures mainly result from an earth- (1970), Richards and Elms (1979), Prakash
quake. Other potential causes of dynamic (1981), and Whitman (1990).
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 383

The method was developed for dry cohe- on dynamic soil-structure interaction by
sionless materials and was based on the Whitman (1990) essentially confirmed the
following assumptions: relative correctness of Mononobe-Okabe
theory. As expressed by Whitman, " ... the
1. The wall yields sufficiently to produce
Mononobe-Okabe equation must rank-
minimum active pressures. along with its ancestral Coulomb equation
and Terzaghi's equation of consolidation-
2. When the minimum active pressure is as one of the most enduring and successful
attained, a soil wedge behind the wall is theoretical contributions to geotechnical
at the point of incipient failure, and the engineering."
maximum shear strength is mobilized
along the potential straight sliding sur-
face.
3. The soil behind the wall behaves as a 4.9.1 Mononobe- Okabe
rigid body so that accelerations are uni-
form throughout the mass.
Formulation
Figure 4-33 shows the configuration of
Although in reality the behavior of soil- forces acting on a gravity wall in the event
retaining structure during earthquake is of an earthquake. If k h and k u are respec-
much more complicated than envisioned in tively the horizontal and vertical accelera-
the model developed by Mononobe and tions in units of the acceleration due to
Okabe, when used with the proper choice of gravity, g (e.g., khg and kug), the modified
input parameters and suitable safety factor, theory essentially adds the inertial forces
their theory provides a sound basis for the Wk h and Wk u to the static case described
design of soil-retaining structures. A review by the Coulomb theory. This results in the
of some important experimental data following equation for the total quasi-active

Figure 4-33. Soil dynamic pressure.


384 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

thrust against the wall (PAE) der conditions assumed by the Mononobe-
Okabe equation the KAE value is very close
(4-31) to that obtained from Eq. (4-32).
For practical purposes, Ichihara (1983)
where recommends the angle of wall friction ( 8) to
y = unit weight of backfill be used as equal to 2j34J for active thrust
H = height of wall calculations. However, the question of
KAE =active lateral soil thrust coefficient whether the wall friction for passive pres-
sure calculation can be relied on during
seismic excitations is a question for which
no experimental data are available to
answer. Thus, it is usually neglected in
cos f/! cos 2e cosUl + e+ f/!) x
design practice.
2
sin(¢+ B)sin(<P- f/!- a) ] 0 "5 ) The Mononobe-Okabe formulation as-
(1 [
x + cos(B + e + f/!)cos(a- e) sumes a linear soil pressure distribution. In
reality, the distribution of static and dy-
(4-32)
namic forces along the wall is not linear, as
assumed for the derivation of Eqs. (4-32)
and (4-34) (Prakash and Basvanna, 1969).
where 8, a, and 8 are as defined in Figure
The latter affects the application height of
4-33 and lj; is the angle between the verti-
the resultant force which is normally higher
cal and the resultant of the weight and
inertial terms (sometimes lj; is called the than that resulting from the static pres-
seismic inertia angle) sures. Wood (1973) found that the resultant
soil dynamic pressure acts approximately at
f/! = arctan(Kh(1 - k)) midheight of the wall.
Seed and Whitman (1970) recommend
Similarly, quasi-passive soil resistance that the resultant of the seismic load be
(PpE) is obtained by applied at a height of 0.6H, whereas
Prakash (1981) recommends that the point
(4-33) of application of this force be 0.45H for a
rigid wall and 0.55H for a flexible wall.
in which However, some more recent experimental
results (Sherif et al. 1982; Sherif and Fang,
1984; Matsuzawa et al., 1985) indicate that
the point of application of the resultant of
cos f/! cos 2e cos(B- e+ f/!) the dynamic thrust lies between 0.45H and
05 2 0.55H and depends on the mode of the wall
X ( 1 _ [sin(¢+ B)sin(<P- f/! +a) ] " ) movement. On the basis of this information
cosUl- e + f/!)cos(a- e)
it is reasonable to assume that for most
(4-34) purposes the height of the dynamic soil
thrust can be taken at midheight of the
As the seismic inertia angle lj; increases, wall.
the values of KAE and KPE approach each As follows from Eqs. (4-32) and (4-34),
other and, for a vertical wall-backfill inter- KAE and KPE values are obviously very
face, become equal when lj; = 4J. sensitive to the values of 4J and kh. For
Experiments by Sherif et al. (1982) and example, for a constant value of 4J, KAE
Sherif and Fang (1984) indicated that un- doubles as k h increases from 0 to 0.35 for
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 385

zero vertical acceleration, and thereafter it 4.9.2 Effect of


increases more rapidly. saturated Backfill
In order to evaluate the increase in soil
active pressure due to earthquake effects The calculation of the hydrodynamic pres-
more easily, KAE can be normalized by di- sure on the backfill side is complex because
viding by its static value KA to give a thrust of the existence of soil particles. The move-
factor (Fr) ment of fluid in the backfill is not free due
to restrictions provided by soil.
(4-35)
When the backfill is submerged, two
modifications to parameters in Eqs. (4-32)
As indicated by AASHTO (1992), Fr in-
and (4-36) must be made. First, as the only
creases rapidly and becomes infinite for
contribution to static lateral earth pressure
specific critical values of cp. The reason for
is the effective vertical stress, the unit
this behavior stems from the nature of eq.
weight, 'Y, of the backfill should be replaced
(4-32) in which the contents of the radical
by the submerged weight. Second, the
must be positive for a real solution to be
quantity tan l/J should be modified to ac-
possible, and for this it is necessary that
count for buoyancy forces. This depends on

:\u)
the soil type. Matsuzawa et al. (1985) show
¢;:::a+lf!=a+arctan( 1 (4-36) that for high-permeability backfill

This condition could also be thought of as tan l/1 = a. (- kh


--- --) (4-38)
specifying a limit to the horizontal accelera- a.- 1 1- ku
tion coefficient that could be sustained by
the gravity-type quay wall supporting any whereas for a low-permeability backfill
given fill.
The limiting condition is that
a.+
tan l/1 = - e ( kh )
-- --- (4-39)
(4-37)
a.- 1 1- ku

For zero vertical acceleration and backfill where G. is the specific gravity of the back-
angle and for a soil friction angle of 35°, the fill soil particles and e is the void ratio of
limiting value of kh is 0.7. This is a figure the backfill.
of some interest in that it provides an abso-
lute upper bound for the seismic accelera-
tion that can be transmitted to any struc- 4.9.3 Hydrodynamic Pressures
ture whatsoever built on soil with the given
strength characteristics. In addition to the active thrust against the
Furthermore, as indicated in AASHTO wall due to the backfill, seismic loads can
(1992), positive values of kv have a signifi- induce hydrodynamic water pressures in the
cant effect for values of k h greater than 0.2. pore water of the backfill.
As is expected from Eq. (4-32), KAE and Fr A reasonable estimate of the hydrody-
are also sensitive to variations in backfill namic pressures may be made for the case
slope, particularly for higher values of hori- of highly permeable soils using solutions
zontal acceleration coefficient when the limit derived by Westergaard (1933), Werner and
implied by eq. (4-32) is approached. More Sundquist (1943), Zangar (1953), and
recent discussion on Mononobe-Okabe Chwang (1978) for the pressures on rigid
theory is given in Fang and Chen (1995). walls due to fluid reservoirs. However, in a
386 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

low-permeability soil, the water experiences that the water can be considered to be part
more interference from the soil particles, of the soil. In this case, the equivalent seis-
resulting in reduced hydrodynamic pres- mic coefficients (k', = 'Ysatkhjy' and k~ =
sures. If the permeability is very low (e.g., 'Ysatku!Y') can be used instead of the origi-
as in clay), there may be very little or no nal coefficients k h and k u. Following this
hydrodynamic pressures. The real situation assumption, it is clear that the water move-
is somewhere between these two extremes. ment is completely restricted by the soil
Matsuo and Ohara (1965) derived the hy- particles (case of a backfill composed from a
drodynamic pressures on vertical walls re- low-permeability material) and the pres-
taining backfill of arbitrary permeability. sure is linearly distributed.
Using this solution, Matsuzawa et al. (1985) Ishibashi et al. (1994) investigated the
derived a generalized procedure, illustrated dynamic lateral pressure due to saturated
with examples, that uses easily measured backfill on rigid nonyielding walls. These
or estimated soil properties to include hy- investigators arrived at the following basic
drodynamic pressures in a rational manner. conclusions:
The distribution of hydrodynamic force
from the water standing against the ex- 1. The dynamic water pressure within back-
posed face of the wall (pw) can be obtained fill material can be either a Westergaard
from the Westerguard formula (1933) type, provided the water can flow
free in a nondeformable backfill soil skele-
ton, or due to deformability of a backfill
(4-40) soil skeleton under undrained conditions.
The latter type of the dynamic water
where pressure distributes in such a way that
there is a peak at an upper section of the
'Yw = unit weight of water
backfill and it decreases to zero at the
h = the depth of water in front of wall bottom of the backfill; the resultant of
y = the depth beneath the water surface this force is applied at approximately 0.6H
from the base of the wall.
Steedman and Zeng (1990) stated that hy- 2. The dynamic total pressure (water plus
drodynamic pressure on a flexible wall such soil) for cohesive soils is nearly twice the
as a sheet-pile bulkhead is higher than that Westergaard' dynamic water pressure,
given by formula (4-40). This is because of and the resulting total force is also lo-
the increased movement of this type of wall. cated at approximately 0.6H from the
For flexible wall analysis, they recommend base of the wall.
a method proposed by Shulman (1987) for
design of flexible dams.
Equation (4-40) has also been used to 4.9.4 Effect of Wall Inertia
estimate the effect of the dynamic water
pressure in highly permeable saturated All of the above formulations ignore the
backfills (e.g., large gravel or rubble) in effect of the inertia of the gravity wall.
which water to some extent is assumed to Richards and Elms (1979) and Elms and
move relatively freely under the inertial Richards (1990) suggested that this is un-
loads. In the latter, the dynamic thrust from conservative and unreasonable because it is
the fill is given by the Mononobe-Okabe the weight of the gravity-type wall that
formulation using the buoyant unit weight provides most of the resistance to move-
of material. ment. By including the inertial effect due to
Amano et al. (1956a, 1956b) suggested the wall, Richards and Elms derived an
that the pore fluid moves with the soil so equation, (4-41), for the mass of the wall
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 387

(Mw) required to prevent sliding along its transmitted to the wall. If the ground accel-
base. They found that the inertial effect of eration exceeds this limit, the wall will slide.
the wall is of the same order of magnitude Richards and Elms suggested that an equa-
as the active thrust due to the backfill. tion empirically derived by Franklin and
Thus, wall inertia cannot be neglected in Chang (1977) for predicting the relative
the seismic design of gravity walls: base displacement of dams and embank-
ments be used to estimate the limiting ac-
0.5yH 2 [cos(8 + fJ) celeration. The equation for the displace-
-sin(8 + fJ)tan ci>b]KAE ment, d, is
Mw = (4-41)
tan cpb - tan 1/J
d = ( i ( _k h ) -
0.087 pgv
4
(in seconds)
where tan cf>b is the base friction coefficient pga pga
and all other variables are the same as in (4-42)
preceding equations.
where pga and pgv are the peak ground
acceleration and velocity, respectively. By
4.9.5 Selection of selecting a maximum allowable displace-
Ground Motions ment and using pga and pgv estimated as
free-field ground motions, the coefficient k h
The accelerations k h and k u are not the may be computed.
free-field ground motions normally derived The vertical acceleration, k u is often
from a seismic hazard analysis. If the back- taken as a fraction, typically j, of the hori-
fill and wall are considered as an above- zontal acceleration, k h (Newmark and Hall,
ground structure, k h and k u are pseudo- 1982). However, Mohraz (1976) showed that
acceleration response spectral ordinates at the relationship between horizontal and
the resonant period of the backfill-wall vertical acceleration depends on site soil
combination. conditions.
The estimation of the resonant period of
the backfill-wall combination ({0 ) requires
a somewhat detailed analysis [see, for ex- 4.9.6 Effect of
ample, Steedman and Zeng (1990)]. How- Wall Movements
ever, because the backfill comprises a sig-
nificant part of the structure, the structure As was noted earlier, a quasi-static dy-
may be considered as a layer of thickness H namic thrust on retaining structures based
overlying a half-space. For details, consult on Mononobe-Okabe equations is primarily
Chapter 3, Section 3.7. Typical values of the a modification of Coulomb's theory of static
resonant period lie in the range 0.25-0.40 s. earth pressure to account for the dynamic
Ground motions may also be selected on condition. In the above formulation, no con-
the basis of maximum allowable displace- sideration is made for the potential effect
ment of the wall relative to its base. for wall movements on soil pressure redis-
Richards and Elms (1979) and Elms and tribution. However, detailed theoretical and
Richards (1990) reasoned that if the wall is experimental studies (for comprehensive list
to move along its base, a shear force must of references, see Bakeer et al. (1990)] on
be developed at the interface between the the effect of wall movements have indicated
wall and its base. The mobilized shear re- that in some cases designs based on the
sistance along this interface places a limit Mononobe-Okabe theory may underesti-
on the force or acceleration that can be mate the magnitude of the dynamic earth
388 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

pressure on retaining walls. Therefore, to Mononobe-Okabe method of analysis origi-


design a soil-retaining marine structure to nally developed for dry soil has been ex-
resist dynamic forces, the expected mode tended for dynamic pressure analysis of
of a wall movement, which can be predic- saturated soil. It proved to be sound, pro-
ted from the field conditions, and the de- vided that the proper choice of input pa-
signed construction sequences should be rameters and· factors of safety has been
incorporated. made.
Displacement-controlled design methods On the other hand, the majority of known
which allow for a small tolerable wall dis- case histories of the dock structure failures
placement may provide a more economical are attributed to liquefaction of foundation
solution than the Mononobe-Okabe analy- and backfill materials. This indicates the
sis method. Some practical recommmenda- importance of the evaluation of site and
tions on this type of design methods are backfill liquefaction potential. The latter
found in AASHTO (1992) and Elms and should be given proper attention during
Richards (1990). dock design and construction processes.
As previously noted, the Mononobe-
Okabe analysis assumes that the wall is REFERENCES
free to yield laterally by an amount suffi- AASHTO, 1992. Guide Specifications for Seismic
cient to mobilize peak soil strengths in the Design of Highway Bridges, American Associ-
soil backfill. For granular soils, peak ation of State Highway and Transportation
strengths can be assumed to be mobilized if Officials, Washington DC.
deflections at the top of the wall are about AMANa, R., RYOKICID, AzuMA., H., HISASID, Ismi,
0.5% of the wall height. For walls which are Y., and YASUMARU, 1956a. "A Seismic Design
restrained against lateral movement by of Quay Walls in Japan." Proceedings 1st
tiebacks or batter piles or gravity walls cast World Conference on Earthquake Engineer-
on bedrock foundation, lateral pressures in- ing, Berkeley, CA.
duced by inertia forces in the backfill will AMANa, R., AzUMA, H., and ISHII, Y., 1956b. "A
be greater than those given by a Seismic Design of Walls in Japan." Proceed-
Mononobe-Okabe analysis. Simplified elas- ings 1st World Conference on Earthquake En-
tic solutions presented by Wood (1973) for gineering, Berkeley, CA.
rigid nonyielding walls also indicate that ANDERSON, R. D., 1974. "New Method for Deep
pressures are greater than those given by Sand Vibratory Compaction." ASCE Journal
the Mononobe-Okabe method. The use of a of the Construction Division, Vol. 100 (CO-l).
factor of 1.5 in conjunction with peak ground BAIKIE, L. D., 1985. "Total and Partial Factors
accelerations is suggested for design where of Safety in Geotechnical Engineering."
Canadian Geotechnical Journal, Vol. 22.
doubt exists that the wall can yield suffi-
ciently to mobilize soil strengths. More in- BAKEER, R. M., BHATIA, S. K., and ISHIBASID, 1.,
1990. "Dynamic Earth Pressure with Various
formation on the seismic displacement re-
Gravity Wall Movements." ASCE Proceedings
sponse of rigid retaining walls is found in a of Speciality Conference on Design and Per-
study by Siddharthan and Norris (1991). formance of Earth Retaining Structures,
In conclusion, it should be noted that at Cornell University, Ithaca, NY.
present all methods of a dynamic soil pres- BARGHOUTHI, A. F., 1990. "Active Earth Pressure
sure analysis involve considerable uncer- on Walls with Base Projection." ASCE Jour-
tainty. This is primarily due to complexities nal of Geotechnical Engineering, Vol. 116,
of the dynamic interaction problems among No.lO.
the retaining structure, saturated soil in BISHOP, A. W. and HENKEL, D. J., 1962. The
both active and passive zones, and water Measurement of Soil Properties in Triaxial
on the seaward side. The simplified Test, Edward Arnold Ltd., London.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 389

BJERRUM, L., CLAUSEN FIRMANN, C. J., and CARCHEDI, D. R. and MORGAN, R. J., 1990. "Geo-
DUNCUN, J. M., 1972. "Earth Pressures on technical Design Considerations," Design of
Flexible Structures-A State-of-the-Art Re- Marine Facilities for Berthing, Mooring and
port." Proceedings of the Fifth European Con- Repair of Vessels, Gaythwaite, J. K (ed.), Van
ference on Soil Mechanics and Foundation Nostrand Reinhold Co., New York.
Engineering, Vol. 2, Madrid. CARDER, D. R. and KRAwcZVK, J. C., 1975. "Per-
BOSSCHER, P. J. and GRAY, D. H., 1986. "Soil formance of Cells Designed to Measure Soil
Arching in Sandy Slopes. ASCE." Journal of Pressure on Earth Retaining Structures." Re-
Geotechnical Engineering, Vol. 112, No.6. port 689, Transport and Road Research Labo-
BOWLES, J. E., 1970. Engineering Properties of ratory, Crowthorne, Berkshire.
Soils and Their Measurement, McGraw-Hill CASAGRANDE, A., 1964. 2nd Terzaghi Lecture.
Book Co., New York. Role of the "Calculated Risk" in Earthwork
BRANDL, H., 1987. "Retaining Walls and Other and Foundation Engineering, ASCE, New
Retaining Structures," Ground Engineer's York.
Reference Book, Bell, F. G. (ed.), Butterworth, CHOW, Y. K., YONG, D. M., YoNG, KY. and LEE,
Boston. S. L., 1992. "Dynamic Compaction of Loose
BRITISH CoDE OF PRACTICE No. 2, 1951. Earth Sand Deposits," Soils and Foundations, Vol.
Retaining Structures, Institute of Civil Engi- 32, No. 7, December, Japanese Society of Soil
neers, London. Mechanics and Foundation Engineering.
CHANEY, R. C., 1981. "Sampling and Preparation
BROMS, B. B., 1974. "Lateral Earth Pressure Due
of Marine Sediments," Chapter 2 in "Founda-
to Compaction of Cohesionless Soils." Pro-
tion Engineering Handbook," Fang, H.-Y. (ed.).
ceedings, 4th Conference on Soil Mechanics
Van Nostrand Reinhold, New York, NY.
and Foundations Engineering, Budapest.
CHUGH, A. K. and VoNTHUN, J. L., 1985. "Pore
BROMS, B. B., 1991. Deep Compaction of Granu-
Pressure Response Analysis for Earthquakes."
lar Soils, in "Foundation Engineering Hand-
Canadian Geotechnical Journal, Vol. 22.
book, edited by Fang, H.-Y., Van Nostrand
CLOUGH, G. W. and DUNCAN, J. M., 1971. "Finite
Reinhold, New York, NY.
Elements Analysis of Retaining Wall Behav-
BROMS, B. B. and INGLESON, T., 1972. "Lateral ior." ASCE Journal S.M.F.D., Vol. 97 (SM12).
Earth Pressure on a Bridge Abutment." Proc.
CLOUGH, G. W. and DUNCAN, J. M., 1991. Earth
5th European Conference on SMFE (1),
Pressures, Chapter 6 in Foundation Engineer-
Madrid.
ing Handbook, 2nd edition. Fang. H.-Y. (ed.).
BROS, B., 1972. "The Influence of Model Retain- Van Nostrand Reinhold, New York, NY.
ing Wall Displacements on Active and Pas- CLOUGH, G. W., IWABUCHI, J., RAD, N. S., and
sive Earth Pressures in Sand." Proceedings KUPPUSAMY, T., 1989. "Influence of Cementa-
5th European Conference on Soil Mechanics, tion on Liquefaction of Sands," ASCE Jour-
Vol. 1, Madrid. nal of Geotechnical Engineering, Vol. 115,
BROWN, R. E., 1977. Vibroflotation Compaction No.8.
of Cohesionless Soils. ASCE. Journal of CoULOMB, C. A., 1776. "Essai sur une application
the Geotechnical. Engineering Division, 103 des regles de maximis et minimis a
(GT-12). guelques problemes de statique relatifs a
BROWN, P. P., 1948. "A Critical Study of Existing !'architecture." Met. Diu. Savants, Acad. Sci.,
Lateral Earth Pressure Theories, Including Paris, Vol. 7.
the Design of a Model Flexible Anchored CoUPLET, 1726-1728. De la pousse des terres
Bulkhead for the Investigation of These Theo- contre leurs revetments et la fource qu'on leur
ries." Master's thesis, Princeton University. doit opposeur, Histoire de L'Academie Royale
BUDIN, A. Y. and DEMINA, G. A., 1979. Quays de Sciences, Paris.
Handbook, Transport Publisher, Moscow (in COYLE, H. M. and BARTOSKEWITZ, R. E., 1976.
Russian). "Earth Pressures on Precast Panel Retaining
CANADIAN FoUNDATION ENGINEERING MANuAL, Wall." ASCE Journal of the Geotechnical
2nd ed., 1985. Canadian Geotechnical Society. Engineering Division, Vol. 102 (GT5).
390 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

D'APPOLONIA, E., 1990. "Monitored Decisions DUNCAN, J. M., WILLIAMS, G. w., SEHN, A. L.,
(The 24th K. Terzaghi Lecture)." ASCE Jour- and SEED, R. B., 1990. "Estimation of Earth
nal of Geotechnical Engineering, Vol. 116, No. Pressure due to Compaction." ASCE Journal
1. Darwin, G. H., 1883. "On the Horizontal of Geotechnical Engineering, Vol. 117, No. 12.
Thrust of a Mass of Sand," Minutes of the DUNNICLIFF, J., 1988. Geotechnical Instrumenta-
Proceedings of the Institution of Civil Engi- tion for Monitoring Field Performance. John
neers, Vol. LXXI, Paper No. 1904, London. Wiley and Sons, New York.
DARWIN, G. H., 1983. "On the Horizontal Thrust
DURANTE, V. A. and VORONKEVICH, S. D., 1955.
of Mass of Sand," Minutes of the Proceedings
"Studies of Sand Densities for Construction
of the Institution of Civil Engineers, Vol.
LXXI, Paper No. 1904, London. Purposes." Annals of Moscow University, No.
2 (in Russian).
DE ALBA, 1984. "Elasticwave Velocities and Liq-
uefaction Potential." Geotechnical Testing EAU (1990). Recommentations ofthe Committee
Journal, No. 7(2). for Waterfront Structures, 5th edition, Erust
& Sotin, Berlin.
DoBRY, R., 1982. Prediction of Pore Pressure
Buildup and Liquefaction of Sands During ELGAMAL, A-W., ZEGHAL, M., and PARRA, E., 1996.
Earthquakes by the Cyclic Strain Method. Na- "Liquefaction of Reclaimed Island in Kobe,
tional Bureau of Standards. Washington, DC. Japan," ASCE Journal of Geotechnical Engi-
DoBSON, T., 1986. "Vibro Techniques for neering, Vol. 122, No.1.
Geotechnical Engineering Practice." Proceed- ELMS, D. G. and RICHARDS, R., 1990. "Seismic
ings of Pennsylvania DOT and Pennyslvania Design of Retaining Walls." ASCE Proceed-
ASCE. ings Specialty Conference on Design and Per-
DUBROVA, G. A., 1947. Methods of Calculating of formance of Earth Retaining Structures, Cor-
Soil Thrust upon Marine Retaining Struc- nell University, Ithaca, NY.
tures, Maritime Transport Publishing House, EvANs, M. D. and ZHOU, S., 1995. "Liquefaction
Moscow (in Russian). Behavior of Sand-Gravel Composites." ASCE
DUBROVA, G. A., 1963. Soil/Structure Interac- Journal of Geotechnical Engineering, Vol. 121,
tion, Inland Waterway Transport Publishing No.3.
House, Moscow (in Russian). FANG, Y-S. and ISHIBASID, I., 1986. "Static Earth
DUBROVA, G. A., 1968. "Determination ofthe Soil Pressures with Various Wall Movements.
Pressure on Retaining Walls (Unsolved Ques- ASCE Journal of Geotechnical Engineering,
tions)." Gidrotechnicheskoe Stroitelstvo, No. 5 Vol. 112, No. 3.
(translated into English).
FANG, Y-S, CHEN, T-Y., and Wu, B-F., 1994.
DUDLER, I. V., DURANTE, V. A., and SMIRNOV, "Passive Earth Pressures with Various Wall
C. D., 1968. Experience Gained in Using the Movements." ASCE Journal of Geotechnical
Penetrometer Probe for Soil Investigation in Engineering, Vol. 120, No.8.
Conjunction with Energy-Related Construc-
tion in the Soviet Union, INFORM-ENERGO, FANG, Y-S. and CHEN, T-J., 1995. "Modification
Moscow. of Mononobe-Okabe Theory," Geotechnique,
Vol. 45, No 1.
DUNCAN, J. M. and SEED, R. B., 1986. "Compac-
tion-Induced Earth Pressures Under k 0 -con- FocHT, J. A., JR., 1994. "Lessons Learned from
ditions," ASCE Journal of Geotechnical Engi- Missed Predictions." ASCE Journal of
neering, Vol. 112, No. 1. Geotechnical Engineering, Vol. 120, No. 10.
DUNCAN, J. M., CLOUGH, G. W., and EBELING, R. FOCHT, J. A., JR. and O'NEILL, N. W., 1985.
M., 1990. "Behavior and Design of Gravity "International State-of-the-Art Practice for
Earth Retaining Structures." ASCE Proc. of Design and Installation of Axially Loaded
Conference on Design and Performance of Piles." Proceedings 11th International Confer-
Earth Retaining Structures. Cornell Univer- ence on Soil Mechanics and Foundation Engi-
sity, Ithaca, NY. neering, San Fransisco.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 391

FRANKLIN, A G. and CHANG, F. K., 1977. Earth- HoLTZ, W. G. and GIBBS, H. J., 1956. "Shear
quake Resistance of Earth and Rock-Fill Strength of Pervious Gravelly Soils." Proceed-
Dams, Report 5: Permanent Displacements of ings ASCE, Paper No. 867.
Earth Embankments by Newmark Sliding HoUGH, B. K., 1957. Basic Soils Engineering,
Block Analysis. Misc. Paper S-71-17, Soils and Ronald Press Co., New York.
Pavements Laboratory, U.S. Army Engineer IcHIHARA, M., 1983. "Remarks on Mononobe's
Waterways Experiment Station, Vicksburg, Earth Pressure Theory" Tsuchi-to-kiso,
Miss. Japanese Society of Soil Mechanics and Foun-
FuKuoKA, M., 1980. "Static and Dynamic Earth dation Engineering, Vol. 31, No. 11 (in
Pressures on Retaining Walls." Proceedings Japanese).
Third Australia-New Zealand Conference of ISHIBASHI, I., OSADA, M., and UWABE, T., 1994.
Geomechanics. Wellington, New Zealand. "Dynamic Lateral Pressures Due to Satu-
GAY, T. E., 1957. "Stone Crushed and Broken." rated Backfills on Rigid Walls." ASCE Jour-
California Division of Mines Bulletin 176, nal of Geotechnical Engineering, Vol. 120, No.
Sacramento. 10.
ITo, T. and MATsm, T., 1977. "Method to Esti-
GERBER, E., 1929. "Untersuchen uber die Druck-
mate Lateral Force Arching on Stabilizing
vertcilung in orflich Belasteten Sand." Disser-
Piles." Soils and Foundations (Japan), Vol.
tation, Technisha, Hochsahule, Zurich.
18, No.2.
GoULD, J. P., 1970. "Lateral Pressures on Rigid IVANOV, P. L., 1967. Compaction of Noncohesive
Permanent Structures." ASCE Proceedings Soils by Explosions, Stroyisdat Publishing
Specialty Conference on Lateral Stresses in House, Leningrad (in Russian): translated
the Ground and Design of Earth-Retaining into English by Indian National Scientific
Structures, Cornell University, Ithaca, NY. Documentation Centre, New Delhi.
GRANGER, V. L., 1965. "Failure of a Reinforced JAKY, J., 1944. "The Coefficient of Earth Pres-
Concrete Reservoir." Proceedings 6th Interna- sure at Rest." Journal of the Society of Hun-
tional Conference SMFE (2). Montreal. garian Architects and Engineers, Budapest.
HARDER, L. F., HAMMOND, W. D., and Ross, P. S., JAMES, H. W., 1973. "Densification of Sand for
1984. "Vibroflotation Compaction of Thermal- Drydock by Terra-Probe." ASCE e1ournal of
ito Afterbay." ASCE Journal of Geotechnical Soil Mechanics and Foundation Division, Vol.
Engineering, Vol. 110, No. 1. 99 (SM-6).
HARDIN, B. 0. and DRUEVICH, V. P., 1972. "Shear JAMES, R. G. and BRANSBY, P. L., 1970. "Experi-
Modulus and Damping in Soils: Design Equa- mental and Theoretical Investigations of Pas-
sive Pressure Problems." Geotechnique, Vol.
tions and Curves." ASCE Journal Soil Me-
20, No.1.
chanics and Foundation Division, SM7.
JAMIOLKOWSKI, M., 1988. "Research Applied to
HAYASID, S., KUBO, K., and NAKASC, A, 1966. Geotechnical Engineering," Proceedings Insti-
"Damage to Harbor Structures in the Niigate tute of Civil Engineers, Part 1, June.
Earthquake." Soil and Foundations, Vol. VI,
JURAN, I. and CHRISTOPHER, B., 1989. "Labora-
No.1.
tory Model Study on Geosynthetic Reinforced
HILF, J. W., 1991. Compacted Fill in "Founda- Soil Retaining Walls," ASCE Journal of
tion Engineering Handbook," edited by Fang, Geotechnical Engineering, Vol. 115, No 7.
H-Y. Van Nostrand Reinhold, New York, N.Y. KANY, M., 1972. "Measurement of Earth Pres-
Ho, C. L., DENBY, G. M., and FRAGASzy, R. J., sures on Cylinder 30 m in Diameter Pump
1990. "Seismic Performance of Tied-Back Storage Plant." Proc. Fifth European Confer-
Walls." ASCE Proceedings Specialty Confer- ence on Soil Mechanics, Madrid.
ence on Design and Performance of Earth KEMPTON, G. T., ENTWISTLE, R. W., and BARCLAY,
Retaining Structures, Cornell University, M. J., 1985. "An Anchored Fill Harbour Wall
Ithaca, NY. using Synthetic Fabrics," Proceedings Institu-
392 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

tion of Civil Engineers, Part 1, Vol. 78, MACKEY, R. D. and KIRK, D. P., 1967. "At Rest,
London. Active and Passive Earth Pressures." Pro-
KEzm, A., 1975. "Lateral Earth Pressure," Foun- ceedings South East Asian Conference on Soil
dation Engineering Handbook, Winterkorn, H. Mechanics and Foundation Engineering,
F. and Fang, H-Y. (eds.), Van Nostrand Bangkok.
Reinhold Co., New York. MARTIN, G. R., FINN, W. D. L., AND SEED, H. B.,
1975. "Fundamentals of Liquefaction Under
KLEIN, G. K. and CHERKASOV, J. J., 1985. Foun-
Cyclic Loading." ASCE Journal of Geotechni-
dations for Urban Transport Structures,
cal Engineering, Vol. 101, No.5.
Transport Publishing House, Moscow (in
Russian). MATSUO, H. and OHARA, S., 1965. "Dynamic Core
Pressure Acting on Quay Wall During Earth-
KosTILYOVA, N. B., 1956. In text "Mechanics of quake." Proceedings III World Conference on
Granular Material," by Klein, G. K., State Earthquake Engineering. Auckland, New
Publishing House of Literature in Civil Zealand.
Engineering and Architecture, Moscow (in
Russian). MATSUO, M., KENMOCHI, S., and YAGI, H., 1978.
"Experimental Study on Earth Pressure of
KOVACS, W. D., EVANS, J. C., and GRIFFITHS, A. Retaining Wall by Field Tests." Soils and
H., 1977. "Towards a More Standardized Foundations, Vol. 18, No.3.
SPT." Proceedings of the Ninth International
MATSUZAWA, H. ISHIBASHI, 1., and KAWAMURA, M.,
Conference on Soil Mechanics and Founda-
1985. "Dynamic soil and water pressures of
tion Engineering, Tokyo. submerged soils," ASCE Journal of Geotech-
LABA, J. T. and KENNEDY, J. B., 1986. "Rein- nical Engineering Vol. 111, No. 10.
forced Earth Retaining Wall Analysis and De- MATTEOTTI, G., 1970. "Some Results of Quay-
sign," Canadian Geotechnical Journal, Vol. Wall Model Tests on Earth Pressure." Pro-
23. ceedings Institution of Civil Engineers,
LAnD, R. S., 1977. "Specimen Preparation and London.
Cyclic Stability of Sands." ASCE Journal of MEYERHOF, G. G., 1970. "Safety Factors in Soil
Geotechnical Engineering; Vol. 103, No.6. Mechanics." Canadian Geotechnical Journal,
LAMBE, T. W., 1951. Soil Testing for Engineers, No.7.
John Wiley and Sons, New York. MEYERHOF, G. G., 1984. "Safety Factors and
LAMBE, T. W. and WHITMAN, R. V., 1969. Soil Limit States Analysis in Geotechnical Engi-
Mechanics, John Wiley and Sons, New York. neering." Canadian Geotechnical Journal,
Vol. 21.
LEE, D. H., 1961. Deep Foundations and Sheet
Piling, Concrete Publications Ltd., London. MITCHELL, J. K., 1976. Fundamentals of Soil
Behaviour, John Wiley and Sons, New York.
LEONARDS, G. A., 1962. Engineering Properties of
MITCHELL, J. K., 1986. "Practical Problems from
Soils, McGraw-Hill Book Co., New York.
Surprising Soil Behavior." ASCE Journal of
LESHCHIUSKY, D. and BOEDEKER, R. H., 1986. Geotechnical Engineering, Vol. 112, No.3.
"Geosynthetic Reinforced Soil Structures," MITCHELL, J. K. and SOLYMAR, Z. V., 1984. "Time
ASCE Journal of Geotechnical Engineering, Dependent Strength Gain in Freshly De-
Vol. 115, No. 10. posited or Densified Sand." ASCE Journal of
LIAO, S. S. C., VENESIANO, D., and WHITMAN, R. Geotechnical Engineering, Vol. 110, No. GT11.
V., 1988. "Regression Models for Evaluating MoHRAZ, B., 1976. "A study of earthquake re-
Liquefaction Probability," ASCE Journal of sponse spectra for different geological condi-
Geotechnical Engineering, Vol. 114, No.4. tions," Bulletin of the Seismological Society of
LOWE, J. and ZACCHCO, P. F., 1991. "Subsurface America, Vol. 66.
Explorations and Sampling," Foundation En- MONONOBE, N., 1929. "Earthquake-Proof Con-
gineering Handbook, 2nd ed., Fang, H-Y. (ed.), struction of Masonry Dams." Proceedings
Van Nostrand Reinhold, New York. World Engineering Conference, Tokyo, Vol. 9.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 393

MONONOBE, N. and MATSUO, H., 1929. "On De- Boston Society of Civil Engineers Section, Vol.
termination of Earth Pressure During Earth- 65, No.2.
quakes." Proceedings World Engineering PILECKI, T. J., 1987. "Discussion on Interface
Congress, Tokyo, Vol. 9. Model for Dynamic Soil-Structure Interac-
MULILIS, J. P., SEED, H. B., CHAN, C. H., tion." ASCE Journal of Geotechnical Engi-
MITCHELL, J. K., and ARULANANDAN, K., 1977. neering, Vol. 113, No.5.
"Effects of Sample Preparation on Sand PITILAKIS, K. and MONTSAKIS, A., 1989. "Seismic
Liquefaction." ASCE Journal of Geotechnical Analysis and Behaviour of Gravity Retaining
Engineering, Vol. 103. Walls-The Case of Kalamata Harbour Quay
NACCI, V., 1969. "Shear Strength Properties of Wall." Soil and Foundations, Vol. 29, No. 1.
Varved Silt." Division of Engineering Re- POLSHIN, D. E. and SINELSCHIKOV, S. J., 1953.
search and Development, Department of Civil "About the Angle of Internal Friction." Pro-
Engineering, Engineering Bulletin No. 11, ceedings of the Research Institute of Soil
University of Rhode Island, Kingston. Mechanics and Foundations, Moscow (in
Russian).
NARAIN, J., SARAN, S., AND NANDAKUMARAN, P.,
PoULos, H. G., 1987. "From Theory to Practice
1969. "Model Study of Passive Pressure in
in Pile Design." Australian Geomechanics, No.
sand." ASCE Journal of Soil Mechanics and
14.
Foundation Engineering Division, Vol. 95,
No.4. PoULos, H. G. and HULL, T. S., 1989. "The Role
of Analytical Geomechanics in Foundation
NEWMARK, N. M. and HALL, W. J., 1982. Earth- Engineering." ASCE Proceedings of Congress
quake Spectra and Design, Earthquake Engi- on Foundation Engineering. Current Princi-
neering Research Institute, Berkeley, Ca. ples and Practices, Vol. 1, Evanston, IL.
OKABE, S., 1926. "General Theory on Earth Pres- PRAKAsH, S., 1981. Soil Dynamics, McGraw-Hill,
sure." Journal Japanese Society of Civil Engi- New York.
neering, Vol. 12, No. 1. PRAKASH, S. and BASVANNA, B. M., 1969. "Earth
PAPPIN, J. W., SIMPSON, B., FELTON, P. J., and Pressure Distribution Behind Retaining Walls
RAISON, C., 1986. Proceedings of Midland During Earthquakes," Proceedings Fourth
Geotechnical Society. Computer Applications World Conference on Earthquake Engineer-
in Geotechnical Engineering, Birmingham, ing, Vol. 3, Santiago, Chile.
U.K. RANKINE, W. J. M., 1857. "On the Stability of
PAPPIN, J. W., BROWN S. F., and O'REILLY, Loose Earth." Transactions of the Royal Soci-
M. P., 1992. "Effective Stress Behaviour of ety, Vol. 147.
Saturated and Partly Saturated Granular REHNMAN, S. E. and BROMS, B. B., 1972. "Lateral
Material Subjected to Repeated Loading." Pressures on Basement Wall. Results from
Geotechnique, Vol. 42, No.3. Full-Scale Tests." Proceedings 5th European
Conference on Soil Mechanics, Madrid.
PECK, R. B., 1969. "Advantages and Limitations
of Observational Method in Applied Soil Me- RICHARDS, R. and ELMS, D. G., 1979. "Seismic
chanics (9th Rankline Lecture)." Geotech- Behavior of Gravity Retaining Walls." ASCE
nique, Vol. 19, No. 2. Journal of Geotechnical Engineering, Vol. 105,
No.4.
PECK, R. B. and MESRI, G., 1987. "Discussion of RocKER, J., 1985. Handbook for Marine Geotech-
Compaction-Induced Earth Pressures Under nical Engineering, U.S. Navy Civil Engineer-
K 0 Conditions." ASCE Journal of Geotechni- ing Laboratory, Port Hueneme, CA.
cal Engineering, Vol. 113, No. 11.
RoWE, P. W., 1954. "A Stress-Strain Theory for
PECK, R. B., HANSON, W. E., and THORNBURN, T. Cohesionless Soil with Applications to Earth
H., 1974. Foundation Engineering, 2nd ed., Pressures at Rest and Moving Walls."
John Wiley and Sons, New York. Geotechniques, Vol. 4, No.2.
PIERCE, F. C. and CALABRETTA, V., 1978. "Unique RoWE. P. W. and PEAKER, K., 1965. "Passive
Challenges Associated with Marine Struc- Earth Pressure Measurements." Geotech-
tures in New England." ASCE Journal of the nique, Vol. 15, No. 1.
394 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

SANDEGREN, E., SAHLSTROM, D. D., AND STILLE, SEED, H. B. and WHITMAN, R. V., 1970. "Design
H., 1972. "Behavior of Anchored Sheet Pile of Earth Retaining Structures for Dynamic
Wall Exposed to Frost Action." Proceedings Loads." Proceedings ASCE Specialty Confer-
5th European Conference on Soil Mechanics, ence on Lateral Stresses in the Ground and
Madrid. Design of Earth Retaining Structures, Ithaca,
SARAN, S., GARG, K. G., and BHANDARI, R. K., NY.
1992. "Retaining Wall with Reinforced Cohe- SEED, H. B., IDRISS, I. M., and ARANGo, I., 1983.
sionless Backfill," ASCE Journal of Geochem- "Evaluation of Liquefaction Potential using
ical Engineering, Vol. 118, No. 12. Field Performance Data." ASCE Journal of
SAXENA, S. K. and REDDY, R. K., 1987. "Mechani- Geotechnical Engineering Division, No. GT3.
cal Behavior of Cemented Sands." Report No. SHERIF, M.A. and FANG, Y. S., 1984. "Dynamic
IIT-CE-8701. Department of Civil Engineer- Earth Pressures on Walls Rotating about the
ing, Illinois Institute of Technology, Chicago. Top." Soils and Foundations, Vol. 24, No.4.
SAXENA, S. K., REDDY, K. R., and AVRAMIDIS, A. SHERIF, M. A., ISHIBASHI, I., and LEE, C. D.,
S., 1988. "Liquefaction Resistance of Artifi- 1982. "Earth Pressure Against Rigid Retain-
cially Cemented Sand." ASCE Journal of ing Walls." ASCE Journal of Geotechnical
Geotechnical Engineering, Vol. 114, No. 12. Engineering Division, Vol. 108, No. GT5.
SCHNABEL, P. B. and LYSMER, J., 1972. "SHAKE SHERIF, M. M. and MACKEY, R. D., 1977. "Pres-
-A Computer Program for Earthquake Re- sures on Retaining Wall with Repeated Load-
sponse Analysis of Horizontally Layered ing." ASCE Journal Geotechnical Engineer-
Sites." Report No. EERC 72-12, Earthquake ing Division, Vol. 103, No. 11.
Engineering Research Institute, University of SHULMAN, S. G., 1987. Seismic Pressure of
California, Berkeley. Water on Hydraulic Structures, AA Balkuna,
SCHOFIELD, A. N., 1961. "The Development of Rotterdam.
Lateral Force of Sand Against the Vertical SIDDHARTHAN, R. and NORRIS, G. M., 1991.
Face of a Rotating Model Foundation." Pro- Japanese Society for Soil Mechanics and
ceedings 5th Intemational Conference on Soil Foundation Engineering, Soils and Founda-
Mechanics and Foundation Engineering, tions, Vol. 31, No. 2.
Paris, Vol. 2. SMIRNOV, G. N., GORUNOV, B. F., KURLOVICH, E.
ScoTT, R. F., 1987. "Failure. The Rankine Lec- B., LEVACHOV, S. N., and SJDOROVA, A. G.,
ture." Geotechnique, Vol. 37, No.4. 1979. Ports and Port Related. Structures,
SEED, H. B., 1987. "Design Problems in Soil Stroyisdat Publishing House, Moscow (in
Liquefaction." ASCE Journal of Geotechnical Russian).
Engineering, Vol. 113, No.8. ., SMITH, I. M., 1988. "Geotechnical Aspects of the
SEED, H. B. and DUNCAN, J. M., 1985. "FE Use of Computers in Engineering: A Personal
Analyses: Compaction-Induced Stresses and View." Proceedings Institute of Civil Engi-
Deformations." Journal of Geotechnical Engi- neers, Part 1, London.
neering, Vol. 112, No. 1. SNIP II-U.10-65, 1966. Retaining Walls for
SEED, H. B. and IDRISS, I. M., 1971. "Simplified Marine Construction, Gosstroy, Moscow, (in
Procedure for Evaluating Soil Liquefaction Russian).
Potential." ASCE Journal of Soil Mechanics SOKOLOWSKI, V. V., 1965. Statics of Granular
and Foundation Division, Vol. 97, No. SM9. Media, Pergamon Press, London (translated
SEED, H. B. and IDRISS, I. M., 1982. Ground from Russian).
Motions and Soil Liquefaction During Earth- SOLYMAR, Z. V., 1984. "Compaction of Alluvial
quakes, Earthquake Engineering Research Sands by Deep Blasting." Canadian Geotech-
Institute, Berkeley, CA. nical Journal, Vol. 21, No. 2.
SEED, H. B. and LEE, K. T., 1966. "Liquefaction SOLYMAR, Z. V. and REED, D. J., 1986. "A Com-
of Saturated Sands During Cyclic Loading." parison of Foundation Compaction Tech-
ASCE Journal of Soil Mechanics and Foun- niques." Canadian Geotechnical Journal, Vol.
dation Division, Vol. 94, No. SM2. 23, No.3.
Geotechnical Aspects of Soil-Structure Interaction Design Considerations 395

SOLYMAR, Z. v., ILOABACHIE, B. C., GUPI'A, R. c., quake Shaking." ASCE Journal of Geotechni-
and WILLIAMS, L. R., 1984. "Earth Foundation cal Engineering, Vol. 113, No. 8.
Treatment at Jebba Dam Site." ASCE TOKIMATSU, K., YAMAZAKI, T., and YosHIMI, Y.,
Journal of Geotechnical Engineering, Vol.llO, 1986. "Soil Liquefaction Evaluations by Elas-
No. GT10. tic Shear Moduli." Soils and Foundations, No.
SONU, C. J., ITo, K., and OISHI, H., 1993. "Harry 26(1).
Seed, Liquefaction and the Gravel Drain." TOWNSEND, F. C., 1977. "Review of Factors
ASCE Civil Engineering, December. Affecting Cyclic Triaxial Tests." ASTM Pro-
SOWERS, G. F., ROBB, A. D., MULLIS, C. H., and ceedings Symposium Dynamic Geotechnical
GLENN, A. J., 1957. "The Residual Lateral Testing.
Pressures Produced by Compacting Soils." TSCHEBOTARIOFF, G. P., 1973. Foundations, Re-
Proceedings, 4th International Conference on taining and Earth Structures. McGraw-Hill
Soil Mechanics and Foundation Engineering, Book Company, New York.
London.
TSINKER, G. P., 1972. "Rational Design of Anchor
SPRANGLER, M. G., 1938. "Horizontal Pressures
Piles with Widenings at the Ends for An-
on Retaining Walls Due to Concentrated Sur-
chored Sheet-Pile Bulkheads." Thesis pre-
face Loads." Iowa Engineering Experimental
sented to Odessa Institute of Civil Engineers,
Station, Bulletin No. 140.
at Odessa, USSR in partial fulfillment of the
STEEDMAN, R. S. and ZENG, X., 1990. "The Seis- requirements for the degree of Doctor of
mic Response of Waterfront Retaining Walls." Philosophy (in Russian).
ASCE Proceedings of Specialty Conference on
TSINKER, G. P., 1983. "Anchored Sheet Pile Bulk-
Design and Performance of Earth Retaining
heads: Design Practice." ASCE Journal of
Structures, Cornell University, Ithaca, NY.
Geotechnical Engineering, Vol. 109, No. GT8.
STOCOE, K. H. and NAZARIAN, S., 1985. "Use of
Rayleigh Waves in Liquefaction Studies." TSINKER, G. P., 1988. "Pile Jetting." ASCE Jour-
Measurement and Use of Shear Wave Velocity nal of Geotechnical Engineering, Vol. 114, No.
for Evaluating Dynamic Soil Properties, GT3.
Woods, R. D. (ed.), ASCE, Denver, CO. TSINKER, G. P., 1995. Marine Structures Engi-
SYMONS, I. F. and WILSON, D. S., 1972. "Meas- neering: Specialized Applications, Chapman
urement of Earth Pressures in Pulverized & Hall, New York.
Fuel Ash Behind a Rigid Retaining Wall." U.S. ARMY CORPS OF ENGINEERS, 1961. Retaining
Proc. Fifth European Conf. on Soil Mechanics Walls, Engineering and Design Manual,
(1), Madrid. EM1110-2-2502, Department of the Army,
TAYLOR, D. W., 1948. Fundamentals of Soil Me- Washington, DC.
chanics, John Wiley and Sons, New York. U.S. ARMY CORPS OF ENGINEERS, 1970. Labora-
TERZAGHI, K., 1932. "Record Earth Pressure tory Soil Testing, EMll0-2-1906, Department
Testing Machine." Engineering News Record, of the Army, Washington, DC.
September 29 (109). VAID, Y. P. and THoMAS, J., 1995. "Liquefaction
TERZAGHI, K., 1934. "Large Retaining Wall Test." and Post-Liquefaction Behavior of Sand."
Engineering News Record, Feb. 1, 22; March ASCE Journal of Geotechnical Engineering,
8, 20; April 19. Vol. 121, No. GT2.
TERZAGHI, K., 1943. Theoretical Soil Mechanics, VAID, Y. P., CHERN, J. C., and Tu~n, H., 1985.
John Wiley and Sons, New York. "Confining Pressure, Grain Angularity, and
TERZAGHI, K., 1948. "Foreword." Geotechnique, Liquefaction." ASCE Journal of Geotechnical
Vol. 1, No. 1. Engineering, Vol. 111, No. GT10.
TERZAGHI, K., 1951. Lecture on Soil Mechanics WANG, W. L. and YEN, B. C., 1974. "Soil Arching
and Foundation Engineering. Given at the in Slopes." ASCE Journal of the Geotechnical
Building Research Congress, London. Division, Vol. 100, No. GT1.
TOKIMATSU, K. and SEED, H. B., 1986. "Evalua- WEILER, W. A., JR. and KULHAWY, F. H., 1978.
tion of Settlements in Sands Due to Earth- "Behavior of Stress Cells in Soil." Geotechni-
396 Geotechnical Aspects of Soil-Structure Interaction Design Considerations

cal Engineering Report 78-2, Niagara Mo- Specialty Conference on Design and Perfor-
hawk Power Corp., Syracuse University, mance of Earth Retaining Structures, Cornell
Ithaca, NY. University, Ithaca, NY.
WEILER, W. A., JR. and KULHAWY, F. H., 1979. WINTERKORN, H. F. and FANG, H-Y., 1975. Foun-
"Factors Mfecting Stress Cell Measurements dation Engineering Handbook, Van Nostrand
in Soil. ASCE Conf., Atlanta. Reinhold Co., New York.
WERNER, P. W. and SUNDQUIST, K. J., 1943. "On
WooD, J. H., 1973. "Earthquake-Induced Soil
Hydrodynamic earthquake effects," Transac-
Pressures on Structures." Report No. EERL
tions American Geophysical Union, Vol. 30.
73-05, Earthquake Engineering Research
WESTERGAARD, H. M., 1933. "Water pressures on Lab., California Institute of Technology,
dams during earthquake." Transactions of the Pasadena.
ASCE, Vol. 98.
Wu, T. H., 1966. Soil Mechanics, Allyn and
WHITENECK, L. L. and HOCKNEY, L. A., 1989.
Bacon, Boston.
Structural Materials for Harbor and Coastal
Engineering, McGraw-Hill Book Company, YoUNG, A. G., 1991. "Marine Foundations Stud-
New York. ies." Handbook of Coastal and Ocean Engi-
WHITMAN, R. V., 1984. "Evaluating Calculated neering, Vol. 2. Herbich, J. B. (ed.), Gulf Pub-
Risk in Geotechnical Engineering." ASCE lishing Company, Houston, TX.
Journal of Geotechnical Engineering Division, ZANGAR, C. N., 1953. "Hydrodynamic pressures
Vol. 110, No. GT2. on dams due to horizontal earthquakes," Pro-
WHITMAN, R. V., 1990. "Seismic Design of Grav- ceedings of Experimental Stress Analysis, Vol.
ity Retaining Walls." ASCE Proceedings 10.
5
Gravity-Type Quay Walls

5.1 GENERAL In general, because of their heavy weight


and the character of the load distribution at
By definition, gravity-type structures are the base, the gravity-type structures re-
those which rely primarily on their weight quire reasonable foundation conditions.
and grip on the foundations to resist any of In the past century, the gravity-quay
the possible adverse load combinations. walls were generally made of rubble ma-
Gravity-type quay walls may be used at sonry (Fig. 5-la) or cast-in-situ concrete
wharves receiving ships of any size and (Fig. 5-lb) built in the dry, provided that
type; from small general cargo vessels to the site could be dewatered and a suffi-
the largest contemporary container ships, ciently strong foundation was available. In
and very large bulk carriers and super- some cases where relatively weak founda-
tankers. These walls are particularly useful tion soils were present, heavy gravity walls
and durable under severe marine environ- built in dry were founded on piles (Fig.
mental conditions, such as salt water, hot 5-2). Because wood completely submerged
and cold temperatures, large waves, and in water does not usually rot, the pile foun-
heavy ice loads. dation was typically made from wooden
A great variety of gravity-type quay walls piles. The top of these piles was usually
has been developed in the past. Typically, embedded in concrete and located below the
the type of walls used for the marine appli- mud line, thus providing protection from
cation was influenced by specific site condi- attack by marine organisms.
tions, such as availability of certain con- Invention of reinforced concrete and im-
structions materials, dry dock facilities, cost provements in concrete technology was fol-
oflocallabor, and others. In the past and at lowed by increased refinements in quay wall
present, in a great many cases gravity-type constructions. Slender cast-in-place L-
structures are used where local founda- shaped walls and the counterfort walls have
tion conditions preclude pile or sheet-pile been introduced (Fig. 5-3). However, con-
driving. struction of the high quay walls in dry at

397
398 Gravity-Type Quay Walls

(a) (b)

Figure 5-l. Dock wall at Greenoch, UK (1980). (a) Portion built of rubble masonry;
(b) portion built of concrete with rubble masonry superstructure.

- ·--., -~ - - ·--

Figure 5-2. Mass concrete wall at Limerick, U.K built on piled


foundation.
Gravity-Type Quay Walls 399

Figure 5-4. Port-of Bongie, Algeria (1940). Typical


cross section through quay wall.

inexpensive locally available rock or coarse


Figure 5-3. Counterfort quay wall at Schepen-Siffer
aggregates. Usually, the blocks for these
Dock, Gent, Belgium.
quays were made from rock obtained from a
nearby quarry and shaped to the required
sites which required cofferdamming was too sizes. Later developments utilized precast
costly and often cost prohibitive. concrete blocks, whenever the necessary
The logical successor of the above de- suitable aggregates and cement were avail-
signs was the construction of quay walls in able and where natural stone blocks were
the wet by employing large masonry and, too costly.
later, massive concrete blocks (Figs. 5-4 The history of the blockwork-type con-
and 5-5) or, at some earlier stages, use of struction is traced back to ports in the
sacks filled with concrete (Fig. 5-6). This Mediterranean, at Marseilles and Algiers,
was not generally a cheap form of construc- with much of this early pioneering work
tion for high gravity walls. However, in some being carried out at the Port-of-Bouqie,
instances it has been justified by the magni- Algeria (Fig. 5-4) where the quay wall com-
tude of the work and by the availability of posed of limestone blocks have been built as

(a) (b)

Figure 5-5. Blockwork quay wall at Mombasa, Kenya (1934). Typical cross sections: (a) wall
built on competent foundation; (b) wall built on weak foundation soil.
400 Gravity-Type Quay Walls

tical cases in the past which is also true


presently, it has been found that built else-
where, then launched and floated in and
installed on prepared foundation, large con-
crete caissons (Fig. 5-7) often represent a
more economical solution to the gravity-type
wall construction than the blockwork
structures.
In modern marine engineering practice
for gravity-type quay wall construction a
number of innovative and economical struc-
tures have been employed. Among them are
large-diameter floated-in or bottomless con-
Figure 5-6. Cross section through pier in Klaksvik, crete cylinders, prefabricated L-shaped
Denmark. !-Concrete in sacks; 2-concrete super-
structure; 3-rubble.
walls of miscellaneous constructions, as well
as others. Some of these structures are dis-
cussed further in this chapter. An economi-
far back as 1840. This wall consisted of five cal solution to the gravity wall construction
tiers of masonry blocks of varying size. in the wet is also achieved by use of
Those in the two lowermost tiers weighed the traditional structures such as large-
about 32 tonnes each. The upper tiers aver- diameter steel sheet-pile cells or timber
aged about 5 tonnes less. When the four cribwork walls with concrete superstruc-
submerged courses had been constructed, ture. The latter is usually used where wood
the wall was loaded with a temporary heavy is locally available.
surcharge, which caused the structure to In general, the gravity-type quay wall
settle to about 1.0 m in a period of 2 months. built in the wet consists of three basic
Then the surcharge was removed and a structural components: a rubble mattress,
coping course installed. an underwater structure, and the above
The wall depicted in Figure 5-5 was built water superstructure.
at Port-of-Mombasa, Kenya in 1934. It was As previously stated, in most practical
built on foundation strata generally consist- cases of gravity quay wall constructions in
ing of hard sand andjor soft sandstone rock the wet, they are placed on rubble mat-
(Fig. 5-5a). At certain locations where foun- tresses. The latter must be solidified either
dation was not good, the wall was placed on by preloading or by vibration. Otherwise,
piles, as indicated in Figure 5-5b. In the despite the presence of a firm foundation
latter case, none of the foundation piles soil material, a considerable settlement may
were connected to the wall. Driving of these be expected.
piles has consolidated the original founda- To achieve better distribution of bearing
tion strata which happened to be too soft. stresses produced by a gravity-type struc-
The blocks were placed on compacted sand- ture, the top of a rubble mattress is usually
fill with bearing piles beneath the wall's covered with a 30-cm-thick layer of thor-
base. It should be noted that improved and oughly leveled gravel or crushed stone. The
more economical blockwork quay walls are front portion of the mattress extending be-
still in use. yond the berthing line is usually protected
Despite the merits of the blockwork walls in one way or the other from the scour
in general, they are very costly and require induced by ship propellers or currents.
heavy lift (sometimes floating) equipment Separation of the quay structure into an
for their construction. In a great many prac- underwater portion and the superstructure
Gravity-Type Quay Walls 401

(a) (b)

Figure 5-7. Quay wall at Port-of-Gdynia, Poland. (a) Typical cross section; (b) method of
launching caissons.

is dictated basically by economies and con- below the design minimum water level. As
struction methods used, as well as by the stated earlier, the type of the underwater
durability requirements applied to both portion of a gravity quay wall is dependent
parts of the structure. on local geotechnical, environmental, and
The superstructure is usually built in dry operational conditions.
conditions and is also readily accessible for The performance of the gravity-type wall
inspection, maintenance, and repair. On the depends heavily on its foundation. Usually
other hand, the underwater portion of the a certain amount of wall movement is ex-
structure can be inspected by divers or with pected. The magnitude of gravity wall
the help of special equipment. Repair of the movements is basically a function of the
underwater portion is usually a quite dif- type of wall, the quality of the foundation
ficult and costly undertaking. This is basi- and backfill materials, as well as the nature
cally the reason why in a great many cases and sequence of application of miscella-
different structural materials are used for neous loads. Furthermore, in the case of
construction of the underwater portion of concrete and concrete masonry structures,
the structure and its superstructure. As the the wall or its elements can shrink or swell
cost of the under-water portion is usually upon loss or gain of moisture; temperature
high, it is customary to limit its height and variations may also cause the structure to
place it at about 0.5 m above the assumed expand and contract in daily or seasonal
construction water level. In case of timber cycles. Finally, a poorly densified rubble
cribworks, the underwater portion of the mattress of substantial depth may also be a
structure is usually placed at about 0.5 m contributing factor to uneven wall move-
402 Gravity-Type Quay Walls

ments in both vertical (settlement) and hor- through a pipeline. Hydraulic fills placed
izontal (displacement) directions. behind walls or bulkheads are placed in
To prevent wall cracking andjor exces- lifts thin enough to permit wash water to
sive overstress due to uneven movements it run off without building up a full height of
is usually provided with controlling expan- hydrostatic pressure.
sion joints installed at suitable intervals. Dredging and placing operations usually
No exact rules can be stated for the location produce significant textural differences be-
of these joints. Each project is site-specific tween original bottom materials and mate-
and must be studied individually for where rials deposited at the fill site. In general,
joints can be placed without compromising there will be an increase in the mean grain
the structural integrity of the wall. How- size as fines are lost and there is decrease
ever, it has been demonstrated in practice in the uniformity coefficient of delivered
that joints in gravity-type walls between versus bottom granular materials. The vol-
adjacent sections can be installed at 20-40- umetric losses resulting from winnowing
m intervals. Smaller intervals are usually associated with this process typically are of
used in the case of relatively soft founda- the order of 10% or more, depending on the
tions such as sand, firm clay, and the like; original bottom material. For detailed infor-
larger intervals are used in the case of firm mation on this subject, the reader is re-
foundations such as different kinds of sound ferred to Whiteneck and Hackney (1989).
rocks. The backfill placement would normally
The width of expansion joint should be proceed landward in order to push bottom
such as to prevent any interaction between mud away from the wall. For detailed dis-
adjacent sections. However, it should not be cussion on this subject the reader is re-
smaller than 15 mm and is usually de- ferred to Chapter 4.
signed to be 20-25 mm. Sometimes it is In cases when soil for backfilling is ob-
smaller at the wall base and wider at the tained from land-based sources and brought
top to allow for some rotation of the adja- to the fill site in trucks, the backfilling oper-
cent wall sections. ation most likely would proceed in a sea-
In the case of granular backfills, the joints ward direction. In this case, to prevent
have to be covered by a filter system to concentration of the bottom mud in large
prevent fines in backfill material from being quantities behind the wall, it should be
washed out. Backfill material can be locally removed before start of backfilling, or alter-
available hydraulically placed dredged natively the structure must be designed for
granular soil or granular soil obtained from the soil pressure with mud geotechnical
land sources typically transported by dump parameters.
trucks or conveyor belts. Soils obtained from Care should also be taken to place select
the harbor basin are normally dredged by fill behind the wall before placing the gen-
dredgers of miscellaneous constructions, or eral fill to ensure that the active pressure
clam shells. It can be transported hydrauli- zone has the shear strength planned for in
cally if in close proximity to the project the design.
location or in barges if the source of fill To prevent excessive settlement, the
material is located far away. In the latter freshly placed granular fill should be densi-
case, it is usually unloaded by clam shells fied to the extent specified in a wall design.
and dumped behind the wall. Methods of granular soil densification and
Typically, hydraulic fills are placed on the resulted effects of soil densification are
land or underwater by pumping material discussed in Chapter 4.
Gravity-Type Quay Walls 403

5.2 BASIC STRUCTURAL signers where local conditions are suitable


ARRANGEMENTS and where walls of blockwork construction
represents an economical alternative solu-
tion to the problem. It is generally believed
With a constant process in improvement
that these walls are relatively immune to
and development of new structural materi-
the various forms of serious deterioration
als, further mechanization of construction
affecting the more sophisticated thin-walled
processes, as well as use of the computer-
concrete structures such as alkali-aggregate
ized design methods, it became possible to
reaction, carbonation, and chloride penetra-
greatly improve the traditional type of grav-
tion. They can also be dismantled relatively
ity quay walls and to develop some . new
cheaply if they should ever become redun-
structural designs. The basic trend in mod-
dant or obsolete.
ern marine engineering practice concerned
Blockwork walls are typically built on
with the design of any kind of structures in
competent foundation soils, or soils whose
general, and gravity-type walls in particu-
bearing capacity can be enhanced if re-
lar, is prefabrication.
quired to meet design requirements.
The prefabrication of structural compo-
The success of blockwork walls in the
nents under plant (shipyard) conditions
past and their continued use at present is
greatly enhances their quality. Further-
based on certain advantages, which may be
more, use of efficient heavy lift equipment
stated as follows:
allows prefabrication and use of large struc-
tural elements which significantly reduces
1. Excellent durability and reliability at-
period of time of above water construction. tributed to the robust nature of the ma-
The following basic types of gravity quay rine concrete blocks
wall (also piers and dolphins) are normally
2. Relatively simple construction technique
considered in modern marine engineering required
practice:
3. Use of basically readily available material
4. Good quality control achieved by the re-
1. Concrete blockwork structures
production process of manufacturing pre-
2. Structures composed of floated-in con- cast concrete blocks
crete caissons
5. Good response to major accidental impact
3. Structures composed of large-diameter by vessels
concrete and steel cylinders
4. Steel sheet-pile cellular bulkheads They may also be adaptable to a variety of
5. Prefabricated concrete L-shaped walls foundation conditions.
6. Timber and concrete cribwork with con- A great many interesting case histories
crete superstructures of miscellaneous on blockwork quay walls built in the past
designs are given in comprehensive works on port
7. Miscellaneous innovative designs and harbor engineering by Cornick (1958)
and Bray and Tatham (1992).
Perhaps, the most notable example of
5.2.1 Blockwork structures blockwork wall construction discussed in
detail by Cornick is the deep water quay
The blockwork walls are by no means a new wall built in 1934 at Port-of-Mombasa,
type of construction. This type of quay wall Kenya (Fig. 5-5). Most of this approxi-
construction has been used in the past and mately 21-m-high quay wall was con-
is still considered by marine structure de- structed on foundation strata consisting
404 Gravity-Type Quay Walls

generally of hard sand or soft sandstone la)


rock (Fig. 5-5a), but at some locations where
the foundation material was not as good, it
was enhanced by piling as indicated in Fig-
ure 5-5b. In the latter case, two rows of
36 X 36-cm timber sheet piles were driven
in front and behind the wall. The wall itself
was founded on compacted sandfill placed
on wooden bearing piles. Blocks about 60
em deep were placed in a pattern as indi-
cated in Figure 5-5. The superstructure was (b)
of a concrete cast-in-situ construction with
a face protected by high-strength concrete
blocks. Rubble backfill comprised of miscel-
laneous granular materials was placed
behind the wall.
Although there have been many varia-
tions on the basic design of blockwork walls
they are generally classified as follows:

1. Bonded construction using solid concrete


blocks (c)
2. Wails formed with hollow or special con-
crete blocks

Both of the above types are usually topped


off with a similar concrete capping block; a
combination of both the above types of wall
(a hybrid construction) is also used.
Type 1 quay walls are used most exten-
sively where foundation conditions are good,
often in rock or cemented conglomerates. Figure 5-8. Examples of typical blockwork walls: (a)
Where foundation conditions are relatively traditional (trapezoidal) cross section; (b) blockwork
weak, particularly in areas subjected to set- wall with lateral soil pressure relieving block a nd
reinforced concrete foundation slab for distribution of
tlement under heavy load, it is advanta-
contact stresses; (c) improved cross section of block-
geous to use type 2 quay walls. Although work wall with later!!). soil pressure r elieving block.
both types of walls will accommodate settle- !-Granular filter material; 2-rubble; 3- reinforced
ment, they differ mainly in that the vertical concrete slab; 4-rubble mattress; 5-concrete block;
independent columns of the type 2 wall will 6-superstructure.
allow substantial amounts of differential
settlement to take place between adjacent
columns prior to casting (or mounting) the the past. The trapezoidal cross section was
upper capping block. Typical examples of dictated by the traditional approach to cal-
type 1 wall construction are illustrated in culation of the wall overall stability. The
Figures 5-8a-8c. improved cross sections are shown in Fig-
The walls of the traditional, trapezoidal ures 5-8b and 5-8c. The former has an
configuration, similar to that depicted in upper block which overhangs the rear line
Figure 5-8a, have been used most often in of the wall, thus reducing lateral pressure
Gravity-Type Quay Walls 405

Figure 5-9. Blockwork wall designed for better distribution of contact pressures at
wall base (EAU, 1990).

on a wall, and the latter represents the slabs was developed in Russia for construc-
further improvement to the blockwork wall tion on inland waterways (Fig. 5-10).
design; it is carefully balanced to provide The quay wall of this design was built at
for the required stability with a minimum Port-of-Malakhovo on the Enisey River,
overall structural weight. This is achieved Russia. There, the relatively light precast
basically by the wall geometry. A relatively solid concrete blocks, each weighing 15
large mass of the wall is located above
water level. The unsubmerged mass of the
wall, which is shifted to the rear, helps to ea. . o ( o _,. -v.)
keep the wall center of gravity to the rear. ~ -
Also, the upper block overhangs the rear
line of the wall which reduces lateral soil
pressure. Both of these design features tend
to reduce maximum bearing stress at the
wall base.
For the same reason, some walls have
been built with a negative slope of a back
line (Fig. 5-9). It should be noted, however,
that although such quay walls are very sta-
ble in their completed condition, they can
be relatively unstable during the temporary
construction stages. To prevent wall failure Figure 5-10. Blockwork wall with prefabricated re-
during the construction stage, the backfill is inforced concrete superstructure and ·pressure reliev-
usually placed incrementally with the ing anchor system. [From Budin and Demina (1979).)
blockwork. 1- Fender system; 2- cast-in-place concrete cope;
3-prefabricated reinforced concrete panel; 4-steel
The standard blockwork wall structure anchor rod; 5-prefabricated foundation slab; 6-rein-
with prefabricated reinforced concrete su- forced concrete pressure relieving anchor slab; 7-solid
perstructure and pressure relieving anchor concrete block; 8- rubble mattress; 9-granular fill.
406 Gravity-Type Quay Walls

tonnes, were used for construction of the of divers by laying of 78 tonnes full width
underwater portion of the quay wall. The keyed precast concrete blocks on a prepared
no-keyed blocks were placed in masonry in situ concrete foundation. The blocks were
wall fashion. To reduce the width of the dry jointed and bonded in brick wall fash-
wall and limit the magnitude of the lateral ion. It must be noted, however, that much
soil pressure, two reinforced concrete slabs of the construction success depended on the
have been placed within granular fill be- divers' skills to carry out all underwater
hind the wall. These slabs also contributed work. The precast blocks were topped off by
to the wall stability by providing an addi- an in situ concrete capping block which con-
tional anchoring force. The granular filter tained all quay services. The experience
was placed behind the blockwork to prevent gained at Tema indicated that the new gen-
the washout of fines through joints between eration of blockwork quay walls offered con-
blocks. The superstructure made from pre- struction which was fast, straightforward,
fabricated reinforced concrete components flexible, and simple.
was erected on the top of the blockwork. A very similar to Tema blockwork wall
The success and continued development 18.5 m high was built in the dry at Port-of-
of the new family of gravity blockwork walls Jebel Ali, Dubai (Loewy et al., 1990). In this
is demonstrated by the construction of a case, solid blocks weighing up to 44 tonnes
quay wall completed in 1965 at Port-of- were utilized. The capping block was a com-
Tema, Ghana. A typical cross section and bination of a precast reinforced concrete
front elevation of this structure is shown in units and a cast-in-situ concrete capping
Figure 5-11. This wall was built with help block.

A-A
(a)
(b )

Figure 5-11. Port-of-Tema, Gana. Quay wall. Typical cross section and elevation. [From Loewy
et al. 1990).] 1- Precast concrete blocks; 2-ashlar blocks; 3-cast-in-situ concrete coping
block Goints at 37m center to center); 4-cast-in-situ concrete foundation; 5-bedrock (gneiss);
6-rubber fender; 7-rockfill; 8-expansion joint; 9-dredged level; 10-service gallery.
Gravity-Type Quay Walls 407

As reported by Loewy et al. (1990), the The experience gained indicated that
wall at Tema was still in an excellent condi- where foundation conditions are less favor-
tion after 30 years in service. A further able for construction of a heavy concrete
reported advantage of the above type of blockwork quay wall (type 1), the wall can
construction is the resilience of the wall and be built by utilizing precast hollow or spe-
its overall stability in the event of acciden- cial concrete blocks (type 2). In the latter
tal damage inflicted by vessels. case, the wall is constructed in separate
The latest example of the type 1 block- vertical columns so that differential settle-
work quay wall is the 360-m-long structure ment of each column may take place before
completed in 1988 at Phuket, Thailand, the capping block is cast (or installed) across
providing this country with her first deep the top of the columns.
water port on the Andaman Sea (Cooke, It must be noted that sometimes the hol-
1990). This quay wall comprises nine low concrete blocks are also used for con-
courses of precast concrete blocks weighing struction of type 1 walls. This is usually
up to 26 tonnes, similar to that built at done to enable the contractor to handle
Tema, was founded on a weathered silt- these blocks by moderate craneage. How-
stone. ever, as soon as blocks are installed, the
The blocks were placed on a concrete voids are filled with tremie concrete, or reg-
base, the principal function of which was to ular concrete in the case of construction in
provide a level base and to serve to dis- the dry. Examples of this type of construc-
tribute the toe pressure from the blocks to tion are given in Cornick (1958). The earli-
the underlying rock. The concrete layer, est type 2 blockwork wall of unusual design
(the Ravier system) has been built at sev-
nominally 0.4 m thick, was actually about
eral ports in North Africa. An example of
0.9 m thick on average except at the deep-
such construction, completed at the port of
ened sections (up to 2.5 m) where badly
Arzen, Algeria in 1936, is depicted in Fig-
weathered rock had to be removed in order
ure 5-12. The wall was constructed from T-
to reach a suitable quality foundation. The
and !-shaped concrete blocks capped by a
tolerance allowed on the finished concrete
decking underneath of which the rubble
was ± 6 mm, with no forward tilt across the filling adapts itself to the natural slope at
foundation. The concrete was delivered un- the angle of repose. This type of construc-
derwater by pumps; the latter significantly tion reduces the quantity of concrete by
reduced time of concreting by proceeding 40-50% in comparison with solid block con-
during sea conditions in which the conven- struction while increasing the stability of
tional tremie method would not have been the wall, as has been proved by field experi-
possible. Another important factor that ments carried out at Port-of-Jibuti (Cornick,
emerged from the use of pumped concrete 1958). Large-scale-model tests also indi-
was the ease with which it was possible to cated that the great stability of such system
stop and restart the flow of concrete, thus obtained was largely attributed to the hold
permitting relocation of the discharge pipe. exerted on the blocks by the rubble fill,
An initial rough screeding of the concrete which surrounded and embedded their rear
was carried out by divers shortly after in- (tail) ends.
stallation. This included removal of the The blocks were cast in metal formwork,
overfillings and poor quality surface con- thus assuring an accurate finish. The wall
crete. After an hour or so, during which the was constructed from separate vertical piers
concrete firmed up but well before any ini- with 5 em clearance between them. The
tial set took place, the concrete was re- capping slab connected the top of all piers
screeded to ensure proper leveling. together, thus helping to distribute opera-
408 Gravity-Type Quay Walls

tional loads (e.g., mooring forces over sev- in each column, between columns. Once the
eral piers). The form of the blocks in T- or blocks were placed in the wall, they were
1-shapes permits the hauling of them by entirely in compression and, therefore, there
grips holding the central portion. was no need for any steel reinforcement.
The most recent construction of type 2 However, in this type of construction a min-
blockwork quay wall at Port-of-Jeddah, imum amount of steel reinforcement may
Saudi Arabia is described by Loewy et al. sometimes by required to resist bending
(1990). The typical cross section and eleva- moments and shear forces induced during
tion of this project are depicted in Figure block transportation and handling.
5-13. The construction of the wall at Jed- The rock:fill behind the wall was placed
dah proceeded in the following stages: once in order to reduce the lateral pressure on
the first two hollow base blocks in each the wall. In order to accelerate the process
column have been placed, all the lower base of wall settlement, preloading in the form of
blocks and part of the upper base blocks solid concrete block kentledge was applied
were filled with tremie concrete. This was to the complete block columns and prior to
intended to produce a full bearing area, to casting the capping block. The kentledge
the underside of the columns at wall base arrangement was such that individual
level. Once a complete column of hollow preloading blocks were placed on individual
blocks was built, the overall voids to the columns such that the settlement could be
rear of the blocks were also filled with accelerated differently. It is worth noting
tremie concrete. Sealing of the rear face of that at Port-of-Jeddah the actual wall set-
the quay wall between adjacent columns is tlement, depending on the specific site
affected by underwater concreting to the geotechnical conditions, varied from 0 (hard
circular void, formed by semicircular voids corals) to 120 em (fine silts).

·99 I-DO ·99

Figure 5-12. Quay wall at Arzen, Algeria (Ravier system).


Gravity-Type Quay Walls 409


I "'-15500 ,,,
\~
(t-o. -

"'"""" llo*
,,..
\~

!US

......

,-11·1

lllflfl - · I lOOM - 1111

Figure 5-13. Port-of-Jeddah, Saudi Arabia. Quay wall; typical cross section and elevation.
[From Loewy et al. (1990).]

In areas where fine silts were encoun- tion of breakwaters in various harbors. This
tered, further quay wall trench dredging happened because 90% or more of the work
was necessary, together with a substantial of building caissons can be executed on-
depth of sandfill cushion between the bot- shore under conditions which normally are
tom of the final dredged trench and the not a subject to the severe amounts of
underside of the rockfill mound supporting downtime experienced offshore. Therefore,
the quay wall structure. In areas where the high cost of equipment and labor during
substantial settlement had taken place, the periods of severe sea conditions was avoided.
resulting difference in level was made up by Furthermore, caisson-type breakwaters
applying in situ concrete to the capping have been built, often where foundation soil
block. More examples are found in Ciortan precluded construction of piled structures.
(1994). Caissons also have been found adaptable to
the loose alluvial soil formation at many
locations in Europe and Japan. Later on,
5.2.2 ouay Walls composed of caisson-type technology has been found
Floated-in concrete adaptable for the construction of quay walls.
Caissons In most cases,_ concrete caissons are built
onshore where launching facilities such as
Advances in concrete technology at the shiplifts, marine rails, slipways, dry docks,
beginning of this century made possible or flooded basins are available. On several
the construction of a relatively thin-walled known occasions, caissons were constructed
cellular reinforced concrete boxes usually at locations with substantial seasonal water
referred to as caissons. In early years, cais- fluctuation (e.g., riverbanks). There they
sons became popular, basically for construe- were built during the low water levels, then
410 Gravity-Type Quay Walls

lifted up by buoyancy and towed away dur- moored at the other facilities (e.g., outfitting
ing a sufficiently high water level (seasonal dock or the like located in sheltered basins).
flood). Thus, construction or use of expen- The caissons are normally constructed in
sive facilities such as dry docks or slipways close proximity to the project site. In some
was avoided. In Eastern Canada caissons cases, however, they are fabricated at loca-
built ashore are typically launched by em- tions within a reasonable towing distance,
ploying the so-called "slipway launch where materials, labor, and construction
method," which includes construction of the and launching facilities are available.
caisson on a tilting platform adjacent to a The methods employed for the construc-
fixed short slipway. There, the partly (or tion of concrete caissons depends largely on
fully) completed caisson slips down the the overall size of the project. For large-size
basin when the platform is tilted, then ac- projects, construction of a special yard for
celerates down-ramp (slipway), and finally caisson mass production with dedicated
splashes down in the basin (Yan, 1983). launching facility is usually economically
As indicated in Figure 5-7, caissons built justified.
ashore at Gdynia, Poland have been cast Caissons can be either slipformed or as-
horizontally at the seashore edge, then sembled from prefabricated elements. Sev-
launched by jetting off part of the shore. In eral typical examples of caisson construc-
this process, caissons are slipped into the tions are discussed later in this section.
water. There are more methods of launch- Recent examples of mass production of
ing caissons which obviously depend on lo- concrete caissons are major developments
cal conditions and available resources. at Dubai, Dumman, and Brighton. At Dubai
The chief advantage of using the floated- (Anonymous, 1977) a special yard was built
in caissons for construction of quay walls is in order to cast 162 caissons weighing be-
the substantial reduction of the working tween 3000 and 3500 tonnes each. A similar
time over water. An additional advantage is arrangement was used at Dumman (Brown,
a minimum number of joints in a quay wall 1978) where 199 caissons 20m long, 14m
structure. The latter requires a minimum wide, and 16m high were cast and launched.
filter material to protect fines in the backfill Structurally concrete caissons may be of
from being washed off. Miscellaneous gran- symmetrical or unsymmetrical configura-
ular materials such as sand or gravel are tions. Typical example of caisson of a sym-
usually used for backfilling of walls com- metrical cross section is shown in Figure
posed of floated-in caisson, as opposed to 5-7. Examples of unsymmetrical configu-
the rubble prism which is usually used in a rations are illustrated in Figures 5-14 and
blockwork wall construction. 5-15.
The caisson-type quay walls can be prac- The structure depicted in Figure 5-7 was
tically of any height and do not require use built at Port-of-Gdynia, Poland in 1924-
of the heavy lift equipment for their con- 1927. The total length of a typical caisson
struction. At the same time, however, in was equal to 18.15 m. It had four internal
some cases use of such facilities like dry walls which divided the caisson into five
docks, slipways, shiplifts, and the like for typical cells. The voids left in internal walls
caisson construction and launch can make were intended to reduce the caisson weight
them cost-prohibitive. That is why in order and also to improve its floating stability
to limit time of use of the aforementioned characteristics. After installation on a rock-
facilities the caissons are normally con- fill mattress, the caissons were filled with
structed just to the depth required for safe sand. Sand was also used for the backfill.
launch only, after which they are towed Caissons of complicated and unsymmet-
away and completed to the full depth being rical cross sections have been built for con-
Gravity-Type Quay Walls 411

The complicated shape of the caisson's


rear wall and the base slab were designed
to reduce the horizontal soil thrust on the
wall and to increase its stability against
sliding. However, the complicated geometry
of the caisson cross section created many
practical problems during fabrication and
towing to site of installation. The caisson
installation was assisted by floating cranes,
which helped to keep it horizontally during
ballasting operation. After installation on a
rubble mattress, the caissons were filled
with concrete; further reduction of soil lat-
eral thrust on the quay wall was achieved
by placing of the rubble mound behind the
wall.
The cross section of the unsymmetrical
Figure 5-14. Port-of-Talkoguanno, Chile. Quay wall; caissons used for construction of the quay
typical cross section. wall at Port-of-Klaipeda, Lithuania in
1960-1963 is shown in Figure 5-15. These
3.4-m-wide, 21.7-m-long concrete caissons
had 10 internal walls which divided the
caisson into 11 typical compartments. The
caisson's structural members were propor-
tioned in a way to keep the caisson strictly
vertical during tow and installation. Conse-
quently, there was no need for the use of
floating cranes. After installation on the
foundation, the caissons were filled with
sand and backfilled also with sand.
Sometimes caissons are partly filled. This
is done in order to reduce the contact pres-
sure on the foundation at the wall toe. In
1 J,lt 2,G some specific cases, the caisson front wall is
7 perfdrated by several rows of holes and left
empty for better dissipation of wave energy.
Figure 5-15. Port-of-Klaipeda, Lithuania. Quay
These holes are properly sealed during cais-
wall; typical cross section. son launch, tow, and installation (Fig. 5-16).
Where tidal or seasonal water fluctua-
tions are substantial, the caissons are usu-
struction of a quay wall at Navy Harbor, ally designed for installation at a low water
Talkoauanno, Chile in 1904-1907 (Fig. level with the superstructure built in the
5-14). The caisson units used there were dry during the low water. The quay wall of
10.35 m high, 6.5 m wide, and 10.0 m long, this type of construction with the super-
with a large buttressed base slab. The lat- structure assembled from prefabricated re-
ter was provided for wall stability. The av- inforced concrete members is illustrated in
erage wall thickness was 0.2 m and each Figure 5-17. The high degree of repetition
caisson weighed approximately 215 tonnes. and, consequently, reduction of the con-
412 Gravity-Type Quay Walls

Figure 5-16. West Port-of-Taranto, Italy. Quay wall; typical cross section. [From Contri et al.
(1986).]

(a ) (b)

Figure 5-17. Port-of-Perm, the Kama River, Russia. Quay wall. (a) Typical cross section; (b)
Cross section through the ballard's stairwell. 1- Caisson assembled from prefabricated compo-
nents; 2- anchor concrete block; 3-,-anchor rod; 4- prefabricated concrete panel; 5- cast-in-place
concrete bock for ballard installation; 6-cast-in-situ capping; 7-bollard; 8-stair; 9-steel
anchor; 10-prefabricated panel; 11- fendering.
Gravity-Type Quay Walls 413

struction time were achieved by fabricating section. The panels were reinforced by webs
the entire structure from standardized pre- 45 em deep and 30 em wide; the steel an-
cast elements. In the latter example, 16 chors were hinged at both the panel and the
caissons 17.4-m long, 4.3-m high, and 8.8-m caisson's diaphragm. The superstructure
wide at the base slab were built from pre- panels were seated on caisson's face panels
fabricated typical elements. The pre- and diaphragms. The latter was particu-
stressed concrete frontal panels, 18 em thick larly important to prevent the soil lateral
( f~ = 30 MPa), were designed to sustain the thrust to be transmitted to the caisson's
severe ice load and cold temperature condi- face panel. After the quay wall was back-
tion (Gurevich, 1969). filled with sand and therefore all steel an-
Each caisson consisted of 52 elements of chors uniformly tensioned, the cast-in-situ
6 typical configurations. The bottom slab capping was installed.
was assembled from standard slabs 6.6 X
Because of substantial water fluctuation
3.0 m (each weighing 10.5 tonnes). Initially,
at the location and for the sake of conve-
all prefabricated elements were joined to-
nience of a ship mooring operation, the
gether by welding special embedded parts
installed at each slab. Then, from both sides mooring bollards were installed at three
of each joint a spiral reinforcement was levels. The access to the lower bollards was
installed and tuck welded to the embedded obtained through a special staircase in-
plates. Finally, all joints were sealed by stalled in the middle of each caisson. The
30-MPa shotcrete (Fig. 5-18). Despite its completed staircase was anchored by an in-
unsymmetrical configuration, the caisson dependent anchor system, comprised of steel
was well balanced during towing and in- rods secured at the prefabricated concrete
stallation procedures. deadmen (Fig. 5-17b).
Mter the caissons were installed and All caissons were assembled at the local
filled with a mixture of sand and gravel, the shipyard, then onshore tested for water-
superstructure was erected. It was assem- tightness and, finally launched via a slip-
bled from prestressed concrete panels 3.15 way. After installation at the project site,
wide, 7.0 m high, and 15 em thick and the space between adjacent caissons was
inclined steel anchors 20 X 150 mm in cross filled with gravel.

l
J

Figure 5-18. Welded joint between concrete slab and diaphragm wall in caisson assembled
from prefabricated components. 1-Base slab; 2-shotcrete; 3-spiral reinforcement; 4-steel-
embedded parts in base slab; 5-anchors of embedded parts; 6-embedded steel in diaphragm
wall; 7 -diaphragm; 8-weld.
414 Gravity-Type Quay Walls

The experience gained at Port-of-Perm The departure from this concept is shown
has indicated that where mass production in Figure 5-20. The concrete caisson illus-
of concrete caissons is required, the prefab- trated in this figure does not have any in-
rication technology helps to significantly re- ternal diaphragms and its structural stiff-
duce the construction time and to achieve a ness depends solely on strength of the ex-
good quality and economy of quay wall con- ternal stiffeners. This type of caisson was
struction. Typical structural elements used used for the construction of several oil ter-
for construction of the quay wall at Port-of- minals on the Kaspian Sea in the 1960s.
Perm are depicted in Figure 5-19. It must After installation on prepared stone bed-
be pointed out that joints are the most criti- ding, the caisson was ballasted with tremie
cal element of caissons assembled from pre- concrete, after which the ballast water was
fabricated components. In a saltwater envi- pumped out and the oil pumping station
ronment where corrosion of embedded steel was constructed in-dry inside the caisson.
may be a problem, the posttensioning tech- Unusual hybrid-type caissons have been
nology used for construction of large float- fabricated and installed off the Australian
ing docks (Tsinker, 1986) may represent a coast for the coal loading terminal usually
more reliable solution for joining of individ- referred to as "Hay Point Terminal" (Fig.
ual prefabricated components than welding 5-21) (Eddie, 1976). Located about 2 km
together the embedded steel. offshore in water averaging 17 m in depth,
The structural stiffness of a typical the terminal was built from three prefabri-
floated-in concrete caisson is provided by cated caissons completed with a steel super-
the internal diaphragms that convert the structure. These caissons were each towed
caisson's internal space into the system of out and installed in a single day; the actual
individual cells. setdown of each caisson took only a few

(a} (b)

Figure 5-19. Concrete caisson assembled from prefabricated components (b) and
concrete panel for the superstructure (a). 1-Side panels; 2-transfer diaphragms;
3-frontal (face) panels; 4-longitudinal diaphragm; 5-rear panels; 6-counterforts;
7-base panel; 8-steel-embedded parts for anchors.
Gravity-Type Quay Walls 415

/1158

Figure 5-20- Concrete caisson with external stiffeners used for construction of an oil
pumping station.

-~--

li+G'
I
I
1 .
0 0 0
l.____________________J
0 0 0 0 0
0
0
0
0
0
0
0
0
0
0 -------~~~~;--~
pttoSGIIIIIulllill ·

Figure 5-21. Hay Point coal loading terminal, Australia. Cross section through
concrete caisson. [From Eddie (1976).] !-Caisson; 2-corner tower; 3-steel super-
structure; 4-bulk carrier 100,000 DWT; 5-pressure-equalizing hatches, removed after
caisson is founded; 6-tidal inlet to tower; 7-grout inlet; 8-grout indicator outlet;
9-grout; 10-screeded crushed rock; 11-crushed rock mattress; 12-concrete block
scour mats; 13-wire rope; 14-surface of rock or hard clay.
416 Gravity-Type Quay Walls

hours. Five more smaller caissons of identi- sate for the narrow width, a cantilevered
cal construction were also used as founda- relieving platform was placed on the top of
tions for approach spans, and two addi- the caisson. The cantilevered portion
tional similar caissons were used as bridged the 4.5-m gap between new caisson
mooring dolphins. Each large caisson (46 X and existing structure. The platform im-
38 X 8 m) was surmounted at each corner proved the system's stability in two ways:
by hollow towers 12 X 12 m in cross section First, it relieved the upper portion of the
and 18 m high. Set in a row, all three soil pressure and eliminated the effects of
principle caissons served as a berth for a surface live loads on a soil lateral thrust,
super coal carrier (100,000 DWT) and sup- and second, it added to the vertical weight,
ported the conveyor belt system and coal thus enhancing the caisson sliding stability
loading machine. Caissons were of cellular- and provided a resisting moment against
type construction. They were built in a dry overturning, resulting in a better distribu-
dock where the concrete base slab and the tion of bearing pressures at the caisson base.
roof were cast in situ, whereas all internal The narrow caisson width required water
and external walls were assembled from ballast to provide for stability during con-
prefabricated elements. struction and while under tow. Caissons
For better performance and watertight- were placed on a mattress made from care-
ness the whole structure was posttensioned. fully densified granular material. The mat-
At the terminal site, a crushed stone mat- tress was screed by a heavy 12-m-wide
tress was constructed. To prevent potential sweep beam suspended from a barge which
problems associated with torsion and stress was moved by winches along the mattress.
concentration, the mattress was screed to The vibrofiotation method was carried out
within ± 25 mm. This was done with help of for mattress consolidation. The method
a custom-built screeding frame. Further- proved to be successful.
more, the gap left between caisson base In general, reduction in caisson width
slab and surface of the mattress was does not save much concrete required for its .
grouted. To minimize grout loss into the fabrication. However, significant saving in
mattress material, the stone size was the total cost of a quay wall may result
adopted to be about 40 mm or less. To avoid from a reduction in volume (and cost) of
an unexpected settlement, the mattress was stone bedding; furthermore, whereas the to-
preloaded by impounding extra water up to tal amount of fill inside and behind the
the high-tide level in the corner towers. caisson will be unchanged, the total cost of
The unusually narrow concrete caissons the fill is expected to be less due to the
have been used for upgrading an existing reduction in caisson inside fill which is nor-
navy pier at Halifax, Canada (Barry, 1983). mally more expensive than regular fill
In this case, the caisson width was actually placed behind the caisson.
dictated by the local operational constraints Usually, caissons are described as struc-
and was limited to 7.0 m with a 8.0-m base tures shaped like boxes and one can wonder
slab. The overall configuration for the typi- whether this view of caissons stems from
cal caisson was 30.0 m long by 7.0 m wide the original meaning of the· french word
by 15.16 m high. (Fig. 5-22). It should be "caisson" to denote an ammunition box,
noted, that under actual design conditions which lent its name to the early caissons
for a similar depth of water, the width of used in civil engineering. In modern engi-
the conventional caisson would be equal to neering practice, however, the term "cais-
approximately 12.0 m in order to provide son" is also applied to floating cylinders
the required factors of safety against slid- used for the construction of miscellaneous
ing and overturning. In order to compen- marine projects.
Gravity-Type Quay Walls 417

t!
~
"""
tn.
.J
-rw. ">000

'61\P naN

Figure 5-22. Navy pier at Halifax, Canada: (a) typical cross section; (b)
caisson plan. [From Barry (1983).]
418 Gravity-Type Quay Walls

Although the circular shape of the cylin- to enhance its stability. An important factor
drical caissons requires use of the rather in caisson design was the anticipated irreg-
complicated formwork for caisson construc- ularities of the rubble mattress potentially
tion, nevertheless, it provides for better use leading to uneven pressure distribution at
of reinforced concrete as a structural mate- the base. In order to avoid base slab over-
rial. In many cases, cylindrical caissons re- stressed conditions, the top of the mattress
sulted in design and construction of a very had been required to be screeded within
economical marine structure. An example of ± 20 mm. The caisson was heavily rein-
the successful use of concrete cylindrical forced in the intertidal and splash zones to
caissons for construction of an important oil meet the required concrete crack width cri-
terminal in Canada is given in Tsinker teria. The wall was completed in the mid-
(1995). 1980s.
A similar kind of technology was used for Recent case histories involve quay wall
the construction of the docking facility at construction at Port-of-LeHavre, France
Matanzas, Venezuela on the Orinoco River comprised of large floated-in cylindrical
(Civil Engineering, 1980). There, 23 cylin- concrete caissons are discussed by Dubois
drical caissons each 20 m in diameter and et al. (1990) and LeLarge and LeBars (1992).
23 m high with wall thickness of 200 mm There, two quay walls, 350 m and 680 m
and weight of about 2000 tonnes were slip- long, were built for container terminals. For
formed on the river bank at low water level. construction of these walls, large floated-in
The caissons were set afloat at a high water concrete cylindrical caissons have been em-
level, then towed away and installed on a ployed. A typical cross section and plan of
1-m-thick rockfill mattress. The caissons these walls is depicted in Figure 5-23.
were filled with granular material, and the Caissons 21.5 m high and 24.0 m in diame-
interstices between adjacent caissons were ter with a wall thickness of 0.5 m and bot-
injected with cement grout. Finally, the 3- tom slab 0.9 m thick have been placed at
m-thick concrete slab with service galleries 25.0 m pitch along berthing line. At quay
was cast on top of the caissons. locations the poor silt at the seafloor has
Recent construction of a 500-m-long quay been replaced by a layer of gravel about 4.0
wall at Jinzhou harbor, People's Republic of m thick, which served as a foundation for
China, consisting of 39, 12-m-diameter the walls. The walls were designed for a
cylindrical concrete caissons is discussed by surcharge load of 40 kNjm 2 extended to
Mornement and Pearce (1987). The cais- some distance from edge of the quay toward
sons, 13 m high, had a 0.6-m-thick octago- the dock storage area and 60 kNjm 2 in
nal base slab and a 0.4-m-thick wall. After cargo storage area. It also was designed
installation on a rubble mattress, the cais- for a load of 700 kNjln m produced by
sons were filled with sand and capped with standard portal container crane 40 tonnes
a heavy reinforced concrete continuous capacity with a 45-m reach. A similar exam-
beam. This beam was designed to support ple is found in Ciortan (1994).
the seaward container handling crane track, An unconventional approach was used
to resist ship impact and mooring loads, for construction of the gravity quay wall at
and to retain the upper portion of the back- Port-of-Voltry (Genoa), Italy (Contri et al.,
fill. The toe of the wall was protected from 1986). There, each caisson unit used for the
scour and erosion by an antiscour apron. wall construction comprised three concrete
The drainage system was installed be- cylinders 9.6 min diameter, 14.25 m high,
tween each adjacent caisson and below the with a wall thickness of 350 mm built on a
base. Good quality granular backfill mate- common slab 1.25 m thick. Thus, the total
rial ( <Pmin = 35°) was placed behind the wall length of the caisson unit was 30.4 m (Fig.
Gravity-Type Quay Walls 419

PLAN

Figure 5-23. Gravity-type quay wall at Port-of-LeHavre, France: 1- cylindrical


caisson 24.0 m in diameter; 2-granular fill; a-concrete superstructure; 4-rockfill
mattress; 5-craneway track; 6-fender system. [From Dubois et al. (1990).]

5-24). The caissons were cast in a locally Such terminals are typically constructed as
available dry dock facility. a single-berth facility. Although these ter-
The most recent example where circular minals have no breakwaters and their
concrete caissons were used is in the availability for loading/unloading opera-
reconstruction of the ferry terminal at tion is limited to about 50-75%, they
Portsmouth harbor in the United Kingdom. can operate at exposed locations quite
There, large caissons, each nearly 20 m effectively.
high and 19 min diameter weighing 2500 Eddie (1986) suggested that the utiliza-
tonnes, were slipformed in a dry dock at tion time of offshore terminals can be ex-
Southampton and then towed 43 km at a tended to up to 90% by providing a partial
speed of about 2 knots across the sea to the protection for ships at terminal location. He
site of deployment. Once sunk into place, stated that this can be achieved by con-
they were filled with dredged ballast. The structing a hybrid-type berth structure that
caissons are used for ship moorings and to will include a sloped caisson-type break-
guide the ship into position (Anonymous, water with a perforated sloping exposed face
1992). and a cargo handling equipment super-
As stated in Chapter 2, the dramatic in- structure with a working deck raised be-
crease in ship sizes led engineers to build yond reach by a maximum design wave.
offshore terminals in exposed but suffi- The conceptual design of such a structure is
ciently deep water areas to serve very large illustrated in Figure 5-25. The structural
vessels with a draft in excess of 20.0 m. design criteria for such offshore installation
420 Gravity-Type Quay Walls

diffraction coefficients at the berth side are


such that safe operation conditions result
when ambient waves are at the 10% ex-
ceedance level. According to Eddie, the
sloped breakwater design in the form of a
concrete floated-in caisson with a perfo-
rated frontal face is the most economical
solution to serve the purpose. This caisson-
type breakwater should be designed to the
following conditions.
SECTION A-A

(i) To prevent overtopping by waves up to


the operating limit required for the berth
-usually 3-4 m maximum (2-2.5 m sig-
nificant wave)
A
(ii) To minimize the horizontal load on the
caisson caused by the waves which are
higher than the berth operating limit
-typically 4-15 m
(iii) To minimize the splash of the overtop-
PLAN ping waves so as to minimize the eleva-
tion of the equipment deck required to
Figure 5-24. Caisson used for construction of dock- clear the highest waves
ing facility at Port-of-Voltri (Genova), Italy. [From (iv) To eliminate scour problems under maxi-
Contri et al. (1986).]
mum wave conditions

as recommended by Eddie should include The advantages and disadvantages of


the following conditions: sloping caissons in general and caissons
with a perforated frontal face in particular
(a) The terminal should provide a safe berth used for the offshore construction has been
on its lee for 90% of the time. discussed by Jarlan (1963), and most re-
(b) To make the structure more economical, it cently extensively discussed by Landgren
must be designed in a way to offer the (1986), Aristaghes et al. (1986), Franco
least resistance to 10% of the highest (1994), Tanimoto and Takahash: (1994), and
storm waves. others. For more discussion on sloping and
(c) The structure must survive the highest perforated face caissons consult Chapter 10.
100-year single wave occurrence. The latter authors concluded that sloping
(d) Construction of such a facility must be caissons such as the one depicted in Figure
feasible and economical in isolation from 5-25 would be able to generate downward
the shore. component of wave load, thus improving
the caisson stability. The value of perfo-
The above objectives could be achieved by rated walls for reducing the wave forces,
preventing overtopping of the breakwater significant reduction of the reflected waves,
portion of the structure in all but 10% of and, consequently, reduction of the wave-
wave activity periods. The latter may be induced scour effects has been amply
accomplished by arranging the breakwater demonstrated in the North Sea and in a
length and geometry in a way that the number of harbors.
Gravity-Type Quay Walls 421

1Z /\_ - -

1 \ -===~
L ~----- I
I --- I I
3"' I I I
............
\
_r----~
I
~jBSiliit;;JIIi:a--fl,... -- - -,
i
...

I
....
0
I
I
--- --- - _ J

Figure 5-25. Offshore terminal with sloping face caisson integrated with superstructure for
load handling equipment. 1-Very large bulk carrier; 2-sloping face caisson breakwater with
performation; 3-cargo handling machinery superstructure; 4-granular fill; 5-energy-
absorbing ballast (if required); 6-steel skirt; 7-riprap; 8-grout; 9-impounded water outlet;
10-temporary buoyancy tank for flotation; 11-design waves. [From Eddie (1986).]

The caisson depicted in Figure 5-25 can of the caissons or consolidation of the sub-
absorb some additional amount of wave en- strata, on the other hand, would not cause
ergy by impounding the water from over- irreparable damage, because the means
topping waves in the caisson cells. This en- of adjusting the supports of the cargo
ergy can then be released in a manner which handling equipment can be incorporated in
will not affect the moored ship, by dissipat- the design.
ing it through openings in the caisson walls, The PIANCs Wave Commission Report
below the level of the vessel's keel. More (1980) stated that caissons are not usually
information on sloping caissons is found in made more than 30 m long, 25 m wide, and
Chapter 10. 20 m high, citing as reasons for these limi-
For the offshore terminal of the type il- tations that there would be problems in
lustrated in Figure 5- 25, it is of paramount launching, towing, stability during towing,
importance that under no circumstances speed and accuracy of installation, and pro-
should there be substantial movement of viding an adequate foundation. It is worth-
the caissons, as this may severely damage while to mention that all of the above prob-
the cargo handling equipment. Hence, in lems have been successfully solved in other
developing a design philosophy for a :fields of marine engineering (e.g., offshore
caisson-type offshore terminal, the defini- platform, submerged tube tunnels, floating
tion of unacceptable failure should be docks, and others).
carefully considered. For example, base fail- As indicated earlier in this section, this
ure of the foundations due to excessive shear limitation was extended far beyond 30.0 m
stress or liquefaction would be clearly unac- at an offshore installation in Australia (Hay
ceptable. However, a small degree of sliding Point Terminal, Fig. 5- 21). Another exam-
422 Gravity-Type Quay Walls

ple is the successful installation of 77.0-m- critical stage in an overall process of a quay
long concrete caissons at Tarsuit explo- wall construction; significant deviations
ration island, Canadian sector of Beaufort from the horizontal plane andjor uneven
Sea (Tsinker, 1995). settlements of the mattress may result in
In his PIANC paper, Eddie (1986) made substantial global or local overstress.
reference to very long floated-in concrete Therefore, the success of the caisson-type
caissons, such as a 77.0-m-long caisson used quay wall construction depends heavily on
for break-water construction and 120.0-m the quality of mattress preparation. Nor-
caissons used for oil storage. mally, it should not be thinner than 0.3 m,
and in most cases is made from rocks 50-85
· 5.2.2.1 Mattress mm in diameter. The mattress must be
Normally, fabrication, launching, towing, carefully densified, which is usually done by
and installation of large (long) caissons have vibration or by preloading with heavy con-
presented no significant technical problems, crete blocks (Fig. 5-26b). Several brands of
and in most practical cases the design vibratory plate compactors able to work un-
length of a caisson of conventional construc- derwater are currently manufactured; they
tion is influenced by tolerances specified for are able to consolidate material ranging
preparation of stone bedding (mattress). The from 75 to 350 mm in diameter in layers up
construction of mattress and specifically its to 4 m thick (Gerwick, 1986).
densification and surface preparation is a As stated earlier, the mattress' upper

(a)

(b)

Figure 5-26. Mattress preparation. (a) Floating mattress


leveling machine: !-pontoon; 2-winch; 3-steel column;
4-blade; 5-guide; 6-steel cable. (b) Preloading of rubble
mattress with concrete blocks. (Photo courtesy of D. Krneta.)
Gravity-Type Quay Walls 423

layer, about 0.3 m thick and made from 160-tonne crawler crane. It operated effi-
gravel, is usually not compacted and must ciently in practically zero visibility, replac-
be carefully screed to the specified toler- ing divers who otherwise would be leveling
ances. If rubble material is to be densified, the mattress using nothing more sophisti-
then initial thickness of the mattress de- cated than conventional equipment. The
pending on the material porosity should be mattress aggregates were fed down to the
assumed to be thicker by 5-20%. In practi- seafloor through the hopper mounted on top
cal cases, the initial thickness is estab- of the machine.
lished by the trial densification. For a bet-
ter result, the depth of the densified layer of 5.2.2.2 Skirt
rubble should not exceed 1.5-2.0 m and the
resulting settlement should not be less than As follows from previous discussions, the
5-8%. The final grade of the rubble in- principle problem with large caissons is
cluded in the mattress structure should be clearly that of providing a suitable founda-
finished within ± 20 em. tion. In large offshore caisson installations,
As stated earlier, the compacted layer of this obstacle was overcome by incorporating
rubble is covered with about 30 em of non- a skirt into the caisson structure. This tech-
compacted gravel or crushed stone. The lat- nique has been used extensively for con-
ter, depending on project requirements, is struction and founding at the North Sea
usually screed within ± 25 to ±50 mm of giant offshore platforms, and the experience
the final grade elevation specified in the gained there is relevant for construction
project. and founding of large conventional caissons
In many cases in North American prac- on an unprepared bed. Skirts generally pro-
tice, the top layer was leveled by dragging a trude under the base slab and are usually
heavy beam suspended from a barge or pon- arranged in a symmetrical pattern directly
toon along the mattress at the design depth. under the perimeter walls and under some
This operation is usually repeated several of the internal bulkheads.
times until the desired result is achieved. A Two types of skirts are usually used: con-
better result can be achieved by using dif- crete and steel. Concrete skirts may be ben-
ferent kinds (usually custom built) of spe- eficial at the perimeter walls when the
cial floating equipment. A sketch of one structure is to be grounded using a tilt-down
such machine is depicted in Figure 5-26a. procedure. Alternatively, corrugated steel
There, the bulldozer-type blade is attached skirts, resembling sheet piling, can be fabri-
to the steel columns, which can be moved cated as assemblies and fixed to the base
vertically by an onboard winch. The barge slab either during caisson bottom slab cast-
(pontoon) is pulled along the mattress, with ing or after caisson launch.
the blade established at the design depth. A steel skirt is able to penetrate the sea
A purpose-built screeder to level off an or river floor by 1-5 m and has a number of
underwater gravel mattress has been re- advantages. These include the prevention of
cently used in the United Kingdom (Annon- inequality in load distribution at the cais-
ymous, 1990). The screeder was designed to son base seated on an uneven surface, bet-
operate in murky waters 16 m deep off the ter protection from the scour effect by the
coast of East Anglia. It was built in the current or ship propellers, and enhanced
form of a space frame 20 x 10 m and 21 m resistance to horizontal loads. Furthermore,
high, installed on four Caterpillar-type free- the skirt may contain weak upper layers of
wheeling tracks, mounted on hydraulic soil, preventing them from spreading and
rams. The machine was lowered into the transferring the surface load to the stronger
sea each working day from a barge-mounted underlaying strata.
424 Gravity-Type Quay Walls

It is important that the skirts do not that reduced C 3 A content increases resist-
penetrate to their full depth; otherwise, lo- ance of concrete against sulfate attack and
calized high bearing pressures may result reduces heat generation, but substantially
on the caisson base slab at the point of increases the probability of reinforcing steel
contact with the bed. For this reason, the corrosion.
skirt design must ensure that a space is left Cement replacement by pozzolan, either
between the bed and the underside of the natural or fly ash, is recommended. Poz-
caisson. This space must be filled with grout.zolan provides a more workable mix, less
As stated earlier, the aforementioned tech- thermal problems and internal heat genera-
nique was successfully applied to the cais- tion, less bleeding, more resistance to sul-
son installation at Hay Point Terminal in fates, and lower permeability. In general,
Australia. Base grouting with sand is dis- silica fume added to concrete mix at a rate
cussed by Buslov and Brakel (1985). of about 12% of the cement weight in
Obviously, the skirt technology requires combination with an appropriate water-
the installation of the grouting system at reducing admixture, is very beneficial to the
the caisson base and the careful control quality desired in concrete exposed to harsh
procedure of grouting the space between seawater environment. This includes a
the caisson base and the seafloor. much lower permeability, high density, and
better protection of reinforcing steel. If pre-
5.2.2.3 concrete stressing is used, the lower creep rate and
higher elastic modulus of silica fume con-
General requirements for the concretes crete can also be beneficial.
used for construction of marine structures Surely enough, the concrete mix should,
are discussed in detail in Chapter 2. Regu- most likely, use admixtures that are based
lar, normal weight structural concrete for on structure specifics as well as on the site
marine application with a minimum crush- environmental conditions. For example, if
ing strength of 30 MPa after 28 days as silica fume is used, then both normal and
determined by a cylinder test is usually high-water-reducing agents are typically in-
used for caisson construction. The concrete cluded. It may well be that during warmer
mix typically has a waterI cement ratio of or daytime placements a different admix-
0.4 or less by weight. ture is used then during cooler or night
The concrete mix and its placement re- placements.
quire proper attention. Typically, mix de- An air entraining admixture may or may
signs should first concentrate on ease of not be appropriate, depending on the ambi-
placement without segregation or rock ent temperature. In general, in a warm cli-
pockets and on mixes with low permeability mate, air entraining admixtures should be
and shrinkage, before worrying about avoided. It may be considered, however, if
strength. Optimizing strength without pay- the mix had a problem of workability and
ing enough attention to these other aspects bleeding that could not be corrected through
of the mix may result in problems. After adjustments in sand gradation, fly ash,
achieving the first goals, adjustments cement coarseness, or other admixtures.
for strength can be made afterward if Corrosion inhibitors are something to
necessary. consider, but should not be necessary with
The selection of cement should be based proper cover over reinforcement and good
on the consideration of a possible sulfate consolidation, especially if a mix with silica
attack, the corrosion potential due to chlo- fume is used. Latex and polymer modifiers
ride penetration, and control of heat gener- are additional items that might be consid-
ated by hydration. It should be recognized ered, but, in general, should not be neces-
Gravity-Type Quay Walls 425

sary if good quality concrete is produced Sea and is used for the construction of the
and practical design section with adequate Hibernia offshore platform to be installed
cover is developed. Retarders are usually offshore Newfoundland and the Heidrun
required; they may be deleted in cold tension leg platform to be permanently
weather and added in warm weather. By moored in the Northern North Sea.
the proper selection of the admixture, an In general, however, if the depth of water
increase in the water-reducing dosage in available at the caisson construction site is
warm weather can also provide retardation. insufficient to complete the structure in the
In the cases where the caisson's draft is dry to its full depth, then the remainder of
limited by shallow navigation channel(s), the construction is carried out in the float-
the high-strength lightweight concrete can ing mode at specifically designated facilities
be used for caisson construction. In general, located at sheltered locations with a suffi-
lightweight concrete offers the following cient depth of water.
advantages: By its very nature the caisson is a cellu-
lar type, usually a rather high, thin-wall
(a) Higher resistance to microcracking due to structure. Its interior walls (diaphragms)
reduced modulus of elasticity of aggre- are usually thin because design computa-
gates. tion allows for it. However, the designer
(b) Lower stress concentrations within the must be sure that project is easily con-
matrix to partially compensate for the re- structible. In reality, too thin walls become
duced aggregate strength. nearly impossible to build on a production
(c) Lower modulus of elasticity, which results basis with consistent quality. Sometimes the
in reduced stresses caused by shrinkage attempt to construct too thin and congested
creep, and the thermal effects. walls containing a double layer of reinforce-
(d) Lower values of thermal expansion, which ment with or without ducts for postten-
provides improved resistance to thermal sioned tendons may result in a much higher
cracking. cost than the construction of walls that pro-
(e) Caissons made from lightweight concrete vide ample thickness for construction.
have much smaller draft than those made Therefore, from the point of view of con-
from normal weight concretes and there- crete placement the minimum thickness of
fore require shallower waterways leading walls equal to 200 mm is usually considered
to site of deployment. for construction of cast-in-situ caissons; it
may be reduced to 150 mm in the case of
Recent research carried out in the United caissons assembled from prefabricated com-
States and Japan indicated that marine ponents. Usually, the thickness of all walls
lightweight concrete having a unit weight is largely dictated by the exposure to ambi-
of 1760-1920 kgjm 3 and compressive ent environment, by way of caisson con-
strength of 62 MPa can be made commer- struction (e.g., cast-in-situ or erection from
cially. For relevant information the reader precast components) and by strength and
is referred to the American Concrete Insti- serviceability requirements.
tute Report ACI 357/22-88 revised in 1993 The crack control requirements are basi-
(American Concrete Institute, 1988). Use of cally as follows: Cracks in the upper portion
5-6% silica fume in lightweight concrete of the face wall within tidal range and/ or
mix could result in a dramatic increase in ice action, in general, are not acceptable;
bond and concrete fatigue endurance. provided that epoxy covered reinforcing bars
Lightweight concrete was used for the are used for wall reinforcement, an accept-
construction of offshore caissons in the able width of cracks in this area can be
Canadian and Alaskan sectors of Beaufort limited to 0.1 mm. In all other parts of the
426 Gravity-Type Quay Walls

caisson structure, the crack width should conventional-type caisson sometimes may
not exceed 0.15 mm. reach up to 25% of the total caisson volume;
From the point of view of structural subsequently, the draft of the cast-in-situ
longevity the minimum wall thickness of caisson usually varies between 0.5 and 0.65
caissons used for permanent installa- of its height. On the other hand, the cais-
tions should not be less than indicated in sons assembled from prefabricated compo-
Table 5-l. nents are about 18% lighter than those cast
in situ. The draft of these caissons would
5.2.2.4 Reinforcement not normally exceed about 0.4 of their
height. The latter is attributed to the fact
Conventional or high-strength steel rein- that the caisson precast components are
forcing bars should be considered; highly typically fabricated horizontally to close tol-
corrodible high-strength wires should be erances at specially equipped facilities,
avoided. In all cases, however, the proper where material, labor, and quality control is
concrete cover over reinforcing steel as readily maintained. Therefore, these compo-
stipulated by recognized codes should be nents can be made thinner than those used
provided. in cast-in-place caissons. Precast panels can
The design of the conventional reinforce- be erected using standard lifting equip-
ment should embody relatively close spac- ment. The closure cast-in-place pours be-
ing and proper distribution of steel to tween prefabricated panels are sized to
minimize structural cracking and ensure allow for splicing of regular and postten-
proper resistance to shear. Prestressing of sioned steels. The pour joins between pre-
steel enables better control of structural cast panels and cast-in-place pour require
cracking. special consideration. The as-cast faces of
· The posttensioned tendons should be the precast panels are normally roughened
grouted according to the procedure that en- by sandblasting before erection. The mating
sures complete fill of ducts; thixotropic,
surfaces of the precast panels are usually
antibleed admixtures should be consid- coated with an epoxy bonding agent prior to
ered. Details of posttensioning anchorage
casting the closure pour. It must be remem-
protection require careful attention.
bered that epoxy exposed to cold tempera-
Normally, the concrete cover over rein- ture may be debonded.
forcement in a perimeter wall in splash and Under cold temperature conditions, the
atmospheric zones and the base slab should use of a cement wash or slurry may be
not be less than 50 mm, it can be reduced to appropriate. In addition to the earlier de-
30 mm in all other structural components of scribed case histories, a number of other
the caisson. recently built floating piers, caissons, and
Depending on the wall thickness and other structures assembled from prefabri-
caisson width, the volume of concrete re-
cated concrete elements are discussed in
quired for fabrication of cast-in-situ American Concrete Institute (1985, 1988)
and by Tsinker (1986). These structures
Table 5-1. Minimum thickness of concrete proved the viability and effectiveness of the
caisson members prefabricating technology.
Minimum Wall Thickness (rom)
Member Sea Water Freshwater
5.2.2.5 Launch, Tow, and
Installation
Face wall 300 150-200
Rear and side walls 200 150 To minimize construction cost, partly
Internaldiaphragr.n 150 150 completed caissons are floated out of dry
Base slab 300 200 docks to be completed at sheltered locations
Gravity-Type Quay Walls 427

while afloat. For this, several important ing to the procedure discussed in Chapter 2.
conditions must be considered: For towing long distances offshore, the cais-
son stability under damaged conditions
(i) The partly completed caisson must main- must also be considered.
tain an adequate freeboard. Forces occurring during caisson launch
(ii) It must be stable under conditions exist- and towing should be checked to ensure
ing during the voyage to the outfitting that they do not exceed loads greater than
facility. those for which it was designed. This should
(iii) It must be able to carry all temporary include both hydrostatic and motion-
construction loads. induced loadings. A suitable design crite-
rion is usually developed on a case-by-case
Completed caissons are towed to a de- basis.
ployment site where they are ballasted Extra care is required during caisson in-
down with water on the prepared stone bed- stallation. The caisson installation and re-
ding. The caisson towing is an important floating is usually accomplished by careful
and sometimes complex marine operation. ballasting of its compartments with water.
It is usually conducted when the environ- During installation, the caisson horizontal
mental conditions, such as wind, waves, and movements are controlled by winches via
currents, are acceptable for the towing op- cables attached to all four corners of the
eration. The presence of ice in a waterway structure. If during the first attempt the
andjor at the deployment site is usually caisson was set down with an unacceptable
considered as unacceptable for caisson deviation from the design position, then it
transportation and installation. The bathy- should be refloated to about 200-300 mm
metry along the tow route should be care- above the mattress level by pumping water
fully investigated. The selected tow route out, then winched into the right position
should ensure that the caisson has an ade- and ballasted down again. Sometimes this
quate underkeel clearance at all times. Due operation is repeated several times until
consideration must be given to the number the caisson is finally placed as designed
and size of tugs required for caisson towing within acceptable limits.
in the given environmental conditions.
It must also be realized that while under 5.2.2.6 Hazards During Caisson
tow, the caisson may sway excessively; Installation
therefore, to prevent caisson flooding, its
freeboard must protrude sufficiently above Lessons learned through experience sug-
the water level. Shallow water, with a small gested that extreme care should be exer-
underkeel clearance, reduces the speed of cised during caisson installation. The fol-
tow. Under such conditions, pusher tugs lowing hazards should be expected during
may be more effective. In cases of large installation of a typical caisson on prepared
caissons where a deep concrete skirt is pro- bedding:
vided at the base perimeter, the compressed
air may be introduced in the skirt compart- 1. Scour under caisson as it nears the
ments to reduce draft where required. This seafloor, due to acceleration of water flow.
operation, however, must be carefully moni- 2. Inability to control caisson's lateral move-
tored to prevent al.r escape due to caisson ments during the last meter of installa-
motion. Furthermore, the adequate skirt tion due to thrust from the escaping wedge
compartments must be provided to reduce of water underneath. This is exacerbated
the caisson instability due to excessive if seafloor is impermeable and if descent
free-surface effect. The required minimum is rapid.
underkeel clearance is determined accord- 3. Slip of anchor during caisson submer-
428 Gravity-Type Quay Walls

gence due to increased area of structure joints are usually keyed to prevent the fines
exposed to current. in backfill from being washed out.
4. Solitary bow waves from passing gauge Typical joints between adjacent caissons
tugs or launches create large dynamic which proved to be practical are depicted in
forces. Figures 5-27 and 5-28. The purpose of
5. Increased sediment content and salinity these joints is to protect the backfill mate-
gradient in water at the bottom causes rial from being washed out and to serve as
increased buoyancy which may exceed ca- a drainage seal. In some cases (e.g., when
pability of offset by ballasting. water can freely move through the mattress
6. Partial flooding of the large or long com- material), the space between caissons may
partment from in-leakage leads to the be jet grouted or closed by tremie concrete
free-surface effect and instability. or by concrete in sacks.
7. High winds that make control of tugboats
difficult.
8. Where the caisson is ballasted with plane
concrete, the heat generated by the large 5.2.3 ouay walls composed of
mass of concrete during hydration, which Large-Diameter Cylinders
will not be dissipated for some time due
to insulation, provided by the caisson's Bottomless large-diameter concrete cylin-
walls may cause substantial pressure ders are essentially an extension of the
against walls that may crack them. The floated-in concrete caissons technology. Ac-
amount of expansion can be minimized by tually, large-diameter cylinders are bottom-
selection of a low-heat-generating admix-
less caissons; following installation on a
ture, such as the blast furnace slag ce-
ment and by precooling of the mix. Alter- seafloor, they are filled with granular mate-
natively, slurried iron ore with a specific rial which, along with weight of a cylinder,
gravity of 2.5-3.0 can be considered. ensure stability of the complete structure.
9. It should be kept in mind that the exces-
In the past 35-40 years, this type of
sive pressure exerted on the base during quay wall construction has been carried out
grouting can lift the caisson. Therefore, at several locations in Western Europe and
the gravity feed is usually recommended in the former Soviet Union.
to keep this pressure under control and is Concrete cylinders may be placed on rub-
also limited. The sealing course of grout ble mattress or submerged below the mud
over the stone bed should contain an anti- line. The latter is usually done to prevent
washout admixture to prevent excessive soil heave in front of the wall and to provide
penetration into the stone bed and to be protection from erosion. The cylinders may
self-leveling. The underbase grout should be installed into prepared cuts in the
have thixotropic, antibleed, and anti- seafloor or can be jetted into the seabed
washout properties. Heat development
due to grout hydration must be consid- with help of the water jet.
ered. Because large-diameter concrete cylin-
ders in general are very heavy there is a
need for heavy lift equipment to handle
5.2.2.7 Joints
them both offshore and onshore. Sometimes
The space between adjacent caissons to mitigate the problems, cylinders are cut
must be given careful consideration. It by horizontal joints into two or more units.
should be such that the expected uneven Alternatively, the size of the cylinders (as
settlement of the caissons when set in place well as weight) may be reduced by installa-
is absorbed without damage during the tion of soil-pressure-relieving platforms
caisson ballasting and backfilling. Caisson which helps to reduce lateral thrust on the
Gravity-Type Quay Walls 429

(8) (b)
Water Side
L t. joint L '"30 0 L t. j oint L

~A / t ys
\ ) \ )

t4ilo0.2
;>

c5"'o.s t3 .. 0 .2

Detail ' A ' Detail 'B'

Figure 5-27. Typical joints between adjacent rectangular concrete caissons: 1-caisson;
2-rubber seal; 3- concrete plank; 4- graded gravel or crushed stone; 5- granular fill;
6- grout.

structure and subsequently allows the use installed to cover the space between the top
of cylinders of a smaller diameter. The of a cylinder and the final grade at the quay
aforementioned platform also helps to dis- edge.
tribute bearing pressures at the cylinder The concrete superstructure can also be
base more evenly. In the case of substantial built in the form of miscellaneous prefabri-
fluctuation of the water level (say more than cated L-shaped walls, or erected from pre-
3.0 m) in front of a quay wall, the h eight of fabricated components secured by conven-
the cylinders may be such as to allow for tional steel tie rods at deadmen.
installation at the low water level. Then the The typical diameter of cylinders used
prefabricated superstructure similar to that for quay wall construction usually varies
discussed in the preceding section may be between 9 and 12 m. Cylinders of this size
430 Gravity-Type Quay Walls

( a) (b)

(b) (c)
t
,
~
.1.1:

~ :;:; .

m ..
~

i:i:i!i!!ii!"::
·:'JI '.!./

Figure 5-28. Typical joints between adjacent cylindrical caissons. !-Cylindrical caisson;
2-concrete diaphragm; 3-rubber seal; 4-grout; 5-tremie concrete; 6-granular fill.

are normally used for construction of quay The deep water quay wall built at Port-
walls of up to 13-16 m high. Larger cylin- of-LeHavre, France in the middle of the
ders would require use of an unreasonably 1950s was, perhaps, among the world's first
heavy superstructure needed to span the major quay walls successfully constructed
space between adjacent cylinders. from large-diameter concrete cylinders (Fig.
In the case of deep water structures 5-30). This 21.5-m-high structure was built
where the depth of the water exceeds 13.0 from concrete cylinders 22.0 m long and
m in addition to the already mentioned 11.0 min diameter with a wall thickness of
pressure-relieving platforms, the combina- 0.92 m. The wall had 40 voids 0.7 m in
tion of two cylinders may also be consid- diameter made around the cylinder perime-
ered. In the latter case the rear cylinder can ter. This helped to reduce the weight of
be not only shorter but also thinner than each cylinder to about 350 tonnes. The can-
the front one. tilever cast-in-situ concrete slab overhang-
Basic types of structural schemes of quay ing the wall by 5.75 m was placed on the
walls constructed from large-diameter top of cylinders. This slab constituted part
cylinders as discussed earlier are depicted of the wall concrete superstructure and
in Figure 5-29. helped to reduce soil lateral thrust on the
The following is a brief discussion of sev- wall. The space between adjacent cylinders,
eral characteristic case histories where 1.5 m, was bridged with special prefabri-
large-diameter concrete cylinders have been cated concrete elements driven in the foun-
successfully employed. dation soil between and in close proximity
Gravity-Type Quay Walls 431

(a) (b) (c) (d)

(e ) (f) (g) (h)

Figure 5-29. Basic type of quay walls, built from large-diameter concrete cylinders. (a)
Cylinders submerged into foundation soil, alternatively installed on a rubble mattress; (b)
cylinders made up from several sections installed on each other in the process of wall
construction; (c, d) cylinders with a prefabricated superstructure; (e, 0 cylinders with a pressure-
relieving platform and slab, respectively; (g, h) double-cylinder walls.

A-A PLAN

Figure 5-30. The quay wall built in Port-of-LeHavre, France. !-Concrete cylinder; 2-fender
system; 3-soil-pressure-relieving platform; 4-filter made from granular material; 5-voids
0.7 min diameter, in a cylinder wall; 6-prefabricated concrete elements used to seal the
space between adjacent cylinders.
432 Gravity-Type Quay Walls

to the cylinders. The space between these


elements and the cylinders was filled with
tremie concrete.
A unique construction method was em-
ployed for fabrication and installation of the
aforementioned giant cylinders. The bottom
part of the 4.0-m-high cylinder was cast at a
yard located in a close proximity to the edge
of the existing quay wall. Next, the com-
pleted wall units were picked up by a 200-
tonne-capacity floating crane and installed
on a berm placed 3.0 m below the water
level. There, cylinders were extended to
10.80 m high, then reinstalled by the afore-
mentioned crane on a new berm, this time
located 9.0 m below the water level.
At the new location, the cylinders were
completed to the full height, and later with
help of a 370-tonne-capacity special pontoon 1
-rr
I I
they were brought to the construction site ~. ,
located about 2 km from the site of fabrica- I
tion. There, the cylinders were sunk into I
A
the foundation soil to the depth of 4.5 m
under own weight; in the process, the foun- .-t.
dation soil was loosened through the voids
located within the cylinder wall, then air-
lifted. Use of the aforementioned soil exca-
vation technology left the soil inside of the
cylinder almost undisturbed. In some in-
stances, the weight of the water pumped
into a special tank placed on the top of a
cylinder was added to the weight of the Figure 5-31. Port-of-Dunkirk (Dunkerque), France.
cylinder to expedite the process of sinking. Typical cross section through berth 6. !-Concrete
In all cases, compressed air was added to cylinder; 2-cope beam; 3- web walls at 19.0-m cen-
ters; 4- grout; 5- sand.
the water jet to facilitate the process of soil
liquefaction. After final positioning, all
cylinders were filled with sand, and all voids
within the cylinder wall on the water side perstructure. The cylinders were provided
were filled with lean concrete and sand. with cutting edges to allow for better pene-
Finally, the wall was backfilled with se- tration into the foundation soil (sand).
lected granular material. The structure was designed for the very
Another quay wall comprised of large- heavy surcharge load of 200 kNjm 2 . Cylin-
diameter concrete cylinders was built in ders were slipformed in the dry to full height
the early 1960s at the French Port-of- at the design location, then sunk under its
Dunkirk (Dunkerque) (Fig. 5-31). This own weight at an average rate of about 2 m
structure was comprised of concrete cylin- per day to the design depth by removing
ders 19 min diameter, 18-23 m high with a soil from inside. The penetration depth of
wall thickness of 0.4 m and a concrete su- 4-9 m varied, depending on local geotechni-
Gravity-Type Quay Walls 433

cal conditions; reportedly, the cylinders had cylinders were filled with a granular fill,
no tendency to drift away from their design the massive concrete cap was cast in situ on
position or to tilt significantly during the the top of cylinders. The space between ad-
process of sinking. The space between adja- jacent cylinders was grouted.
cent cylinders was grouted and the cylin- The quay wall comprising of two cylin-
ders filled with sand. ders installed back to back was built at Port
Two unusual quay walls were built at La Rochelle (Fig. 5-33). The front cylinders,
Port Fos-Marseilles, France for the mar- about 11.8 m in diameter and 18.0 m high,
ginal wharf and the pier at oil terminal had uneven wall thicknesses; the front por-
(Gorunov et al., 1974) (Fig. 5-32). These tion of these cylinders was 0.5 m thick,
walls were composed of concrete cylinders whereas thickness of the rear part was 0.3
with unequal wall thicknesses; the front m. Furthermore, the front wall was 1.5 m
wall of these cylinders facing the basin had thick at the top of the cylinder to accommo-
a wall thickness equal to 0. 7 m, and on the date the powerful fenders and to be able to
fill side it was equal to 0.4 m. The cylinders, absorb the heavy ship impact load. The rear
10.5 m in diameter and 16.0 m high, had cylinder was only 13.0 m high and had even
been in-situ assembled from two sections wall thickness of 0.3 m. Both cylinders were
each 8.0 m high. The upper section had a joined together as indicated in Figure 5-33
horizontal diaphragm which cantilevered and installed on concrete foundation slabs;
1.15 m from the back of the cylinder. This space between these slabs was filled with
diaphragm helped to reduce the lateral soil crushed stone. Finally, cylinders were filled
thrust on the wall and to equalize the bear- with granular material and capped with a
ing pressure at the cylinder base. After concrete superstructure.

Figure 5-32. Quay wall at Port Fos-Marseilles, France built from large-
diameter concrete cylinders with unequal wall thicknesses.
434 Gravity-Type Quay Walls

8
-+

s-s A-A

Figure 5-33. Quay wall comprised of double cylinders built at Port La Rochelle, France.

Obviously, to erect the wall composed of


large-diameter concrete cylinders, heavy lift
equipment is required. As stated earlier,
the required capacity of the lifting equip-
ment can be reduced by splitting the cylin-
ders into two or more units. Sometimes de-
signers try to reduce the weight of the wall
units by reducing the wall thickness. An
example of the thin-walled concrete cylin-
ders used for construction of the quay wall
at Port-of-Novorossiysk, Russia is depicted
in Figure 5-34 (Smirnov et al., 1979). There,
concrete cylinders 11.1 m in diameter and
12.0 m high with a wall thickness of only
0.2 m have been used. Each completed Figure 5-34. Quay wall built at Port-of-Novorosiysk,
cylinder was comprised of two units, each Russia. 1, 2-Concrete cylinders; 3, 4- cast-in-situ
6.0 m high and weighing 100 tonnes. The components of the concrete superstructure; 5- front
top unit was guided in place by a special track of the craneway; installation guide.
guiding system installed at the top of the
bottom cylinder. In conclusion of this section, it should be
The guiding system was left inside of the noted that in some cases the large-diameter
complete cylinder and kept both units firmly concrete cylinders used for quay wall con-
together. Finally, the cylinders were filled struction have been built from prefabri-
with granular material, capped with a con- cated curved panels joined together by vari-
crete superstructure and backfilled with se- ous methods, and some walls of limited
lected granular material. Another example height (up to 10 m) have been built from
of a similar wall is given in Ciortan (1994). polygonal "cylinders," assembled from flat
Gravity-Type Quay Walls 435

concrete panels up to 2.8 m wide, 7.5-8.5 m level. This type of construction adapted well
long, and 20-23 em thick (Smirnov et al., to the physical condition in harbors and
1979). rivers, being resistant to the effects of ice,
There is a number of factors that influ- currents, waves, and foundation scour, and
ence design of individual panels; for exam- at the same time possessing the merits of
ple, minimum wall thickness requirements, ease and speed of construction similar to
number of panels per cylinder and method that characteristic of concrete caissons and
of joining of these panels together, and large-diameter cylinders.
weight of the individual panel and its im- In most cases, timber cribs have asym-
pact on construction procedures and equip- metrical cross sections which are basically
ment requirements. The obvious advantage dictated by the economy of ·construction;
of the flat panels as opposed to curved units asymmetrically shaped cribs required less
is the use of a very simple formwork. timber.. The maximum height of timber
Large-diameter cylinders fabricated from cribwork walls is limited by the strength of
steel were also used for marine application. the timber and usually does not exceed
An example of a 10-m-diameter steel cylin- 15-17 m. A typical cross section of the tim-
der used for harbor construction in Japan is ber cribwork quay wall 2960 m long built in
given in Bruun (1989). Another example of the former Soviet Union (Smirnov et al.,
the gravity-type wall constructed in Japan 1979) is depicted in Figure 5-36. This crib
from large-diameter steel cylinders is illus- of solid wall construction was built from
trated in detail in Figure 5-35. logs 22-25 em in diameter. The length of
the individual crib unit was 43.0 m. The
concrete superstructure was built on the
top of a crib at about 0.5 m above the lowest
5.2.4 Cribwork Quay Walls
water level in the river. This wall proved to
be resistant to the heavy ice effects.
5.2.4.1 Timber Cribwork Walls
As stated earlier, timber as structural 5.2.4.2 Concrete Cribwork Walls
material has proved to be extraordinarily
durable when submerged below water level These are composed from precast units
and, in fact, appeared to be almost inde- erected in situ. Concrete cribs naturally
structible when submerged in fresh water provide a substitute for timber structures
or cold water. It is also inexpensive where where wood is scarce or otherwise unsuit-
available locally. In the latter case, timber able for marine application (e.g., where mol"
cribwork may represent a competitive alter- lusks are present). To date, various concrete
native for construction of the quay wall. units have been used for crib constructions:
These structures, although requiring a great from simple square or rectangular in cross
deal of manual work, generally prove to be section, to sometimes quite complicated
practical and economical, particularly for shapes.
construction at remote locations. The basic principles of concrete crib de-
Timber cribs are normally built ashore, sign are similar to that used for the design
then launched, floated, and sunk into posi- of timber cribs. In general, a crib filled with
tion, either directly on prepared stone bed- granular material is designed to perform as
ding or on the natural bottom. Normally, a gravity retaining wall. An example of a
cribs are filled with stone rubble to form a 6.5-m-high concrete crib quay wall built on
gravity-type base for mass concrete or con- the Angara River, in the former Soviet
crete masonry superstructure walls which Union is depicted in Figure 5-37 (Gurevich,
are placed on the cribwork at a low water 1969).
436 Gravity-Type Quay Walls

(a) (b)

(c)

Figure 5-35. Construction of a gravity-type wall comprised of large-diameter steel cylinders.


(From commercial brochure by Sumitomo Metals.) (a) Transportation of prefabricated steel
cylinders from the fabrication yard to the site of installation. (b) Installation at predetermined
location. After installation on the soft soil bed, the cylinders are submerged to a certain depth
with. the help of a set of powerful vibrators. (c) After installation, the cylinders are filled with a
good quality granular material. (d) Installation of a steel arc. Arcs that are installed on both
sides of the wall are joined with the cylinders via an interlocking system as illustrated in (e) and
(0. (e) Interlock. (0 Concrete grout.
Gravity-Type Quay Walls 437

Figure 5-35. Continued.

Similar cribs, although of different type of crib proved to be time-consuming.


heights, have been built at some other loca- The latter was basically attributed to the
tions in Russia. The weight of the crib's need of handling the large number of mis-
individual components typically do not ex- cellaneous crib components.
ceed 1.5 tonnes and, therefore, they can be A simple concrete cribwork structure is
easily handled elsewhere, without the need illustrated in Figure 5-38 (Tsinker, 1965).
of heavy lifting equipment. The face wall of Several small passenger docks of this type
the concrete crib illustrated in Figure 5-37 have been built on the Dnieper River,
was constructed from T -shaped elements. Ukraine. Cribs of this type are generally
The gravel filter was placed behind the front suitable for construction where foundation
wall to prevent the granular fill material soil permits pile-driving. The piles are used
from being washed out. Construction of this as guides for speedy installation of crib
438 Gravity-Type Quay Walls

Figure 5-35. Continued.

2.

Figure 5-36. Timber crib quay wall built in Russia. [From Smirnov et al. (1979).]
1-Timber crib; 2- concrete superstructure; 3-rubble mattress; 4-riprap;
5- gravel ffiter; 6- hydraulically placed sand.
Gravity-Type Quay Walls 439

9 6 5.2.5 steel Sheet-Pile Cell


Bulkheads
Generally, gravity-type quay walls con-
structed from steel sheet-pile cells due to
their simplicity are quite popular among
designers of ports and harbors. These walls
are commonly used where the depth of the
water does not exceed about 15.0 m, and
the bottom conditions are suitable to sup-
port this type of structure.
~~&~~~ffi~;;~~~7~
1 "
The cellular-type bulkheads have a long
history of successful use as construction
cofferdams and waterfront retaining walls.
The subject of design and construction of
cellular-type retaining structures has been
extensively discussed by Terzaghi (1944),
Cummings (1957), Tennessee Valley Au-
thority (1957), White et al. (1961), Swatech
(1967), Lacroix et al. (1970), Tschebotarioff
(1973), Soros and Koman (1974), Schroeder
et al. (1977), Schroeder and Maitland (1979),
Rayburn (1981), Sorota and Kinner (1981),
Carchedi and Porter (1983), Schroeder
l>ETAI.L 'A' (1987), Sekiguchi et al. (1992), and others.
Some cases of failure are discussed by
Figure 5-37. Concrete cribwork, Angara River, Rus-
sia. [From Gurevich (1969).] 1-4-Cribwork elements; Tscheboterioff (1973), Thomas et al. (1975),
5-concrete cap; 6-gravel filter; 7-fender; 8-rubble and the U.S. Army Corps of Engineers
mattress; 9-sand. (1974).
Sheet-pile cells are typically constructed
from flat web steel sheet piles and are usu-
ally filled with granular soil or rubble. When
fill is placed inside the cell, the sheet piling
acts in tension. The pile-fill system forms a
components underwater; however they also gravity-type structure able to resist lateral
provide the cribwork with a better resist- and vertical loads and forces. Cells used for
ance to horizontal forces (e.g., ship impact dock structures may be circular or have
and ice load). Timber guide piles driven circular ends and straight webs (Fig. 5-39).
from inside of a cribwork are used for pre- In the former case, a continuous wall is
cise installation of the crib components. constructed from cells connected together
The completed crib is filled with rock. by arcs of smaller diameter located on both
This type of cribwork construction proved to sides of the wall. In cases where the stabil-
be economically viable and particularly ity of the bulkhead does not present a prob-
suitable for construction of docks in small- lem, internal arcs can be eliminated (Fig.
craft harbors, fishing ports, and for small 5-39b). Arcs generally intercept the circles
passenger terminals, specifically those lo- at a point making an angle of 30° or 45°
cated in remote areas. with the longitudinal axis of the wall, and
440 Gravity-Type Quay Walls

(a)

,,,,
z----_tl
~}

-+
A

+-
A
-+--9-...-r:r- --rr-.-.-d- - - - - ---i.-- -''----l:'G_

PLAN
(b)

Figure 5-38. Concrete cribwork. [From Tsinker (1965).] (a) Artistic impression.
(b) Typical cross section and plan: !-concrete pile 30 X 30 em; 2-concrete pile
25 X 25 em; 3-timber piles; 4-crib components; 5-rubble; 6-cast-in-situ
concrete superstructure; 7-approach trestle; 8-timber fender; 9-catwalk;
10-concrete slab.
Gravity-Type Quay Walls 441

the minimum radius of arc (as well as of a


4 2 cell) is determined by the sheet pile maxi-
(a) mum rotation angle at the interlocks (Fig.
5-40). Depending on the pile fabrication,
this angle typically varies between 10° and
12°. When a smaller arc or cell radius are
required, special bent piles are also avail-
able.
Junction points in cellular quay walls
require special prefabricated pieces, com-
monly 90° T's and 30°-50° and 120° Y's
(b) (Fig. 5-41).
Besides continuous quay walls, free-
standing sheet-pile cells are also used in
port and harbor engineering. These struc-
tures are basically used as mooring andjor
breasting dolphins (Fig. 5-42), or as foun-
(c) dations for the heavy cargo handling equip-
ment.
The cell diameter or width of the bulk-
head basically depends on the depth of the
water, the type of fill placed inside and
outside of the cell, and the maximum live
load anticipated. The amount of steel per
unit length of a quay composed of circular
cells is almost independent of the cell diam-
eter and, therefore, the economy of the
6 structure is basically tied to the difference
(d) in the cost of placing fill in the cells and
placing of backfill material. However, an
increase in cell diameter would result in an
increase in sheet-piling interlock tension
and also in a heavier superstructure.
It should be noted that circular cells are
used more often than diaphragm cells, as
each individual circular cell may be filled to
the top and be stable in itself. In the case of
straight web (diaphragm) cells for which
straight transverse walls are not designed
TYPICAL CROSS SECTION to withstand substantial lateral load, the
Figure 5-39. Cellular quay wall; typical arrange- filling operation is usually conducted on
ments. (a) Circular cells, connected by semicircular several cells simultaneously, in which case
arches on both sides; (b) circular cells, connected by the difference of fill level in adjacent cells
semicircular arches on external only; (c) cells with
circular ends and straight webs; (d) typical cross sec-
should usually not exceed 1-2 m.
tion. !-Circular cell; 2-arc; 3-straight web cell; The straight web cells filling usually re-
4- granular fill; 5-backfill; 6-concrete superstruc- quires more time than that used for filling
ture; 7 -fendering. of circular cells. The latter is a very impor-
tant consideration where the construction
442 Gravity-Type Quay Walls

L L

. .-_~~~--111111111;~-
1 I
Figure 5-40- Steel sheet-pile cellular bulkhead composed of circular cells. Typical layout,
basic parameters, and detail.

time is limited by the weather window or by placements in the field during andjor fol-
a short construction period, such as may be lowing construction without sustaining ap-
the case in cold climate regions. preciable damage. This is discussed later in
Another important consideration is quay Section 5.8. Because sheet-pile cellular walls
serviceability. Because circular cells are are normally treated as being rather flexi-
self-supporting and therefore independent ble, their design usually includes limita-
of adjacent cells, the damage of one cell will tions for allowable horizontal displacement.
not affect the integrity of the adjacent cells; Several factors can affect cell deflection.
however, damage of a straight web cell may These include the shape of a cell andjor
produce a domino effect on adjacent cells interconnecting arcs, the elevation differ-
due to the failure of straight (diaphragm) ences around the cell, the amount of dredg-
walls. The latter type of construction how- ing required inside the cells, the sequence
ever, although requiring some sheet piling of dredging and filling operations, and the
per unit length of the wall, can be widened type of backfill and its densification.
easily by increasing the length of the The construction sequence is very impor-
straight walls. This increase will not raise tant because the cell acts in hoop tension,
the interlock tension, which is a function of and any pressures opposed to those causing
the radius of the arc portion of the cell. hoop tension before the cell is filled can
Obviously, the interlock stress in straight cause it to collapse. The construction
web wall construction is smaller than sequence usually involves the following
that at the joints of a circular cell of equal operations.
design.
All types of cellular wall can undergo (a) Preexcavation (predredging) typically in-
relatively large vertical and horizontal dis- volves removal of silt, debris, and excess
Gravity-Type Quay Walls 443

J>

~ - 0

Figure 5-41. Typical components of cellular quay wall.

Figure 5-42. Steel sheet -pile cell used as a mooringj breasting dolphin.
444 Gravity-Type Quay Walls

overburden that might prevent the proper ~,---,,~


driving of sheet piles. The extra cost of I \
-preexcavation is generally compensated I
I
\
I
by savings in driving the sheet piles. \ ,__.oj
\ I
(b) Setting the template is the next step. A ~"'- _... ~
template is a rigid frame stiffened by at - L"2
least two horizontal rings placed at least (a) (b)

3.0 m apart. The template has the same


3
plan shape as the cell. A clearance of
about 100 mm is usually set between the
perimeter of the template and the inside
face of the cell. When installed, the tem-
plate is secured firmly in place by sturdy
spuds.
(c) Setting and driving sheet piles. During
this stage, all sheet piles are inspected for
(c)
straightness, lamination, and interlock
shape. Next, the sheet piles are set and
closure made prior to any driving, in or-
der to maintain the design shape. Closure
5
of the cell is often made with a three- or
four-sheet pile set. At this point, all sheets
must be plumb and interlocked with each
other. Then the sheets are driven usually
in pairs to their design position. Typi-
cally, the pair of sheets are not driven
more than 1.5-2.0 m in front of adjacent (d)
sheets. If underlaying rock is located close
Figure 5-43. Typical sequence of cellular wall con-
to the design dredge line, then sheet piles
struction. [From Lacroix et al. (1970).] (a) Install tem-
are driven individually into the rock or to plate (1) and set connection sheet pile T or Y (2). (b)
refusal on rock. Mter all sheets are driven continue setting piles (4) proceeding away from TorY.
to the grade, they are trimmed to their (c) close first cell (4) and set about three sheet piles in
final elevation. As it is difficult to handle each arc wall (3); drive sheet piles of first cell and arc
sheet piles longer than 25.0 m without walls (3) in steps. The first cell may be filled. Install
deforming them, they can be spliced to- arc cells in second cell (5)similarly to that which was
gether if longer sheets are required. Sheet done in cell (4). (d) close, drive, and fill arc cell.
splices would normally be required to be
staggered at least 1.0-1.5 m vertically.
Experience indicates that the sheet ened to provide better resistance against
piles used in cellular wall construction cell sliding on the base.
should not be driven to more than 5-10 m (d) Filling Cells. After all sheet piles are
into the foundation soil. It is preferable to driven to the design depth, the cell is
maintain this value at 3-5 m. Excessive filled with free-draining granular mate-
sheet-pile-driving may cause the sheets to rial; the excessive presence of fines in the
split out of the interlocks and curl the fill may prevent rapid drainage that, in
toes. The typical sequence of sheet-pile- turn may be the cause of excessive inter-
driving is depicted in Figure 5-43. lock tension. Circular cells are normally
In the case where the bedrock founda- filled after adjacent connecting arcs are
tion is not covered with overburden soils, installed. This is because during filling,
the sheet piles are set directly on the rock process circular cells would be slightly
surface, without any driving. If practical, deformed in both the vertical and hori-
the surface of the rock should be rough- zontal directions, which could sometimes
Gravity-Type Quay Walls 445

make installation of arcs difficult or in In some cases, depending on local condi-


some cases even impossible. tions the sequence of cell construction can
Fill is usually placed in the middle of be somewhat different; for example, it can
the cell in order to prevent the concentra- be as follows:
tion of material at one side only. The
latter may cause serious cell distortion.
Normally, fill is placed with a clam shell (a) Adjacent cells are set and installed.
or hydraulically. (b) Interconnecting arcs are placed and
Loose, submerged granular fill has liq- driven.
uefaction potentials. Therefore, to prevent
fill liquefaction, it must be properly densi- (c) Poor soil inside and between cells is
fied. Compaction of a granular fill such as dredged (excavated).
sand is best achieved by pumping water (d) Adjacent cells are filled.
into the cell when filling above the water (e) Space between adjacent cells is filled.
level. Experience indicates that this
method helps to consolidate material (f) The superstructure is constructed.
within the cell without overconsolidation. (g) Backfilling is accomplished.
Granular fill can also be consolidated by
just the cyclic rise and fall of the water
level in the harbor. Fill overdensification This construction procedure, however, may
may result in heavy interlock tension and not be suitable for the construction of the
in some cases, can burst the cell open large-diameter cells because a deep excava-
(Brown and Forrest, 1983). tion inside the cell may cause the cell to
It is generally difficult to control fill collapse inward or to change shape under
densification. In the current practice, fill active lateral pressures produced by an
density is usually determined by the overburden from outside of the cell. The
Standard Penetration Test (SPT) or by innovative procedure was utilized for con-
taking measurements of the volume of fill struction of the artificial transshipment is-
placed and documenting the compaction land (terminal) in the Atlantic Ocean, 8
effort necessary to obtain the specified
densities for the follow-up work.
miles off the coast of Brazil near Areia
The fill density may also be determined Branca (Soras and Koman, 1974).
by measuring the amount of energy and There, the island's 12.5-m-high perime-
penetration of the vibrating probe itself ter wall consisted of 26 steel sheet-pile cells
(Brown and Forrest, 1983). Some authori- 17.55 m in diameter topped with a concrete
ties, however, believe that considerably seawall. Cells were preassembled around
more field and theoretical documentation the template on the deck of a barge and
is required before the latter method can installed into position by a 50-tonne-
be proved reliable (Sorota et al., 1983). capacity floating derrick crane. Then sheet
(e) Construction of the superstructure. This is piles were driven to a penetration of ap-
usually accomplished by construction of a proximately 4.6 m. As soon as each cell was
concrete wall similar to that used in large driven, the template was removed and the
concrete cylinder walls. In some cases, the cell was filled with suitable material
continuous concrete slab is placed on the
top of cells. A stiff concrete slab placed
dredged from the seafloor. Construction of
within the cell may contribute greatly to the perimeter wall was completed by driv-
minimizing differential movement when ing connecting arcs, filling the space inside
large total movements are experienced the perimeter wall with dredge from the
(Schroeder, 1987). seafloor material, placing an extensive
(f) Backfilling is the final operation. Good riprap protection in front of sheet-pile cells,
quality granular material is usually used and constructing the seawall on the top of
for the backfill. the cells.
446 Gravity-Type Quay Walls

5.2.6 Quay Walls


composed of L-Shaped
Prefabricated Units
This type of construction has been devel-
oped as an alternative to other quay wall
structures (e.g., blockwork wall or walls
(a) (b) (c)
composed of concrete floated-in caissons).
L-shaped walls are basically used where Figure 5-44. Typical precast units used for con-
foundation conditions do not permit pile- struction of L-shaped quay walls: (a) simple cantilever;
driving. These walls are usually founded on (b) one counterfort; (c) two (or more) counterforts.
a thoroughly leveled rubble mattress under-
laid by a competent foundation material
(e.g., bedrock, dense sand, or hard clay). wall height exceeds 7.0 m it must be stiff-
When built in the wet, the L-shaped walls ened by the counterforts.
are erected from prefabricated concrete The obvious disadvantage of counterfort
units. The sizes and, particularly, the length walls is the need to use rather complex and
of the precast units are basically governed expensive (although reusable) forms for
by the availability of the heavy lift equip- casting of wall units. Counterfort units as-
ment. sembled from precast components (e.g.,
A variety of different types of L-shaped foundation slab, front panel, and counter-
concrete wall has been constructed in the fort) (Fig. 5-45), have been developed and
wet in recent years. All of them were made practically implemented relatively recently
of units precast onshore, then transferred in Russia. This type of wall unit is typically
and installed underwater by floating crane. fabricated with one or two counterforts.
The advantages of L-shaped walls are as Precast elements are joined together by
follows: welding together steel-embedded parts
placed in these elements. Subsequently,
(i) L-shaped walls require less concrete than joints are concrete grouted. These elements
blockwork or caisson walls. can also be joined together by posttension-
(ii) Wall units are typically precast in a con- ing.
trolled environment at yards with estab- As stated earlier, the length of the pre-
lished quality control procedures. Hence, cast wall unit depends on the capacity of
the quality of these units are under strict available cranage and it normally varies
control. between 3 and 12 m. Where crane capacity
(iii) Precast units are typically cast in is limited, the wall units may be composed
reusable formwork. of prefabricated elements such as base slab
and vertical panel assembled in place (Fig.
An obvious disadvantage of this type of 5-46). Here, the vertical panel can be an-
construction is the need to use heavy lift chored either by a stiff metal anchor bar
cranes both onshore and offshore. hinged at the base slab and field welded at
Typically, L-shaped walls, depending on the upper end of the face panel (internal
their height, are configured as simple can- anchorage) (Fig. 5-46b) or by a conven-
tilever or counterfort walls (Fig. 5-44). The tional tie rod system secured at the anchor
height of simple cantilever walls usually slab similar to that used in conventional
does not exceed 5 to 7 m and depends heav- sheet-pile bulkheads (external anchorage)
ily on wall loading conditions. When the (Fig. 5-46a).
Gravity-Type Quay Walls 447

Figure 5-45. L-shaped counterfort wall composed of prefabricated


units: (a) Typical cross section; (b) prefabricated components. !-Cast-
in-place part of the counterfort; 2-prefabricated counterfort; 3-
geosynthetic filter; 4-prefabricated front panel; 5-wall coping;
6-rubble mattress; 7-gravel filter; 8-granular backfill.

In the case of a wall with external an- (ii) Next, the frontal panels (1) and anchor
chorage, the erection process includes the blocks (11) are installed and firmly fixed
following basic phases (refer to legend for in place. It is usually done with the help
Fig. 5-46 for numbers in brackets): of temporary props installed between the
base slabs and frontal panels, and by
(i) First the base slabs (2) are placed on a short timber piles driven in front of the
thoroughly screed rubble mattress (5). anchor slabs.
448 Gravity-Type Quay Walls

II

(a)

,· ~ •.: . '.

8
7

~1_·7--~-.~,.,1~;0e:rv=·?.2£21~~ ,_
I
A-"
---

5
(b)

Figure 5-46. L-shaped quay wall erected from prefabricated components.


Typical cross sections: (a) with external anchorage and (b) internally anchored.
1-frontal panel; 2-base slab; 3-steel anchor bar; 4-concrete coping
beam; 5-rubble mattress; 6-gravel filter; 7-geosynthetic filter; 8-granular
backfill; 9-timber pile; 10-steel tie rod; 11-prefabricated concrete anchor
block; 12-timber log.

(iii) Operations (i) and (ii) are followed by the dure. The advantage of the wall with an
installation ofthe tie rods (10) and gravel external anchorage is that the bearing
(6) and synthetic (7) filters. stresses at the interface of the base slab
(iv) The operation is completed by placement and stone bedding are almost uniform. This
of the granular backfill (8) which pro- type of construction is usually recom-
ceeds from the anchor slabs toward the mended where the foundation soil is rather
frontal panels. weak and thus sensitive to heavy bearing
stresses.
As follows from the above, the obvious Wails with internal anchorages do not
disadvantage of this type of construction is need anchor slabs; steel anchor bars used in
its quite complicated construction proce- this type of construction, although heavier
Gravity-Type Quay Walls 449

in cross section, are much shorter than those


used in walls with external anchorages. The
disadvantage of the internal anchorages is
that anchor bars can be easily damaged
during the wall backfilling operation; the
latter is particularly true if rockfill-type
backfill material is used. In general, con-
struction of several quay walls with inter-
nal anchorage in the former Soviet Union
proved to be less (by 10-12%) expensive
than walls with external anchorages
(Gurevich, 1969).
The construction sequence of L-shaped
quay wall typically involved the following
phases:

(i) Manufacturing of wall units. Essentially,


all structural elements must be propor-
tioned in a way to resist all design loads
and particularly those attributed to lift- Figure 5-47. L-shaped counterfort wall built at
ing, transferring, and installation. Inkoo, Finland. Typical cross section. (From Pitkiilii
(1986).] 1-Double counterfort unit; 2-cast-in-situ
(ii) Preparation of rockfill mattress. Typi- concrete cope wall; 3-concrete slab for scour protec-
cally, the top level of stone bedding tion; 4- mattress; 5-originallevel ofbedrock; 6-ex-
is screed to an accuracy of ± 50 em cavated level of bedrock; 7-rubble fill; 8-pavement;
measured from the design elevation. 9-craneway track.
(iii) Installation of wall units side by side.
The gap between walls should not exceed
50 mm and maximum deviation from the
ements. The quay was founded on a layer of
vertical typically should not exceed 1 : 50.
crushed stone, placed over partly excavated
(iv) Installation of filters at joints between
bedrock. The bedding was protected from
adjacent wall units to prevent granular
the scour induced by ship propellers by pre-
backfill from being washed out. In some
fabricated concrete slabs placed along the
specific cases, joints may also be cement
grouted. seaward base of the wall. The wall was
backfilled with a quarry run material of
(v) Backfilling. This is usually done with a
mixed co:mposition.
good quality granular material or rock-
fill. Another example of a very high counter-
fort wall built at the port located on the
(vi) Cast-in-place or installation of prefabri-
cated cope structure. northern foreshore of Bantry Bay, Australia
is discussed by Moss-Morris and Hodge
Pitkalii (1986) described an example of (1981) (Fig. 5-48). In this case, 368 single
successful construction of the coal wharf counterfort, 360-tonne, 18.65-m-high, 6-m-
built in 1981-1982 at Inkoo, Finland. This wide concrete units have been used to build
240-m-long, 21.0-m-high quay wall was a 2.2-km-long quay wall. Units were precast
composed of double counterfort 18.6-m-high in three separate concrete pours in steel
concrete units (Fig. 5-47). These 450-tonne formwork which was straddled by a large
units were installed in water 17.5 m deep. gantry crane.
Joints between adjacent units were over- The latter wall units were installed side
lapped by special prefabricated concrete el- by side on a crushed stone mattress 16.74 m
450 Gravity-Type Quay Walls

below sea level. Joints between adjacent experimental rapidly assembled wall
units were subsequently cement grouted by structure that could be used basically in
utilizing special techniques. The cast-in-situ military applications. Because of the experi-
capping beam was installed on top of the mental nature of this project as well as due
units after allowing the wall to settle under to local site conditions and specific technical
the influence of backfill and sand sur- requirements, the structure, 120 m long,
charge, which simulated design live load was only 4.5 m high and all structural com-
conditions. The backfill was made from the ponents were designed in a way to be as
reclaimed sand. Scour protection was placed light as practical from a fabrication view-
along the seaward base of the wall to pro- point. It consists of floated-in units com-
tect it from the erosion induced by ship posed of three basic structural elements:
propellers and side thrusters. concrete floating foundation base, T-shaped
As stated earlier, the use of prefabricated concrete face panels, and steel anchors. The
wall units is limited by the capacity of the face panels are hinged at the lower end to
available lifting equipment and particularly the foundation base and are supported by
by capacity of the floating cranes. To over- the anchor tie rod at the top end. The an-
come this limitation, Tsink.er (1958) pro- chor rod has a turnbuckle and is hinged at
posed a hybrid-type structure which at- both ends to a vertical panel and the foun-
tempted to combine the advantages of the dation base. In the middle of its length, the
L-shaped walls (smaller amount of con- anchor rod has a double pin hinge which
crete) and floated-in caissons (no need in allows the anchor to be folded inside the
heavy lift floating cranes) (Fig. 5-49). wall base as shown in Figure 5-49.
This structure was built in the early Before launch, each wall section was as-
sixties on the Dnieper River, near City-of- sembled and panels were verified for verti-
Kherson, Ukraine. It was designed as an cality; next, the panels were rotated about

Figure 5-48. Quay wall at Bantry Bay, Australia. Typical cross section.
[From Mass-Morris and Hodge (1981).]
Gravity-Type Quay Walls 451

the bottom hinges and placed horizontally a prepared gravel mattress; this was done
on vertical ribs inside the base. The open with the help of a winch installed on floats,
space between the panels and the base as illustrated in Figure 5-49c. The installa-
perimeter as well as between adjacent pan- tion process was controlled by divers, who
els was sealed to prevent the water from also unsealed the installed sections under-
flooding the base during launch and towing water after they were placed to specified
operations. The floating base is designed to tolerances on a rubble mattress. Then the
stay afloat, with a buoyancy sufficient first vertical panel was raised and secured
enough to support the weight of the assem- by a cable to an offshore anchor. This opera-
bled structure with a minimum freeboard of tion was followed by gradual raising of all
about 100 mm only. At the construction other panels, the position of which was
site, the wall units were ballasted down on marked by small marker buoys.

(b)
... ~
., lj
"'; J£
=F= F=
I= f= ~

3~ 3E
~

""'~ Yf~ ~I l
3.E 3t
nr

PLAN

(c)

Figure 5-49. L-shaped quay wall composed from floated-in sections. [From Tsinker
(1958).] (a) Typical cross section; (b) floating base; (c) sequence of installation of wall
sections on gravel mattress; (d) lifting front panels in vertical position. 1-Floating base;
2-front panel; 3-steel anchor rod 40 mm in diameter; 4-cope beam; 5-fender;
6-axle of front panel; 7-double pin hinge; 8-seal around base perimeter;
9-turnbuckle; 10-gravel mattress; 11-front panel in horizontal position; 12-bolts for
installation of fender timber frame; 13-concrete mooringjbreasting piers installed at
24.5 m center to center along the quay; 14-access bridge; 15-slot for front panel axle;
16-embedded steel pipe for installation of anchor rod pin; 17-marker buoys; 18-as-
sembledsections of quay wall installed on a mattress; 19-surface of bedrock; 20-anchor-
agefor the first raised front panel; 21-floating platform with winch; 22-floating derrick.
452 Gravity-Type Quay Walls

/0
(a)

Figure 5-49. Continued.

In the process of installation, all raised Under certain local conditions, walls
panels were joined to each other by tuck similar to those to be discussed may repre-
welds; this allows the use of only one off- sent a viable solution to port waterfront
shore anchor per each wall unit. Verticality construction.
of frontal panels, where required, have been A number of heavy buttressed and arched
adjusted by turnbuckles. The wall backfill- walls which may be defined as a gravity-
ing was done with hydraulically placed type structure have been constructed in the
sand. Finally, the capping concrete beam past 25-30 years in waterfront develop-
was cast. Installation of 11 wall units, which ments. All these structures had their merits
included 2 wing wall units, and all associ- and disadvantages and, in fact, were used
ated operations to complete the 120.0-m- to fit certain site conditions.
long wall took about 2 weeks.
5.2-7.1 Walls with Fins
Basic structural schemes of these walls
5.2.7 Gravity-Type Walls are depicted in Figure 5-50. Although some
of Miscellaneous of these walls will usually be analyzed as a
unconventional Designs stiffened free-standing cantilever structure
and not necessarily. as a gravity wall, they
In this section, a few examples of gravity- still belong to a class of a gravity-type walls
type walls of unusual designs are discussed. because their stability, to some degree, de-
Gravity-Type Quay Walls 453

pends on side shear (friction or adhesion) at Depending on site conditions, heavy but-
the interface with the surrounding soil. The tressed and arched walls can be constructed
latter may have considerable impact on the from cast-in-situ concrete, precast compo-
analysis of wall stability, because the weight nents, and steel sheet piles. Under specific
of the soil column between buttresses at the conditions, slurry wall techniques can also
fins should be taken into consideration in be successfully employed (Xanthakos, 1979).
calculating the loads acting against the wall. The continuous series of T sections as
Naturally, the stability of these walls can depicted in Figure 5-50a, depending on the
be enhanced by installation of soil or rock space between buttresses versus length and
anchors. depth qf these buttresses, may be analyzed
as a stiffened cantilever wall or a gravity-
type wall. In this type of construction, the
(a)
front wall will be analyzed as a continuous
slab loaded with horizontal soil pressure.
The stability of wall depicted in Figure
I 5-50b will be analyzed in the same way as
I
I in the previous case. However, in this case
____ _, I
I
the front arch wall will not have tensile
stresses. The wall depicted in Figure 5-50c
is usually analyzed as a gravity-type struc-
(b) ture because its stability largely depends on
the weight of the wall and its interaction

:ITrT
with the foundation material. The wall is
usually provided with heavy deck structure
and massive buttresses to develop sufficient
._ ____ j 'resistance to lateral thrust.
B- B Finally, the wall shown in Figure 5-50d
is a hybrid structure which comprises some
properties of all previously discussed
schemes. The following is a discussion on
two case histories which have been con-
structed according to the schemes depicted
in Figures 5-50c and 5-50d.
,_ ____ J The arch-type gravity wall with massive
c-c buttresses was built in the late 1960s for
extension of the Ballard pier at Port-of-
Bombay, India (Fig. 5-51). This 330-m-long,
lI 17.4-m-high wall was designed and built
according to the scheme depicted in Figure
I
I 5-50c. This wall was constructed from pre-
I
'l-"V"'''F'"
.._ _ _
.'
~--...J cast hollow concrete blocks which formed
buttresses spaced at 12m (center to center)
and hollow concrete arches which formed a
Figure 5-50. Gravity-type walls that include but- retaining wall supported on buttresses. The
tresses and arches; typical structural arrangements. buttresses were fixed in place by concrete
(a) Wall composed from buttresses and continuous
frontal wall; (b, c) walls comprised of buttresses and
plugs placed on a bedrock floor through
arches; (d) hybrid-type construction that includes cavities left in the buttresses. The latter
arches and fins at rear of the wall. provided a reliable foundation base for the
454 Gravity-Type Quay Walls

INJO

Figure 5-51. Ballard pier extension, Bombay, India. Arch-type gravity wall; plan and
typical cross section. !-Precast unit of concrete block buttress with cavities filled with
colloidal grout concrete; 2-precast unit of concrete arch; 3-precast prestressed concrete
beams; 4-precast concrete slabs and walls; 5-tremie concrete plug; 6-concrete in sacks for
shattering; 7-rubble bank; 8-colloidal grout concrete; 9-cement grout; 10-cast-in-si tu
concrete deck; 11-fill; 12-rubber fender.

buttresses and ensured good cohesion with to the scheme depicted in Figure 5-50d.
the underlaying bedrock. The design of this arch-type gravity wall
In order to prevent the tremie concrete with fins on a rear side (Fig. 5-52) used a
from coming out and spreading beyond the novel combinatio n of techniques tailored to
buttress' outside perimeter and to form a suit unusual site conditions; a high ground
backfill impervious retaining wall, the open level over mixed soils included dense sand
space between bedrock and buttresses and at an upper layer underlaid by a consider-
arches was shuttered with concrete in sacks. able deposit of boulder clay with sandstone
The cavities inside the buttresses were filled at low level.
with high-streng th colloidal grout concrete The site was dewatered for excavation of
which turned these elements into the mono- the dock basin, but before this was done the
lithic concrete. The keys between the arches wall was built. For this, a slurry wall tech-
and buttresses were cement grouted. Fur- nique was employed; the concrete was
thermore, to provide for additional stability, placed in trenches filled with heavy
the rubble bank was placed inside the space thixotropic suspension of natural clay (ben-
between adjacent buttresses. Left-in-place tonite) which was gradually displaced and
precast prestressed slabs and beams have substituted by concrete. To enhance the
been used as a formwork for the cast-in-situ performanc e of the structure at its front
heavy deck structure. corners, this part was designed and built in
The next example is an unusual con- the form of bored piles that penetrated the
struction of the east wall for Seaforth dock, bedrock. The fin wall on a rear wall en-
Liverpool, United Kingdom (Agar and hanced the stability of the wall by adding
Irwin-Child s, 1973). This 1677-m-long and weight to the structure, increasing the arm
about 18.0-m-high wall was built according of the resisting moment, and adding friction
Gravity-Type Quay Walls 455

t
A
iA
I 6

I
EL.-.rm
5-j
2 .

TYPICAL CROSS SECTION

A-A
Figure 5-52. Seaforth dock wall, Liverpool, United Kingdom.
!-Concrete arch keyed into sandstone to 0.90 m (min);
2-rear concrete fin keyed into sandstone 0.9 m (min);
3-concrete deck; 4-compacted sand fill; 5-2000-kN-capacity
rock anchor; 6-bored pile 1.4 m in diameter at 6.5 m center
to center; 7-30-mm-diameter anchor rod.

forces attributed to the interaction between anchors. As indicated by Agar and Irwin-
fins and surrounding soil which kept it from Childs (1973), because of the fairly good soil
moving upward. As a matter of precaution condition at the site, the use of ground
and in order to develop additional resisting anchors was not required. The heavy deck
forces, the wall was posttensioned by rock · was cast in situ on the top of arches and
456 Gravity-Type Quay Walls

fins. The voids in the deck structure were


filled with densified sand. The wall of simi-
lar nature however constructed from steel
sheet piling is discussed by Nottingham
(1995).
It should be noted that interaction of this
type of wall with soil is very uncertain;
current theories cannot provide a close esti-
mate of earth pressures resulting from a
complex three-dimensional action. There-
fore, care should be exercised when design-
ing this type of wall. A conservative design
or a design based on a large-scale-model
test would be an appropriate approach to
the problem;
More examples of gravity-type walls built
with the help of the slurry wall techniques
is provided by Irwin-Childs and Macknight
(1977) and Xanthakos (1979).
Figure 5-53. Wall at Port-of-Copenhagen, Denmark.
1-Front wall; 2-pressure-relieving slab; 3-counter-
5.2. 7.2 Walls with fort 1.2 m thick at 5.0 m center to center; 4-timber
Soil-Pressure-Relieving sheet piling; 5-fill; 6-bedrock.
components
The wall of this type, which calls for Several practical examples of such type of
construction, one of which is depicted in
attention, was built at Port-of-Copenhagen
figure 5-54, are discussed by Tsagereli
(Fig. 5- 53). This concrete wall was built in
(1969). The wall as depicted in this figure
the dry and its stability was enhanced by
was built in the dry as a bank protection
the soil-pressure-relieving slab supported by
structure on the Kura River in the city of
counterforts. The wall geometry and its in- Tkibuli, Georgia (former Soviet Union). Ac-
teraction with soil produced uniform distri- tually this type of design replaced the one
bution of bearing pressures on the founda- originally considered for construction: a
tion at the wall base. As discussed earlier in solid masonry gravity wall as outlined in
this chapter, in modern marine engineering Figure 5-54.
practice, the idea of using soil-pressure- The new wall was built in the following
relieving slabs for reducing soil lateral sequence. First, the foundation block with
thrust on retaining walls has been modified two recesses in it was cast in situ. This was
and adopted to different types of dock followed by casting of the bottom part of the
structure. wall and installation of the precast frame
A variety of examples of quay and dry supported by temporary anchors secured at
dock walls similar to that built at Port-of- the wall. The frame's legs were installed
Copenhagen is given in Co:ffiick (1958) and in leftover recesses. Next, the precast
Bray and Tatham (1992). This type of soil- pressure-relieving slabs were installed on a
pressure-relieving technique is also dis- wall and support frame. Next, the recesses
cussed in Tsinker (1995) and in Chapter 4. were grouted with concrete, the wall was
The soil-pressure-relieving platform may completed to the designed elevation, and
significantly reduce the size of the retaining granular material was placed behind it. Ac-
structure and make it more economical. cording to Tsagareli construction of 540 lin-
Gravity-Type Quay Walls 457

(a)

s
(b)

Figure 5-54. Bank protection wall at Tkibuli, Georgia (fonner


Soviet Union). [From Tsagareli (1969).] (a) Typical cross section and
details; (b) sequence of construction. 1-Mass concrete wall; 2-
precast concrete frame; 3-precast concrete soil-pressure-relieving
slab; 4-granular fill; 5-backfill.

ear meters of this wall proved to be 36% 5.2.7.5 Gravity-type Pier


less expensive than the originally consid-
ered massive masonry wall. All earlier examples of gravity-type quay
A wall of similar design was built on the walls discussed in this chapter are related
same river in the city of Gori, Georgia. The to soil-retaining structures. Gravity-type
typical cross section and back-view eleva-· dock structures can also be built in a form
tion of this 8.2-m-high, 624-m-long wall and of an open type of construction, such as the
the construction sequence used for wall con- pier built in 1957 at Ilo, Peru, for the
struction are illustrated in Figure 5-55. Southern Peru Copper Company (Quinn,
Here, the precast concrete frames were em- 1972). The typical cross section of this pier
bedded into the wall and prefabricated is illustrated in Figure 5-56. The free-
pressure-relieving slabs were placed hori- standing pier structure was designed to re-
zontally on them. ceive bulk carriers of up to 40,000 DWT
458 Gravity-Type Quay Walls

I
!•
1.72 10.61·•
>I
I I

(a)

(b)

Figure 5-55. Bank protection wall at Gori, Georgia (former Soviet Union).
[From Tsagareli (1969).] (a) Typical cross section and elevation; (b) sequence
of construction. 1-Mass concrete wall; 2-prefabricated concrete frame;
3-prefabricated concrete slab; 4-re-bars; 5-granular fill.
Gravity-Type Quay Walls 459

Figure 5-56. Ilo, Peru. Pier for the southern Copper Company; typical
cross section. 1-Steel cylinders filled with concrete; 2-stiffeners; 3-
concrete deck; 4-ballast; 5-timber fender; 6-pavement; 7-railway
track; 8-steel bracing; 9-capping girder.

docked at both sides of the pier. Each bent ders, 152.5 rom in diameter, with a 6.5-mm
of this pier was designed as a rigid gravity wall thickness in the straight section, flar-
frame proportioned to take the substantial ing out into a bell section 335 rom in diame-
force of the earthquake (0.1 X g) as well ter with a 9.5-mm wall thickness, were in-
as to resist ship berthing impact and moor- stalled on a rock foundation. For this, the
ing forces. The seafloor at the pier location seafloor contours were accurately deter-
was a granitic rock with some small pockets mined and the bottom of each column was
and pinnacles; the floor was bare in most cut to fit the contours of the rock bottom at
areas except for some thin sand and shell a particular spot. When the cylinders were
deposits. Because use of the conventional lowered in their designated position on the
piling was not feasible at the site and the seafloor, the remaining gaps between the
design earthquake forces were large, the cylinder bottom and the rock were closed
most economical solution to the problem with sandbags by divers.
was found in the installation of heavy Next, each cylinder was filled with tremie
concrete-filled steel columns, three at each concrete to approximately one-half its
bent, capped with a large concrete girder, height. When this concrete gained sufficient
on the top of which a heavy concrete deck strength, the cylinder was dewatered, the
was built. For additional stiffness in the · laitance removed, a key cut in the surface of
transverse direction, the colu:rims were previously placed concrete, and the remain-
braced by heavy 1-beams. ing portion of the cylinder was filled in the
The pier construction proceeded in the dry simultaneously with concreting of the
following sequence. First, the steel cylin- capping girder. It should be noted that the
460 Gravity-Type Quay Walls

•A •A
.
c

B D

t. .t

A-A c-c
Figure 5-57. Rapidly assembled gravity-type structures. [After Tsinker (1994).] (a) L-shaped
wall; (b) breakwater or pier. 1-Floating base (concrete or steel); 2-vertical buoyant component;
3-vertical panel; 4-sand or concrete fill; 5, 6-hinges; 7-capping superstructure; 8-stone
bedding (mattress); 9-granular fill.

concrete placed inside of cylinders was structural components: floating base (1);
heavily reinforced. Finally, the pier was vertical load-carrying components (2) that
completed with a heavy concrete deck de- are attached to the floating base (1) through
sign to support four railway tracks in addi- hinges (5) and (6); intermediate panel (3)
tion to other live loads. The pier was con- that is installed in between vertical compo-
structed by the overhead method; the job nents (2); capping structure (7). These
advanced offshore, as work progressed. Pre- structures are usually designed for rapid
viously installed bents were used to support construction as may be required either in
a large crane used for pier construction. military application or where the overwater
Several more interesting examples of construction time is limited by the weather
free-standing, open, gravity-type marine conditions. They can be used for the
structures are given in Cornick (1958) and gravity-type quay walls or dolphins, as well
Smirnov et al. (1979). as the construction of breakwaters. As is
common for the modern offshore construc-
5.2.7.4 Floated-in Structures tion practices, these floated-in structures
can be installed either on previously con-
One structure of this nature is illus- structed stone bedding or directly on the
trated in Figure 5-49. Two more examples seafloor; in the latter case they should have
are illustrated in Figure 5-57. These struc- a concrete or steel skirt, and the space be-
tures are comprised of the following basic tween the seafloor and the underside of the
Gravity-TypeQuayWalls 461

structure's base needs to be grouted as pre- Establishing of


viously discussed. Structural components of wall design conditions
these structures are prefabricated at the
construction yard.
The basic construction stages for these Assumptions on wall
structures are depicted in Figure 5-58. The structure and dimensions
floating base can be launched by utilizing
any available techniques as discussed ear-
lier. Then sealed and therefore watertight Determination of external
vertical components are hinged to the base's loads and forces
deck through hinges (5). These hinges along
with hinges (6) are designed to resist all
kinds of lateral load that are specific for the Evaluation of
particular structure (i.e., soil lateral pres- wall sliding stability
sure, unbalanced hydrostatic pressure,
mooring and ship impact loads, wave-in-
Evaluation of
duced loads, etc.). At the deployment site,
wall overturning stability
the base is flooded and the base ;vertical
component assembly is sunk on the pre-
pared foundation base or, as discussed ear- Evaluation of
lier, on the seafloor. In the process, the foundation bearing capacity
buoyant vertical component rotates about
hinges (5) to its vertical position, after which
it is firmly fixed in the design position by Evaluation of
the pins at hinges (6) that are installed by wall global stability
divers. This is followed by grouting the base
and the vertical component either by pump-
ing in sand or concrete, or a combination of
Detail design
both. Then, vertical panels are installed be-
tween vertical elements and the structure
is completed by the granular fill. Finally,
the capping blocks (7), either prefabricated
or cast in situ, are installed.
5.3.1 Loads and Forces Load
combinations
5.3 BASIC DESIGN The quay wall and its components must be
CONSIDERATIONS designed to resist all permanent and tempo-
rary loads and forces without being over-
stressed and without appreciable displace-
Gravity-type walls design is usually con-_ ments and deformation. All basic loads and
ducted in the sequence indicated in the fol- forces acting on port related marine struc-
lowing diagram. The design process is es- tures are discussed in detail in Chapter 3.
sentially of a trial-and-error nature during As was stated earlier, all loads acting on
which the best and most economical solu- the quay wall are usually defined as.perma-
tion should be found. nent and temporary. The former include the
462 Gravity-Type Quay Walls

Note: In some cases, soil lateral thrust from


live loads which by its very nature is a
temporary load may also be treated as a
permanent load because it may not disap-
pear with removal of live load. The explana-
tion of this phenomenon can be found in
Chapter 4.
The temporary loads are classified as
regular (normal) and special (extreme)
loads. The former include the following:

(a) Live loads such as weight of stored cargo


and cargo handling and hauling equip-
ment
(b) Differential hydrostatic pressures at-
tributed to water fluctuation in the basin
or to wave action
(c) Effects of environmental forces such as
waves, wind, current, and ice
(d) Ship impact and mooring forces
(e) Miscellaneous loads during construction
(f) Loads attributed to changes in ambient
L------~
~------------------
temperature

Special loads include loads associated with


natural disasters such as earthquakes, hur-
ricanes; and abnormal flooding.
The structure should be designed to
safely respond to the worst combination of
all the above loads. In formulating the load
combinations and allowable stresses, the
designer should be guided by existing local
regulations and experience. For economical
Figure 5-58. Installation phases of rapidly assem-
design, only realistic loads should be con-
bled gravity-type L-shaped wall. [After Tsinker (1994).] sidered. Those loads which have a low prob-
!-Floating base; 2-vertical buoyant component; ability of occurrence should not be com-
4-sand or concrete grout; 5, 6-hinges; 7-capping bined with earthquake loading, or at least
block; 8-stone bedding. should be reduced significantly. Typical load
combinations are suggested in Chapter 3.
Normally, the gravity-type wall is analyzed
following: as a two-dimensional structure in which
case the wall design scheme is represented
1. The weight of the wall and weight of by its typical cross section with an indica-
permanently installed equipment and tion of all associated loads.
structures For specific design conditions such as
2. The weight of soil sliding, overturning, or overall stability and
3. The lateral active and passive soil pres- for evaluation of bearing stresses at the
sures wall base, live loads should be placed ac-
Gravity-Type Quay Walls 463

cording, but not limited to, the design mally used for the determination of normal
schemes indicated in Figure 5-59. Obvi- bearing stresses at the base of practically
ously, these loads must be used in combina- all gravity-type walls, and also at the inter-
tion with other loads such as ship impact, face between the stone bedding and founda-
mooring forces, environmental loads, arid tion soil. Along with the load distribution
all other relevant loads. according to scheme (d), it is also critical for
In Figure 5-59, loads q 1 , q 2 , and q 3 rep- the design of the capping wall.
resent design surcharge load as discussed L-shaped walls with external anchorages
in Chapter 3, Pc represents either concen- are usually designed to resist loads as pre-
trated or line live load produced by gantry sented in schemes (a), (c), and (e); scheme
or mobile cranes, and qeq is the equivalent (a) is used for the review of wall sliding
of Pc and may be used instead of it. stability at the base and at the interface
Typically, design load schemes (a), (b), between the rubble1gravel mattress and the
(c), and (d) are used for the design of block- foundation soil; scheme (c) is used for the
work and crib-type structures. In blockwork determination of maximum bending and
wall design, scheme (a) is basically used for shear stresses in the face panel and anchor
the evaluation of wall sliding stability at system, and for calculating maximum bear-
the base level, along with a rubble mattress ing stresses at the base slab and at the
and at the base of each course of blocks. It interface between the rubble I gravel mat-
is also used for the evaluation of the posi- tress and the foundation soil; scheme (e) is
tion of the resulting load acting at the base used for the review of anchor slab
of the wall, as well as at the bottom of each (deadman) stability. The strength and sta-
course. If required, this scheme is also used bility of L-shaped walls with internal an-
for the review of the wall overturning sta- chorages and counterfort walls are usually
bility at each critical design level. calculated according to schemes (a) and (c).
Load distribution according to scheme (b) For the evaluation of sliding stability of
is the most unfavorable for performance of L-shaped walls with any type of anchorages
the upper block of the blockwork wall. Load and counterfort walls, the live load shall be
distribution according to scheme (c) is nor- placed within surface boundaries of the

Figure 5-59. Basic schemes for placing design live loads.


464 Gravity-Type Quay Walls

sliding wedge behind the wall. For details, tice has been considered by many marine
see Chapter 3. structures designers as, although conserva-
With due consideration given to soil fric- tive, satisfactory. This, however, is relevant
tion on concrete walls, the soil pressures on mostly for the design of the gravity walls
the frontal panel of counterfort walls as constructed on competent foundation soil.
determined by standard methods can be On the other hand, the stability of gravity
reduced by 20-30%; the minimum reduc- walls to be built on soft foundation strata
tion is used when the distance between ad- must be thoroughly examined, and neces-
jacent counterforts is about 4 m, and maxi- sary redundancy to the wall structure must
mum when this distance is about 2 m or be provided.
less. As stated by Osterberg (1989) in his
The horizontal component of the mooring twenty-first Karl Terzaghi lecture, the
force normal to the face of the quay is typi- key to all successful civil-geotechnical-
cally distributed along the capping super- foundation engineering projects is neces-
structure. The distance to which this force sary redundancy, and in order not to over-
is distributed depends on the type of bollard look any serious condition which may exist
foundation and the type of capping super- in the foundation, the best tools of the trade
structure. in drilling and testing earth formation
should be used and highly experienced and
knowledgeable specialists must be in charge
of all important aspects of the project devel-
5.3.2 Basic Static Principles opment. Foundation and structure redun-
dancy is of paramount importance, particu-
As previously mentioned, there are still larly where rapid changes in water eleva-
many questions about marine structures tion and, therefore, forced water seepage
performance and their interaction with en- under and around a structure can occur.
vironmental forces, foundation soil, and Under the aforementioned conditions, at-
backfill materials. Despite all the science tention to minor details could be fundamen-
available and finite element calculations via tal to the success of the project. As stated
computers, at present many of these ques- by Osterberg, the necessary redundancy in-
tions and specifically those related to the cludes requirements for problem analysis
gravity-type structures are best answered by more than one method, project peer re-
with somewhat simple, time-tested practi- view, establishing good means of communi-
cal formulas. cation between owner (developer)-
Again, the reader must be aware that the engineer-contractor as well as having a
design procedures that are discussed in this good monitoring and surveillance program
and in the following chapters are intended in place. All of the aforementioned helps to
to point out the complexity of the particular detect and correct hidden poor foundation
problem and illustrate factors that should formations such as weak soils or open joints
be considered and included in an appropri- and seams in rock, and therefore prevent
ate design scenario. Therefore, these design structural failure or excessive deformations.
procedures should not be used indiscrimi- In the case of heavy gravity walls built
nately and particularly for the detail on relatively soft foundation soils, staged
design. construction using controlled rates of load-
It should be noted that, in general, the ing to enable the soil to strengthen via
design of the gravity earth-retaining struc- consolidation in order to increase founda-
tures has not undergone significant changes tion stability should be considered (Ladd,
in the past 20-30 years. The existing prac- 1986).
Gravity-Type Quay Walls 465

A recent overview of the behavior and several options are usually considered that
design of gravity earth-retaining structures include: (1) increase the base of the struc-
along with a comprehensive list of refer- ture, (2) replacement of local bottom soil by
ences is given in Duncan et al. (1990) and good quality granular material, (3) reloca-
Rourke and Jones (1990). tion of the structure to an alternative site,
Because all types of gravity quay wall or (4) performance of additional soil investi-
are, in general, considered as a solid block, gation to justify possible reassessment of
the basic static principles are similar for all soil parameters used in analyses.
of them. The design process normally starts In the following sections, the conven-
with a tentative dimensioning of the wall, tional analytical procedures commonly used
which is followed by an analysis for stabil- for design of the gravity type quay walls are
ity and structural requirem.ents; this is ba- described in detail.
sically a trial process during which several
alternative solutions are analyzed in an at- 5.3.2.1 stability Against Sliding
tempt to obtain the most economical and at Failure
the same time reliable solution to the prob-
The wall must provide adequate stability
lem. As a rule of thumb, the width of the
against the sliding mode of failure which is
base of a gravity-type wall basically de-
determined by the following ratio.
pends on the type of wall construction, kind
of the backfill material, and the properties
Sum of resisting forces
offoundation soil; the width is typically (0.5 F = -------,---
to O.B)H, where H is the height of the wall. "1 Sum of driving forces
In general, the design process can be pro-
grammed easily for the computer which may where F51 is the factor of safety against
help to find the most economical solution to sliding.
the specific problem. Normal driving loads typically include
The gravity-type wall is routinely ana- soil lateral thrust, unbalanced hydrostatic
lyzed for the following conditions. pressure, and mooring forces (Fig. 5-60).
The former and the latter are discussed in
1. Sliding stability at the base level details in Chapter 3.
2. Sliding stability of the wall-mattress sys- The unusually low and rapidly developed
tem at the interface between the mattress water level in front of a quay wall that can
and the foundation soil be the result of a full-moon tide, wind ef-
3. Acceptability of bearing stresses at the fect, or severe storm waves, and the
base level and at the interface between groundwater level behind the wall remains
the mattress and the foundation soil substantially higher produce unbalanced
4. Overturning stability hydrostatic pressure on the wall, which
5. General or global stability should be taken into account. Furudoi and
6. Settlement when constructed on com- Katayama (1971) suggested that in general
pressible foundation the design unbalanced hydrostatic load can
be considered as equal to about one-third of
Normally, all the above analyses are con- the tidal range above the low water level
ducted on a typical cross section of the wall. (LWL).
Several interactions of these analyses The effect of an unbalanced hydrostatic
usually produce a balanced design with re- pressure is less pronounced when the wall
spect to economy and safety level. If poten- i~ placed on permeable rubble bedding and
tial problems are identified with respect to is backfilled by coarse granular material
any one of the above design conditions, then such as rubble or gravel. This enables free
466 Gravity-Type Quay Walls

Figure 5-60. Design schemes for evaluation of wall sliding and over-
turning stabilities. (a) Wall comprised of floated-in caisson, large-diameter
cylinders, and the like; (b) blockwork wall, or walls with pressure-relieving
platforms.

flow of water in and out, thus minimizing it should also be routinely reviewed at the
the possibility of unbalanced hydrostatic lowest water level, where unbalanced hy-
pressures. drostatic pressure can reach its maximum
If a rapid water-level equalization in front value.
and behind the wall is ensured by the proper The wall-resisting forces typically in-
choice of the backfill and mattress materi- clude the buoyant weight of the structure
als and a proper drainage system is in- and the weight of the backfill material above
stalled at wall expansion joints, then the the projected width of the wall.
unbalanced hydrostatic pressure on the When the sliding stability at the inter-
quay wall due to wave action can be re- face between the wall mattress and the
duced to only one-half of the height of the foundation soil is evaluated, the weight of
maximum design wave. Otherwise, the un- the mattress material in boundaries
balanced water pressure should be taken as 1-2-3-5-4 and 1-2-3-5-6-4 as outlined
equal to a maximum height of the design in Figures 5-60a and 5-60b should be in-
wall. cluded in the calculation of resisting forces.
On the other hand, where the structure Furthermore, the wall sliding potentials
is embedded in foundation soil (e.g., cellular at plane 1-3, particularly in the case as
bulkheads or large-diameter cylinders depicted in Figure 5-60b, should be also
driven into the harbor bed, or caisson-type reviewed.
structures with skirts) or when structure is Extreme driving loads include unusually
backfilled with regular sand and an efficient severe loads generated by earthquakes,
drainage system is not provided, the effect heavy waves, and very strong wind. Some-
of the unbalanced hydrostatic pressure on a times, the effects of rarely used very_ heavy
structure may be dramatic. Normally, un- cargo handling and hauling equipment is
balanced hydrostatic pressure due to wave also considered as an unusual and, there-
action is not combined with mooring forces. fore, extreme load. Normally, extreme loads
In most cases, the wall sliding stability is are not combined with each other.
critical at maximum water level. However, Numerically, the wall sliding stability is
Gravity-TypeQuayWalls 467

expressed as ally taken as equal to 2.0, and for


extreme loading it is reduced to
O::V- L.U)f
F., = --L.::--:-H-- (5-1)
1.5.
Mr and M 0 =resisting and overturning mo-
ments, respectively, about the
where wall toe (point 4 in Fig. 5-60).
F81 = safety factor; F.1 should not be less than
1.5 for normal loading and no less than The evaluation of the wall stability against
1.25 for an extreme load combination overturning mode of failure is usually not
L. V = sum of all vertical loads acting on wall required where the resultant vertical force
base; L. V includes a vertical component is within the middle third of the wall base,
of the lateral soil thrust or where e :::;; b j6, where e is the eccentric-
r. U = uplift (bouyant) force ity and b is the width of the wall base (Fig.
L. H = sum of all horizontal driving forces; L. H 5-60).
basically includes horizontal component
of soil lateral thrust, unbalanced hydro- 5.3.2.3 contact Pressure
static load, and mooring force
f = coefficient of friction. For a concrete Contact stresses at the wall base and at
structure placed on a rockfill mattress, the interface between the mattress and the
the coefficient of friction f of 0.5-0.65 is foundation soil as well as at any critical
usually used. The friction coefficient be- horizontal section in the wall structure a
tween rubble mattress and foundation soil are determined from the following (Fig.
is usually assumed to be tg(j¢), where 4> 5-61):
is the angle of internal friction of mate-
rial used for the wall bedding or founda- r.v-r.u Mo
±- (5-3)
tion soil, whichever produces the smaller amaxjmin =
A w
value of f. The upper limit of the coeffi-
cient of friction f = tg ¢. where
L. V and L. U = as in formula (5-1)
Note that 0::: V- L U) represents the effec- A = base (section) area
tive weight of the structure. M 0 =moment of all loads about the
It should also be noted that in most cases geometrical center of the wall
the passive pressure is not included in the base or any critical section
calculation of resisting forces. W = base (section) sectional modulus
about its longitudinal axis.
5.3.2.2 stability Against
overturning Subsequently, A = b per linear meter of
This is determined from the following wall; M 0 = (LV - L U)e; W = b 2/6 (all in
ratio: appropriate metric units), and
Sum of moments to resist overturning
amaxjmin =
L.V-EU(
b 1±
6e)
b (5-4)
Sum of overturning moments

(5-2) where e is the eccentricity of the resulting


load 0:::: V - L: U) with respect to geometri-
cal center of the wall base or any critical
where section. For the rectangular base of the wall,
F0 = factor of safety against overturn- e = 0.5b - e1 , where e 1 is the distance from
ing; for normal loading, F 0 is usu- the wall toe (front edge of the section) to the
468 Gravity-Type Quay Walls

(a) (b)

Figure 5-61. Design diagram for calculating contact stresses at the wall base
and bearing stresses at the mattress interface with the foundation soil.
(a) Resulting force acts within middle one-third of the wall base; (b) resulting
force is acting beyond middle one-third of the wall base.

point where the vertical component of the tions or in the cases where the wall is built
resultant force acts at the wall base (critical on sound bedrock foundation or on founda-
section): tions which include dense granular materi-
als. In the case of bedrock foundations, the
M
el= - - - - (5-5) value of e should not exceed 0.25b, and in
EV-I:U the latter case it should not exceed 0.2b.
Naturally, the maximum contact stress
where M is the moment of all loads about
as determined from equations (5-4) and
the wall toe (or edge of the critical section).
(5-6) should not exceed the allowable value
Relationship (5-4) is used when e ::;;; b j6;
in other words, when the vertical compo-
of the bearing stress, a,.
The contact
stresses at interface between the stone bed-
nent of the resulting force is acting within
ding and foundation soil, a ', under the e ::;;;
the middle one-third of the base (or middle
b j6 condition are determined with due con-
one-third of the wall critical section) (Fig.
sideration given to the stress distribution
5-61a). In the latter case, both amax and
through the mattress material at 45° (Fig.
amin cannot be negative.
5-61). Hence,
When e > b j6, amin will have a negative
value, and because the soil cannot take ten-
b
sile stresses, the "effective" bearing area a,;,axjmin = amaxjmin b + 2h + Yrhm :=:; at
per linear meter (ln. m) of wall will be m

somewhat smaller than b X (1.0 m) m 2 • In (5-7)


conventional analysis, this effect is ac-
counted for by reducing the bearing area of where
the footing according to empirical guide-
lines. According to the latter, the maximum Yr = buoyant weight of the mattress ma-
contact stress can be determined from the terial
following formulation (Fig. 5-6lb): amaxfmin =bearing stress level at the wall base
hm =mattress thickness
2 I:V- I:U at = allowable stress on foundation soil;
(5-6)
for preliminary design, the data pre-
sented in Table 5-2 may be used.
Normally, the e > b j6 condition would
be acceptable for the extreme load combina- Accordingly, the minimum thickness of the
Gravity-Type Quay Walls 469

Table 5-2. Allowable pressure on foundation soils ( ur) for preliminary design (kPa)

Type of Foundation Normal Loading Extreme Loading


Sound rock One-seventh of ultimate One-fifth of ultimate
strength in water bearing strength in water bearing
condition condition
Weathered rock 600-1500 900-2100
Marl and heavy chalky clay 250-750 350-1000
Well-compacted crushed stone,
dense gravel bedding 600 850
Dense sand
Coarse-grained with gravel 350-450 500-650
Medium 250-350 400-500
Fine 150-250 200-350
Sandy clays
Void ratio 0.5 250-300 330-400
Void ratio 0.7 150-250 200-330
Void ratio 1.0 100-150 150-200
Dense clay 300-500 400-700
Soft clay 50-200 100-300

stone bedding, hmin' can be obtained from analysis (Pirah, 1975; Penzien and Tseng,
1976; Prevost et al., 1981, and others).

5.3.2.4 Ultimate Load on


Foundation (Bearing
b(umax- u,)
0.5
Capacity)
]
2yr In foundation engineering, it is generally
assumed that the failure of a loaded shal-
(5-8)
low foundation would occur when the peak
ultimate load is reached simultaneously
Vesic (1963) has proposed various bearing with the appearance of slip lines at the
capacity factors for the determination of the ground surface, which is followed by foun-
foundation soil bearing capacity and the dation collapse and a considerable bulging
"effective" contact area at the wall base. of the soil mass on the side of the footing.
Vesic's method is discussed in detail in the Vesic (1963) defines the ultimate load
sections that follow. criterion as the point where the slope of the
In more sophisticated analyses, the foun- load-settlement curve first reaches zero or a
dation is treated as an elastic half-space in steady, minimum value. Another consistent
which the foundation stiffness characteris- ultimate load criterion defines the ultimate
tics are represented by replacing the soil load at the point of break of the load-
mass with linear spring and dashpot ele- settlement curve in a logjlog plot (De Beer,
ments. Sophisticated analytical methods 1967).
based on the upper bound analysis are dis- In general, computation of the ultimate
cussed by Murff and Miller (1977). In recent load represents a problem of elastic-plastic
years, finite element analyses are more fre- equilibrium which can be solved in plane-
quently used for soil-foundation interaction strain and axisymmetric geometries. In the
470 Gravity-Type Quay Walls

case of gravity-type quay wall the resulting posed to get the bearing capacity of footing
force is inclined in relation to the wall base subjected to the eccentricalload.
and is eccentric. Therefore, failure can oc- There are several proposed methods for
cur either by sliding along the base or by analysis of the bearing capacity of eccentri-
general shear of the underlying soil. At the cally loaded footings. The following are rec-
verge of sliding, the total (ultimate) load ommendations on the determination of
horizontal component I: E is basically re- foundation soil ultimate bearing capacity as
lated to the vertical component I: V by recommended by the American Petroleum
Institute (API) (1984) for the design of off-
L Emax = LV tan <P + A' ca (5-9) shore structures.
The bearing capacity equations provided
by API are limited to shallow foundations
where and are based on the assumption that the
A' = effective bearing area of the wall soil is a rigid, perfectly plastic material that
ca and <P =the adhesion and the angle of inter- obeys the Mohr-Coulomb yield criterion.
nal friction between the soil and wall Note that, by general definition, shallow
base, respectively foundations are those for which the depth of
embedment is less than the minimum lat-
It should be mentioned that one of the eral dimension of the foundation element.
earliest set of recommendations on soil The API method is strictly applicable to
bearing capacity was proposed by Terzaghi idealized conditions of uniform soil strength.
(1943) whose theory has been further devel- Where use of these equations is not justi-
oped by Meyerhof (1951, 1963), Hansen fied, a more refined analysis should be con-
(1957, 1970), and others. A good discussion sidered.
on the subject matter is given in Bowles
(1988). For a comprehensive list of refer- 5.3.2.4.1 Undrained bearing
ences on the subject matter, the reader is capacity
referred to Saran and Agarwal (1991).
If loading occurs rapidly enough so that
The quay wall, which is essentially a
no drainage and hence no dissipation of
retaining wall, is subjected to moments and
excess pore pressure occurs, then an "un-
shears in addition to vertical load. These
drained" (or "short-term") analysis is per-
forces are usually replaced by an eccentric-
formed. In this case, the soil is treated as if
inclined load action at the wall base. As
its friction angle <P equals 0, such that the
indicated earlier, the usual practice in
stability of the foundation is controlled by
the design of such walls is to resolve the
an appropriate undrained shear strength, c.
eccentric-inclined load into two parts,
The maximum gross vertical load Vu
namely: (1) an eccentric vertical load and
which a foundation can support under
(2) a central oblique load. Unless a heavy
undrained conditions is computed from
gravity wall is placed on a competent
bedrock foundation the bearing capacity of (5-10)
the footing must be evaluated.
In conventional practice. the bearing ca- where
pacity of the footing is obtained by analyz- Vu = maximum vertical load at failure
ing the problem in two separate parts: (1) c = undrained strength of soil
the bearing capacity of footing subjected to Nc =a dimensionless constant; Nc = 5.14 for
the central vertical load and (2) the bearing undrained conditions (</J = 0).
capacity of footing subjected to the central y = total unit weight of soil
oblique load. The two values are superim- D = depth of embedment of foundation
Gravity-Type Quay Walls 471

A' = effective area of the foundation depending cp' = effective friction angle of the Mohr
on load eccentricity envelope
Kc = correction factor which accounts for load y' = effective unit weight
inclination foundation shape, depth of em- q = y 'D, where D is the depth of the
bedment, inclination of the base, and incli- embedment of the foundation
nation of the ground surface; this factor is b =minimum lateral foundation di-
discussed below mension
A' = effective area of the foundation de-
5.3.2.4.2 Drained bearing capacity pending on the load eccentricity
If the rate of loading is slow enough such Kc, Kq, K"' = correction factors which account for
that no excess pore pressures are developed load inclination, footing shape,
(i.e., complete drainage under the applied depth of embedment, inclination of
stresses) and sufficient time has elapsed base, and inclination of the ground
since any previous application of stresses surface, respectively; the sub-
scripts c, q, and y refer to the
such that all excess pore pressures have
particular term in the equation;
been dissipated, a "drained analysis" is per- these correction factors are dis-
formed. The stability of the foundation is cussed below
controlled by the drained shear strength of
the soil. The drained shear strength is de- 5.3.2.4.3 Effective area (A' J
termined from the Mohr-Coulomb effective
stress failure envelope (i.e., the cohesion Load eccentricity decreases the ultimate
intercept c' and friction angle ¢'). vertical load that a footing can withstand.
The maximum net vertical load V~ which This effect is accounted for in bearing ca-
a foundation can support under drained pacity analysis by reducing the effective
conditions is computed from area of the footing as discussed earlier or
according to other available empirical
V~ = (c'NcKc + qNqKq + 0.5y'bN1 K)A' guidelines; for example, guidelines given in
(5-11) American Petroleum Institute (1984). The
relevant example is given in Chapter 8.
where
5.3.2.4.4 correction factors
v~ = maximum net vertical load at fail-
ure The correction factors Kc, Kq, and K"'
c' = effective cohesion intercept of the are usually written
Mohr envelope
Nq = [exp('lT tan cf.>')][ tan 2 (45o + cp' j2], a (5-12)
dimensionless function of cp'
Nc = (Nq - 1) cot cp ', a dimensionless (5-13)
function of cp'
NY = empirical dimensionless function of (5-14)
cp' that can be approximated by
2(Nq + 1) tan cf.>'; Kumbhojkar
where i, s, d, b, and g are individual cor-
(1992) evaluated the magnitude of
N"' and suggested some corrections rection factors related to load inclination,
to this factor that are based on the foundation shape, embedment depth, base
shape of the base, the depth and inclination, and ground surface inclination,
tilt of the footings, as well as the respectively. The subscripts c, q, and 'Y
rigidity and layering of the soil be- identify the factor (Nc, Nq, or N"') with
low footing which the correction term is associated.
472 Gravity-Type Quay Walls

The recommended correction factors for accordingly, for a continuous wall,


Nc and Nq that account for variations in L' = 1.0
loading and geometry not considered in the H = horizontal component of resultant
theoretical solutions are obtained from the load
expressions for Nc and Nq as suggested by m =a dimensionless function of b' jL'
DeBeer and Ladanyi (1961) [as cited by
Vesic (1975)]. Letting kq represent some The general expression for m is given as
general individual correction factor for the follows:
Nq term (for example, iq, which accounts
for load inclination), the relationship be- (5-19)
tween Nc and Nq suggests that
where
(5-15)
() = the angle between the long axis of the
footing and H; essentially in the case of a
Using Nc = (Nq- 1)cot <P and solving for continuous quay wall, () = 90°.
k c in terms of Nc yields 2 + L'jb'
1 +L'b'
1-k
k = k - q (5-16) 2+b'jL'
c q Nctan¢ (5-20)
1+b'jL'

Thus, the appropriate correction factors for Shape factors for rectangular geometry
the Nc term can be determined once it is are determined as follows.
known for the Nq term. Most expressions
for correction factors for Nq and N-y are
determined empirically. Following are the (5-21)
recommended expressions for the correction
factors.
Inclination factors are computed from Sq = 1 + (~:)tan¢ (5-22)
the following.
When <P > 0 8=1-0.4(~)
"~ L'
(5-23)

iq = ( 1-
H )m Depth factors are computed from the
Vu + b 'L' c cot ¢ following expressions:
H )m+l
(5-17)
i'Y = ( 1-
Vu + b 'L' c cot ¢ dq = 1 + (2tan ¢)(1- sin ¢) 2 ( ~) (5-24)

. • 1- i q
l = l - (5-25)
c q Nctan¢
1-d
d = d - q (5-26)
When <P = 0 c q Nctan¢

mH It should be emphasized that the effect of


ic = 1 - - - - (5-18)
b'L'cNc foundation embedment is very sensitive to
soil disturbance at the soil-structure inter-
where face along the sides of the embedded base.
b' and L' =the width and length of the founda- Where significant disturbance is expected,
tion effective area A', respectively; it may be prudent to reduce or discount
Gravity-TypeQuayWalls 473

entirely the beneficial effect of the over- diate, long-term, and differential settlement
burden shear strength. (wall tilt) are conducted when required to
assure that the wall displacements are
5.3.2.4.5 Base and ground surface within tolerable limits for the overall satis-
inclination factors factory performance of the structure.
For a horizontal base and a horizontal Walls built on dense granular soils would
ground surface, inclination factors are usu- undergo most of the expected settlement by
ally taken as equal to zero. the time its construction and backfilling are
Because in most practical cases of quay completed; there, the long-term settlement
wall construction the base of the wall is is negligible because settlement immediate
horizontal, the base inclination factors bq = due to the rapid dissipation of pore pres-
be = 1.0. For sloped ground surfaces, incli- sures. In the case of cohesive soils with
nation factors are computed from the fol- consolidation potential, the wall will con-
lowing expressions: tinue to settle for some time after the com-
pletion of construction, because excess pore
gq = g 1 = (1- tan ,8) 2 when 4> > 0 (5-27) pressure dissipates very slowly in highly
1- gq impermeable soils. When long-term settle-
gc =gq- Nc tan</> when 4> > 0 (5-28) ment is expected, then in order to keep the
settlement relatively uniform, the resultant
2,8 force, R, should be kept within the middle
gc = 1- N when 4> = 0 (5-29)
c of the wall base.
A reliable prediction of the various
where f3 is the ground inclination angle in settlements requires a thorough knowledge
radians. of the soil properties and subsurface varia-
Numerical values of Nc, Nq, and N 1 are tions along the wall. Comprehensive labora-
tabulated in Table 5-3 and are also pre-
tory testing of quality samples is required
sented in Figure 5-62.
to better understand the stress history,
time-rate consolidation characteristics,
5.3.2.4.6 Safety factor
Young's modulus, and the effects of cyclic
The quay wall foundation should have an loadings on the engineering properties of
adequate margin of safety against bearing the foundation soils.
failure. Hence, values of ultimate bearing Where the foundation soil varies greatly
capacity determined by the described along the wall, a differential settlement may
method are normally reduced by a factor of result. In this case, the wall performance
safety of 2.0 for normal loading and may be can be improved by replacing, compaction,
reduced by 1.5 for extreme load conditions. or stabilizing the relevant soils along the
These values should be used after cyclic wall or by reducing the contact stresses on
loading effects (if any) have been taken into the wall base and, therefore, on the founda-
account. Where geotechnical data are sparse tion. The latter can be achieved by increas-
or site conditions are particularly uncer- ing the width of the wall base andjor by
tain, an increase in values of factor of safety increasing the depth of the stone bedding
may be warranted. (mattress).
A certain amount of wall tilt should be
5.3.2.5 settlement and Tilt expected when it settles. When wall tilt is
Appreciable settlement of a gravity wall not associated with a foundation failure, its
built on nonbedrock foundations may be ex- rotation toward and away from the fill is
pected; therefore, analyses to predict imme- most likely to occur. The problem with wall
474 Gravity-Type Quay Walls

Table 5-3. Bearing capacity factors Nc, Nc Q, and Ny

NY
8d Caquotand
(degrees) Nc Nq Kerisel (1953) Hansen (1970)

0 5.14 1.00 0.00 0.00


1 5.38 1.09 0.07 0.00
2 5.63 1.20 0.15 0.01
3 5.90 1.31 0.24 0.02
4 6.19 1.43 0.34 0.05
5 6.49 1.57 0.45 0.07
6 6.81 1.72 0.57 0.11
7 7.16 1.88 0.71 0.16
8 7.53 2.06 0.86 0.22
9 7.92 2.25 1.03 0.30
10 8.35 2.47 1.22 0.39
11 8.80 2.71 1.44 0.50
12 9.28 2.97 1.69 0.63
13 9.81 3.26 1.97 0.78
14 10.37 3.59 2.29 0.97
15 10.98 3.94 2.65 1.18
16 11.63 4.34 3.06 1.43
17 12.34 4.77 3.53 1.73
18 13.10 5.26 4.07 2.08
19 13.93 5.80 4.68 2.48
20 14.83 6.40 5.39 2.95
21 15.82 7.07 6.20 3.50
22 16.88 7.82 7.13 4.13
23 18.05 8.66 8.20 4.88
24 19.32 9.60 9.44 5.75
25 20.72 10.66 10.88 6.76
26 22.25 11.85 12.54 7.94
27 23.94 13.20 14.47 9.32
28 25.80 14.72 16.72 10.94
29 27.86 16.44 19.34 12.84
30 30.14 18.40 22.40 15.07
31 32.67 20.63 25.99 17.96
32 35.49 23.18 30.22 20.79
33 38.64 26.09 36.19 24.44
34 42.16 29.44 41.06 28.77
35 46.12 33.30 48.03 33.92
36 50.59 37.75 56.31 40.05
37 55.63 42.92 66.19 47.38
38 61.35 48.93 78.03 56.17
39 67.87 55.96 92.25 66.75
40 75.31 64.20 109.41 79.54
41 83.86 73.90 130.22 95.05
42 93.71 85.37 155.54 113.95
43 105.11 99.01 186.54 137.10
44 118.37 115.31 224.64 165.58
45 133.88 134.87 271.76 200.81

Source: From Det Norske Veritas (1977).


Gravity-Type Quay Walls 475

tilt may be alleviated by using a pressure- parameters such as time-dependent consoli-


relieving platform at the wall interior, dation characteristics, Young's modulus,
thickening the rubble mattress at the toe, and Poisson's ratio that are required for the
and using lightweight backfill material (e.g., aforementioned analyses can be obtained
slag and the like). from standard penetration tests (SPTs).
The rate at which settlement and However, for either type of field data, one
settlement-induced tilt of the wall will oc- must make some adjustment for overconsol-
cur can be determined by methods de- idation, overburden pressure (if very low),
scribed in standard texts on soil mechanics and the presence of compressible contami-
and foundation engineering. As concluded nants of miscellaneous nature.
by Bowles (1987), the use of the conven- A conceptual framework for better un-
tional elastic settlement equations can pro- derstanding the effects of overconsolidation
duce the credible result for analysis of wall in reducing the compressibility of all types
settlement on sand deposits (depth of sand of soils has been presented by Leonards and
is equal to or more than 5b, where b is the Frost (1988). These authors indicated that
width of the foundation base). Elastic soil the stress range within which the soil re-

1000
800 I I
SELECTED TABULATED VALUES
600 r-
<!>' Nq Nc N~ /
400 ,... 0 1.00 5.14 0.00
5 1.56 6.40 0.45 /
200 -
10 2.47 8.33 1.22
/ f
i,/
15 3.94 10.97 2.65
20 6.39 14.81 5.38
25 1\).66 20.71 10.87
100 - 30
80 -
18.40 30.14 22.40 I /I
"£ 60 -
35 33.29 46.11 48.02 / I
.I//
40 64.19 75.31 109.40
0
z 134.87 133.87 271.74
<( 40 - 45
;l 50 319.05 266.87 762.84
// j;foq
<"'" 20
/ '/

..- /
/
""I
/
10
8 ..-
. - Nc /
6 ..- / /1
1--
/;
V,/
4

/ IN~

VI t/
2

10 2C 30 40 50
SOIL FRICTION ANGLE (<!>), DEGREES

Figure 5-62. Values of Nc, Nq, and NY as a function of


the soil internal friction angle. [From Det Norske Veritas
(1977).]
476 Gravity-Type Quay Walls

mains overconsolidated is dependent on the The value of allowable tilt is basically


stress path, both during unloading and governed by the type of equipment operat-
reloading. ing on a wall grade. For example, the wall
Although, in general, the complexity of tilt may affect the front rail of the crane
the stress field beneath the wall base makes track located within the sliding wedge,
analytical methods extremely difficult to de- which can be displaced both horizontally
velop, the time-settlement and initial set- and vertically due to wall tilt. Thus, in the
tlement relationships proposed by Terzaghi case in question, the wall tilt must not
(1943) and Poulos and Davis (1974) gener- exceed the allowable tolerance for the crane
ally give reasonable predictions, provided rail displacements.
that accurate soil data and soil stratigraphy Gurevich (1969) and Brum et al. (1959)
are available. Settlements caused by cyclic recommend the values of allowable settle-
loading are much more difficult to predict ment, tilt, and horizontal displacement of
due to the generation of excess pore pres- quay walls of different constructions as pro-
sures and the reduction in soil stiffness. vided in Table 5-4. For more information,
The initial settlement is the immediate re- consult Young (1992).
sponse of the soil to the loading that occurs
from the elastic deformation of the soil, and 5.3.2.6 Global (Overa/IJ stability
the long-term vertical wall settlement de-
The overall stability of the wall will be
pends on the soil type and the magnitude of
compromised if the mass of soil located
imposed load.
around the wall fails in shear along a slip
An estimate of vertical settlement for
surface. It involves a downward and out-
each soil layer included into the wall foun-
ward movement of the entire mass of soil
dation (u) can be obtained from
that participates in the failure. This type of
failure may happen when soil underlying
(5-30) the wall base is stratified with poor mate-
rial. A circular slip failure mode is normally
used for the analysis of wall overall stabil-
where ity. The analysis proceeds in the following
uu = vertical consolidation settlement sequence (Fig. 5-63):
h' = layer thickness
e0 = initial void ratio of the soil 1. Draw the walljsoil system.
Cc = compression index of the soil 2. Draw a circle about point 0 with a radius
u; = initial effective vertical stress sufficient to penetrate the poor soil stra-
tum (if present).
1!. uu = added effective vertical stress
3. Divide the system outlined by slip circle
on segments with width b and compute
The total consolidation settlement is the all forces associated with each individual
sum of the settlements computed for each segment.
individual soil layer. In the above calcula- 4. Compute both driving (Md) and resisting
tions, the soil compressibility characteris- moments (Mr) about point 0, and deter-
tics should 'be based on one-dimensional mine the safety factor against sliding
standard consolidation tests on high-quality along the slip circle as
undisturbed samples. Thick layers of homo-
geneous clay should be subdivided in the
analysis. When the added vertical effective
stress varies greatly through a layer, the
stresses at the layer midpoint are used. Because in the conventional slip circle
Gravity-Type Quay Walls 477

Table 5-4. Allowable displacements for gravity quay walls

Mode of Displacement (em)


Uniform Settlement (em) Tilt (radian) Horizontal (em)
Type ofWall Gurevich Brumet al. Gurevich Brum et. al. Gurevich Brumet al.

Blockwork 10-15 0.01


Floating-in caisson 12-15 15-20 0.005-0.008 0.015 5-8
L-shaped with
internal anchorage 10-12 0.005-0.008 4-6
L-shaped with
external anchorage 10-12 0.005 4-6

Notes: 1. Greater values are applicable to walls with no portal or gantry cranes installed.
2. Where dense granular foundation soil is present, the wall tilt practically is governed by live loads only. When the wall
is founded on consolidated-type soils, the effects of the long-term wall settlement should be acounted for, in addition to
the live load.
Source: Data from Gurevich (1979) and Brum et al. (1959).

Figure 5-63. Design diagram for calculating wall global stability.

method the effect of shear force existing In typical calculations, the weight of each
between adjacent segments is not taken into segment gi is resolved into normal, gn(i)'
account, this method, in general, is treated and tangential, g t(i)' components, relative
as conservative. Therefore, a factor of safety to the slip circle; essentially,
equal to 1.1 is usually considered as satis-
factory. Essentially, the slip circle method
is of a trial nature; all the above calcula-
tions should be repeated by drawing new where a.; = arcsin(rJR). The shear force
slip circles about the points 0 2 , 0 3 , .•. , 0; in along the slip circle area for each segment,
an attempt to find out the smallest value of t;, of the cohesionless soils is computed as
F,. This type of analysis is usually con- t; = g n(i) {, where the friction coefficient f =
ducted with the help of a computer. tan c/J; and c/J; is the angle of internal fric-
478 Gravity-Type Quay Walls

tion of soil at each segment, i, of a circle. factor of safety while, simultaneously, satis-
When the soil has both cohesion and inter- fying all three limiting equilibrium equa-
nal friction, an additional shear force at the tions for the entire sliding mass. Conse-
slip circle equal to cJi, where ci is the soil quently, for a prescribed sliip surface the
shear strength at particular location and l corresponding factor of safety can be
is the length of a shear plane, at a particu- estimated without resorting to statistical
lar segment can be added to ti, which will assumptions.
then be expressed as ti =gicos ai tan cf>i + More recently Espiuoz~ et al. (1994) pro-
cJi· posed a unified stability analysis based on
Subsequently, the resisting moment at limit equilibrium concepts and slice dis-
segment i will be cretization. The proposed approach is based
on a concept of noncircular slip surfaces
and it is also incorporating the most cur-
rent methods of stability analysis. The
accordingly, the driving moment will be ex- aforementioned authors have developed
pressed as computer program SLOPAS for stability
analysis.
In conclusion of this section it is worth
The safety factor against sliding along mentioning that wall movements to a
the slip circle will be expressed as greater or lesser degree have always been
experienced and should be expected unless
it is founded on hard rock. To date, not
many failure cases have been reported, and
in nearly every case of those reported, the
R I:i(gi cos ai tan cpi + cJ) cause can be traced to a bad foundation. A
(5-31)
few notable examples of gravity wall global
failures are given in Cornick (1958). Almost
The effective shear parameters cf>i and ci all of them indicate either neglect in obtain-
of a cohesive soil which are necessary to ing sufficient site geotechnical data or not
calculate the wall stability are determined appreciating the significance of some soil
by standard methods used in modern soil characteristics.
mechanics and foundation engineering. To
account for inaccuracies in the laboratory
tests, the values of ci obtained from testing 5.4 DESIGN OF BLOCKWORK
undisturbed samples are typically reduced QUAY WALLS
by a factor of 1.3. In the case of the presence
of unbalanced hydrostatic pressure, the
mass (weight) of water should be included 5.4.1 Basic Design Principles
in the g i calculations.
Leshchinsky (1990) presented a general- As stated earlier in this chapter, the block-
ized slope stability method, which is based work walls are the oldest known type of
on the variational limiting equilibrium structure used for quay wall and breakwa-
analysis, introduced by Baker and Garber ter constructions. Because a substantial
(1978). The differential equation describing part of these structures is usually built un-
the normal stress distribution, provided by derwater by divers, as well as the need to
the former authors, determined analytically use very heavy marine construction equip-
through a process of minimization of the ment, sometimes makes blockwork quay
Gravity-Type Quay Walls 479

walls quite expensive. However, where site As stated earlier, blockwork walls pro-
conditions are favorable (e.g., long quay wall duce substantial pressure on foundation
to be founded on a hard, competent founda- soils and normally the most favorable con-
tion strata), a scarcity of skilled workers dition for this type of construction exists
and foreign exchange and the availability of where blocks can be laid directly on rock
cheap labor to cast a large number of con- leveled with stone bedding or with in-situ
crete blocks the blockwork wall can be very concrete footing; where bedrock does not
successful and present a competitive con- exist, the blockwork wall must be founded
struction alternative. on a carefully graded, well-compacted, and
The size of individual blocks is usually thoroughly screed rubble mattress. When
determined by wall stability requirements the underlying foundation soil is not a good
and by the capacity of the available block quality rock, the minimum thickness of a
handling equipment both onshore and off- rubble mattress should be no less than 1.0
shore. The weight of concrete blocks may m. The surface of a mattress must be care-
vary from 5 to more than 200 tonnes. When fully cleaned from any sediments before the
local conditions so demand, the blocks can first course of blocks is placed, so that the
be cast with large pockets or voids to suit interface between blocks and mattress does
available handling equipment. If these not become a sliding failure plane.
blocks are too light to resist the design When fine-grained granular material un-
loads, the pockets or voids can be filled with derlies the rubble mattress, then to prevent
concrete or iron ore. Sometimes if the occa- the mattress from settling under the heavy
sion so demands, the interconnecting block load, the voids in rubble must be filled with
reinforcement is provided in order to turn suitably graded granular material. Other-
the wall into a monolithic structure. This wise, the gravel filter must be placed be-
type of block anchoring is particularly bene- tween the foundation soil and the mattress.
ficial where frequent and severe stress re- The quay wall design is usually based on
versal is expected (e.g., in the case of a quay static analysis conducted on the basis of
wall used as a breakwater). If practical, at-site geotechnical, hydraulic, wind, and
individual blocks should cover the entire ice conditions as well as dock operation cri-
width of the wall; they have to be keyed to teria. The latter typically includes complete
provide for continuity along the wall. The information on a design vessel, cargo
blocks are to be shaped and placed in a way handling and hauling equipment, and mis-
to provide for the least horizontal soil thrust cellaneous live loads associated with dock
and for the best distribution of bearing operation.
stresses at the wall base. This may be The above information along with care-
achieved through balancing the vertical and fully prepared structural design criteria is
horizontal forces acting on the structure. used for wall design. The wall design crite-
An example of walls with a well-balanced ria normally includes the following:
cross-section geometry are depicted in
Figures 5-8b, 5-8c, and 5-9. 1. Definition of all design loads and load
The concrete used for fabrication of blocks combinations
should be dense and resistant to effects of 2. Establishment of the most unfavorable
local water conditions. Typically, for better cases of load-structure interaction
interaction, blocks are placed in a way to 3. Conditions (factors of safety) for wall sta-
overlap each other; however, as indicated in bility (e.g., stability against sliding, over-
Section 5.2.1, in some cases they are placed turning, contact pressures, foundation
in the form of individual columns. bearing capacity, overall stability, and al-
480 Gravity-Type Quay Walls

lowable vertical and horizontal displace- being distributed from the bollard founda-
ments) tion block down the wall along vertical joints
4. Requirements for materials to be used for between adjacent blocks, as indicated in
fabrication of wall structural components Figure 5-64. At each design level i, the
and for the backfill material mooring force per wall unit length is
5. Recommended sequence of wall construc- equal to
tion and construction tolerances
(5-32)
Typical loads and load combination cases
used for the design of gravity-type struc-
tures are discussed in Section 5.3 (Fig. 5-59) Accordingly, the moment at level i per
and in Chapter 3. unit length of wall added by the mooring
force to moments created by other loads is
computed as follows:
5.4.2 Design Phase 1 (5-33)
At this design phase the quay is loaded
according to the load case (a), Figure 5-59. where z i is the arm of the mooring force at
As stated earlier this load case is critical for level i.
wall stability against sliding at the base
and at the interface between stone bedding 5.4.2.2 Soil Lateral Thrust
and foundation soil; it is also critical for
This is determined according to the rec-
wall stability against sliding at horizontal
ommendations provided in Chapter 4. It is
joints between each coarse of blocks.
calculated with due consideration given to
This load case is also critical for wall
load sheltering effect produced by overhang
overturning stability and is used for evalua-
blocks. Normally the assumption of the ac-
tion of contact stresses at the wall base.
tive pressure is acceptable provided that
wall movement away from fill is granted;
5.4.2.1 Mooring Force otherwise soil pressure 'at rest' should be
The mooring force (HM) must be added considered.
to loads as defined in aforementioned load With generally insignificant blockwork
case (a). The mooring force is considered as wall movements at the base as well as usu-

st El<.P. :JOINT t
+-·-----'---------+

Figure 5-64. Distribution of mooring force vertically along the blockwork quay
wall.
Gravity-Type Quay Walls 481

ally shallow foundation depth the effect of sure, then associated uplift force I: U must
passive soil pressure, if any, is typically also be considered. The diagram ofthis force
neglected. An example of soil pressure dia- will have a trapezoidal shape (Fig. 5-66).
gram is illustrated in Figure 5-65b. Note that both lateral and uplift pressures
can be determined more accurately by the
5.4.2.3 Effects of Unbalanced use of an equipotential flow net; the actual
Hydrostatic Pressure pore pressures derived from a flow net can
be used in the calculations.
The unbalanced hydrostatic water pres- The full uplift force acting on a wall in-
sure, I: Hh, due to differential water levels cludes the uplift force at the wall base and
in front of and behind the wall can be repre- uplift forces acting on parts of the wall
sented by the trapezoidal diagram as shown projected beyond its base, as indicated in
in Figure 6-65. As stated earlier, the value Figure 5-66. Naturally, when sliding and
of I: Hh depends on local wave climate and overturning stabilities or contact stresses
tide conditions, permeability of backfill and are reviewed at any wall level i, the full
mattress materials, as well as permeability uplift force at this level should be included
of the wall itself. In general, blockwork quay in the calculations.
walls are permeable enough to allow for
relatively free flow of water in and out.
5.4.2.4 vertical Loads
Hence, the effects of the unbalanced hydro-
static pressure can be ignored or at least In the case in question, the vertical loads
significantly minimized. represent the effective weights of the struc-
However, if there is a good reason to ture and of the associated backfill material.
consider the unbalanced hydrostatic pres- When friction between the wall and the

<nl (b)

£ 1-
(c )

--,--
0,
-2 -
bl/2
Vs

3-

4-
's
va
~ v7

7-
28- 30"
Figure 5-65. Loads for calculating blockwork quay wall stability. (a) Typical cross section; (b)
soil pressure diagram; (c) unbalanced hydrostatic pressure diagram.
482 Gravity-Type Quay Walls

Figure 5-66. Uplift forces due to the effect of an unbal-


anced hydrostatic pressure.

backfill material is considered, the addi- cific experimental data, a friction coefficient
tional vertical force at the interface be- between dry concrete blocks fc(dr ) = 0.6 can
tween the wall and the backfill, VH, should be used, which may be considered to be
be added. fairly conservative; correspondingly, for un-
Where the fill grade behind the wall is derwater blocks, the value of the friction
horizontal, this force can be determined as coefficient is reduced to 0.5. The most recent
VH(i) = Hs(i) tan 8i, where 8i is the angle of discussion on concrete-to-concrete friction is
wall friction. Therefore, the total vertical given in Tassios and Vintzeleon (1987) and
load at any horizontal section of the wall Braestrup (1988).
can be expressed as I:: v = I:: cv; + VH( i)).
5.4.2.6 stability Against
5.4.2.5 stability Against Sliding overturning
This is evaluated in accordance with the This is evaluated according to the gen-
procedure discussed in Section 5.3. Stability eral procedure discussed in Section 5.3.
against a sliding mode of failure should be Similar to stability against sliding, the
evaluated at each potential horizontal slid- overturning stability should also be evalu-
ing plane, such as at the interface between ated at each critical horizontal plane.
the superstructure and the upper course of The overturning moment at any horizon-
blocks, at joints between all courses of tal plane i across the wall Mo(i) is deter-
blocks, at the base of the wall, and at .the mined as the sum of the moments produced
interface between the stone bedding and by lateral soil thrust I:: Hs( i )• unbalanced
the foundation soil. In the absence of spe- hydrostatic pressure I:: Hh( i )• uplift force
Gravity-Type Quay Walls 483

I: ll;,, and mooring force HM(i)· Hence, this, the live load acting on the wall is
placed as postulated in Figure 5-59b; at
MO(i) = L: Hs(i)hs(i) + L: Hh(i)hh(i) this stage the mooring force is not included
(5-34) in the load combination. To avoid the "open
joint" condition at Section 2-2 on the har-
bor side, the load eccentricity should not
Correspondingly, the resisting moment at
exceed l ::;; b j6. Normally, the resultant
any design horizontal plane MR(i) is ob-
force at Section 2-2 is shifted from the
tained from
center of the section toward the fill, there-
(5-35) fore, the value of e is negative. If minimum
contact stress on the harbor side at Section
The effective weight of each individual 2-2 is negative, then the overhang of the
wall element V; is concentrated at its cen- upper block should be reduced accordingly.
ter of gravity, and the corresponding lever- At this design stage the contact stresses
age r; is taken from the point of rotation to and sliding stability at the base of all other
the element's center of gravity. courses of blocks should also be evaluated
Again, all calculations on overturning and where e; is found in excess of bJ6, th~
and resisting moments can be easily com- wall overturning stability at this level
puterized. should be evaluated.

5.1.2.7 contact Pressures 5.4.3.1 strength of the Upper


Block
Analysis of contact pressures should be
carried out for each course of blocks, as well There is always the possibility of having
as at the wall base and at the interface a small gap between the upper and under-
between the wall mattress and the founda- laying course of blocks on the backfill side.
tion soil. It is typically carried out according Therefore, in practical design, the over-
to the procedure discussed in Section 5.3. hanging length of the upper block l 0 is
As stated earlier, in order to prevent major usually increased by about ~ b 1 , where b 1 is
tilt of any type of gravity wall (the block- the width of the underlaying block. Hence,
work wall in particular), the eccentricity of the design length ZO(d) is computed from
the resultant force at the wall base in gen- lo(d) = Zo + ~b 1 (Fig. 5-67).

eral should not exceed e ::;; b j6; it may be The cantilever is then designed for a uni-
increased up toe ::;; b/ 5 in the case ofwalls form distributed load, q 0 , which is com-
founded on dense resilient soils and to e <
b/4 for walls founded on sou~d rock. The
latter criterion may also be used for evalua-
tion of stability of concrete blocks included
into the wall structure.
In all design cases, compliance with the
required safety factors is a must.

5.4.3 Design Phase 2


At this design stage the stability of the
upper course of blocks at Section 2-2 (Fig. Figure 5- 67. Upper coarse block-loads and design
5-65) and their strength are evaluated. For scheme.
484 Gravity-Type Quay Walls

puted as follows: q 0 = ( Ych 2 + 'Yc(b)ha) + capping superstructure is usually designed


Yrh 1 + q 1 , where Yc and 'Yc(b) are dry andjor in the form of an L-shaped wall.
buoyant unit weights of concrete, respec- The wall stability against sliding toward
tively, and Yr is the unit weight of fill. the harbor is evaluated according to the
Tensile strength in concrete per linear load distribution indicated in Figure 5-59d
meter of block along the quay, feu)• is com- with the addition of the mooring force H M.
puted from The strength of the cantilever part of the
wall at section a-a should be routinely re-
(5-36) viewed for the effects of ship impact force
H; transmitted to the wall through its
fender system (Fig. 5-68a) and for the com-
Because assumptions for this type of analy-
sis may be considered rather conservative, bined effects of the mooring force, H M, and
fc(t) can be taken as equal to or less than
soil lateral pressure E with the live load
80% of the concrete ultimate flexural ten- located as indicated in Figure 5-68b.
sile strength.
If the space between adjacent blocks is
grouted, then flexural tensile stresses per
linear meter in the upper block is deter- 5.4.5 Design Phase 4
mined from
At this stage the maximum contact stresses
at wall base, as well as at interface between
(5-37)
the stone bedding and the foundation soil
are evaluated. For this, the design loads as
indicated in Figure 5-59c with the addition
5_4_4 Design Phase 3 of mooring force (HM) and unbalanced hy-
drostatic pressure (if any) are considered.
At this design stage the stability of the Contact stresses are computed according to
capping superstructure is evaluated and its the procedure discussed earlier in Section
structural members are . proportioned. The 5.3.

-=:::-
-+"=='"-'.,.;:.:.==~----r-
1 - i --;.z::........"""""-'-"-':.;.;.:.,.....:..=,____+--
I
b

(a) (b)

Figure 5-68. Capping wall-design loads.


Gravity-Type Quay Walls 485

R
Finally, where required, the foundation
bearing capacity is evaluated. Assumption on caisson dimensions.

t
Calculation of stability during launch while
5.5 DESIGN OF QUAY WALLS buoyant, and during installation
COMPRISED OF FLOATED-IN
CONCRETE CAISSONS j_
Structural design:
(a) While Buoyant
5.5.1 Basic Design Principles 1. Determination of design loads.
2. Design perimeter wall and base slab.
There is no conceptual difference in ap- (b) During Installation upon foundation
proach to the design of the quay wall com- 1. Determination of design loads.
posed of floated-in caissons and a wall com- 2. Design of diaphragm wall. r-
posed of large concrete blocks as discussed 3. Review of previously design perimeter
~S~tion5A · wall and base slab.
Load cases (a), (c), and (d) as indicated in (c) While Completed and in Operation
Figure 5-59 are basically used for evaluat- 1. Determination of design loads.
ing wall slid~g, overturning and overall 2. Review of previously design perimeter
stabilities, and reviewing contact stresses wall and base slab.

t
at the wall base and at the interface be-
tween the rubble mattress and the founda-
tion soil. Where the wall is provided with a Evaluation of wall sliding and overturning
pressure-relieving platform or anchor slab, stability, and footing design

i
the design scheme as ~dicated in Figure
5-55d is also used. For a discussion on
basic design principles and procedures, the Design of superstructure, mooring and
reader is referred to Section 5.3. fender systems, built-in devices
A discussion of the design of a concrete for caisson control
capp~g wall is provided in Section 5.4. Soil during towing and installation
lateral pressures, unbalanced hydrostatic
pressure, and uplift load is determined as
discussed in Section 5.4 and ~ Chapter 4.
If the caisson is provided with an ex-
tended base slab, such as that illustrated in
Figure 5-19, then the lateral soil pressure 5.5.2 Buoyancy and Buoyant
on this wall should be determined in a man- stability of a Caisson
ner similar to that used for cantilever walls
with a short foundation slab (Chapter 4). Buoyancy and buoyant stability are princi-
In addition to static analysis, the caisson pal properties of a body immersed in a fluid.
is designed for buoyancy and stability dur- The amount of fluid displaced by the body
ing launch and during the voyage from the and the upward force (which is equal to the
fabrication yard to the deployment site weight of that volume of fluid) are com-
and during installation. The sequence of monly known as the buoyancy, and the vol-
caisson-type quay wall design is depicted in ume of the immersed art of the body is
the following diagram. known as the volume of displacement and is
486 Gravity-Type Quay Walls

usually used to describe the mass of water hatched area in Fig. 5-69b). The total
displaced rather than the volume. shaded area is the new displacement, which
The force opposing buoyancy is the weight must be matched by the new gravity force if
of the floating body, which, for equilibrium, the system is to remain in equilibrium.
must equal the buoyancy. These forces act If the weight (P), instead of being added
as ifthey were concentrated at points known at the axis of flotation, had been applied at
as the center of gravity (CG) and the center one side of the floating body, the body would
of buoyancy (CB) (Fig. 5-69a). Because tilt. Because the total weight of the floating
gravity acts downward and buoyancy up- system would remain the same, as in the
ward, for equilibrium the two centers must case illustrated in Figure 5-69b, its buoy-
lie in the same vertical line. This line is ancy also would not change. However, the
called the axis of flotation. In order for the system's new center of gravity will shift
body to stay afloat, its immersed part must toward the added weight (P), and to meet
have enough volume to develop sufficient conditions of equilibrium, so will the center
buoyancy. For the condition of equilibrium, of buoyancy (Fig. 5-70). An increase in
weight (P) naturally would increase the
B = G = yV (5-38) system's tilt and its buoyancy, and contin-
ued increases in (P) could eventually turn
where over the floating body.
B = buoyancy of the floating body Considering a small tilt angle of a con-
G = weight of the floating body ventional caisson (which practically does not
y specific weight of the water
= affect the center of gravity) for the purpose
V = volume of the immersed portion of the of further discussion, assume that the axes
floating body of the caisson rotation AB and CD (Fig.
5-70a) always run through the center of
The stability of a body immersed in a fluid gravity of the caisson water plane area and
is the body's ability to balance an action of therefore are the principal axes of this area.
moments that try to turn it over. Adding a
weight (P) at the axis of flotation will in-
crease the floating body's buoyancy (cross- (a I

d,
Pl•n

Section c-o
Figure 5-69. Buoyancy of a body immersed in a
fluid. Figure 5-70. Caisson stability.
Gravity-Type Quay Walls 487

The caisson tilt could be caused by the where


action of eccentrically applied force (P) y x ep( dF) = buoyancy force
which is the case when one of the compart- I = moment of inertia of waterplane
ments is filled with ballast. Because, practi- area in relation to axis of tilt AB.
cally, caisson inclination is limited and the
angle of tilt cp is rather small, the inclina- The weight of the system along with the
tion could be expressed as cp = tan cp = reaction of water also produces the pair of
sin cp. Note that cp is the angle between the forces. Assuming sin cp = tan cp = cp, the
caisson original waterline W0 L 0 and the moment of this pair M" can be described as
waterline W1 L 1 after caisson tilt (Fig.
5-70b). The action of the force that tends to M" = yVaep (5-40)
tilt the caisson could be assumed to be the
combined action of force P, applied at the where V is the volume of displaced water.
caisson's center of gravity and the bending The sign of this moment depends on the
moment M = Pl, as shown in Figure 5-70b. distance a. Under a> 0, this moment ro-
There is now a leverage, or moment, trying tates the system in the same direction as
to bring the caisson back to the upright moment M does, thus adding additional tilt
position, which is known as the righting to the system. For a = 0 and M" = 0, and
moment. A caisson that, when displaced by when a < 0, M" produces a stabilizing ef-
reasonable amounts from the upright posi- fect. Under conditions of equilibrium,
tion, possesses the ability to return to the
M' +M" +M=O
upright is said to be stable; the capability is
known as stability. and therefore
Assume that the distance between the
caisson center of buoyancy (CB) and its cen- M = yep(] - Va) (5-41)
ter of gravity (CG) is equal to a. There are
three practical positions of CG and CB: from which

1. When CG is located above CB (a > 0) M M


ep= - - - - - (5-42)
yV(IjV- a) yV( p- a)
2. When CG is located below CB (a < 0)
3. When CG and CB coincide (a= 0). Equation (5-42) can be rewritten as

When the caisson tilts under the action of ep = Mjyla (5-43)


the bending moment M, the forces of grav-
where
ity and buoyancy are positioned on a new
axis of flotation, and a new pair of forces a=l-aVjl
(wedges as are shown in Fig. 5-70) will
produce a stabilizing moment M 1 • Equation (5-42) is known as the metacen-
Because the value of angle cp is small tric formula of floating body stability. In
enough, the moment M could be presented
1
this equation, p = I jV is known as the
as metacentric radius above the center of
buoyancy; note that p is not a visible part
of a floating object.
M' = - JF yxep(dF) x = -yep JF x 2 dF = yep] Considering the caisson's rectangular
form at the water plane area, the trans-
(5-39)
verse metacentric radius (about axis CD,
488 Gravity-Type Quay Walls

Fig. 5-70a) is equal to its center of gravity and increase displace-


ment, thereby reducing the caisson's meta-
f2lb 3 b2 centric radius.
(5-44)
p = lbd = 12d Caisson draft at a distance x from the
axis of tilt is expressed as
where
l =length dx = Mxjaly (5-46)
b =width
d =draft; d = Gjlby, where G is the caisson As discussed in Chapter 3, the floating body
weight
has six degrees of freedom in movements.
However, for sake of convenience and sim-
As can be seen from Eq. (5-43), the caisson's
plification assume that the caisson is ro-
length l does not affect its transverse meta-
tated about two principal axes x-x and y-y
centric radius.
only as indicated in Figure 5-71. Assume
A typical caisson is compartmentalized
now that caisson is exposed to a concen-
by watertight bulkheads. When these com-
trated load P with coordinates lx and ZY.
partments are filled with ballast water dur-
Then formulation (5-4 7) can be used to
ing caisson installation, its metacentric
determine the amount by which the caisson
radius is expressed as
may sink at different locations:
(I- Ei)
p' = (5-45)
v
in which E i is the sum of the moments of
inertia of the water surface in each com-
where
partment. The stability of a floating body is
determined by the value of its metacentric de = the caisson sinking at point C due
height above the center of gravity, which is to the effects of load P only
expressed as p - a. A = caisson area of water plane
With Eq. (5-42), one can judge the effect Ix and Iy = moments of inertia of the caisson
of p- a on caisson stability. A necessary water plane area at axes x-x and
condition of caisson stability is that p - y-y
a > 0, which means that the caisson meta-
center must be located above the center of Correspondingly, ax= 1- aVflx and ay =
gravity. For p- a = 0, an infinitively large 1- aVjiY.
tilt is likely. If p - a < 0, the floating object
is unstable and the smallest unbalanced y
force could cause it to overturn.
As a minimum requirement for the cais-
son stability while in still water, p - a ;;:::: 0.2
m. This, however, does not consider the x---t--
wave and wind effects which can tilt the
caisson substantially. Where a strong cur-
rent or substantial waves are expected dur- 1..
ing caisson installation, a minimum value
y
of p - a = 0.5-1.0 m is recommended. Fur-
thermore, heavy equipment or another live Figure 5-71. Plan of caisson eccentrically loaded
load placed on a floating caisson will change with load P.
Gravity-Type Quay Walls 489

5.5.3 Buoyancy and stability accidental damage; hence, the impact of this
of a Damaged Caisson on the caisson buoyancy and stability must
be evaluated.
If the caisson has to be towed a long dis- Depending on the watertight-bulkheads
tance, then accidental damage to the cais- system, the caisson flooding could be sym-
son hull should be considered. If the hull is metrical or asymmetrical. Symmetrical
damaged, water may enter the caisson, and flooding occurs if the caisson has transverse
if unrestricted flooding is permitted, the watertight bulkheads only (Fig. 5-72a);
caisson will eventually sink. To prevent this, asymmetrical flooding is typical for caissons
or at least reduce the probability, conven- divided into a series of watertight compart-
tional caissons are usually divided into a ments by transverse and longitudinal
series of compartments by watertight bulk- watertight bulkheads (Fig. 5-72).
heads, which typically extend from wall to Assume that two compartments of dam-
wall in caisson transverse and longitudinal aged caisson (shown shaded in Fig. 5-72)
directions. The compartmentalization of the are open to the water. Hence, the buoyancy
caisson can localize and contain accidental of the caisson between the bulkheads
damage and thus avoid uncontrolled sink- bounding these compartments is lost and
ing or capsizing. Two important effects of the caisson must sink in the water until it
caisson flooding must be considered: picks up buoyancy from the undamaged
compartments to maintain equilibrium. At
1. The caisson must sustain loss of buoyancy the same time, because of the loss of buoy-
of a limited magnitude. ancy, the overall position of the caisson's
2. The caisson must remain stable in a dam- center of buoyancy is altered. If the caisson
aged condition. flooding is symmetrical (Fig. 5-72a), the
center of buoyancy is shifted ·longitudinally
If a caisson is designed for a long journey, only; if flooding is asymmetrical (Fig.
then it is usually assumed that two adja- 5-72b), the center of buoyancy is altered
cent compartments may be flooded due to both longitudinally and transversely. In or-

a-symmetrical flooding b-asymmetrical flooding


(a I IbI

Section A- A

Figure 5-72. Stability of a damaged caisson. !-Caisson; 2-watertight bulk-


head; 3-flooded area; 4-additional buoyancy to compensate for lost buoyancy.
490 Gravity-Type Quay Walls

der to restore the center of buoyancy to its effect of the water in the damaged compart-
original position, the caisson must be ments. The value of new metacentric height
trimmed by rearranging the ballast. ( pa) has to be determined, given the new
The caisson that was floating originally position of the caisson's center of buoyancy
at waterline W0 L 0 will now float at water- corrected for the new waterline (W1 L 1).
line W1 L 1 • It is usually considered that the Asymmetrical flooding would cause the
new waterline W1 L 1 is in no case higher at caisson to heel. It is important to limit the
any point than the top of the caisson. For angle of heel, and if it becomes excessive,
calculating the position of waterline W1 L 1 , the corresponding compartments on the op-
the added-weight method is commonly used. posite side of the caisson should be flooded.
In this method, the water entering the Although this cross-flooding would make the
damaged compartments up to the original caisson sink deeper into the water, it would
waterline can be regarded as an added eliminate the unacceptable angle of heel.
weight P = y bld and the values of a cais-
son sinking in water, d 1 , d 2 , d 3 and d 4
(Fig. 5-72) can be determined from Eq. 5.5.4 Caisson Launch
(5-47). It is obvious that the new average
draft is more than the original draft· d as- A caisson is typically launched from a slip-
sumed for the initial calculation of the added way, dry dock, shiplift, submersible barge
weight P. Hence, additional approxima- or pontoon, or from other facilities as dis-
tions in the calculation of P are necessary cussed earlier. Besides, it can be launched
to obtain the desired degree of accuracy. by method employed at Port-of Gdynia (Fig.
This calculation can be programmed for the 5-7). Similar case history is discussed by
computer, and any number of iterations Chung-Chuang Wu (1987). The following is
could be made. Once the new buoyancy con- discussion on slipway launching that may
dition is established, the caisson's stability have pronounced impact upon caisson
should be reinvestigated. structure. The design procedure have been
Naturally, the effect oflost buoyancy will developed by Acres International Ltd. (1987)
result in the rise of the caisson center of for Public Works, Canada.
buoyancy and, subsequently, in the reduc-
tion in its metacentric height. The magni- 5.5.4.1 5/ipway Launching
tude of the caisson's new metacentric This is generally considered in six stages
radius above its new center of buoyancy (Fig. 5-73).
should be determined by calculating a new
value of water plane area moment of inertia (i) Translational sliding on the dry part of
(I) that is accounting for the free.-surface the slipway

STAGE I

Figure 5-73. Stages of caisson slipway launching. [After Acres International Ltd. (1987).]
Gravity-Type Quay Walls 491

(ii) Translational sliding on the submerged about the y-axis plus the added moment of
part of the slipway inertia
(iii) Translation and rotation about the trail-
ing end of the caisson (pivoting) For each of the six stages of the caisson
(iv) Translation and rotation about the lead- motion along the slipway, Eqs. (5-48a)-
ing end of the caisson (5-48c) take different forms, as do the cor-
(v) Translation and rotation about the end responding equations of constraints.
of the spillway (tipping) Stage 1-Caisson sliding along the dry
(vi) Free caisson floating part of the slipway. The forcing parts of
Eqs. (5-48a)-(5-48c) are
In all stages, the caisson's equations of mo-
X 1 = Fv(sin (}- p. cos 0)
tion are integrated using the fourth-order
Runge-Kutta-Gill method. The time step Z1= -w (5-49)
recommended for simulation is about 0.1 s. Z 2 = F.(cos (} + p. sin 0)
Transition from one stage to another is
determined by the dynamic reaction of the with the initial conditions at t = 0
slipway on the bottom of the caisson. The
x(O) = x(O) = 0
location and magnitude of the reaction de-
pends not only on the external forces acting z(O) = z(O) = 0
on the caisson during launching but also on and the caisson mass mx = mz = m =
variables such as location, velocity, and ac- constant.
celeration. Formally, the caisson motion is The velocity of the caisson in the direc-
described using three degrees of freedom tion parallel to the slipway is
(DOF): translation in the x direction, trans-
lation in the z direction, and rotation 1/J u= [2gS 0 (sin (}- p. cos 0)] 0 "5 (5-50)
about the caisson's gravity center. As the
caisson is in contact with the slipway, the where
three DOFs have to be reduced using con- 8 0 = distance down the slipway from the leading
straints imposed on the caisson's motion. end of caisson to the water (m)
From the equations of constraints, one can g = 9.81 mjs 2 , the gravity acceleration
derive the unknown dynamic reaction of the
slipway and thus make possible the inte- Therefore,
gration of the caisson's equation of motion. X= U COS(}
The general form of the Newton's equa- (5-51)
z=usin(}
tions of motion applied to caisson launching
are as follows: These values are used as initial conditions
for the caisson entering the water.
mxx = r:x, (5-48a) Stage 2-Caisson sliding on the wet part
of the slipway. The forces and moments
mzz = L:z, (5-48b) acting on the caisson are (Fig. 5-74a)

JYJ, = L:M, (5-48c) I: X, = F:f cos (} + F.(sin (} - p. cos (}) (5-52)

where
L Z, = -(W- B)- F:f sin(}

mx = m + D.mx =total mass of the caisson plus + Fv(cos (} + p. sin(}) (5-53)


added mass in the x direction
mz = m + D.mz =total mass of the caisson plus
added mass in the z direction
JY = J + t:.JY =caisson's moment of inertia =0 (5-54)
492 Gravity-Type Quay Walls

r
(a}

(b)

Figure 5-74- (a) Stage 2-caisson sliding along wet part of slipway; (b) stage
3-caisson rotation about the trailing end while sliding. [After Acres International Ltd.
(1987).] Note: CG = center of gravity and CB = center of buoyancy.

where Substituting the accelerations obtained


rw =vector of the center of gravity from the caisson's equations of motion
rb = vector of the center of buoyancy (5-48a)-(5-48c) into constraints (5-55), the
r:J: =vector of the point of application of drag formula for the force Fv can be obtained:
force F:t
rv = vector of the point of application of reaction [W- Bt(l- e)F:! sin 0] cos()
of slipway Fv F
v
= --~----~-----------------
cos 20 + e sin 20 + p,(l - e) sin () cos ()
(5-56)
At this stage, the motion of the caisson is
restricted to the slipway surface and the
caisson cannot rotate. From these condi- where E = mzfmx is the ratio of the caisson
tions the equations of constraints are de- mass plus added mass in the z and x
rived: direction.
The location of the force Fv is determined
using Eq. (5-54) and the relation between
ji = -.X tan() and ;[! = 0 (5-55) the . x and y coordinates at the point of
Gravity-Type Quay Walls 493

application where
Mb =moment of buoyancy force
M:f =moment of drag force F:f
M:J =moment of drag force FJ
so that Mf = moment of unknown force Fv
Md. = moment of rotational drag

All moments are taken with respect to the


caisson's center of gravity. The motion of
the following end of the caisson is restricted
where to the slipway surface at this stage; thus,
Mw =moment of the caisson's weight W the following equation of constraint applies:
M b = moment of buoyancy force B
(5-59)
M:f =moment of drag force F:f
Mr =scaling moment of the force Fv obtained
assuming x f = 1
The point of application of force (x,, Yr) is
located at the following end of the caisson.
All moments are calculated with respect to Relating the coordinates (x,, Yr) to the
the global origin (x = 0, z = 0). If the point coordinates of the center of gravity (x 0 , y 0 ),
of application of the slipway reaction Fv is x 0 =xr+rcos'¥
located on the caisson base, then the motion (5-60)
is stable and the caisson slides on the slip- Yo =yr+ rsin'¥
way surface. If, however, the point of appli-
cation (w,, Yr) moves beyond the trailing where r is a distance between point (x,, Yr)
end or the leading end of the caisson, the and the center of gravity (x 0 , y 0 ), and sub-
additional degree of freedom, rotation, has stituting the caisson's acceleration into Eq.
to be added to the equations of motion and (5-59), the formula for Fv can be obtained,
the caisson enters into Stage 3 or Stage 4,
A+ B tan 0
as described below. The caisson also can F = ------ (5-61)
move directly to Stage 6 (free floating) if Fv v U+Vtan(}
becomes less or equal to zero. where
Stage 3-Caisson sliding on wetted part
of slipway and rotating about the trailing A= Q + r(!/1 2 sin a- r cos a)
end. The forces and moments acting on the B = P + r(r/1 2 cos a + r sin a)
caisson can be determined from the follow- P = (F:f cos r/J- FJ sin r/J)/mx
ing formulations (Fig. 5-74b): Q = [ -(W- B)+ F:f sin r/J + FJ cos rfJ]/mz
R = (Mb + M:f + M:J + Md)jJY
U =(cos(}+ J.L sin 0)/mz - G cos a
V =(sin(}- J.L cos O)fmx + G sin a

+ Fv(sin (}- J.L cos 0) (5-58a) G = rMr/Jy


M f = scaling moment for Fv = 1
L, Z, = -(W- B)+ F:f sin r/1 + FJ cos r/J
From Stage 3, the caisson can move to Stage
+ F/cos 8 + p, sin(}) (5-58b) 6 (free floating) when Fv becomes less than
or equal to zero.
L, M, = Mb + M:f + M:J + Mr = Md. (5-58c) Stage 4-Caisson sliding on wetted part
of slipway and rotating about leading end.
494 Gravity-Type Quay Walls

The routine to determine the force Fu fol- The basic difference between the equations
lows the same steps as during Stage 3. The of motion for Stage 2 and the preceding
only difference is that the location of the equations is that at this stage the direction
application point (xf, Yt) is at the leading of the force Fu is changing along with the
end of the caisson. caisson's angle of rotation 1/J. Also, the point
Stage 5-Caisson sliding along and ro- of application (xf, Yt) is moving along the
tating about the end of the slipway. The caisson's bottom.
following forces and moments are acting on The equation of constraint for this stage
the caisson (Fig. 5-75): can be derived from the fact that the height
of the center of gravity with respect to the
E X, = F:J cos 1/1 - F:J sin 1/J end of the slipway remains constant and
equals Ag. This relation can be expressed
as an angular relation (Fig. 5-75):
(5-62a)

E Z, = -(W- B)+ F:J sin 1/J + F:J cos 1/J 1/J=a+'l! (5-63)

(5-62b) Equation (5-63) has to be expressed in


terms of the caisson's state variables in
EM, = Mb + M:t + MJ + M:i_ + Mr (5-62c) order to be applicable as a constraint. This
relation is

(a)
arctan(
x cos 1/J
Za .
+ y sm 1/J
) =arctan(~)
x
+ 1/J
(5-64)

Differentiating Eq. (5-64) twice and substi-


tuting caisson state variables into the re-
EQUATION OP'
CONSTIIAIHT
sultant expression gives the formula for the
W'•a++ ·force Fu:

D -AP-BQ- CR
F = --------------- (5-65)
(b)
v AU+BV+ CW

where
A = - y + Z 0 cos 1/1
B = x + Z 0 sin 1/J
C = Z 0 (y cos 1/J- x sin 1/J)- r 2
r2 = x2 + y2
D = [2(ri- yy)jr]j[Ai +By+ (C + r 2 )1/J]
- 2Z0 (.i: sin 1/1- y cos 1/J )1/J + Z 0 (x cos 1/J +
y sin 1/1)1/1 2
u = (-sin 1/J + JL cos 1/J)fmx
u =(cos 1/J- JLSin 1/J)fmx
Figure 5-75. Stage 5-sliding and rotation of the
caisson over the tip of the slipway: (a) initial stage; (b) w = Mr/Jy
final stage. [After Acres International Ltd. (1987).] Mr =scaling moment obtained assuming Fv = 1
Gravity-Type Quay Walls 495

The caisson can move from this stage to the caisson is buoyant. The buoyancy and sta-
floating stage when Fu :$ 0 or when the bility of the buoyant caisson are computed
caisson leaves the slipway. as discussed earlier in this section.
Stage 6-Caisson free-floating. The free
floating of the caisson is described by the 5.5.4.3 Launch from
same equations of motion as for Stage 2, 3, Submersible Barge
or 4 assuming Fu = 0. Note that at no time
This launch process assumes that the
(specifically at Stages 4, 5, and 6) should
caisson is resting on a specially designed
the caisson's free board be less than 0.5 m.
barge and that the barge and caisson as-
The caisson buoyancy and stability parame-
semblage initially has sufficient floating
ters when buoyant are computed as dis-
stability. The assemblage is gradually sunk
cussed earlier.
by uniform flooding of the barge, and at
For a further discussion on caisson slip-
each draught increment, the new stability
way launch, the reader is referred to Yan
parameters (buoyancy, freeboard, metacen-
(1983).
tric heights; reaction between barge and
caisson) are evaluated. The caisson floating
5.5.4.2 Dry Dock or Shiplift
stability parameters are evaluated when it
Launching
becomes buoyant, as discussed earlier in
This launch process assumes that the this section.
caisson is initially resting on a dry dock An example of a large (28,000 DWT) sub-
floor or shiplift platform. The water depth is mersible vessel is shown in Figure 5-76. If
incremented by a specified amount until the a conventional ballasted pontoon is used for

Figure 5-76. Submersible heavy lift barge "Mighty Servant" 28,000 ·DWT. (From
Wijsmuller, Holland Bulletin.)
496 Gravity-Type Quay Walls

INITIAL POSITION FINAL POSITION

WATER PLANE
AREA

INTERNAL
WATERPLANE
AREA

Figure 5-77. Launch of caisson from submersible barge (pontoon).

caisson launching, then the critical phase of of four units separated by floating fenders,
the pontoon-caisson system sinking is when placed between individual units, or in train
the pontoon's deck crosses the water sur- of up to six units linked with each other by
face while the caisson is still sitting on a towline).
pontoon deck. At that moment, the water The tow boat minimum required power
plane area rapidly decreases (Fig. 5-77) Ne can be computed from the following em-
while the internal water plane area re- pirical formulation:
mains constant. This causes a rapid de-
Nt = FhV /75 (in horse power) (5-66)
crease of the system's metacentric heights,
which may result in loss of the floating where
stability of the assemblage. To overcome Fh = resistance to caisson tow (in kg)
this problem, the submerged pontoon should V = speed of tow in still water (in mjs)
have a sufficient margin of stability so that
the metacentric heights about its principal The resistance to the caisson tow includes
axis remains positive during the launching the resistance of water R w in combination
operation. with wind, current, and wave effects. The
The latter can be achieved by providing former, however, typically constitutes the
the pontoon with supplemental buoyancy major resistance factor. It can be deter-
tanks, extending up through the water mined from the following empirical formu-
plane. These tanks installed at each corner lation:
of the pontoon act as stabilizers, transform-
ing the pontoon-caisson system into a Rw = (k 1 A1 + k 2 A 2 )V 2 (in tonnes) (5-67)
semisubmersible vessel. For example con- where
sult chapter 8 (Figure 8-12).
k 1 = resistance factor of the front submerged
area k 1 ""0.06 tonness 2 jm4
A1 = submerged area of front leading wall of
5.5.5 Towing and Sinking individual caisson or cluster of caissons
(in m 2 )
Caissons are usually towed a certain dis- k 2 = friction of caisson submerged area k 2 ::::
tance from the site of fabrication to the 0.00015-0.0002 tonnes s 2 jm4
deployment site under no substantial wind A 2 =caisson submerged (wet) area which in-
and/ or wave conditions. Depending on size cludes both perimeter walls and bottom
and geometry, they can be towed either slab (in m 2 )
individually or in a group (e.g., in clusters V = speed of tow in still water (in mjs)
Gravity-Type Quay Walls 497

The current and wave effects on caisson teristics are changing during sinking pro-
towing are determined on a case-by-case cess and have to be calculated along with
basis as discussed in Chapter 3. the environmental forces. The direction of
The optimum length of towline Lt which the wind, wave, and the current is specified
provides the best caisson control while un- with respect to the long direction of the
der tow can be determined from the follow- caisson. The effects of environmental forces
ing empirical formulation: are discussed in Chapter 3.

0.67aL(20 + NP)
L = (in m) (5-68)
t 120 + Nt 5.5.6 structural Design
where The caisson structural elements such as
a= empirical coefficient a= 9.5 to 10.5 perimeter wall, bottom slab, and dia-
NP = tow boat actual registered power (in horse phragms should be proportioned in a way to
power) resist all critical loads which may act upon
L = length of individual caisson or cluster of the caisson during construction, launching,
caissons (in m) towing, installation backfilling, and while
in operation. Normally, the most critical
The tug must be equipped with a pump of loading conditions would exist during cais-
sufficient capacity and be ready to pump son tow and installation operations. The
water out of a leaking or damaged caisson. hydrostatic loading diagram of a floating
A more sophisticated approach to caisson caisson while in still water is shown in
towing is given in Acres International Ltd. Figure 5-78a. Sometimes however, the crit-
(1987) and Arctec Canada Ltd. (1984). ical loading condition may exist during slip-
The caisson positioning and sinking re- way launching, when the hydrodynamic
quires the evaluation of environmental loads due to caisson movement through the
forces acting on the caisson. The caisson's column of water should be added to the
draught as well as exposure to wind and normal hydrostatic loads.
current areas are changing during the cais- Naturally, the effects of caisson move-
son sinking; thus, the forces have to be ments and its exposure to environmental
evaluated for every step of sinking from the forces, if any, should be added to normal
towing conditions down to the final in-situ design hydrostatic loads. For the purpose of
position. Also, the caisson stability charac- installation on the foundation, the caisson

(a) (b)

r ~~~
.JIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII~

Figure 5-78. Caisson loading diagrams: (a) while afloat in still


water; (b) during installation with no fill.
498 Gravity-Type Quay Walls

should be ballasted. This is usually done by lowable stresses used for the normal load
pumping in water; at this stage, to keep the condition design.
caisson horizontally it should be carefully Structurally, the caisson cell may be con-
balanced by water or by the other means. sidered as a silo (bin), and therefore pres-
Obviously, in the case of a symmetrical sure produced by fill material on caisson
caisson, the required minimum volume of walls and the bottom slab is computed ac-
ballast water (Acd where Ac is the total
1
), cordingly. A simplified approach to caisson
area of the compartments filled with ballast wall design is usually used. Generally, the
water and d is the depth of water in the
1
wall is usually divided into two parts. The
compartments, should be somewhat more lower part is assumed to be extended from
than the volume (weight) of water displaced the base slab up to 1-1.5lc, where lc is the
by the caisson while sitting on the surface distance between adjacent transverse di-
of the foundation (Hbl), where H is the aphragm walls placed inside the caisson
depth of water at the site during caisson (Fig. 5-78b). Normally, the bottom portion
installation and l is the length of the cais- of the wall is designed as a plate fixed at
son, minus the weight of the caisson (G). three edges and free at the top edge; the
Thus, d can be computed from the follow-
1
upper slab is usually divided into strips 1.0
ing expression: m wide, each of which is treated as a con-
tinuous slab supported on internal di-
d' > (Hblyw- G)fAc'Yw (5-69) aphragms and fixed at the side walls. In
this method, the base slab is usually consid-
where 'Yw is the density of water. ered to be a series of individual slabs, each
The hydrostatic pressure diagram re- of which is fixed at all four edges.
lated to the caisson installation stage is In some instances, the maximum contact
shown in Figure 5-78b. pressure from the foundation [load case (c),
As stated earlier, when a caisson is slip- Fig. 5-59] may be the most critical for the
way launched, it is exposed to an additional base slab design. In practical design, the
hydrodynamic load which is usually treated potential surface unevenness of the stone
as quasi-static. The value of this load (in bedding (Fig. 5-79) should be accounted for
kNjm 2 ) can be estimated as cV 2 , where c by computing the local stresses in the base
is the empirical coefficient conservatively slab. This means that the maximum shear
estimated to be equal to 0.85-1.0 kN s 2 jm 4 and flexural tensile stresses can be concen-
and V is the speed of caisson movement (in trated at locations projected up from the
mjs). In a preliminary design, V = 5.0 mjs average mattress grade (x 1 and x 2 in Fig.
is usually assumed. This will produce about 5-79) and at voids located underneath the
20 kPa of additional pressure on the ex- base slab.
posed caisson wall. Where applicable, this Obviously, the load distribution scenario
load should be added to normal hydrostatic depicted in Figure 5-79 differs substan-
pressure to design the caisson's leading wall.
In addition to hydrostatic loads, caisson
walls should be designed for the load pro-
duced by fill material, and where applicable
for wave, ice, and ship impact loads.
Typically, the stresses in caisson struc-
ture related to launch, tow, and installation
due to the temporary nature of loads exist-
ing at these stages can be treated as those
related to extreme loading and therefore Figure 5-79. Effect of poorly leveled mattress on the
may be increased by one-third of basic al- local stress concentration.
Gravity-Type Quay Walls 499

tially from that where a full contact be- Gurevich (1969). The description of this wall
tween the base slab and foundation is de- is given in Section 5.2.2.) According to
veloped. In a practical design, the latter is Gurevich, the initial effective contact area
considered to be the normal load condition between the unloaded (just filled with bal-
and the former is assumed to be an unusual last water) caisson and the rubble mattress
situation and, therefore, is treated as an was about 30% of the full base area. How-
extreme load case. ever, the contact area was increased to
An excessively high local contact pres- about 80% of a full base area after the
sure may rupture the base slab, causing caisson was filled with granular fill. The
corrosion problems. Clausen and Lunue increase of the "effective" contact area was
(1979) indicate that in the case of offshore most likely influenced by the approximate
gravity platforms, reasonable estimates of 8% consolidation of the mattress material.
the contact pressures can be predicted with Consolidation was practically completed
classical bearing capacity theories and when the maximum contact stress at the
elastic-plastic analysis using reasonable caisson base reached about 320 kPa.
stress-strain-strength properties of the Furthermore, during the initial stages of
foundation material and seafloor topogra- wall backfilling the structure relative dis-
phy. Kvalstad and Dahlberg (1980) sug- placement (rotation about the toe and away
gested a detailed semiempirical procedure from the fill) was less than 1/1000. At this
to compute local contact pressures based on stage, the coefficient of the soil lateral pres-
combined elastic-plastic analysis. In the sure was close to K 0 = 0.5 (pressure "at
proposed procedure, the local contact pres- rest" condition). At later stages of wall
sure is computed as the sum of the incre- backfilling and loading, the relative move-
mental stresses based on an assumed incre- ment has reached the value (1j4700)H. At
mental stress distribution. this point, the coefficient of soil lateral
To avoid heavy loading on the internal thrust was close to KA = tan 2 (45°- ¢/2)
diaphragms cells they should be loaded uni- and the total measured area of soil pressure
formly on both sides during ballasting or diagram was equal to the value obtained by
backfilling operations. Essentially, the Coulomb's theory. The soil pressure inside
whole process of caisson design can be com- the caisson compartments was very close to
puterized. One currently available software that obtained by Janssen's theory (for de-
on caisson design was developed by Acres tails see Section 5.6).
International Ltd. (1987). This program
treats the concrete caisson as a three- 5.5.6.2 Skirt Design
dimensional body exposed to all potential
loads which may exist during construction, As previously stated, the wall sliding sta-
launch, towing, and operation, and it in- bility as well as contact between the caisson
cludes hydrostatic and soil pressures, ice base and the foundation soil can be en-
impact, effects of. earthquakes, and other hanced greatly by the use of a steel or
loads. concrete skirt. To serve this purpose the
skirt should penetrate to the design depth.
5.5.6.1 Caisson Field Therefore, prediction of the forces necessary
Investigation to allow the skirt to penetrate into the foun-
dation soil is an important and critical
The interesting, however limited results analysis in the design of the caisson
of full-scale investigation of the quay wall equipped with a skirt.
at Port-of-Perm, Russia composed of In general, the total skirt penetration
floated-in concrete caisson are discussed by resistance is defined as the sum of the skin
500 Gravity-Type Quay Walls

friction and point resistance, similar to that 5.6 DESIGN OF QUAY


present during pile-driving. The preferred WALLS COMPOSED OF
method of determination of skirt penetra-
LAROE-DIAMETER CYLINDERS
tion resistance is based on the results of
standard cone penetration tests (CPTs).
Based on the CPTs, the total skirt penetra- As stated earlier, large-diameter cylinders
tion resistance, R., may be computed as according to their interaction with the foun-
follows (Foss and Dahlberg, 1979): dation can be classified as those submerged
into the foundation soil or placed on pre-
pared rubble mattress.
Normally, when the ratio of cylinder di-
ameter to its height does not exceed 0.7-0.9
and it is submerged into the foundation
where soil to 0.2H or less (H is the height of
the cylinder), the cylinder is considered a
KP and gravity-type structure. In other cases,
Kr = empirical coefficients; KP values vary it may be considered a bottom-fixed or a
from 0.4 to 0.6 for clay formations and
semigravity-type structure.
from 0.3 to 0.6 for sand formations; K f
The depth to which cylinders are usually
values range from 0.03 to 0.05 for very
stiff silty clays and from 0.001 to 0.003 submerged is determined by the provision
for dense to very dense silty fine sands for sufficient protection against scour;
(Clausen and Lunue, 1979) otherwise, the wall should be protected by
other means.
d 8 = skirt tip penetration
When large-diameter cylinders are placed
q c = cone point resistance
on a rubble mattress, gravel or other filter
At and material should be installed inside of the
A 5 =tip and side areas, respectively, on both cylinders to prevent fines in the granular
sides of the skirt per unit penetration
fill from penetrating out of the cylinder and
x = depth below the soil surface. down into the rubble and eventually being
washed off through the rubble.
The best result of skirt penetration forces The volume of concrete per linear meter
is obtained from a scale-model test. It must of wall, V, composed from large-diameter
be pointed out that an accurate prediction cylinders with relatively thin walls does not
of skirt penetration resistance is very im- actually depend on the cylinder diameter.
portant; an underestimate of the resistance The latter may be concluded from the fol-
will result in a smaller penetration of the lowing: V =:: 'TTDtHjD =:: 'TTtH, where D is
skirt into the foundation soil and, therefore, the cylinder diameter, t is the thickness
in less caisson resistance to sliding. On the of the wall, and H is the height of the
contrary, an overestimate of the skirt pene- cylinder.
tration resistance may cause excessive con- Practical application of quay walls con-
tact pressure on the caisson base. structed from large-diameter concrete cylin-
For more information on skirt design, the ders indicates that they may be economi-
reader is referred to Federation Interna- cally feasible for locations where the water
tionale de la Precontrainte (FIP) (1979), Det depth is less than 20 m. The optimum foun-
Norske Veritas (DNV) (1977), American dation conditions for this type of construc-
Petroleum Institute (API RP 2A) (1984), tion are bedrock or foundations that include
Buslov (1986) and Young (1992). hard materials difficult for pile-driving. It
Gravity-Type Quay Walls 501

may be ideal where a relatively thin stra- pressure where applicable; as usual, the
tum (3-4 m) of weak soils is underlain by resisting forces include the weight of the
bedrock or other competent materials such cylinder, the fill inside of the cylinder and
as hard clay or dense sand. In this case, a above it, and the concrete capping struc-
weak soil should be removed from the cylin- . ture.
der interior after its installation upon com- Live load scheme (a) (Fig. 5-59) is used
petent foundation material. for the evaluation of the structure sliding
The quay wall constructed from large- and overturning stabilities. In the latter
diameter cylinders (concrete or steel) is de- case, the driving forces which tend to over-
signed to resist the same load combinations turn the cylinder about point 0 (Fig. 5-80)
which are used for the design of walls com- are the same as those used for the evalua-
posed from heavy blocks or floated-in cais- tion of sliding stability. However, unlike the
sons. The basic design scheme of quay wall case of the evaluation of the wall sliding
composed from large-diameter cylinders is stability, the resisting forces should be re-
illustrated in Figure 5-80. duced by the weight of the fill, V5 , included
As is the case for other gravity-type re- inside the cone at the cylinder base as indi-
taining quay walls, the driving loads acting cated in Figure 5-80; V5 ::::: 1.05Rhs, where
on the wall composed from large-diameter R i is the cylinder intemal radius and 'Ys is
cylinders include the lateral soil thrust, the unit weight of the submerged fill. The
mooring forces, and unbalanced hydrostatic presence of a cone inside the cylinder · is

Figure 5-80. Large-diameter cylinder; design scheme.


502 Gravity-Type Quay Walls

explained by the ability of the soil to form following expression (Smirnov et al., 1979):
perfect arches inside silos and bins. The
analysis of the cylinder sliding and over- (5-71)
turning stability may be conducted either
as per linear meter of the wall, or the sta- where
bility of a whole unit may be evaluated. The havr =average depth of compressed fill material
latter, however, is less confusing and thus inside the cylinder
recommended. f3 = coefficient which takes into account limits
Udovichenko and Yakovlev (1976) stated for soil lateral expansion, f3 = 1 - 2 JL 2I
that the weight of the backfill material in- (1 - JL), where JL is the Poisson ratio
cluded between adjacent cylinders may be <Tavr = average value of bearing pressure at
added to forces contributing to the wall sta- cylinder base
bility against sliding, and the soil lateral E 0 = soil modulus of deformation
thrust against the wall could be determined
by conventional methods as acting on the The average depth of compressed material
wall at the rear plane AB (Fig. 5-80). The is computed from the equilibrium condition
bearing stresses at the cylinder's base are between bearing stresses at the base and
calculated as discussed in the previous sec- frictional forces at the interface between
tions. However, in the case of walls com- the cylinder interior area and the backfill
posed from large-diameter cylinders, the material (Fig. 5-81). In the preliminary de-
minimum bearing stress should never be sign, havr may be computed from the follow-
negative. Similar to blockwork walls or ing formulation recommended by Smirnov

r
caisson-type structures, a linear stress dis- et al. (1979):
tribution at the cylinder base is usually
considered acceptable. This, however, does
not account for the fact that actual bearing havr = [( 1 +3 m:avr
5
- 1] 2~ (5-72)
stresses at the concrete cylinder itself is
four to six times more than average as de-
termined by the conventional method where
(Smirnov et al., 1979). m = 1.7K(tan ¢)/D;, where K is the coefficient
The heavy bearing pressure produced by of the internal soil pressure determined for
the cylinder shell creates a "cutting wedge"
effect under which the structure tends to
settle and tilt; the settlement and tilt pro-
cess will continue up to the point when the
compressed fill material inside the cylinder
pick up the unbalanced load. Eventually,
the overall average settlement of the struc-
ture, S, consists of two parts: One is at-
tributed to full weight of the structure, S w,
and the other is attributed to compression
of the fill material inside the cylinder in a
process of tilt andjor settlement, Sc. The
former can be determined by conventional
methods found in standard texts on soil
mechanics and foundation engineering, and Figure 5-81. Tilt oflarge-diameter cylinder:
the average value of soil compression inside 1-before tilt; 2-final position; a-consolidated fill;
the cylinder, Sc, may be computed from the 4-design bearing pressure diagrams.
Gravity-Type Quay Wails 503

the 8 = 4> condition, where 8 is the wall where


friction and 4> is the angle of soil internal I: V = sum of all effective vertical loads with the
friction exception of V5 (Fig. 5-80)
y = unit weight of soil M 0 =overturning moment about longitudinal
axis at the cylinder base
Assume that the compression (consolida- Ac = area of cylinder at the base Ac =
tion) of soil inside the cylinder is propor- 0.257T(D 2 - Dl)
tional to the bearing stresses at the cylin- W = cylinder base sectional modulus W =
der's base. Then the cylinder tilt angle, a, 1r(D 4 - Df)j32D
attributed to soil compression may be com-
puted as a= arctan(Sc(max)- Sc(min))jD. Structurally, large-diameter cylinders
The local and global stabilities of the made from steel are usually designed for
wall constructed from large-diameter cylin- hoop stress only. Unlike an external hori-
ders are evaluated as discussed in Section zontal soil pressure, the internal pressure
5.3. used for computing the hoop tensile stresses
The cylinder's structural design is typi- in the cylinder wall is not proportional to
cally carried out in the following sequence: the height where this pressure is consid-
ered. The latter is usually determined by
(a) All loads acting on the cylinder during Janssen's (1885) method. According to this
fabrication, transportation, installation, method, the soil lateral pressure inside a
and while in operation are used to com- silo or bin at depth y is determined by
pute the most critical moment, the shear ah = avKA, where av is the vertical stress
force, and the hoop tensile load at each at depth y and K A is the coefficient of
characteristic section.
active pressure KA = tg 2 (45°- 4>/2). The
(b) The wall thickness and amount of rein- vertical stress av at any depth y is com-
forcement are proportioned according to puted from
the design parameters as noted in (a).
(c) If the cylinder is to be placed on bedrock au = yem + q(l - m) (5-74)
or a rubble mattress made from hard
rocks, then the strength of concrete at the where
cylinder's base should be evaluated.
y = unit weight of soil inside the cylinder
e = AciKAuf, where Ac is the internal area of
In case (c), the maximum stress umax should the cylinder cross section; u is the inter-
not exceed the allowable compression stress nal perimeter of the cylinder; q is the
for concrete that is determined from the uniform distributed load on the top of the
following: cylinder; f is the friction coefficient be-
tween the cylinder wall and fill, f = tg 8,
where 8 varies from 0.5¢ to cf>. Note:
amax = ( L V)/Ac + Mo/W (5-73) Large-scale investigations carried out at

Table 5-5. m = f(y je) in Eq. (5-74)

yje m yje m yje m yje m yje m yje m

0.1 0.095 0.6 0.451 1.1 0.667 1.6 0.798 2.1 0.877 2.6 0.926
0.2 0.181 0.7 0;503 1.2 0.699 1.7 0.817 2.2 0.889 2.7 0.933
0.3 0.259 0.8 0.551 1.3 0.727 1.8 0.835 2.3 0.900 2.8 0.939
0.4 0.330 0.9 0.693 1.4 0.753 1.9 0.850 2.4 0.909 2.9 0.945
0.5 0.393 1.0 0.632 1.5 0.777 2.0 0.865 2.5 0.917 3.0 0.950
504 Gravity-Type Quay Walls

Port Le Havre, France indicated that de- ternal and external anchorages. As with
pending on fill compaction the angle of any type of gravity wall, the design of a
wall friction 8 is 18° to 28° (Gurevich, L-shaped wall includes the evaluation of its
1969) sliding, overturning, and local and global
m = f(yje) is obtained from Table 5-5 stabilities, that is, essentially a trial-and-
error process. The design usually proceeds
Another method of determining the soil with the selection of wall dimensions, fol-
pressures inside a silo was developed by lowed by structural analyses and revision
Reimbert and Reimbert (1976). Both meth- as required according to the following
ods are discussed in detail by Turitzin sequence:
(1963), Safarian (1969), and Briassoulis
(1991). The latter author provides the most
recent comprehensive bibliography on the Assumption on type of the wall
subject matter. Several codes (e.g., ACI and its structural dimensions.
313R-77 revised in 1983) utilize both classic
silo theories as unconditional alternatives
for the calculation of the theoretical static Determination of design
pressures. An in-depth discussion on pres- loads and forces.
sures exerted by materials stored in silos is
given in Blight (1986, 1990, 1991). Finally,
it should be noted that where required the Evaluation of sliding, overturning
ambient temperature effects on stress in and global stabilities, review of
concrete silo must be considered. contact stresses and their impact
As in the design of concrete caissons, the on foundation soil.
minimum wall thickness of the concrete
cylinder should not be less than 300 mm in
a saltwater environment. This can be re-
duced to 200 mm when cylinders are used Design of structural components.
in freshwater. The crack control criterion
for cylinder wall is similar to that applied
for concrete caissons; in general, cracks are In some instances, to reduce the width of
not acceptable within the tidal zone andjor the wall base designers use an underbase
the zone of ice action. Provided that the key or introduce a sloped base. The design
wall's reinforcement in the above zones is procedure on a wall with a key is given in
protected by epoxies or by other credible Concrete Reinforcing Steel Institute Hand-
means, cracks of up to 0.1 mm wide may be book (CRSI) (1984) which includes tables
acceptable. In all other locations, the width for the design of the kay. It should be noted
of cracks should not exceed 0.15 mm. that in most practical situations under-
water construction of a key may be neither
practical nor economical. On the other hand,
5.7 DESIGN OF L·SHAPED WALLS the sloped base wall may be constructed
underwater, where a reduction in wall width
is required. A discussion of the design of
5.7.1 Basic Requirements retaining walls with a sloped base is given
in Elman and Terry (1987) as well as in
As stated earlier, this type of quay wall most standard texts on a subject of design
construction includes simple cantilever and construction of soil-retaining struc-
walls, counterfort walls, and walls with in- tures.
Gravity-TypeQuayWalls 505

Tentative dimensions of L-shaped compo- slab. This reduces the weight of the block
nents used for construction of a quay wall without reducing its length. The base slab
are shown in Figure 5-82 as a function of dimensions should be sufficient to satisfy all
wall height, H. stability and contact stresses requirements
as discussed in Section 5.3.
5.7.1.1 cantilever Wall
Similar to previously discussed struc- 5.7.1.2 counterfort Wall
tures, the stem thickness in this type of
construction, in general, should not be less Similar to the cantilever wall system, the
than 300 mm when installed in a saltwater length of the individual prefabricated blocks
environment; it may be reduced to 200 mm in the counterfort wall is designed to meet
when used for construction in freshwater. the capacity of available onshore and off-
Naturally, the lower stem part must be shore lifting equipment. Accordingly, coun-
strong enough to satisfy the shear force terfort blocks are designed with one, two, or
requirements without the use of shear rein- more counterforts. To save on the cost of
forcing steel. However, where the weight of block forming, the face wall and the base
the individual prefabricated unit is critical slab are usually fabricated with uniform
for wall erection, the shear reinforcing steel thicknesses; normally, the length of the base
can be introduced in both the stem and base slab across the wall is proportioned simi-

(a) b)
II 0.2MIN A
tl 0.2MIN

Ill
0

(d )
E (c)
Ill
0
g

I
N

I 0.75 to o.asl-i 0.4 to 0.45H


0

Figure 5-82. L-shaped wall; basic structural arrangements. (a) Cantilever wall; (b) counterfort
wall; (c) wall composed of prefabricated elements with internal anchorage; (d) wall composed of
prefabricated elements with external anchorage.
506 Gravity-Type Quay Walls

larly to that considered in the cantilever ceased. This results in uniform tensioning
wall design. of the anchor bars.
The distance between adjacent counter-
forts is usually obtained by a trial process 5.7.1.4 wall constructed from
to find the most economical solution for the Prefabricated
wall. As a rule of thumb it may be consid- components with
ered as equal to one-third the height of the External Anchorage
wall, but, in general, it should not exceed All basic principles used for the design of
4.0-5.0 m. the previously discussed walls are applica-
As stated earlier in this chapter and dis- ble for this type of construction. The differ-
cussed in detail in Chapter 4, under certain ence, of course, is that base slab of this wall
conditions, backfill soil may create arches is somewhat shorter. This is attributed to
between adjacent counterforts. This may the fact that a substantial portion of the
have the dramatic impact of substantially horizontal thrust against this type of wall is
reducing the soil lateral pressures on the absorbed by the independent anchor sys-
front wall. tem. Furthermore, because the base slab of
the wall with an external anchorage usu-
5.7.1.3 Wall constructed from ally is not exposed to the vertical compo-
Prefabricated nent of the anchor load, the minimum
components with thickness of this slab can be smaller than in
Internal Anchorage the case of the previously discussed wall
with an internal anchorage. However, in no
The overall dimensions of this type of case should it be thinner than 300 mm
wall are similar to those previously dis- when used in a saltwater environment, and
cussed. Prefabricated components such as 200 mm when installed in freshwater.
face panels and foundation slabs, depend-
ing on wall height, may be cast as a flat 5.7.1.5 crack Control and
slab of uniform thickness or where dictated concrete cover over
by the economy, more complex shapes such Reinforcement
as slabs with vertical andjor horizontal These are essentially the same as ap-
stiffening ribs can be utilized. plied in the design of any type of thin-walled
In no case, however, should the thickness marine structure as discussed in Sections
of the face panel be less than 300 mm in a 5.5 and 5.6.
saltwater environment and 200 mm in a
freshwater environment. The width of the
base slab may be as great as practical; how- 5.7.2 Design of cantilever
ever, the width of the vertical panel should Walls
be such to provide for installation of no
more than two anchors. The latter is usu- The stability of cantilever wall against slid-
ally attributed to the fact that in the struc- ing and its global stability are evaluated for
tural system in question, it is very difficult effects of live loads from case (a) (Fig. 5-59)
to balance tensile loads in the anchor sys- in combination with the mooring force and
tem where the number of anchors exceeds the unbalanced hydrostatic pressure; the
two. maximum bearing pressure at the wall base
A monolithic concrete capping beam is and on the foundation soil are evaluated for
usually installed after the wall is backfilled the loading similar to the load case (c) (Fig.
and all basic wall displacements have 5-59) in combination with the mooring force
Gravity-Type Quay Walls 507

and the unbalanced hydrostatic pressures. where


The load case (c) in combination with the <Tmax and
mooring force and unbalanced hydrostatic <Tmin = maximum and m1rumum values, re-
pressures will also yield maximum shear spectively, of contact pressures at the
force and bending moments in wall ele- wall base
ments. E = design lateral load
Forces acting on wall stem and base slab b = length of the wall base across the wall.
are shown in Figure 5-83. Here, the stem is
designed as a cantilever beam loaded by the
associated lateral and vertical forces. Its
upper portion should also be evaluated for 5.7.3 Design of counterfort
the concentrated load due to ship impact. Wall
Forces included in shaded areas in Figure
5-83 are used for computing shear forces The method and sequence of the wall stabil-
and bending moments at any critical sec- ity and the contact stress analysis are simi-
tion y-y at the base slab. lar to that discussed in the preceding sec-
For the above analysis, the shear force tion. It differs, however, in the computation
along the wall base, {, is usually propor- of shear forces and moments at wall compo-
tioned to the design values of contact pres- nents. As stated earlier, the horizontal soil
sures at the interface of the wall base slab thrust on a face slab can be reduced due to
and the stone bedding. In the latter case, soil arching effects, or by the soil friction
the maximum and minimum values of shear force on the faces of adjacent counterforts.
force at the wall base are computed from Quantitatively, for preliminary calcula-
the following expressions: tions this reduction in soil pressures may
be assumed as equal to 20% to 30% of
2( <Tmax/ <Tmin)E the total soil thrust on a face slab when
(5-75)
fmax = b(1 + <Tmax/<Tmin) counterforts are placed at 4 to 3 m center
to center respectively (Udovichenko and
{, . =
2E ( 1 -
-
<Tmax/ <Tm in )
(5- 76)
Yakovlev, 1976). Soil pressure reduction
mm b 1 + <Tmax/ <Tmin may be represented by the triangular dia-
gram. A simplified approach, although
somewhat conservative but less time-

--
~..,
consuming, is commonly used for comput-
ing shear forces and bending moments in
the face slab. Where two or more counter-
forts per wall unit is used, the upper por-
tion of the face wall is treated as a continu-
ous slab supported on counterforts with
cantilevered parts at both ends.
The bottom part of the face slab between
counterforts at the junction with the base
slab is analyzed as a slab fixed at three
edges and free at the upper one. In a simi-
rT"r-r.,..,.~rr-fv-rT,.,., c:;,..,"' 0 .......... ,. lar manner, the bottom part of the face slab
G,....._,.. between counterforts and the end of the
block may be analyzed as a slab fixed at the
counterfort and the base slab and free at
Figure 5-83. Cantilever wall; design diagram. the two other edges.
508 Gravity-Type Quay Walls

The base is usually analyzed in a man- 5.7.4 Design of Wall


ner similar to that used for the face slab. constructed from
Furthermore, the base slab should be evalu-
ated for the condition similar to that de-
Prefabricated
picted in Figure 5-79. components with
The counterfort is usually treated as a Internal Anchorage
cantilever beam fixed at the base slab and
loaded by the reaction loads transmitted The basic methods and sequence of stability
from the face slab. A proper amount of ten- and contact stresses analysis used for this
sile steel should be provided at the junction type of wall design are similar to those used
of counterfort, at the face and base slabs, in for the cantilever and counterfort walls. In
order to keep these elements firmly to- the case in question, the face panel is ana-
gether. The wall block with only one coun- lyzed as a single span beam with cantilever,
terfort is analyzed by similar methods used loaded with an active soil pressure, mooring
for the end portion of the block with two or force, and unbalanced hydrostatic pressures
more counterforts. where applicable.
Bending moments and shear forces in Two live load conditions are normally
structural members of the counterfort wall considered for the face panel analysis (Fig.
can be determined with the help of a series 5-84a); to obtain the maximum bending
of tables, produced by the U.S. Bureau of moment in the face panel, the live load is
Reclamation; these calculations can be eas- placed at a distance b = tan(45° - cp j2)
ily computerized. A solution for the treat- from the quay edge. To compute the maxi-
ment of counterfort wall problems is also mum anchor load and maximum bending
given in Huntington (1957). moment and shear force at the cantilever,

(b)

Figure 5-84. Wall composed from prefabricated elements with internal anchorage; Design
diagram. (a) Full contact between base slab and foundation; (b, c) cases of incomplete contact
between base slab and foundation.
Gravity-Type Quay Walls 509

the uniform distributed load should be ex- f is the shear force at interface between the
tended further to the edge of the wall and base slab and the mattress.
the mooring force should be added. In most A more accurate solution is obtained from
practical cases, the face panel is treated as calculations when the base slab is analyzed
a "flexible" wall anchored by a relatively as a beam on an elastic foundation. Exten-
stiff anchor system. The latter implies that sive treatment of beams and slabs on a
the actual bending moment in the panel continuous elastic support is given in
span will be smaller and loads in the an- Hetinyi (1946).
chor bar will be greater than those obtained Gurevich (1969) reported results of nu-
from the conventional analysis of a single merous full-scale and laboratory experi-
span beam with a cantilever. For more in- ments on the walls in question. The main
formation on this subject, the reader is re- findings from these experiments may be
ferred to Chapters 4 and 6. The tensile load summarized as follows:
in the anchor bar, R a, is obtained from
1. The wall placed on hard (bedrock, dense
R
R =~l (5-77) gravel, or dense sand) foundation tends to
a cosa a rotate about the toe and away from the
fill. The average wall settlement is basi-
where cally attributed to consolidation of the
Ra(h) =horizontal component of design anchor mattress material which, in some cases,
load can reach a value of up to 8% (maximum
la =space between adjacent anchors contact stress, amru<' equal to about 320
a = angle between anchor bar and horizontal kPa) of the total thickness of the rubble
(gravel) mattress.
Where "softer" foundations such as not
Anchors are typically made up from metal
very dense sands or relatively soft clays
strips hinged at special embedded parts in are present, the initial (during the back-
both the face panel and foundation slab. fill placements) wall displacement is char-
Alternatively, anchors may be hinged at the acterized by rotation about the base slab
base slab and field welded at the face panel. heel toward and away from the fill. How-
Vertical component of the anchor load, ever, under the effect of live loads, the
Ra(v)' must be considered in the face panel wall tends to rotate about its toe and
design. Similarly, action of the horizontal away from the fill.
component of anchor load on the base slab 2. The face panel-soil interaction exhibited
must be considered in the base slab design. typical phenomena characteristic of
Three principle design schemes for the base "flexible" retaining walls, as discussed in
slab are depicted in Figure 5-84. Chapters 4 and 6. The total soil lateral
The potential gap between the base slab thrust was about equal to that obtained
and the mattress is usually accounted for, by Coulomb theory; however, the soil
as illustrated in Figures 5-84b and 5-84c; pressure distribution along the panel,
in case (b), the length of the gap x can be similar to that observed in sheet-pile
bulkheads, was not linear; it was concen-
assumed to be equal to about 0.2B, where
trated at supports and reduced at the
B is the length of the base slab, and the midspan. The latter affected the value of
situation as shown in case (c) is typically the anchor force, which was about 20%
considered for evaluating the strength of greater than that obtained from Coulomb
cantilever portion of the base slab lc. In theory. The same is applied to the reac-
Figure 5-84, q is the load which represents tion force at the lower end of the face
the weight of the soil and the live load, and panel at the base slab.
510 Gravity-Type Quay Walls

The value of bending moments were both the face panel and anchor slab. Fur-
from 35% to 50% less than those obtained thermore, because the face panel is not fixed
from Coulomb theory. The reduction in at the base slab and therefore does not
bending moment was dependent on panel transmit any moment to this slab, the bear-
stiffness (EI), . where E is the Young's ing pressure on the rubble mattress is prac-
modulus and I is the modulus of inertia.
tically uniformly distributed along the base
3. It was very difficult to achieve full contact slab. The length of the base slab B = llB +
between the face of the rubble mattress B' (Fig. 5-85) is determined from the condi-
and the base slab installed underwater.
tion of wall stability against sliding at the
The contact pattern found was very simi-
lar to that discussed earlier in Section 5.5. base:
4. It has been found that friction between
the concrete face panel and granular fill (5-78)
material, o, was equal to about ¢,where
¢ is the angle of internal friction of the where
fill material.
5. Although the standard value for the fric- F.1 = factor of safety against sliding
tion coefficient at the interface between f = friction coefficient; in most practical de-
the base slab and the rubble mattress is sign cases f = 0.5 to 0.65 is used
usually taken to be equal to 0.5-0.65, in E V = sum of all effective vertical loads E V =
real field conditions under bearing pres- Y1 + V 2 + Y3 , where Y1 is the weight of
sures at base slab varying between 200 the face panel, V 2 is the weight of the
and 300 kPa it was equal to (1.05 to 1.1) base slab, and Y3 is the weight of the fill
tg ¢. material included in prism abed.
R~(h) =horizontal reaction force at the base slab

5.7.5 Design of Wall Stability against sliding at the interface be-


tween the rubble mattress and foundation
constructed from soil is evaluated as discussed in Section 5.3.
Prefabricated The wall anchor system is designed to
components with resist the maximum anchor load Ra(h) as
External Anchorage determined from the previously outlined
load conditions. The value of Ra(h) is deter-
The face panel in this type of construction mined by methods similar to that usually
is designed similar to that in walls with an used for the design of the flexible bulkheads
internal anchorage, as discussed in Section (consult Chapter 6). The anchor system de-
5.3.4, with the exception that the axial load sign process typically includes the determi-
produced by the anchor bar is usually very nation of space between adjacent anchors,
small and in most practical cases equal to the length and required diameter of the
zero. anchor rod, the height and length of the
The wall anchor system is designed for anchor block, and the evaluation of the an-
the load condition case (e) as indicated in chor block stability.
Figure 5-59 plus mooring force and unbal- The length of the tie-rod, LA, is usually
anced hydrostatic load where applicable. determined from the following expression
This load combination produces a maxi- (Fig. 5-85):
mum load on the anchor system.
Because the anchor tieback is usually LA= B' + Htan(45°- ¢/2)
placed horizontally or with a gentle slope, it
does not produce a vertical component at + t tan(45o + ¢/2) (5-79)
Gravity-Type Quay Walls 511

The height of the anchor block, ha, is usu- NAVFAC DM-7.2 (1982), and by other au-
ally recommended as equal to (0.3 to 0.5)t. thorities. In further discussions this method
will be referred to as a "conventional
method"; in the literature it is also referred
5.8 DESIGN OF CELLULAR· TYPE to as the "Terzaghi Method." Cummings
STEEL SHEET·PILE (1960) proposed the so-called "horizontal
BULKHEADS shear" method, which is also used in North
American marine structures design prac-
tice. A more complicated approach to the
design of cellular bulkheads proposed by
5.8.1 Introduction Hansen (1953) is basically used in Europe.
The design of cellular bulkheads is, to a
Cellular-type sheet-pile bulkhead by defi-
great extent, affected by the site's geotech-
nition is a column of granular soil enclosed
nical condition (e.g., construction on rock) or
by a steel shell, consisting of interlocking
foundation soil (e.g., sand or clay which per-
sheet piles. Essentially, the steel sheet-pile
cellular bulkheads are nonrigid gravity-type mit sheet-pile penetration). The type of
structures, the stability of which depends foundation material affects the cell interac-
on a balance of external driving forces and tion with both the foundation and backfill
internal forces which are the result of the soil and greatly influences the vertical and
sheet-piling cell-fill interaction. horizontal displacements of the cell. Unfor-
Currently, there are several essentially tunately, none of the design methods cur-
empirical methods available for analysis of rently in use include a criterion for vertical
the cellular bulkheads. Dismuke (1975) pro- and horizontal displacements and their ef-
vided a summary of several design methods fects on soil-structure interaction. It is most
used in North America. However, the most likely that in future analyses of cellular
popular in North America is the design structures the finite element technique will
method described by Pennoyer (1934) and become increasingly commonplace. An at-
later modified by Terzaghi (1944), Ten- tempt at such an analysis has already been
nessee Valley Authority (TVA) (1957), made by Guillermo and Lawrence (1987).

t
1-
a

t=
-

Figure 5-85. Wall composed of prefabricated elements with external


anchorage; design scheme.
512 Gravity-Type Quay Walls

The complexity of the soil-nonrigid cell in- The following is a discussion on conven-
teraction in future analytical design models tional design methods of cellular bulkheads
most likely will be taken care of by using placed on rock or driven into the foundation
field data from full-scale instrumented soil.
structures. In order to correctly calculate
the response of the nonrigid cell,. it will be 5.8.2.1 Design of Cellular
necessary to adjust its elastic properties Bulkheads Placed
until the response will match field observa- on Rock
tion data. Valuable, although limited, data
In current conventional design practice,
are already available. A list of the most
the bulkhead placed on a rock foundation is
recent and relevant references is provided
typically evaluated against the following
by Schroeder (1987).
failure modes: (1) sliding on the base, (2)
overturning, (3) loss of internal stability,
and (4) excessive interlock tension.
5.8.2 conventional Design Although standard design methods are
Method applied to the design of a cellular structure,
one must realize that a sheet-pile cell is a
The conventional design of steel sheet-pile nonrigid structure, and to prevent any of
cellular wall (bulkhead) typically proceeds the above-noted modes of failure, the mech-
in the following sequence: anism of the sheeting fill and the sheet-piie
cell-backfill interaction must be correctly
understood.
Assumption on cell arrangement In a practical design, the cellular struc-
ture is usually replaced by a fictitious rect-
angular bulkhead with equivalent width, B
(Fig. 5-86). This width is usually defined as
Calculation of external forces the width of an equivalent rectangular sec-
tion having a section modulus equal to that
of the actual bulkhead. For circular cells,
the relationship between B and D, where
Assumption of cell dimensions and
D is the cell diameter, may be used as
spacing between adjacent cells
follows (Fig. 5-86):
B = 0.785D for a= 30° (90° T)
Calculation of cell stability against B = 0.818D for a = 30° (30° Y)
sliding on the base and overturning B = 0.875D for a = 45° (90° T)
T and Y are jointly illustrated in Figure
Calculation of shearing stability at 5-41. Note that for diaphragm-type cells,
cell centerline and between sheeting an equivalent width, B, may be taken as
and fill the ratio of the area enclosed by the cells to
the distance between diaphragms. For de-
sign purposes it can be taken as equal to 0.9
times the total dimension of the cell from
Calculation of hoop tension front to back.

5.8.2.1.1 Sliding on the base


Design of structural components The bulkhead must provide adequate re-
sitance to sliding on the base caused by
Gravity-Type Quay Walls 513

(b)

__ __
~v
,
I
0

+0 ~~ -Q :

/ I
A
"'Y2
I
I
.._,-.,...,,..,.....,-:<-+..,..,.,..,,..,..( I/ ",t
r tJ
~0 --

(c) (d)

,--
1

- E; I
I
I
l
I
I

Figure 5-86. Stability of steel sheet-pile cell; conventional design method.

lateral soil thrust and, in some cases, by the and the unbalanced hydrostatic pressure
unbalanced hydrostatic pressure. Because where applicable; F.1 = 1.25 to 1.50.
of low wall permeability, the latter is usu-
ally used as equal to two-thirds of the tidal Fsl = o::: V)f!E (5-80)
range above the low water level (LWL). where
The factor of safety against sliding on the
I: V = effective weight that includes the weight
base, F.1, is defined as the ratio of the wall of the superstructure and the weight of
shearing resistance at the base to the driv- the fill above the water table line plus the
ing force. The driving force basically in- submerged weight of the fill below the
cludes the soil thrust, the mooring force, water table; usually the seepage force (if
514 Gravity-Type Quay Walls

any) and the weight of sheet piles are due to slipping between sheeting and the
neglected cell fill; in such a case, lifting the outboard
f = friction coefficient; for granular fill placed sheeting will cause the cell to lose the fill,
on rock, the conservative value f = 0.5 is which will run out the heel of the cell. The
often used; the upper theoretical value of safety factor against such a mode of failure,
f is equal to tan ¢, where ¢ is the angle F8 , represents the ratio of the resisting mo-
of internal friction of the granular fill.
ment due to the fill friction on outboard
E = lateral load which includes soil lateral sheeting to the driving moment discussed
pressures, unbalanced hydrostatic pres-
earlier:
sure, where applicable, and mooring force
BE tan 8 B tan 8
Passive pressure from the usually shallow F.=--- (5-82)
EhE
depth of soil in front of the cell founded on a
rock is commonly neglected. where tan () is the coefficient of friction be-
tween steel sheet piles and cell fill. The
5.8.2.1.2 overturning stability values of tan () are given in Table 5-6.
The cellular bulkhead must be stable From Eq. (5-82) follows that from the
against overturning. To prevent this failure slipping condition between sheeting and the
mode from occurring, the bulkhead must fill the required cell width can be computed
have sufficient width to ensure that the as
bearing stress of the heel, computed assum- (5-83)
ing that the structure is "rigid," is not nega-
tive. Similar to the method described in The value of F8 = 1.25 is commonly used.
Section 5.6, the factor of safety against Lacroix et al. (1970) recommended that
overturning about the bulkhead toe (point 0 shear force at crosswalls and at interludes
in Figure 5-86a). F 0 is the ratio of the should be included in the above calcula-
resisting moment Mr = 2: V'(B /2) to the tions. In this case, the factor of safety F8 =
overturning moment M 0 = Eh E; in the for- 1.5 is recommended.
mer, l:V' = 2: V- 0.5B 2y8 , where the sec-
ond member of equation represents the 5.8.2.1.4 Internal stability at
weight of submerged fill included in the cell centerline
triangular prism with a height of B j2 (Fig. Soil shear resistance along the plane
5-86a). F 0 is commonly assumed to be equal through the cell centerline (internal stabil-
to 1.5. ity of the cell) is another possible mode of
To avoid overturning and reasoning that failure. For cell internal stability, the
negative pressure at the base is avoided, shearing resistance along its centerline
the resultant weight l:V' should lie within plane, that is, the sum of the fill shear
the middle one-third of the base and there- · resistance and resistance in the interlocks,
fore l:V' eccentricity is computed from the must be equal to or greater than the shear
following: due to the overturning effects. With refer-
B EhE B ence to Figure 5-86a the total shearing
e=----<- (5-81) force on the centerline plane of the cell is
2 I: V' - 6
computed as

5.8.2.1.3 Slipping between Q = 0.5(!) ( 6; 0 ) = l.~Mo (5-84)


sheeting and cell fill
When a cellular bulkhead is subject to a where Q and M 0 are respectively the total
large overturning load, failure can occur shearing force and net overturning moment
Gravity-Type Quay Walls 515

per unit length of cellular bulkhead. The depends on the kind of fill material and
force Q is resisted by vertical shear within methods of its placement and densification.
the cell fill and friction in the interlocks. Attempts to obtain precise mathematical
The vertical resisting shear within the cell solution for computing soil pressure inside
fill along the centerline of the bulkhead, R s the cell may sometimes be misleading;
(Fig. 5-86b) is obtained from therefore, the conservative approach to
computing the value of K that is repre-
(5-85) sented by Krynine's formula is recom-
mended.
where The friction resistance in the sheet-pile
Ec = soil horizontal pressure interlocks per unit length of bulkhead is
4> = angle of internal friction equal to the total interlock tension times
the coefficient of friction {;; for practical
The coefficient of soil lateral pressure, K, purposes, a steel-on-steel friction coefficient
which is used for computation of the soil may be assumed to be equal to {; = 0.3. The
lateral pressures at the cell centerline, has interlock force T is computed from the con-
1

a magnitude that is somewhat higher than ventional hoop tension formulation T = 1

KA" Terzaghi (1944) stated that K could not EiR, where Ei is soil pressure against the
be predicted on a rational basis and sug- sheet piling as outlined in diagram abed in
gested that it must be equal to a value Figures 5-86c and 5-86d, and R is the cell
between 0.4 and 0.5. Krynine (1944) pro- radius. For the purpose of evaluating the
posed the following expression obtained cell's frictional resistance at the interlocks,
from Mohr's circle construction for comput- the soil pressure E i is usually assumed to
ing the magnitude of K: be equal to active soil pressure EA" As indi-
cated in Figures 5-86c and 5-86d, the soil
(5-86) pressure is reduced to zero at point d. This
is due to the following reasons: When the
This formula is recommended in the USS cell is placed on the rock (Fig. 5-86c), the
Steel Sheet Piling Design Manual, Bowles lower end of the piling bites into the rock,
(1988), and Navdocks (1962); NAVFAC thus reducing the hoop tension practically
(1982) with reference to Lacroix et al. (1970) to zero; when the cell is driven into sand or
recommends K = 0.5 to 0.6, which actually a clay foundation strata (Fig. 5-86d), the
corresponds to 4> = 35° to 30° in Krynine's hoop tension is reduced to zero at the sheet
formula. pile fixity plane x-x. In the case of a sand
Experience indicates that the soil pres- foundation, the fixity plane may be as-
sure inside the cell is quite erratic and sumed as located at a depth x = 0.1H from
varies from project to project; it basically the dredge line, where H is the height

Table 5-6. Friction coefficient tan 8 for various soils against steel sheet piles

8 (degree) tan 8

Clean gravel, gravel-sand mixtures, well-graded


rock fill with spalls 22 0.40
Clean sand, silty sand-gravel mixture,
single-size hard rock fill 17 0.30
Silty sand, gravel or sand mixed with silt or clay 14 0.25
Fine sandy silt, nonplastic silt 11 0.20

Source: From Navdocks (1962).


516 Gravity-Type Quay Walls

of the bulkhead; if sheet piles are driven Calculations of the maximum soil pres-
into the relatively hard clay, the fixity point sure that is used for the evaluation of the
is usually assumed at the dredge line level. critical tension at interlocks requires the
The peak of soil pressure inside the cell use of a coefficient of soil lateral pressure,
is assumed to act at a distance (i to t)hc K;, which is not necessarily the same as is
from the sheet-pile fixity point as indicated used for the calculation of the interlock fric-
in Figures 5-86c and 5-86d. This usually tion force as discussed earlier. From field
corresponds to the maximum deflection, dx, observations, it became apparent that K;
of the sheet piles. has greater value than K A (Shannon and
Hence, the sheet pile's resistance at the Wilson, Inc., 1982; Sorota and Kinner,
interlocks per unit length of cellular bulk- 1981).
head, T, is computed from For an isolated cell, Terzaghi recom-
mended K; = 0.4 as a reasonable choice,
. T= 2T'fJL (5-87) and Schroeder (1990) stated that K; = K 0
may be considered. During the actual cell
where L is the center-to-center distance be- filling, values of K; as great as 0.66 have
tween adjacent cells. Thus, the total shear- been calculated from interlock force obser-
ing resistance along the centerline of the vations (White et al., 1961). Field data sug-
cell, S, per unit length of the bulkhead is gest that K; = 0.4 to 0.5 may be considered
computed as follows: as a reasonable value for amax calculations.
The bursting force per unit of cell height
S = (Rs + T) = Ec tan cp + 2T'fJL (5-88) Tmax with the amax located as indicated in
Figure 5-86c and 5-86d is calculated from
Accordingly, the safety factor against the Tmax = amaxR, where R is the cell radius.
fill shearing mode of failure Fsr is obtained The value of Tmax must not exceed the al-
from lowable value for interlock tension provided
by the sheet-pile manufacturer.
S (Ec tan cp + 2T'fJL)B Most of cellular cofferdam failures re-
FST = Q = M (5-89) ported to date resulted from failure of the
1.5 0
connecting tee from either a fabrication fail-
ure or interlock failure (Swatek, 1967;
The minimum value of Fsr = 1.5 is recom-
Grayman, 1970; Tschebotarioff, 1973).
mended for normal load combinations, and
The Tennessee Valley Authority (TVA)
Fsr = 1.25 may be considered for extreme
(1957) suggested that the maximum inter-
load combinations or for the design of tem-
lock tension can be computed from the free
porary structures.
body of the cell as follows:
5.8.2.1.5 Bursting stability (5-90)
Tmax = amax(0.5L)jcos a
The cell must withstand maximum inter-
nal pressure, a max' without bursting at the where a is the angle as indicated in Figure
interlock joints. 5-40.
Maximum earth pressure at the inter- Sorota et al. (1981) compared the inter-
locks, a max' is typically located as discussed lock tension calculated by various design
earlier in this section and indicated in procedures to actual interlock tensions
Figures 5-86c and 5-86d. If the cell is filled measured in the field. They concluded that
hydraulically, then an unbalanced hydro- the TVA formula (5-90) produces conserva-
static pressure must be added to the soil tive results. Lacroix et al. (1970) stated that
pressure. Eq. (5-90) contributes little accuracy in view
Gravity-Type Quay Walls 517

of the mechanism controlling the interlock In addition, several other requirements


tension. The latter authors pointed out that must be satisfied to ensure cell stability
the tensile forces in the sheet piles of the [e.g., stability with respect to bearing capac-
adjacent arcs and the bending stresses in ity failure of the underlying soil strata and
the connection sheet pile produced by this global stability (Oresen, 1962)]. Bulkhead
force must be given proper consideration in sliding at the base, in general, is not a
the design of the connection sheet pile. practical problem and the designer's atten-
Lacroix et al. cited an equation recom- tion should be concentrated on the investi-
mended by Swatek (1970) for the common- gation of the bulkhead's internal stability.
wall interlock force The latter is usually done by methods simi-
lar to these discussed earlier for the case of
(5-91) cellular bulkheads on rock. However, de-
pending on sheet-pile penetration depth and
properties of the underlying soil, the sheet-
Rossow (1984) presented an analytical
pile restraint as indicated in Figure 5-86c
method for the calculation of interlock ten-
may or may not be achieved. In most cases,
sion in a circular sheet-pile cell. His ap-
engineers use full soil active pressure to
proach is based on two critical assumptions:
compute interlock tension.
(1) The sheet pile wall has no resistance to
In general, penetration of sheet piles on
bending about its vertical axis and (2) the
the cell unloaded (harbor) side must be
sole action of the fill against each wall is a
sufficient to prevent further penetration.
uniform pressure of known magnitude.
The minimum value of the factor of safety
Rossow's method is basically a generalized
against sheet-pile penetration, defined as
version of the Swatek (1970) method which
the ratio of shear resistance on both sides
takes into account information on soil pres-
of the embedded portion of the sheet piles
sures in both the arc and the main cell. In
on the unloaded side to the internal down-
Rossow's opinion, if such information is not
ward shear force on the same side, should
available, then for the common-wall situa-
be equal to 1.5. The maximum internal
tion, Swatek's formula (5-91) is sufficiently
downward shear force is equal to EA tan o.
accurate, although quite conservative. Stan-
The shear resistance on both sides of the
dard strength interlocks are most com-
embedded portion of sheet piles depends on
monly used in cellular bulkhead construc-
the coefficient of lateral soil pressure K. For
tion. However, sheet piles with high and
a dense sand foundation, the preliminary
lower strength interlocks are also available
pile penetration depth on the cell unloaded
from sheet-pile manufacturers in the United
side can be assumed as equal to one-third of
States and overseas.
the total height of the bulkhead.
On the loaded (backfill) side, penetration
5.8.2.2 Cellular Bulkheads Driven
of sheet piles could be much shorter. How-
into Foundation Soil
ever, sheet-pile penetration into the founda-
The design procedure for large-diameter tion strata on the backfill side must be
cellular bulkheads founded on foundations adequate to ensure bulkhead stability with
that permit sheet-pile-driving to a substan- respect to pull-out of the outboard sheeting
tial depth is similar in principle to that due to cell rotation (tilting).
used for the design of cellular bulkheads The harbor side of a bulkhead must be
placed on rock. Almost all requirements dis- protected from scour; otherwise, sheet piles
cussed in the previous section 5.8.2.1 are on this side must be driven into the founda-
directly applicable to the design of cellular tion to at least 1.0-1.5 m below the ex-
bulkheads founded on soil deposits. pected level of erosion.
518 Gravity-Type Quay Walls

AB with any gravity wall, cellular bulk- 5.8.3 Horizontal Shear


heads impose a substantial bearing load on (Cummings•> Method
the supporting soil stratum; this load is
particularly heavy at the wall harbor face. A horizontal shear analysis of cellular
The allowable bearing stress on the founda- sheet-pile structures was proposed by
tion soil is calculated as discussed previ- Cummings (1960). This method is based on
ously (Section 5.2), and in the case of cellu- the premise that the fill will resist lateral
lar bulkheads, the safety factor is usually distortion of the cell through the buildup of
taken to be equal to 2.0, where medium and soil resistance by sliding on horizontal
coarse granular soils are present, and planes, as illustrated in Figure 5-87.
2.5-3.0 where fine-grained material consti- The proposed design model is based on
tute the foundation soils. observations of scale-model tests conducted
The foundation soil condition (e.g., the with cells placed on rock. Here, Cummings
presence of soft clay) could limit the wall concluded that the shear resistance within
height. With respect to the ultimate bearing the fill is developed only in a triangle, form-
capacity of this kind of foundation, the USS ing an angle f/J to the horizontal. Accord-
Steel Sheet Piling Design Manual recom- ingly, he proposed a design model which is
mends the following empirical formulation based on the premise that the cell fill within
for computing the maximum possible height the aforementioned triangle resists the lat-
of the wall, Hmax, for construction on clay eral distortion of the cell through the
foundation (safety factor of 1.5 is included): buildup of soil resistance by sliding on hori-
zontal planes. The soil within the triangle,
Hmax = 3.8C/'Yc (5-92) which produces a resisting pressure on the
outboard sheeting, is considered as being in
where a passive pressure state and the weight of
c =cohesion the overlaying soil is assumed to act as a
'Yc = effective unit weight of cell fill surcharge load. The following equations de-
rived by Cummings summarize his method
Because of the nonrigid nature of the cellu- of computing the resisting moment due to
lar structure, stability is practically inde- this pressure. The variables in the following
pendent of the strength of the cell fill; as equations are as shown in Figure 5-87:
the shearing resistance through a vertical
section offered by the cell fill cannot be R = W tan cp = yBH tan cp = yac + c 2y (5-93)
mobilized without overstressing the inter-
locks, cellular bulkheads are almost never where R is the ultimate shear resistance of
built on compressible materials (e.g., soft to cell at the interface with rock. This equa-
medium clay) where substantial structural tion is illustrated graphically by the pres-
distortion can be expected. sure diagram shown in Figure 5-87 d. The

l•l (b) (d)

Figure 5-87. Stability of steel sheet-pile cell; Cummings' method. [From Cummings (1960).]
Gravity-Type Quay Walls 519

resisting moment is calculated as M, = Rh,. 5.8.5 Seismic Design of Cellular


An interlock friction, T', also provides some Bulkheads
shear resistance. The value of T' is com-
puted as previously discussed. The friction To date, little discussion has been offered in
force at the interlocks due to cell tilting is the literature regarding the seismic design
assumed to act equally at all individual of cellular structures. Among published re-
sheet piles. ports are a manual compiled by the Japan
Accordingly, the resisting moment due to Society of Civil Engineers (1977) and pa-
interlock friction, Mi, is computed from pers by Chakrabarti et al. (1977) and
Schroeder (1990). Chakrabarti et al. con-
(5-94) cluded that an approach to cell seismic de-
sign may lay in the extension of static de-
sign methods that incorporate a consistent
where L is the center-to-center distance be- pseudo-static mechanism. For this the ex-
tween adjacent cells. The safety factor ternal loads such as active and passive soil
against tilting is calculated as follows: pressures and hydrodynamic pressure can
be computed as discussed in Chapters 4 and
(5-95)
6.
Under seismic conditions, the sheet-pile
cell placed either on rock or driven into the
In Cummings' method, the stability against foundation strata must be evaluated for the
sliding is computed as discussed in Section same failure modes as discussed in previous
5.8.2.1. sections. The soil liquefaction potential must
be considered while evaluating sliding on
the base, overturning, interlock tension,
and the loss of intemal stability modes of
5.8.4 Brinch Hansen Method failures.
Seismically induced liquefaction of the
Hansen (1953) proposed the extreme fill or foundation andjor backfill materials
method of cell stability analysis based on a could ultimately lead to either local or global
simple concave or convex surface of rupture failure of the structure. Therefore, it is pru-
approximated by a logarithmic spiral. This dent to remove loose foundation soils that
is based on observations of model tests of are potentially liquefiable and replace them
cells installed on rock and embedded in with well-densified, free-draining materials.
sand. Furthermore, to prevent prohibitively ex-
In Hansen's method, the factor of safety cessive interlock tension due to fill liquefac-
against failure is defined as the ratio of tion and to limit earthquake-induced settle-
moments due to the effective weight of the ments, the required relative density of the
cell fill above the failure surface (intemal fill must be established. The fill compaction
gravity force) to the overtuming forces procedure must be carefully monitored. In
about the pole of the logarithmic spiral this case, conventional dynamic soil densi-
which approximates a surface of rupture. fication is not possible; thus other methods
The pole of the logarithmic spiral is found of soil improvement (e.g., grouting or chemi-
by trial until the lowest value of safety cal stabilization) should be considered. For
factor is calculated. For a detailed explana- better densification, the granular material
tion of this method, the reader is referred to used for fill should have a limited percent-
Hansen (1953) and Oresen (1962). age of gravel and large boulders. This per-
520 Gravity-Type Quay Walls

mits the full-depth penetration of vibratory tion of soil liquefaction potential is also
probes used for compacting the fill. For bet- required.
ter wall permeability, the fill should have a The latter authors concluded that be-
minimum percentage of fines. cause excessively large deformations may
A discussion on soil liquefaction is given be required to mobilize the full static pres-
in Chapter 4. According to Chakrabarti et sure and because the seismic passive com-
al. (1977), the seismic safety factor against ponent is uncertain, prudent practice dic-
cell sliding on the base may be determined tates that only a partial (say about 50%)
as the ratio of the base shearing resistance, passive pressure be considered in seismic
defined as (V' - ku V)tan cp, where V' is the stability analysis.
effective weight of the cell, V is the total The interpretation of the aforementioned
weight of the cell, k u is the vertical seismic method for the design of cellular bulkheads
coefficient as defined in Chapter 4, to total is given in Schroeder (1990), who also pro-
driving force causing the sliding. The latter vided the comprehensive list of references
includes soil seismic pressures and hydro- on subject matter.
dynamic water pressure (for details, consult As pointed out in Chapter 4 the granular
Chapter 4). fill liquefaction potentials can be greatly
The stability against overturning is de- affected in a positive way by installation of
termined by a method similar to that used a crushed stone (gravel) drain (Figure 4-2).
for the static load condition. For stability,
the resultant of the vertical and horizontal
forces should pass through the middle third
of B. 5.8.6 Deflection of Cellular
The maximum interlock tension due to Bulkhead
seismic loading is calculated for the maxi-
mum horizontal force (k h V), where k h is The deflections of cellular bulkheads at fail-
the horizontal seismic coefficient as defined ure obtained from model tests corresponded
in Chapter 4. The latter includes soil and to 3-6% of the cell free height (Maitland
water pressures inside the cell acting on the and Schroeder, 1979). Relatively recent field
outboard sheet-piling. It is also assumed observation at the Seagirt Marine Terminal
that the horizontal inertia force (kh V) and Fulton Terminal (Clough and Martin,
makes no contribution to centerline shear- 1988) indicated that these cellular bulk-
ing resistance. heads have experienced lateral deflections
In a seismic loading condition, from about 1% to an extreme value of about
Chakrabarti et al. (1977) recommend that 13% of their free height.
no change in value of K be used in comput- Measurements made at Long Beach
ing the internal shearing resistance of the (White et al., 1961) indicated that maxi-
cell at its centerline, where K is deter- mum deflections of about 2-4% of cell free
mined by the Krymine (1944) formulation. height occurred. In another example ob-
If a low factor of safety is obtained, then the served by this author in the mid-1980s,
internal stability may be improved by pro- substantial distortion (up to 10-15%) in
viding more rigidity to the cell through the almost all directions occurred in the in-
welding of the interlocks. harbor breakwater constructed in Resurrec-
As noted by Chakrabarti et al., the prob- tion Bay, Alaska (Fig. 5-88a). This break-
lem of soil liquefaction may also exist on water comprised of steel sheet-pile cells
the "passive side" of a cell driven into foun- filled With a sand-gravel mixture was
dation strata. Therefore, a careful evalua- founded on dense gravel strata underlain
Gravity-Type Quay Walls 521

(a)

(b )

Figure 5-88- Deformations sustained by in-harbour breakwater,


constructed in Resurrection Bay, Alaska. (a) General view; (b) over-
stressed interlocks.

by loose granular soil. The sheet-pile inter- It should be noted that there is often some
locks of this structure have been severely concern among designers andjor contrac-
overstressed (Fig. 5-88b). tors that fill overdensification may increase
Deflections are usually associated with sheet-pile interlock tension to the point that
cell settlement; by common logic, cells are it burst the cell (Schroeder et al., 1977;
expected to tip forward due to a higher toe Sorota et al., 1981, and others).
compression settlement, which is the case Therefore, the fill placement and its den-
in almost all reported field observations. sification to specified relative density must
The cell deflection is usually influenced by be carefully monitored. Cell deflection can
two parameters: foundation soil compress- be reduced by improvement of soil at the
ibility and density of fill within the cell. base of the structure andjor around it. The
Larger deflections were observed where cells discussion on some methods of soil improve-
were placed on compressible foundations ment is given in Chapter 4. An in-depth
and where fill within the cell was not prop- discussion on cell deflection is given in
erly densified or was not compacted at all. Schroeder (1987, 1990).
522 Gravity-Type Quay Walls

5.8.7 Effects of concentrated Naturally, stronger or stiffer fill will offer


Horizontal Loads on better resistance to concentrated forces. In
Sheet-Pile Cell cases where large concentrated forces on
the cell are expected, the area where these
forces are expected to act should be rein-
As stated earlier, a typical cellular bulk-
forced by placing concrete diaphragms in-
head is described as a column of granular
side the cell.
soil enclosed by a steel shell consisting of
The top of a cell may be particularly
interlocking steel sheet piles. This shell is
vulnerable to loads such as ice or ship im-
usually considered as a relatively stiff
pact transmitted to the cell via a fender
structure, and therefore effects of locally
system. In most cases of marine dock con-
applied concentrated loads such as ice or
struction, the top of a cell is protected
ship bow impact must be considered in
either by a concrete slab placed on the top
bulkhead design. Calculations show that
of a fill or by concrete coping wall with a
these loads in most cases are much larger
stiffening horizontal diaphragm placed in-
than the local resistance offered by the cell.
side the cell. It must be pointed out that
Finite element analysis by Guillermo and
these structures must be able to settle along
Lawrence (1987) indicated that the deflec-
with the cell fill without transmitting verti-
tion profiles of a steel sheet-pile cell, ex-
cal loads to the sheeting. An example of this
posed to the heavy concentrated ice load, type of construction is depicted in Figure
differ very much from those characteristic 5-89.
to cells loaded by soil andjor hydrostatic
water pressures. In this case, the sheet-pile
movement occurs in the form of a localized
bulge, and the lateral movements of the cell 5.9 DESIGN OF CRIBWORK·TYPE
walls diminish sharply below the point of OUAY WALLS
the concentrated load application. Natu-
rally, this type of cell deflection may affect As discussed earlier in this chapter, the
interlock stresses. As indicated by Guillermo cribwork-type quay wall is usually built
and Lawrence, the location of the maximum from interlocking timber logs and some-
interlock force occurs at the same place as times from prefabricated reinforced con-
the point of maximum bulge of the cells. crete elements of miscellaneous configura-

Figure 5-89. Steel sheet-pile cellular bulkhead; concrete superstructure. 1-Steel sheet pile;
2- concrete capping wall; 3-concrete diaphragm; 4-ice, lJ is the space between the top of
sheet pile and the capping wall.
Gravity-Type Quay Walls 523

tions. When completed and installed on a may cause the crib member overstress,
foundation, cribs form a cell-type structure specifically at the intersections of the mem-
typically filled with a coarse granular bers. Fill which has not been properly den-
material. sified may also be a cause of greater wall
Where cribs are designed with a solid settlement due to load concentrations at the
front face (which is usually done when cribs crib's bottom perimeter. The crib members
are installed in ice-infested water), they are must be designed to resist the lateral
filled with sand or a sand-gravel mixture; loads which are determined as discussed in
in all other cases, a rockfill of miscellaneous Section 5.6. Typically, the size of the cell
composition is generally used. In the latter cross section varies from 2.5 X 2.5 m to
case, the open space between horizontal 4.0 X 4.0 m.
members will usually govern the size of As stated earlier, the crib wall is often
rocks which should be large enough to be placed directly on a natural bedrock foun-
reliably retained within the crib structure. dation. Therefore, the lower part of the crib
A common assumption in crib design is frame (specifically in the timber crib con-
that a crib wall acts like a massive gravity- struction), which is made on land, is tai-
type wall. Subsequently, the width of the lored to fit the contours of the rock surface.
crib wall is determined in the same way as Consequently, when calculating the wall
indicated earlier for other types of gravity stability against sliding, the composite ef-
wall (e.g., floated-in caisson walls or L- fect of wall shear on rock is considered.
shaped walls). An example of a bottomless crib wall de-
As stated earlier, crib structures and sign is illustrated in Figure 5-90. In the
particularly those made from timber compo- above example, most of the wall's weight,
nents easily adjust themselves to deforma- I: V', which includes the weights of the crib
tions in all directions; therefore, the length frame, fill, and concrete superstructure, is
of an individual crib is practically dictated transmitted to the foundation via the lower
by the onshore construction facilities and part of the crib frame. The remainder of the
launch and towing conditions. Naturally, weight is from the fill cones, I: q 1 , as indi-
cribs made from the reinforced concrete ele- cated in Figure 5-90. Therefore, the wall's
ments are erected in situ. These cribs are
more sensitive to deformations; therefore, qI
reinforced concrete members that are used J7_ .f. .U._U J.UJ
for crib construction must be short enough ..,E
r-- ~~
I
to avoid any additional stress due to crib M.W.L.o
0 I
I
deflections and settlement and particularly -- I
i'

I
I


to deflections due to transverse differential I
settlements (Tschebotarioff, 1965). ~ I

Normally, if correctly proportioned, the


crib structures made from reinforced con-
crete elements are well adapted to with- L L. t -
!e

""
/r',
I 2

stand any damage from the settlement of


J'' /i' '
J

the foundation. The crib is a cellular type of 9


/ / /

structure; therefore, the fill pressure within q, q

the cell is affected by arching, or by the


so-called bin effect as discussed in Section 6min.
5.6.
The fill within cells must be carefully
densified to prevent any settlement which Figure 5-90. Timber crib quay wall; design scheme.
524 Gravity-Type Quay Walls

factor of safety against sliding can be com- termined to ensure crib integrity during
puted from launching, sinking, filling, and while in op-
eration.
Fsl = ( E V'fc + E q'fr )/E E (5-96) It is important to realize that during crib
installation on a foundation, a substantial
uplift load acting on the crib structure must
where be considered in joints design.
fc = friction coefficient between the foundation Last, but not least, the bearing stress
and the crib frame material(s); fc = 0.6 produced by a concrete superstructure on
for timber on rock or crushed stone is the upper part of the crib frame at plane
normally used B-B (Fig. 5-90) must be evaluated. For
fr = friction coefficient between the fill mate- this, the full live load plus the design moor-
rial and the foundation depending on the ing force should be added to the weight of
fill and foundation materials; normally,
the concrete (or other) superstructure with
f = tan ¢, where cp is the angle of inter-
nal friction of fill material
all relevant backfill loads, as required.
I: E = sum of driving forces

When a timber crib is placed on a bedrock 5.10 REINFORCED EARTH QUAY


foundation, the lower part of its frame must WALLS <BY D. WEINREB AND
be evaluated for compressive stresses, 0"0 , P. WU)
produced by I: V' and overturning moment
Mo:
5.10.1 General concept
(5-97)
Massive gravity-type retaining walls where
where soil and reinforcement interact to form a
A = bearing area of the crib frame coherent structure are the common applica-
we = sectional modulus of lower part of a crib tion of Reinforced Earch technology. The
frame in contact with the foundation concept is simple and has been used for
centuries. By placing layers of reinforce-
Contact stresses on the foundation are de- ment between layers of compacted frictional
termined as for any gravity wall; as for the backfill, a composite material is formed,
large-diameter cylinder design, the nega- whereby the soil provides the compreS,sion
tive stresses at the wall base are not and shear strength, and the reinforcements
acceptable. provide tensile strength. Transfer of stresses
Sometimes, for economical reasons, in from the soil to the reinforcing strips is
crib-type construction, the rear cells are cut done by friction at the point of contact be-
short, as indicated in Figure 5-90. When tween the two. The mass of soil and rein-
this type of construction is considered, the forcement forms a continuous and homoge-
stability of the upper part of the structure neous block which transfers the effects of
above plane A-A should be evaluated. earth pressures, surcharges, and its own
Where relevant, the shear strength of the weight to the foundations. Because the
vertical members may be added to the structure has a relatively wider profile than
structure's stability against sliding. Ele- the classical reinforced concrete cantilever
ments of the timber crib are typically joined wall, the bearing stresses transferred to the
by means of bolts and spikes. The number foundations are distributed more evenly
and size of these fasteners are usually de- over the entire width of the base of the
Gravity-Type Quay Walls 525

structure. The typical Reinforced Earth re- Interchange in Great Britain, and the
taining wall is illustrated in Figure 5-91. Dufferin-Montmorency Autoroute in
Reinforced Earth is a well-known pat- Quebec City, Canada, all successful first-
ented soil reinforcement technology which time extensions of the concept in practice in
was developed in France by engineer1 these countries. Since 1972, scientific and
architect Henri Vidal in the mid-1960s. (Re- engineering communities worldwide have
inforced Earth, a term that has become scrutinized the concept and its application,
generic in many countries, is a Registered which has become an ~cceptable standard
Trade Mark of the Reinforced Earth Com- construction practice as an alternate to
pany Ltd.) Vidal (1969) researched and other methods for resisting earth forces.
tested the concept, the related method of The introduction of the Reinforced Earth
design, and the materials from 1957, in structures by Vidal (1969) led to rapid de-
collaboration with the Laboratoire Central velopment of this type of structure. Much
des Ponts et Chaussees and the French fundamental work was sponsored by vari-
Road Research Laboratory, effecting the ous national bodies, such as U.S. Depart-
first successful full-scale Reinforced Earth ment of Transportation (Walkinshaw, 1975),
retaining wall built in Pragneres in the U.K. Department of Transport (Murray,
French Pyrenees in 1966. 1977), and California Department of Trans-
Earlier constructions of Reinforced Earth portation (Forsith, 1978).
retaining wall internationally outside of A typical section of Reinforced Earth is
France included Highway 39 in Los Angeles illustrated in Figure 5-91. The monolithic
in the United States, Route 42 in Sete reinforced soil mass is typically faced with

Horizontal Jotnt - -f>o..h


Vertical Joint - - + - t
Preceat tonerflte feeing Penela

lh""elnforcea tanct"tte Levelling Pea


1151 n x 3ee at
t 25 .. ~lpe
I
I t 19 D Pi n
I

Figure 5-91. Conventional Reinforced Earth wall; typical elements.


526 Gravity-Type Quay Walls

prefabricated-type elements, either in the engineered to balance the frictional and


form of precast concrete panels, which is tensile resistance generated in the soil mass
commonly used in urban areas, or portable against internal and external earth pres-
thin steel elliptical skin elements for struc- sures, and other imposed forces to the
tures at remote sites. structure. External forces can include a uni-
In a typical construction, facing elements formly distributed surcharge load, point
are erected progressively with earth and loads from cargo handling and hauling
strips placement. The integral nature of the equipment, seismic forces, and other loads
interlocking facing along with its joints in of miscellaneous nature. Steel strips are
reinforced earth structures provides several connected to the facing by structural bolts,
features: also with zinc-coated protection. In salt
water and other aggressive environments,
1. Inherent flexibility of the structure for its thicker sections of the reinforcement are
application at sites of significant settle-
included in the design to allow for the antic-
ment.
ipated reduction of the reinforcement sec-
2. Drainage is allowed.
tion due to corrosion.
3. Premanufactured components enable
rapid assembly and eliminate the need The Reinforced Earth technology has
for site formwork. been used worldwide since its introduction
in many diversified applications, such as
The embedded metal strips are manufac- conventional retaining structures for grade
tured of different sections of hot-rolled separations for highway and urban environ-
structural steel of tested tensile strength ment, load-supporting structures for bridge
and are protected from corrosion by hot-dip abutments, facilities with cranes sur-
galvanized zinc coating. These strips are charges, industrial applications for reten-
placed in successive layers within the soil tion dikes and ponds, and hydraulic and
mass in the linear direction where the soil marine structures such as coastal protec-
is stressed. The sectional dimensions, the tion walls, marinas, wharfs, and docks; the
length, and the frequency of the strips examples are shown in Figures 5- 92 and
within a defined geometric soil section are 5- 93.

Figure 5-92. Reinforced Earth seawall, Gaspe, Quebec, Canada.


Gravity-Type Quay Walls 527

Figure 5-93. Reinforced Earth piers, Vancouver, British Columbia,


Canada.

The joints of the Reinforced Earth struc- front of the wall. Adequate protection in the
ture are designed to prevent the flow of form of riprap and filter fabric material or
backfill material from within the structure; other materials are provided to prevent the
however, they allow the free flow of water loss of fines underneath the structure due
in and out to minimize the effects of the to turbulence caused by wave action or ship
unbalanced hydrostatic pressures. propellers.
The bottom of the wall is embedded at The Reinforced Earth principle and con-
least 10% of its height below the mudline in struction system has been successfully

Figure 5-94. Construction of a seawall above the water level, Gaspe,


Quebec, Canada.
528 Gravity-Type Quay Walls

Regardless of the installation method, the


various influences of the marine environ-
ment on the behavior of Reinforced Earth
walls are generally well understood and are
incorporated in the structure design. Many
Reinforced Earth marine structures in
service today have been designed to accom-
modate the effects and behavior of the
Reinforced Earth mass under submerged
conditions, hydrostatic pressures, wave,
current, and ice forces, drainage, scour, and
impacts of marine crafts. These structures
are in service in many countries worldwide
and proved to be reliable and durable.
Several types of marine walls are cur-
rently used in Canada, the United States,
and in other countries around the world. To
date, Reinforced Earth technology has been
applied in various types of seawalls and
waterfront structures built in the marine
environment. Typically, the seawalls are
designed as grade separations to provide
the additional space needed for the develop-
ment of coastal highways and seafront or
Figure 5-95. Seawall, Reunion Island, Indian Ocean. riverfront properties. Examples of these
structures are the coastal highway seawall
at Reunion Island in the Indian Ocean (Fig.
adopted in its basic usage from land struc- 5-95), the seawall in Swansea, Great
tures to applications to river and marine Britain (Fig. 5-96), and the over 30-km-long
environments as early as 1975. Where the seawall supporting the major highway in
Reinforced Earth concept is similarly ap- Canada's Gaspe Peninsula along the St.
plied to structures in the marine environ- Lawrence River (Fig. 5-97). More examples
ment, the basic construction technique is are found in Gegnon (1979).
varied to suit the site conditions. The con- Although the Reinforced Earch seawalls
ventional method of "construction in the could basically be constructed with a con-
dry'' was used in river and coastal struc- ventional earthwork technique, quays and
tures where placement of leveling footing, piers require a different approach, with
facing, strips, and earth fill proceed without consideration given to both design and con-
restrictions from the submerged condition struction methods, where the installation of
during the structure construction (Fig. the prefabricated components of the Rein-
5-94). On the other hand, many marine forced Earth wall and the earth fill can be
structures such as some seawalls, piers, and placed directly from above the water level
quays, face the need of installation through and yet result in an eventual monolithic
existing depths of water. A method of in- structure. Two such structures were suc-
stallation of these underwater walls is de- cessfully completed in 1986 and 1987, using
vised to either construct these structures the latest developed technique; the wall in
from above the water or to install these the Solomon Islands (Fig. 5-98) was built
walls with the aid of divers. with substantial involvement of the divers,
Gravity-Type Quay Walls 529

Figure 5-96. Seawall, Swansea, United Kingdom.

~r=-"""·~r'--~~""co-_-_-_-_-T"---~~~----~~~----~~~------.:..1-
~.~\_______~
___R_e_in_f_o_rc_t_n_g_s_t_r_tp_s______~J

~
:: I ~ l
---=~v=Avg. Hill ~~ I Reinforced Earth
-:- f- 1 Volume
·~~------------------------------~
Cork ~~r~--f-- Filter Fabric

Avg. LIIL ~·~: ~~----------------------------~,


--=~== '•'1"-tf--------------------------__,1
.:~I J

Leve 111 ng Ped

Figure 5- 97. Seawall, Gaspe, Quebec, Canada; typical cross section.


530 Gravity-Type Quay Walls

Figure 5-98. Reinforced Earth dock, Honiara Copra Wharf, Solomon


Islands.

and the other wall was built at Prescott, nance facility for the "ice boom," which are
Ontario with the minimum assistance of logs attached end to end, placed in the win-
divers (Fig. 5-99). ter months for the protection of the dam
The dock at the Honiara Copra Wharf, against pack ice. Placement of this Rein-
completed in August 1986, was built for the forced Earth quay wall in 1988 involved the
Solomon Islands Port Authority. The Rein- guidance of a diver during the grouting of
forced Earth quay wall built underwater the foundation base course only. All other
varied between 3 and 6 m in height. The materials such as concrete facing panels
wall was 85 m long and consisted of 342 m 2 and strip reinforcements were preassem-
of wall facing. The project was constructed bled on land and installed from above
to allow small ships access to the quay front water. The wall was completed by place-
storage sheds which stored copra, one of the ment of the backfill.
country's principal exports. Placement of
concrete face panels and backfill was 5.10.1.1 Longevity
achieved from above the water and the re-
inforcement was lowered into the water Service life in the marine environment is
along keyways in the face panels with the a primary concern in the design of any
guide· of a diver. structure and in the selection of any con-
The Reinforced Earth wall at Prescott struction material. The corrosion character-
built for Ontario Hydro was designed as the istics and the methods of protecting the
"ice boom" maintenance dock. The struc- material have been researched extensively
ture, located near Cornwall, Ontario, was 4 (Darbin et al., 1978). For more details on
m high and 52 m in length, supporting a material corrosion in marine environment
concrete load distribution slab for heavy the reader is referred to Tsinker (1995).
maintenance equipment. The dock, located Two basic methods of designing steel struc-
35 km upstream from a power generating tures and allowing for corrosion are by pro-
project, provides periodic dockside mainte- viding protection against corrosion or by
Gravity-Type Quay Walls 531

(a)

(b)

..
0
0
"'

2 5 2

Figure 5-99. Reinforced Earth wall at Prescott, Ontario, Canada. (a)


General view; (b) typical cross section. 1-wall panels; 2-base panels;
3- reinforcing strips; 4- reinforcing strip supporting cage; 5-tremie
concrete; 6- fill; 7- scour protection.

providing more steel than is required. This example of this is cathodic protection, which
is done by adding a layer of steel which is can be applied by either an electric direct
not included in the actual strength calcula- current or by the use of sacrificial anode
tions and which is sacrificed. In the design such as zinc coating. Galvanization is the
of Reinforced Earth marine retaining walls, most popular form of cathodic protection.
both of these techniques are used. The process used for galvanizing reinforc-
Protectors of steel in the marine environ- ing strips is by hot dipping the processed
ment fall into two categories: passive and fabricated steel. In this process, the steel
active. Examples of passive protectors are strips are first cleaned by submergence in
paints and organic coatings. acid. Following this, they are dipped in
The active protectors function by chang- molten zinc for a specified period of time to
ing the nature of the chemical reaction. An acquire a minimum thickness of 80 J.L per
532 Gravity-Type Quay Wails

side. Zinc provides a sacrificial anode, yet is such as external, internal, and global
tough enough to withstand abrasion due to instabilities.
construction activities. This is an important
property, as the strips may be damaged by
5.10.2.1.1 External stability
handling or equipment during backfilling
and compaction. The structure is evaluated for the insta-
In combination with galvanization, the bility resulting from overturning, sliding on
concept of sacrificial thickness is used. This the base, tilting, rotational slip, and failure
concept is to provide a thicker section of of the foundations by breaking into the un-
steel than required by design. The thick- derlying soil.
ness of this layer depends on the desired For practical design purposes the Rein-
service life and the corrosiveness of the forced Earth blo(!k is usually assumed to
environment. perform as a gravity-type retaining struc-
Zinc coating and sacrificial thickness are ture; therefore, the external stability of this
used in design of structures in a freshwater block and its overall global stability is eval-
environment such as rivers and lakes. For uated using a procedure similar to that used
marine structures in a saltwater environ- for the analysis of a conventional gravity-
ment, zinc coating is ineffective and only type quay wall, as described elsewhere. Ac-
added sacrificial thickness is used. A sacri- cordingly, the factors of safety used are also
ficial thickness of 1.5 mm is usually used similar. For more details on the evaluation
for freshwater environment for a design life of the Reinforced Earth wall external stabil-
of 70 years. ity the reader is referred to Johes (1985).
For more details on corrosion protection
Relevant information is also found in many
in marine environment the reader is re- recent texts on foundation engineering.
ferred to Tsinker (1986, 1995).
Because of the relative flexibility of the
It must be pointed out that Reinforced
reinforced block, the lateral soil pressure
Earth walls may be vulnerable to ship im-
against it is usually taken as "active," re-
pact forces, particularly to those with bul-
gardless of foundation material. A trape-
bous bows. Therefore, the wall must be
zoidal bearing pressure distribution be-
reliably protected by the properly design
neath the structure is usually assumed.
fender system.
Schlosser (1972) suggested that Meyer-
hoffs pressure distribution (Meyerhoff,
1963) can also be used for Reinforced Earth
5.10.2 Design of Reinforced structures analysis. During normal working
Earth Marine Structures conditions, the state of stress within and
outside of the structure is different from
In this section, only the basic design princi- that prevailing at failure and, therefore, the
ples are discussed. For a detailed discussion normal working condition that may be
on the design of the Reinforced Earth struc- equated with an elastic condition. Under
tures, the reader is referred to Jones such a condition, two finite element ap-
(1985) or to any recent text on foundation proaches to the Reinforced Earth structures
engineering. analysis are possible:
5.10.2.1 Basic Design Principles
1. The Reinforced Earth wall may be ideal-
Essentially, the Reinforced Earth wall is ized as a unit cell or composite structure
a gravity-type structure. It is analyzed to in which the reinforcement system is
withstand three distinct modes of failure modeled as a locally homogeneous ortho-
Gravity-Type Quay Walls 533

topic material (Hermann and Yassin, Instrumentation of actual structures as


1978). well as finite element analysis indicated
2. The Reinforced Earth wall is considered that the vertical stresses in the soil within
as a heterogeneous system in which the the Reinforced Earth block is close to uu(z)
soil and reinforcement are represented and is slightly higher toward the facing.
separately (Al Hussaini and Johnson, The maximum tensile force in the strips
1978). If the finite element method is con- Tu(max) occurs at the intersection of these
sidered, then the analytical system should strips and the failure surface, and at any
model accurately the following character- given reinforcing strip level, it is computed
istics (Naylor, 1978):
from the following:
(a) The. longitudinal stiffness of the rein-
forcing elements
(b) The transfer of shear stress between (5-98)
the reinforcing elements and the soil
(c) The transfer of shear through the soil
in the vertic~ plane containing the where
reinforcement au(z)=vertical stress
K A = earth pressure coefficient
5.10.2.1.2 Internal stability A = tributary area (or volume) per reinforcing
strip
The structure is evaluated for the insta-
bility resulting from the slip out of reinforc- At any given reinforcing strip level, the
ing strips (adherence failure), or yield of the maximum tensile force for a given tributary
reinforcing strips (structural failure), or area is distributed among the number of
both. the reinforcing strips in that area.
Having satisfied external stability, the From observations made on instru-
structure is evaluated for the internal sta- mented full-scale Reinforced Earth struc-
bility, namely for the capacity of the rein- tures as well as from the laboratory tests, it
forcing strips in tensile rupture and their follows that at the connection point be-
adherence (through friction) to the soil. tween the reinforcing strip and the facing,
Laboratory tests, finite element analysis, the tensile force, Tuf varies between 85%
and measurement on an actual full-scale
structure (Schlosser and Juran, 1978) con-
firm that the shape of the failure wedge
inside a Reinforced Earth structure is in-
deed curvilinear, as suggested by Terzaghi
and Peck (1967).
The curvilinear failure surface divides
the internal mass of Reinforced Earth into
two distinct zones: the "active zone" where
the shear stresses in the soil are directed '
toward the facing, and the "resistance zone"
where the shear stresses are directed away
from the failure surface toward the fill (Fig. '
I
5-100). The resistance zone is, in fact, where ~----_j
the reinforcing strips adhere to the soil. The
whole essence of the Reinforced Earth
structure is the holding of the active zone Figure 5-100. Failure wedge inside a Reinforced
by the resistance zone through the strips. Earth wall.
534 Gravity-Type Quay Walls

and 100% of Tu(max)> depending on the level is determined by


in the structure (Fig. 5-101).
The capacity of the reinforcing strips is Tf<•l = 2bf*(L- D)[ 'Y 1 (Z- Z 1 ) + yz] (5-99)
checked at the point of maximum tensile
force as well as at the connection (net sec- where
tion) as a function of the yield stress, factor L - D = length of the reinforcing strip in the
of safety, applicable codes, and the reduc- resistance zone (Fig. 5-102)
tion of net section by a sacrificial thickness b = width of the strip
pertaining to the expected service life. The f* = apparent friction coefficient
assumed length L, width b, and the num- z = distance from the top of the wall to the
ber of reinforcing strips are then checked strip level
with respect to the minimum available z = distance from the top of the wall to the
1

frictional capacity. Under compacted and water level


unsaturated conditions, the coefficient of y = density of dry soil
apparent friction f* between earth and y 1 = buoyant density of soil
high-adherence reinforcements decreases as
a result of dilatancy from a maximum value Studies on fills placed underwater indicate
of f* = 1.5 at the surface to tan cp at a that clean uniform crushed stone would un-
depth of 6 m and beyond. However, in case dergo the least postplacement settlement
of the structure built underwater with little due to consolidation and would be the ideal
or no compaction to the backfill below the choice for backfill in the Reinforced Earth
water level, f* is considered to be equal to structure placed underwater due to the low-
tan cp below the water level (Fig. 5-102). est relative movement between the fill and
The maximum frictional capacity of the re- the wall.
inforcing strip to pullout Tf(z) at a depth z Figure 5-103 shows the internal settle-

I
0.6h

\ J
. 0.2
I
•0
.t:.

. 1~
1.0h
~
; -------
z
S2
I l
J
Figure 5-101. Tensile stress distribution in reinforcing strips.
Gravity-Type Quay Walls 535

~------~:-----
/ I
I

I
I
N ,. .....!----
, __J_ ..,., 8
I --=-
I " '
I
I --.....,=~ t .....--='--o.~l I
I
z +-----L-J-=--- - -
f• -
I J! ! +! i J i Gvm
1
I I 1
I
, f -- - -
6 m t - - -- T-- I
1 I
I I
-S-c 19-c
~I!!

I
~~
2
1-1 --~-- j
L
z
Figure 5-l02. Variation of apparent friction coefficient f*.

Tap or veil II

t :::::,_:::::::::=-=::::::::------ Col!lr..,tcn ot cork


bet-- edjiCIIrt pntll
sz I
I I - to be 21oo
per hal;"I zontol Jotnt

Too or ..u
111 151 211
Intrnel aettl..,t rolottvo too rtgld foctng lool

Figure 5- 103. Estimated internal settlement of different kind of fill materials used
for Reinforced Earth structures. (A) Well-graded sand and gravel (maximum compres-
sion = 7%); (B) poorly graded gravel (maximum compression = 5%); (C) clean uniform
stone (maximum compression= 2.5%).

ment characteristics of three types of fill Because the reinforcing strips are con-
relative to a theoretical 10-m-high wall with nected to rigid concrete panels which can-
rigid facing. The fill is assumed to be placed not compress (other than elastic shorten-
in layers 1 m thick and is uncompacted. The ing), artificial flexibility is built into both
values of relative movement between the the horizontal joints between the panels
rigid wall and the fill indicate that the most and into the strip-to-panel connection; 40-
critical location is approximately halfway mm-thick compressible resin bonded cork is
up the wall. normally used in the horizont al joints. Ob-
536 Gravity-Type Quay Walls

servation of the in-service Reinforced Earth


structures show that cork usually com-
presses to about 50% of its initial thickness.
This added compressibility of 20 rom per
cork layer is shown in Figure 5-103 as
cumulative cork compression in a stepped
line. The difference between lines A, B, and
C, representing the relative movement of
the different backfill materials, and the cork
compression envelope is the residual rela-
tive settlement between the fill and the
rigid facing, which should be taken by a Figure 5-104. Connection between face panel and
sliding connection mechanism. reinforcing strip: 1-panel; 2-reinforcing strip;
An example of a strip-tie strip sliding 3-embedded tie strip; 4-styrofoam block; 5-bolt.
connection detail is shown in Figure 5-104.
A gap between the lugs of the tie strip
allows the strip to slide on a bolt. A styro-
Estlooted settleoent of
foam block is placed between the reinforc- clean uniform stone
ing strip and the lugs of the tie strip to
prevent backfill from jamming the strip's 3. 62· ..... ... . .. .............. .... ..... . ... .. ... ........ ..... ~t_rlp__ level

downward movement. !..sz_ 3

Figure 5-105 illustrates the calculated


'j ~ 2 . 78 ..... . ........ . . ~ ..... g .............................~:tf:!P. )~vel
~ Heesured compression
versus measured relative movements of fill ~ 2 16 of cork

in an actual monitored underwater wall Re- .i 1.62 ..... .... H.......1;.....................................~~.1.P.J.evel


inforced Earth structure. The backfill used
in that structure was similar to type C with
gradation as shown in Figure 5-106.
In'tllrlwl eettltM~Rt " 101
nlotlve to e rigid rac:tnu C•l

5_10.2.1.3 overall lgloba/J stability 0 Heosured Internal aettlaoent I location 1 l

f:> Measured Internal settleoent llocotlon 2 l


The structure is evaluated for the overall
instability resulting in a slip circle type of Figure 5-105. Estimated and measured internal
failure of the whole embankment beyond settlement on actual structure.
the limits of the structure.

methods of construction and placement of


5.10.3 construction of backfill are considerably different. The un-
derwater visibility where backfilling has
Reinforced Earth Walls taken place is very limited; therefore, the
Underwater use of divers for attaching the reinforcing
strips to the facing panels underwater usu-
5.10.3.1 Basic Principles ally is not considered. Guidance for vertical
The assembly of basic components of the and horizontal alignment is done from above
Reinforced Earth wall used for underwater the water level. Hence, the design should
construction is illustrated in Figure 5-107. take into account the irregular bottom and
Although, in principle, performance of the caters for it. The components of the wall
Reinforced Earth structures placed under- lowered from the water level are designed
water are similar to those built in the dry, to be self-adjusting and self-leveling.
Gravity-Typ e Quay Walls 537

Sieve sizes 1200 1111 lUI #1 3/8 3/1


100

1
v
80

.s. 60
."'
Cll

c.
....c:

I
10
.,
.,....<>
D..
20
1/
0
10 .. 10.. 10"" 10 10 2

Grein die~eter lm1l

to e1lt

unified so11 clessificetio n

Figure 5-106. Reinforced Earth wall placed underwater ; backfill gradation.

Figure 107. Reinforced Earth wall for underwater construction : 1-base panel; 2-first course
panel; 3-standard panel; 4-reinforci ng strips; 5-reinforci ng strip support cages; 6-guide
pile; 7-coping; 8-filter fabric; 9-tremie concrete.
538 Gravity-Type Quay Walls

An assembly of two concrete beams (typi- Each panel is reinforced to withstand the
cally 8 m long) and guide piles are lowered relevant earth and hydrostatic pressures
below the water level. The base panel is behind it as well as ship impact load, ice
embedded in a dredged trench and provides loads where applicable, and other relevant
a "levelling pad" for the first course panels. loads.
The side walls of the base panel provide a
formwork for rapid-hardening concrete 5.10.3.2 construction sequence
which is tremied below water. Filter fabric,
draped between the sides of the base panel, The basic construction stages of under-
prevents the tremied concrete from leaking water reinforced earth walls include the fol-
out and hence seals the bottom. The trem- lowing sequence.
ied concrete while still in a plastic condition
provides a horizontal level between the 5.10.3.2.1 Site preparation
guide piles. As soon as the first course of To prepare the site, an appropriate sur-
wall panel is lowered and placed on the vey is normally carried out to identify the
leveled surface between the piles, the pivot- proper location of the structure and to es-
ing piles become vertical. tablish the alignment reference for the
The base panel initially supports the ec- construction.
centric load produced by guide piles and General excavation is then performed.
wave action against their protruding por- For typical sites, this involves removal of
tion above the water level. Subsequently, soft surfacial marine deposits to the floor
the structure is held in position by the rein- lines and grades within the Reinforced
forcing strips-panels-backfill interaction Earth structure. More refined trimming is
and the base panels are no longer struc- needed along the front face of the structure
turally required. The base panel does, how- where a trench will be excavated longitudi-
ever, continue to provide a sealant for the nally to accommodate the base panel. This
structure against scour. The reinforcing trench, which runs along the toe at the
bars in the base panel are basically de- front face of the structure, requires a more
signed to withstand the lifting and trans- detailed sounding verification. It is neces-
portation loads and some wave loading sary to remove obstructions encountered in
transferred through the guide piles at the the course of trenching and to trim the
initial stage of construction only. trench so that it is free of irregularities,
The height of the side walls at the base is lumps, or projections. After excavation,
determined to contain the tremied concrete backfill with granular materials, such as
and depends on the slope of the foundation. gravel or crushed stones, is needed.
The width of the base assembly should pro-
vide for sufficient stability against over-
5.10.3.2.2 Premanufactured
turning.
components delivery
Light, wide flange steel !-sections are
normally used for guide piles. These guide Premanufactured component materials,
piles are designed to provide a vertical link including the precast base panels and wall
between concrete panels and to prevent loss panels, steel guide piles and pile connection
of backfill material through the vertical accessories, steel reinforcing strips and their
joints. The guide piles do not, however, support frames, fasteners and joint fillers,
play any structural role in the in-service are trucked to the job site, stored there, and
structure. made ready for on-site assembly. The con-
Concrete panels together with the rein- . tractor needs to plan his proper equipment
forcing strips are installed between piles. and materials schedule at this stage, to be
Gravity-Type Quay Walls 539

readily available for the placement of tremie intervals (Fig. 5-108) and are supported on
concrete for the base panels, and the subse- the aforementioned cages.
quent backfilling operations progressively
with the wall panel and strips placements. 5.10.3.2.4 Underwater installation
The sequence of the structure installa-
5.10.3.2.3 Wall components tion can be summarized in the following
preassembty basic stages.

To assemble the delivered components 1. Installation and aligning of the base pan-
prior to installation, a lay-down area on the els (Fig. 5-109). With the aid of an align-
site is first cleared for the assembly staging. ment beam, the hollow base panel ar-
Two rectangular side beams form units rangement is positioned longitudinally at
which are connected with the supplied the structure front face and lowered into
the water. The base panel can be seated
hardware of pipes and bolts to the guide
on the marine bottom up to a 10% irregu-
piles to provide a temporary formwork as- lar incline. The loosely suspended geotex-
sembly, referred to as the base panel. The tile fabric takes up the irregularities of
guide piles are free to pivot to allow for the seating surface. Once the first base
vertical positioning on irregular marine bot- panel is aligned and positioned, the align-
tom. Geotextile is anchored onto the side ment beam is extruded.
form and loosely suspended through the Subsequently, the base panel units are
bottom of the base panel. The ends of the installed by plumbing the free-pivot guide
pile at the leading end of the previous
unit is blocked by styrofoam inserts.
units and threading onto the alignment
Concurrent to the base panels assembly, beam which is now attached to the unit
vertical support steel cages are prepared. being installed. This panel is then low-
The front end of the reinforcing strips are ered in a similar fashion into the water.
then connected to the panels at prespaced By using this technique, the alignment

Figure 5-108. Reinforced Earth quay wall; preassembly of reinforcing


strips and support cages.
540 Gravity-Type Quay Walls

Figure 5-109. Reinforced Earth quay wall; installation of the base panel.

Figure 5-110. Reinforced Earth quay wall; installation of first course


panel.

beam serves as a target to conform closely ment of this concrete is to provide a level
with the preestablished surveyed design platform or footing within the base panels
location of the wall facing. on which the overlying wall panels will
2. Installation of tremie concrete. Upon com- sit. This operation uses the conventional
pletion of the installation of all the base tremie method and is the only construc-
panels, inspection, and cleaning, if neces- tion stage which requires the service of a
sary, the placement of tremie concrete diver.
can proceed. The purpose of the place- Tremie concrete is allowed to cure to a
Gravity-Type Quay Walls 541

minimum period of 12 hours; the main Compaction of the gravel or crushed


operations of the wall panel installation, stone fill in the underwater portion is not
strips and backfill placement can now required because the design already in-
commence. corporated the considerations of the fill
3. Installation and aligning of the first section, its characteristics, and behavior.
course panels (Fig. 5-110). Installation The density of lower-grade fill can be en-
of the first course panels and the cage- hanced if desired by various vibro ground
supported reinforcing strips are rather improvement techniques when the fill
straightforward. The free-pivoted guide reaches above the water level.
piles at this point are now "pinned" in the The simple combined sequence of
fixed location at the base panels and en- panel-strips arrangement and backfill
sure the horizontal spacing for the first placement as described is repeated until
course panels once they are lowered to the wall height increases to above the
the top of previously installed tremie water level.
concrete. Finally, the above-water portion of the
Precast face panels are lowered be- wall with backfill compaction is installed.
tween adjacent guide piles; these piles are It can be of the same type as used for
roughly vertical until the panel is well underwater construction or, alternatively,
seated. A custom-fabricated lifting beam a conventional Reinforced Earth struc-
is used to hoist the panel and strip as- ture. In any event, compaction of the
sembly together in a balanced fashion for backfill above water is required using the
this operation. Hoisting cables are re- conventional fill placement method.
leased with the help of a remote release 5. Installation of other finishing external fea-
system controlled from above the water. tures. The overall underwater wall panel
When the first course panels are in place, facing, by design, is composed of individ-
they serve as a template for subsequent ual columns of panels between the guide
panel installation. At this point, the guide piles. The elevation of the top of each
piles are primarily vertical due to the panel column, which is governed by the
support provided by the panel templates. tremied concrete elevations within the
The next steps of construction will then base panel, may exhibit some small vari-
follow a repeated cycle of progressive ance. Therefore, there will be a need for
backfilling and adding the panel courses aesthetic reasons to install a capping
as the wall goes up toward the water beam longitudinally on top of the entire
level. facing to provide a uniform finish appear-
ance. The capping beam is made up of
4. Backfill below water. Backfilling below
water is normally carried out using stan- prefabricated or cast-in-place concrete
dard marine construction equipment such where other marine accessories such as
as clamshells or excavator buckets. small-size mooring accessories or pedes-
Backfill is placed in even distribution; trian fencing for marinas could be incor-
porated.
first along the locations of the cages to
stabilize the strip arrangement and re- The bollards used for mooring of large
vessels should be located at a safe dis-
duce excessive deflection of the strips, and
tance from the edge of the wall in order to
then in a direction laterally away from
the facing. Sounding of the fill level is prevent large mooring loads from being
transmitted to the wall facing. In some
required to guide the fill placement so
cases, the bollard foundation located in
that progressive fill layers are roughly
close proximity to the wall can be an-
horizontal. Longitudinally, fill placement
chored by the independent anchor system.
should also be controlled such that open
slopes would not impede the placement of 6. Scour protection. As required in marine
the wall panels and strip cages in adja- environment this will be the final opera-
cent areas. tion to complete the structure.
542 Gravity-Type Quay Walls

7. Monitoring and quality control. Because AMERICAN CONCRETE INSTITUTE, 1985. "State-of-
the major portion of the materials used in the-Art Report on Offshore Concrete Struc-
the Reinforced Earth type of structure is tures for the Arctic." ACI357-1R-85.
premanufactured in an in-plant con- AMERICAN CONCRETE INSTITUTE, 1988. "State-of-
trolled environment, the structural qual- the-Art Report on Barge-Like Concrete Struc-
ity of the materials is assured during tures." ACI357-2R-88; revised in 1993.
production. The remaining monitoring re- AMERICAN PETROLEUM INSTITUTE, 1984. "Recom-
quired is to verify the proper assembly of mended Practice for Planning Designing and
the materials at site, to establish and Constructing Fixed Offshore Platforms." API
achieve correct structure location, eleva- RP2A.
tions, and facing alignment, and to verify ANoNYMous, 1977. "Dubai Dry Dock Slides to
the proper source, quality, and placement Success." Construction Industry Interna-
condition of the backfill materials to con- tional, December.
form with the design assumptions. ANoNYMous, 1990. "Screeding on the Sea Bed."
Construction Weekly, July 25.
All materials, which include all prefabri- ANoNYMous, 1992. Plain Sailing in Portsmouth
cated elements, are manufactured and Harbor. Construction Weekly, January.
tested in accordance with the local codes ARCTEC CANADA LTD., 1984. "Study of Caisson
and standards applied to the materials used Towing and Sinking Methods." Report pre-
pared for Public Works Canada, Vols. I and
in marine environments. Prefabricated
II, September.
structural components are supported with
ARISTAGHES, P., LANNuZEL, J. M., and BoUTTES,
standardized testing and certificates. The F., 1986. "Caissons peu Reflechissants."
owner or the contractor is required to test, PIANC, Bulletin No. 50.
inspect, and certifY the backfill used. BAKER, R. and GARBER, M., 1978. "Theoretical
Field control routines include surveys Analysis of the Stability of Slopes." Geotech-
at different construction stages, alignment, nique, Vol. 28, No. 4.
installation inspection, scheduling and BARRY, P. J., 1983. "Jetty and Renovations CFB
control, and recording as-built structure Dockyard, Halifax, Nova Scotia." ASCE Pro-
geometry. ceedings Special Conference PORTS '83. New
Orleans, LA
REFERENCES BLIGHT, G. E., 1986. "Pressures Exerted by
Materials Stored in Silos: Part 1-Coarse
ACRES INTERNATIONAL LTD., 1987. "Caisson-An Materials." Geotechnique, Vol. 36, No. 1.
Integrated Soft-wave System for the Analysis BLIGHT, G. E., 1990. "Defects in Accepted Meth-
and Design of Marine Caissons." Report pre- ods of Estimating Design Loading for Silos."
pared for Public Works Canada, Ottawa. Proceedings Institute Civil Engineers, Part 1,
AGAR, M. and IRWIN-CHILDS, F., 1973. "Seaforth Vol. 88, December.
Dock, Liverpool; Planning and Design." Pro- BLIGHT, G. E., 1991 "Defects in Accepted Meth-
ceedings Institute Civil Engineers, Vol. 1. ods of Estimating Design Loading for Silos."
Discussion by J. Y. Osi, G. Mathieson, and M.
AL HUSSAIN!, M. M. and JoHNSON, L. D., 1978.
J. Blackles. Proceedings Institute Civil Engi-
"Numerical Analysis of Reinforced Earth
neers, Part 1, Vol. 90.
Wall," ASCE Proceedings Symposium Earth
Reinforcement, Pittsburgh, PA. BOWLES, J. E., 1987. "Elastic Foundation Settle-
ments on Sand Deposits." ASCE Journal of
AMERICAN CONCRETE INSTITUTE, 1977. "Recom- Geotechnical Engineering, Vol. 113, No.8.
mended Practice of the Design and Construc- BRAESTRUP, M. W., 1988. "Discussion on Paper
tion of Concrete Bins, Silos, and Bunkers for by Tassious and Vintzleon, 1987." ASCE
Storing Granular Materials." ACI373R; re- Journal of Structural Engineering, Vol. 114,
vised in 1983. No.12.
Gravity-Type Quay Walls 543

BRAY, R. N. and TATHAM, P. F. B., 1992. "Old CIVIL ENGINEERING, 1980. Venezuelan Dock Fea-
Waterfront Walls: Management, Maintenance tures Unconventional Design, Quick Set-Up,
and Rehabilitation." E & FN Span, London, July.
U.K. CLAUSEN, C. J. F. and LUNUE, T., 1979. "The
BRIASSOULIS, D., 1991. "Limitations in the Range Application of Soil Investigation Data to the
of Applicability of the Classic Silo Theories." Design of Offshore Gravity Platforms." Pro-
ACI Structural Journal, Vol. 88, No. 4, ceedings International Conference on Off-
July-August. shore Site Investigation, London.
BROWN, K., 1978. "Caissons at the Dumman CLOUGH, C. W. and MARTIN, J. R., 1988. "Cellu-
Site." Construction News, May. ral Cofferdams-Developments in Design and
BROWN, P. P. and FORREST, J. B., 1983. "Discus- Analysis." Proceedings 2nd International
sion on Paper by Sorota, Kinner and Haley, Conference on Histories in Geotechnical Engi-
1981." ASCE Journal of Geotechnical Engi- neering, St. Louis, MO.
neering, Vol. 109, No.4, April. Concrete Reinforcing Steel Institute Handbook,
BRUM, A. J., ET AL., 1959. Seaports and Port 1984.
Related Structures, 'Sea Transport," Pub- CoNTRI, G., NocE, G., and FEDOLINO, M., 1986.
lisher, Moscow (in Russian). "Precast Elements in Harbour Works."
BRUUN, P., 1989. Port Engineering, Vol. I, 4th PIANC, Bulletin No. 56.
ed., Gulf Publishing Co., Houston, TIC CooKE, A. R., 1990. "Underwater Concrete Con-
BUDIN, A. Y. and DEMINA, G. A., 1979. Quays, struction of Phuket Port Quay Wall Founda-
Design guidelines, Stroyisdat, Moscow (in tion." The Dock and Harbour Authority,
Russian). September.
Buswv, V. M., 1986. "Base Skirts for Arctic CORNICK, H. F., 1958. Dock and Harbour Engi-
Offshore Drilling Platforms." Proceedings 5th neering, Vol. I, The Design of Docks, Charles
International Offshore Mechanics and Arctic Griffin & Co. Ltd. London, U.K.
Engineering (OMAE) Symposium, Vol. 4, CUMMINGS, E. M., 1957 "Cellural Cofferdams and
Tokyo. Docks." ASCE Journal of the Waterways and
BUSLOV, V. M. and BRAKEL, J., 1985. "Base Un- Harbor Division, Vol. 83, No. WW3.
derfilling for Arctic Platforms." Proceedings of CUMMINGS, E. M., 1960. "Cellural Cofferdams
the 4th International Offshore Mechanics and and Docks." Transactions ASCE, Vol. 125.
Arctic Engineering (OMAE) Symposium, Vol. DARBIN, M., JAILLOUX, J. M., and MONTUELLE, J.,
II. 1978. "Performance and Research on the
CAQUOT, A. and KERISEL, J., 1953. "Sur la terme Durability of Reinforced Earth Reinforcing
de surface dams le calcul des foudations en Strips." ASCE Symposium on Earth Rein-
milien pulverult." Proceedings 3rd Interna- forcement, Pittsburgh, PA.
tional Conference on Soil Mechanics and DE BEER, E. E., 1967. "Proefondervindelijke b:rj-
Foundation Engineering, Vol. 1, Zurich. drage tot de studic van het gransdraagrermo-
CARCHEDI, D. R. and PORTER, D. L., 1983. "Marine gen van zand ouder funderingen op staal.
Terminal Modifications. Port of Albany, New Bepalin von der vormfactor sb." Anales des
York." Proceedings ASCE Specialty Confer- Travanx Publies de Belgique, Vol. 68, No. 6;
ence PORTS '83 (March), New Orleans, LA. Vol. 69, Nos. 4-6.
CHAKRABARTI, S., HUSEK, A. D., CHRISTIANO, P. DE BEER, E. E. and LADANYI, B., 1961. "Etude
P., and TROXELL, D. E., 1977. "Seismic Design Experimentale de la Capacite Portante du
of Retaining Walls and Cellular Cofferdams." Sable sons des Foudations Circulaires
ASCE Proceedings Specialty Conference on Etablies en Surface." Proceedings 5th Inter-
Soil Dynamic and Earthquake Engineering, national Conference on Soil Mechanics and
Vol. I, Pasadena, CA. Foundations Engineering. Paris.
CIORTAN, R., 1994. "Efficient Solutions for Grav- DET NORSKE VERITAS, 1977. "Rules for the De-
ity Quay Construction." Proceedings 28th sign, Construction and Inspection of Offshore
Congress PIANC, S:II-3, Seville, Spain. Structures." Det Norske Veritas, Oslo.
544 Gravity-Type Quay Wails

DISMUKE, T. D., 1975. "Cellural Structures and GEGNON, G., 1979. "Sea Wall Constructed in Re-
Braced Excavations." Foundation Engineer- inforced Earth." Comptes Rendus Colloguim
ing Handbook, Van Nostrand Reinhold, New International Reinforcement des Sols. Paris.
York. GERWICK, B. C., 1986. Construction of Offshore
DUBOIS, M. J., 1990. "Politique suivie en France Structures, John Wiley & Sons, New York.
en matiere de conception, construction et
entretien d'ouvrages portuaires." PIANC GoRUNOV, B. F., 1974. Operation and Mainte-
Proceedings 27th Navigation Congress, nance of Port Related Marine Structures (in
Osaka. Russian).
DUNCAN, J. M., CLOUGH, G. W., and EBELING, R. GRAYMAN, R., 1970. "Cellural Structures Fail-
M., 1990. "Behavior and Design of Gravity ures." Proceedings Specialty Conference on
Earth Retaining Structures." ASCE Proceed- Design and Installation of Pile Foundations
ings Conference on Design and Performance and Cellural Structures, Lehigh University.
of Earth Retention Structures, Ithaca, NY. GUILLERMO, E. C. and LAWRENCE, S. A., 1987.
EAU, 1985. Recommendations of the Committee "Design of Aida Seaport of the DeLong Moun-
for Waterfront Structures, Ernst and John, tain Regional Transportation System at
Berlin. Chukchi Sea, Alaska." Proceedings 9th Inter-
EDDIE, A. G. F., 1976. "Concrete Replaces Steel national Conference POAC, Fairbanks. AK.
in Offshore Coal-Loading Terminal." Civil GUREVICH, V. B., 1969. Port Related Marine
Engineering, April. Structures for Construction of Inland Ports,
EDDIE, A. G. F., 1986. "Modern Developments in Transport, Moscow (in Russian).
Caissons and their Impact on Future Possibil-
HANSEN, J. B., 1953. Earth Pressure Calcula-
ities in the Design and Construction of Break-
tion, The Danish Technical Press, Copen-
waters." PIANC, Bulletin No. 50.
hagen.
ELMAN, M. T. and TERRY, C. F., 1987. "Retaining
Walls with Sloped Base." ASCE Journal of HANSEN, J. B., 1957. "}?oundations of Structures,
Geotechnical Engineering, Vol. 113, No.9. General Report." Proceedings 4th Interna-
tional Congress on Soil Mechanics and Foun-
ESPINOZA, R. D., BOURDEAU, P. L., and
dation Engineering, Vol. 2.
MUHUNTHAN, B., 1994. "Unified Formulation
for Analysis of Slopes with General Slip Sur- HENSEN, J. B., 1970. "A Revised and Extended
face," ASCE Journal of Geotechnical Engi- Formula for Bearing Capacity." The Danish
neering, Vol. 120, No. 7. Geotechnical Institute, Bulletin No. 28,
FEDERATION INTERNATIONALE DE LA PRECON- Copenhagen, Denmark.
TRAINTE (FIP), 1979. "State-of-the-Art Report; HERMANN, L. R. and YASSIN, A. L., 1978.
Foundations of Concrete Gravity Structures "Numerical Analysis of Reinforced Earth Sys-
in the North Sea," London. tems." ASCE Proceedings Symposium Earth
FORSITH, R. A., 1978. "Alternative Earth Rein- Reinforcement, Pittsburgh, PA.
forcements." ASCE Proceedings of Sympo- HETINYI, M., 1946. Beams on Elastic Founda-
sium on Earth Reinforcement. Pittsburgh, PA. tions, University of Michigan Press, Ann
Foss, I. and DHALBERG, R., 1979. "FIP State of Arbor.
the Art Report: Foundations of Concrete
HUNTINGTON, W. C., 1957. Earth Pressures and
Gravity Structures in the North Sea." Foun-
dation Design Methods for Gravity Structures, Retaining Walls, John Wiley & Sons, New
York.
Federation Internationale de la Precon-
trainte, London. IRWIN-CHILDS, F. and MACKNIGHT, A., 1977. "Ap-
FRANco, L., 1994. "Vertical Breakwaters: the plications of Diaphragm Walling to Port
Italian Experience," Coastal Engineering, Vol. Structures." ASCE Proceedings Specialty
24. Conference PORTS '77.
FURUDOI, T. and KATAYAMA, T., 1971. "Field Ob- JANSSEN, H. A., 1885. "Versuche Uber Getreide-
servation of Residual Water Level." Technical druck in Silozellen." VDI Zeitschrift, Vol. 39,
Note of PHRI, No. 115, Japan. August. Dusseldorf, Germany.
Gravity-Type Quay Walls 545

JAPAN SOCIETY OF CML ENGINEERS, 1977. Spe- MAITLAND, J. K., and SCHROEDER, W. L., 1979.
cial Publication. Earthquake Resistant Design "Model Study of Circular Sheetpile Cells."
for Civil Engineering Structures, Japan Soci- ASCE Journal of Geotechnical Engineering,
ety of Civil Engineers, Tokyo. Vol. 105, No. GT 7.
JARLAN, G. E., 1963. Canadian Patent on Perfo- MEYERHOF, G. G., 1951. "The Ultimate Bearing
rated Breakwater Structure, National Re- Capacity of Foundations." Geotechnique, Vol.
search Council of Canada, Ottawa. 2, No.4.
JONES, C. J. F. P., 1985. Earth Reinforcement MEYERHOF G. G., 1963. "Some Recent Research
and Soil Structures, Butterworths, London. on the Bearing Capacity of Foundations."
KRYNINE, D. P., 1944. "Discussion on Stability Canadian Geotechnical Journal, Vol. 1, No.1.
and Stiffness of Cellural Cofferdams." Trans- MORNEMENT, P. C. and PEARCE, R. W., 1987.
actions ASCE, Vol. 110. "Jiuzhou Harbor. Zhuhai Special Economic
KUMBHOJKAR, A. S., 1992. "Numerical Evalua- Zone PRC." Proceedings Coastal and Port En-
tion of Terzaghi's Ny." ASCE Journal of gineering in Developing Countries, Vol. I,
Geotechnical Engineering, Vol. 119, No. 3. China Ocean Press.
KVALSHAD, T. J. and DAHLBERG, R., 1980. "Cyclic Moss-MoRRis, A. and HODGE, N. M., 1981. "De-
Behaviour of Clay as Measured in Laboratory sign and Construction of Quay Walls for a
Soils Under Cyclic and Transient Loading." Container Terminal Founded on Fissured
International Symposium on Soils Under Clays." PIANC Proceedings 25th Congress,
Cyclic and Transient Loading, Vol. 1. SII-4, Edinburgh.
LAcROIX, Y., ESRIG, M. I., and LuscHER, U., 1970. MURFF, J. D. and MILLER, T. W., 1977. "Stability
"Design, Construction and Performance of of Offshore Gravity Structure Foundations by
Cellural Cofferdams." ASCE Proceedings of the Upper Bound Method." Ninth Annual
the Soil Mechanics and Foundation Division. Offshore Technology Conference, Houston,
Specialty Conference on Lateral Stresses in TX, Vol. 3, Paper No. 2896.
the Ground and Earth Retaining Structures.
MURRAY, R. T., 1977. "Research at TRRL to De-
LADD, C. C., 1991. "Stability Evaluation During velop Design Criteria for Reinforced Earth."
Staged Construction ..The Twenty-Second Varl Proceedings Symposium on Reinforced Earth
Terzaghi Lecture." ASCE Journal of Geotech- and Other Composite Soil Techniques,
nical Engineering, Vol. 117, No.4. Heriot-Watt University, TRRL, Sup. 457.
LANDGREN, H., 1986. "Development of Caisson NAVDOCKS, 1962. Design Manual DM-7, Depart-
Breakwater Technology." PIANC Bulletin No. ment of Navy, Alexandria, VA.
50.
NAVFAC DM-7.2, 1982. Foundation and Earth
LELARGE, P. and LEBARs, X., 1992. "Design and
Structures, Design Manual 7.1, Department
Computation of Caisson Type Wall in Port of
of Navy, Alexandria, VA.
LeHavre." Proceedings lOth International
Harbour Congress, Antwerp, Belgium. NAYLOR, D. J., 1978. "A Study of Reinforced
Earth Walls Allowing Slip Strip." ASCE Pro-
LEONARDS, G. A. and FROST, J. D., 1988. "Settle-
ceedings Symposium Earth Reinforcement,
ment of Shallow Foundations on Granular
Pittsburgh, PA.
Soils." ASCE Journal of Geotechnical Engi-
neering, Vol. 114, No. 7. NOTTINGHAM, D., 1995. "Open Cell Bulkheads."
LESHCHINSKY, D., 1990. "Slope Stability Analy- ASCE Proceedings Specialty Conference
sis: Generalized Approach." ASCE Journal of PORTS '95, Tampa, FL.
Geotechnical Engineering, Vol. 116, No. 5. 0RESEN, N. K., 1962. "Cellural Cofferdams, Cal-
LOEWY, E. I., PANNETT, R. J., and COLLIUGE, culation Methods and Model Tests." Danish
1990. "Structures in Commercial and Fishing Geotechnical Institute Bulletin 14, Copen-
Ports." Part 1: Blockwork Quay Walls Design, hagen, Denmark.
Construction and Utilization in Developing OsTERBERG, J. 0., 1989. "Necessary Redundancy
Countries. PIANC, Proceedings 27th in Geotechnical Engineering." ASCE Journal
Congress, SII-3. Osaka. of Geotechnical Engineering, Vol. 113, No. 11.
546 Gravity-Type Quay Walls

PENNOYER, R. P., 1934. "Gravity Bulkheads and SARAN, S. and AGARWAL, R. K, 1991. "Bearing
Cellural Cofferdams." Civil Engineering, Vol. Capacity of Eccentrically Obliquely Loaded
4, No.6. Footing." ASCE Journal of Geotechnical En-
PENZIEN, J. and TSENG, W. S., 1976. "Seismic gineering, Vol. 117, No. 11.
Analysis of Gravity Platforms Including SCHLOSSER, F., 1972. "La Terre Armee-
Soil-Structure Interaction Effects." Eighth Recherdres et Relisations." Bulletin de Liai-
Annual Offshore Technology Conference, son LCPC, No. 62.
Houston, TX, Vol. 3, Paper 2674. SCHLOSSER; F. and JURAN, I., 1978. "Theoretical
PIANC WAVE COMMISSION, 1980. Report on Wave Analysis of Failure in Reinforced Earth Struc-
Studies, PIANC, Brussels. tures." ASCE Symposium on Earth Reinforce-
PIRAH, I. C., 1975. "Applications of Finite Dif- ment, Pittsburgh, PA.
ference Techniques to Geotechnical Prob- ScHROEDER, W. L., 1987. "Wharf Bulkhead Be-
lems." Analysis of Soil Behavior and havior at Fulton Terminal 6." ASCE Journal
Its Application to Geotechnical Structures, of Geotechnical Engineering, Vol. 113, No. 6.
Proceedings of the Technical Session of the
ScHROEDER, W. L., 1990. "Cellural Sheetpile
Symposium University of New South Wales,
Bulkheads." ASCE Proceedings of Conference
Australia, July 14, 18, 1975. Vallipan, S.,
on Design and Performance of Earth Retain-
Rain, S., and Lee, I. K (eds). Kensington,
ing Structures, Ithaca, New York.
Australia: Unisearch Ltd.
PITKALA., L., 1986. "Retaining Wall Made of SCHROEDER, W. L. and MAITLAND, J. K, 1979.
L-Elements as a Quay Structure." PIANC "Cellular Bulkheads and Cofferdams." ASCE
Bulletin No. 54. Journal of the Geotechnical Engineering Divi-
sion, Vol. 105, No. GT 7.
POULOS, H. G. and DAVIS, E. H., 1974. Elastic
Solutions for Soil and Rock Mechanics, John ScHROEDER, W. L., MARKER, D. K, and KHuAY-
Wiley & Sons, New York. JARERNPANISHK, T., 1977. "Performance of a
Cellural Wharf." ASCE Journal of Geotechni-
PREVOST, J. H., 1981. "Offshore Gravity Struc-
cal Engineering, Vol. 103, No. GT 3.
tures Analysis," ASCE Journal of the
Geotechnical Engineering Division, Vol. 107, SEKIGUCHI, K, RoWE, R. K, Lo, K Y., and
No. GT 2. OGAWA, T., 1992. "Hoop Tension Analysis of a
QUINN, A. DE F., 1972. Design and Construction Steel Cell Using Optimization Method for
of Ports and Marine Structures, 2nd ed., Pseudo-Viscoplastic Parameters." JSSMFE,
McGraw-Hill Book Co., New York. Soils and Foundation, Vol. 32, No.3.
RAYBURN, L., 1981. "River Cellural Structures SHANNON & WILSON, INc., 1982. "Lock and Dam
-Design-Built Concept," Presentation at 26(R). Analysis of Instrumentation Data, Vol.
March GEO-Pile Conference, Associated Pile I, Interim report. St. Louis District, October."
& Fitting Corporation. Atlanta, GA. U.S. Army Corps of Engineers.
REIMBERT, M. L. and REIMBERT, A. M., 1976. SMIRNOV, G. N., 1979. Ports and Port Related
Silos-Theory and Practice, Trans Technical Marine Structures, Stroyisdat, Moscow (in
Publications, Clamsthal-Zellerfeld, Germany. Russian).
Rossow, M. P., 1984. "Sheet Pile Interlock Ten- SOROS, P. and KoMAN, B., 1974. "Artificial Island
sion in Cellural Cofferdams." ASCE Journal Trans-Shipment Terminal." Proceedings OTC,
of Geotechnical Engineering, Vol. 110, No. 10. Paper 2099, Dallas, TX.
ROURKE, T. D. and JONES, C. J. F. P., 1990. SoROTA, M. D. and KINNER, E. B., 1981. Cellural
"Overview of Earth Retention Systems: Cofferdam for Trident Drydock: Design; Per-
1970-1990." ASCE Proceedings Conference on formance." ASCE Journal of the Geotechnical
Design and Perforri:lance of Earth Retention Engineering Division, Vol. 107, No. GT 12.
Structures, Ithaca, NY. SOROTA, M. D., KINNER, E. B., and HALEY, M. X.,
SAFARIAN, S. S., 1969. "Design Pressure of Gran- 1983. "Closure on Discussion by Brown and
ular Materials in Silos." ACI Journal, Forrest, 1983." ASCE Journal of Geotechnical
August. Engineering, Vol. 109, No.4.
Gravity-Type Quay Walls 547

SWATEK, E. D., 1967. "Cellural Cofferdam Design lished report, Tsinker & Associates, Inc.,
and Practice." ASCE Journal Waterways and Niagara Falls, Ontario.
Harbors Division, Vol. 93, No. WW3. TSINKER, G. P., 1995 Marine Structures Engi-
SWATEK, E. P., 1970. "Cellural Structures Design neering: Specialized Applications, Chapman
and Installation. Proceedings of the Confer- & Hall, New York.
ence on Design and Installation of Pile Foun-
TuruTZIN, A. M., 1963. "Dynamic Pressure of
dations and Cellural Structures, Envo Pub-
Granular Material in Deep Beans." ASCE
lishing Co., Pennsylvania.
Journal Soil Mechanics and Foundation En-
TANIMOTO, K. and TAKAHASHI, S., 1994. "Design gineering, Vol. 89, No. ST 2.
and Construction of Breakwaters," Coastal
Engineering, Vol. 22. The Sea and Inland Waterway Hydrotechnical
Structures. "Transport" Publisher, Moscow,
TASSIOS, T. P. and VINTZELEON, E. N., 1987.
USSR.
"Concrete-to-Concrete Friction." ASCE
Journal of Structural Engineering, Vol., 113, U.S. ARMY CORPS OF ENGINEERS, 1974. Cellural
No.4. Sheetpile Cofferdam Failures," Office of the
TENNESSEE VALLEY AUTHORITY, 1957. Cofferdams Chief of Engineers, Washington, DC.
on Rock, TVA Technical Monograph 75, Vol. U.S. DEPARTMENT OF THE INTERIOR BUREAU OF
1, Tennessee Valley Authority, Knoxville, TN. RECLAMATION. "Moments and Reactions for
TERZAGHI, K., 1943. Theoretical Soil Mechanics, Reactangular Plates. Denver, Colorado.
John Wiley & Sons, New York. VESIC, A., 1963. "Bearing Capacity of Deep
TERZAGHI, K., 1944. "Stability and Stiffness of Foundations in Sand." Highway Research
Cellular Cofferdams." ASCE Transactions, Record. National Academy of Sciences NRC.
Vol. 110, 1945, Paper No. 2253. VESIC, A. S., 1975. "Bearing Capacity of Shallow
TERZAGHI, K. and PECK, R. B., 1967. Soil Me- Foundations." Foundation Engineering Hand-
chanics in Engineering Practice, 2nd ed., John book, Winterkorn, H. F. and Fang, H. Y. (eds.),
Wiley & Sons, New York. Van Nostrand Reinhold, New York.
THOMAS, H. E., MILLER, E. J., and SPEAKER, J. J., VIDAL, H., 1969. "The Principles of Reinforced
1975. "Difficult Dam Problem-Cofferdam Earth." Highway Research Record 282.
Failure." ASCE Civil Engineering. Special
WALKINSHAW, J. L. 1975. "Reinforced Earth Con-
publication.
struction." Department of Transportation,
TSAGARELI, Z. B., 1969. New type of Light Re- FHWA Region 15, Demonstration Project No.
taining Walls (in Russian). 'Stroyisdat', 18.
Moscow, USSR.
WHITE, A., CHENEY, J. A., and DUKE, C. M.,
TSCHEBOTARIOFF, G. P., 1965. "Analysis of a High
1961. "Field Study of a Cellural Bulkhead."
Crib Wall Failure." Proceedings 6th Interna-
ASCE Transaction, Vol. 128, Part 1, Paper
tional Conference on Soil Mechanics and
No. 3426.
Foundation Engineering, Vol. 2, Montreal.
TSCHEBOTARIOFF, G. P., 1973. Foundations, Re- WHITENECK, L. L. and HacKNEY, L. A., 1989.
taining and Earth Structures, 2nd ed., Structural Materials for Harbor and Coastal
McGraw-Hill Book Co., New York. Engineering, McGraw-Hill Book Co., New
York.
TSINKER, G. P., 1958. "New Type of Quay Wall
for Construction on Rock." Inland Transport, XANTHAKOS, P. P., 1979. Slurry Walls, McGraw-
No. 12 (in Russian). Hill Book Co., New York.
TSINKER, G. P., 1965. "Optimization of Passenger YAN, H. T., 1983. "Concrete Crib Structures."
Traffic on the Dnieper River." Unpublished Canadian Journal of Civil Engineering, Vol.
report (in Russian). 10.
TSINKER, G. P., 1986. Floating Ports, Gulf Pub- YoUNG, A. G., 1992. "Marine Foundation Stud-
lishing Company, Houston, TX. ies." Handbook of Coastal and Ocean Engi-
TSINKER, G. P., 1994. "Gravity Type Marine neering, Vol. 2, Herbick, J. B. (ed.), Gulf Pub-
Structures for Rapid Deployment," Unpub- lishing, Houston, TX.
6
Sheet-Pile Bulkheads

6.1 INTRODUCTION tion of harbor related structures, and for


the execution of deep foundations.
Sheet-pile bulkheads are frequently used in The use of timber sheet-piling was also
marine engineering. These structures have known to the ancient Egyptians. Various
proven to be a viable and economica! solu- types oftimber sheet-piling techniques (e.g.,
tion to waterfront construction, where the tongue and groove, laminated, and various
wall height does not exceed 18-20 m and other types of sheet piles) were used exten-
the foundation soil permits pile-driving. sively in earlier centuries and specifically in
These walls are usually treated as flexible the Roman Empire. Descriptions of differ-
and are normally distinguished by the ma- ent methods of sheet-pile use and handling
terial of sheet piles (e.g., wood, steel, or are found in treaties by Roman architects
concrete), method of sheet-pile support, and (Leimdorfer, 1979). In principal, the timber
sequence of wall construction. sheet piles used by ancient engineers for
marine application were flat piles driven
successively edge to edge to form a vertical
6.1.1 Sheet-Piling- Background wall for the purpose of preventing the re-
tained materials from spreading and also
6.1.1.1 Timber Sheet Piles for protecting the retained material from
Timber sheet-pile walls are perhaps the undermining by the action of water. Differ-
oldest known structures with close connec- ent types of timber sheet-piling techniques
tions to port and harbor construction. Use found in ancient Egypt and Mesopotamia
of wood sheet-piling for construction of har- are illustrated in Figure 6-1.
bors along the Mediterranean coast line was Under certain site conditions (e.g., when
quite common thousands of years ago. The timber piles are fully submerged in water),
Phoenicians, for example, knew how to use they would not decay and are likely to last
the Lebanon cedars for creating long sheet- for centuries. However, timber piles usually
piling planking for building ships, construc- deteriorate rapidly if subjected to alternat-

549
550 Sheet-Pile Bulkheads

above the mudline is subject to possible


(a) extensive mechanical damage by ice, ship
impact, drifting sediments, and so on. Fi-
nally, it should be mentioned that the me-
chanical properties of timber do not allow
for the construction of bulkheads of sub-
stantial height. Ali of the above reasons
explain why wooden sheet-pile bulkheads
are usually not used for construction of per-
manent or retaining structures higher than
3-4 m, particularly in marine environ-
ments.
Timber sheet-pile bulkheads roade from
treated wood have been constructed in the
United States, particularly in coastal re-
gions, as a secondary line of shore defense
in ocean-exposed locations, where they are
typically built in combination with groins.
Timber sheet-pile bulkhead are also used in
Figure 6-1. Miscellaneous sheet-piling techniques the construction of small-craft harbors.
used in ancient Egypt Mesopotamia. [From Leimdorfer
(1979).]
6.1.1.2 cast-lron Piles
Cast-iron piles became complementary to
ing drying and wetting. Different forms of
timber piling in the early part of the 19th
impregnation treatment of the entire mass
century. Perhaps one of the first discussions
of the pile timber by various chemical
on the use of iron sheet piles for marine
preservatives usually just delay the process
application was offered by Borthwick (1836).
of decay but cannot prevent it entirely. For
example, in some tropical harbors, chemi- The earliest reported use of iron sheet
cally well-treated timber piles did not last piles in the United Kingdom was construc-
more than 14-15 years on average when tion of the N orth Pier of Bridlington harbor
exposed to seawater. For an extensive dis- in the early 1820s (Mackley, 1977). Various
cussion of the causes of wood decay in ma- types of sheet-pile sections were available
rine environment, the reader is referred to at that time, and a considerable amount of
Chapter 1 and Tsinker (1995). exploratory work was carried out in order
Where economically feasible, extremely to develop the most economica} profile. Cast
durable and strong tropical woods can iron, however, as a material had limitations
be used in worm borer-infested waters. primarily because of its vulnerability to
Mechanical properties of several kinds of brittle fracture during driving in hard soils.
such woods are given in EAU (1990) and Typically, the wrought-iron piles were used
Chapter 1. in a composite riveted form and were based
In wooden sheet-pile bulkheads, mois- primarily on the fitting of plates between
ture trapped in the vicinity of steel fasten- suitable guides or against supports.
ers and stringers and on top of piles usually In 1897, Larssen, a Danish engineer, rev-
become a source of fungal attack, and olutionized the use of iron sheet piles by
stringers and caps very often sustain exten- introducing a new pile section which was
sive rot. Additionally, the exposed part of developed from a rolled trough section plus
wooden bulkheads in a marine environment a riveted "z"-section, to form an interlock;
Sheet-Pile Bulkheads 551

this shape is very familiar in modern con- marine application for construction of
struction. In 1914, Larssen also introduced cellular-type bulkheads. These piles were
the first deep-arch section in which inter- first manufactured and used in the United
locks were situated in the neutral axis of States in 1908-1909.
the complete section; thus, their material For more details on the history of iron
bulk did not in:fluence the bending moment and steel sheet-pile development, the inter-
to be absorbed. ested reader is referred to Mackley (1977)
Larssen's inventions and modifications and Leimdorfer (1978, 1979). Details on
helped to substantially increase the capac- rolled steel sheet piles currently used in
ity and effectiveness of sheet piles in their marine engineering practice are discussed
ability to resist earth and water pressures. in Chapter 2 and later in this chapter.
lncreased pile stiffness enabled it to be It should be noted that steel pipe-type
driven without buckling or springing under sheet piles are now coming into widespread
the blow of the driving hammer, increased use for deep water construction where a
watertightness of the sheeting prevented sheeting of greater strength is required.
seepage through the wall, and, most impor- An obvious advantage of steel sheet pil-
tantly, the efficient use of rolling mills pro- ing is its universality and ease of installa-
duced an economica! section with inter- tion. Furthermore, steel sheet piles have an
locks. advantage in their ability to absorb local-
ized overloading without damage. The dis-
6.1.1.3 steel Sheet Piles advantage of any steel structure operating
in a marine environment and particularly
The first sheet pile ever made from a in a warm saltwater environment is its sus-
rolled section was used in Chicago in 1901; ceptibility to corrosion. In an aggressive
it was called the Jackson pile. This was marine environment, the steel sheet piling
followed by the rather fast development of is usually protected by miscellaneous ca-
numerous straight or trough sections of thodic protection systems, otherwise, corro-
steel sheet piles that were produced around sion allowance is specified.
the world either with integral locking ar-
rangements or with a separate interlocking 6.1.1.4 Reinforced concrete
member. For details, the reader is referred Sheet Piles
to Chapter 2.
The first steel "z"-shaped sheet piling Reinforced concrete sheet piles have been
known as the Hoesch system was intro- used in harbor construction since the begin-
duced in Belgium in 1913. In comparison ning of the century. They are usually con-
with other sections, this type of piling was sidered as relatively maintenance-free com-
stiffer and had a higher section modulus at ponents of a sheet-pile wall. Although many
equal weight with other piling systems. different design types have been developed
Since this time, various combinations of and used in the past 50 or so years, the
different piling systems and various types straight web piling bar provided with a
of box piles (H-section) have been intro- tongue and groove, similar to that used on
duced in North America and Europe. At timber piles, is the most commonly used.
present, a variety of high-strength sections Prestressing of concrete sheet-pile rein-
are available from different pile manufac- forcement may be an advantage, especially
turers; some piles are strengthened by in a seawater environment, as cracking of
adding a reinforcing plate to the flanges. concrete in the tension zone is thereby
In addition to the above-mentioned piles, largely eliminated and the danger of cor-
relatively low sectional modulus straight rosion of reinforcement is correspondingly
web steel sheet piles are often used in the decreased.
552 Sheet-Pile Bulkheads

Concrete sheet-piling is a relatively thin Steel sheet piles, because of their econ-
type of construction; therefore, from a dura- omy and availability in many different sec-
bility point of view, the quality of concrete tions and differing strengths and stiffnesses
is of paramount importance. Today's struc- which allow for driving practically without
tural concrete is indeed a mixture of admix- buckling and springing under the blows of
tures. Products such as superplasticizers, the driving hammer at present are the most
retarders, accelerators, air entrainers, and commonly used in waterfront construction.
others allow concrete pours in cold temper- Concrete sheet-piling is relatively main-
ature or hot weather. A denser and higher- tenance free and is gaining more recogni-
quality product is obtained. As discussed in tion among port designers and contractors.
Chapter 2, the use of silica fume in a slurry As stated earlier, the state of the art of
or powdered form can increase the strength modem concrete technology allows fabrica-
and density of concrete dramatically and tion of very cost-effective and durable con-
make it virtually impervious to chloride crete sheet piles.
penetration in a harsh marine environ- The type of sheet piling selected for a
ment. Furthermore, the use of corrosion in- particular project is usually based on struc-
hibitors may slow down the potential onset tural as well as economic aspects. It is also
of corrosion in steel reinforcing bars. greatly influenced by factors such as driv-
Last but not least, another concrete prod- ing conditions, acceptable deflection, re-
uct which is common in marine environ- quired wall watertightness, required dura-
ments, which can be demanding to say the bility, and resistance to loads due to ship
least, are miscellaneous antiwashout ad- berthing maneuvers. In some cases, where
mixtures. Unlike a regular tremie which is required and economically justified, the
poured underwater in large quantities, the bulkhead consisting of mixed sheet-piling
antiwashout admixtures allows the con- can be used.
crete to be installed under water in small
quantities. An example of their utilization
are at locations such as in interlocking sys- 6.1.2 Anchoring svstems
tem between large sheet-pile units.
Appropriate consideration of the above Bulkheads are typically distinguished as
allows the fabrication of high-quality, rela- freestanding (cantilever) or anchored. The
tively thin walled, durable components for latter is usually classified according to the
construction of sheet-pile bulkheads (e.g., type of anchor system used to support the
sheet piles, anchor blocks, anchor piles, and sheeting. The following basic types of an-
cope beams). choring system are usually used in conven-
tional sheet-pile bulkhead construction
(Fig. 6-2).
6.1.1.5 summary
In summary of this section it should be 6.1.2.1 Conventional Tie-Rod
pointed out that timber sheet piles are usu- System
ally used for temporary sheeting or where This system (Fig. 6-2a) is usually com-
the height of sheet pile wall does not exceed prised of steel tie-rods secured at an anchor
2-3 m. It may also be part of the perma- concrete wall or block (plate or deadman).
nent structure, where the top of the sheet Altematively, piles, sheet piles, and prefab-
piles is always located under the lowest ricated steel or concrete panels are also
water level and where there is no marine used. Normally, anchors are placed back far
organisms (e.g., borers, which can destroy enough from the sheeting to ensure that
wood). their passive resistance zone in the backfill
Sheet-Pile Bulkheads 553

(a) (b) (c)

-.---
~ .
~ MIN.W.L.
:z:

SHEET PILE

(d) (e) (f) (g)

GROUND OR
ROCK ANCHOR
ANCHOR
PILE

Figure 6-2. Anchored sheet-pile bulkhead structural arrangements. [From Tsinker (1983).]

does not encroach appreciably on the active To protect the anchor system from addi-
soil wedge behind the wall. tional loading caused by the weight of the
settling backfill soil, the tie-rods have to be
6.1.2.2 oouble lor MoreJ Tie-Rod placed with the negative sag or chamber
Anchorages (normally about 5-10 cm) compatible with
the expected settlement of the backfill (Fig.
These (Fig 6-2b) are usually used when 6-4a). Alternatively, they may be placed in
a single anchor is inadequate to provide protective pipes, the diameter of which is
acceptable stresses in the sheeting. The dis- somewhat larger than the expected settle-
advantage of this system is that the an- ment of the fill (Fig. 6-4b). Although in
chors tend to yield independently. This is most cases the anchors adjust themselves
attributed to the elastic deformation of the without failure by yielding of the metal, it
tie-rods and also to the deformation of the is highly desirable to equalize the initial
anchoring component (e.g., deadman, plate, tension in the tie-rods by having them
pile, or the like). Because the density of the slightly posttensioned.
backfill is practically not uniform, it is dif- In some special cases, the tie-rod can be
ficult to account for the yield of the wall- secured at the structure located in the im-
anchoring component. The latter inevitably mediate vicinity of the sheet-pile wall. A
leads to uncertain performance of a sheet- relevant example is illustrated in Figure
pile · wall. To overcome the aforementioned 6-3, where the retaining wall is part of the
problem, both tie-rods can be secured at a in-ground storage facility for coarse granu-
single anchor point, as depicted in Figure lar materials discharged from the railway
6-2c, or the sheet-pile wall can be split into track. Here, the 9.75-m-high sheet-pile wall
two parts, as illustrated in Figure 6-2d. is composed of 1.0-m-wide T-shaped rein-
554 Sheet-Pile Bulkheads

Sand

Figure 6-3. Port-of-Kiev, Ukraine. Bulk materialloading berth: 1-T-shaped reinforced con-
crete sheet piles; 2-reinforced concrete counterfort superstructure; 3-fendering; 4-anchor tie
made up from two back-to-back channels C160; 5-gravel (crushed stone) filter; 6-track for
portal crane, 100 kN capacity; 7-hinge; 8-retaining wall; 9-pier; 10-train track; 11-con-
crete piles, 0.35 X 0.35, L = 8.5 m; 12-riprap.

(a) (b)

ANCHOR
WALL

Figure 6-4. Typical installation of steel anchor tie-rod: (a) with negative sag and, (b) with
protective plastic pipe or a similar protection. [From Tsinker (1983).]

forced concrete sheet piles which is tied to sion piles are usually provided for each
the retaining wall by stiff anchors (two compression pile because, in most cases,
channels placed back to back) hinged at piles of equal length and similar cross sec-
both ends. tion can carry about twice as much load in
compression as in tension.
6.1.2.3 A-Frame Type of Pile Due to their large carrying capacity, A-
Anchorage Systems frame-type pile anchors could be used in
Tschebotarioff (1973) recommends this cases whenever the capacity of other types
type of anchorage to be used in soft com- of anchorages is not sufficient. Because of
pressible soil which cannot be conveniently its small yield, an A-framed pile anchor
removed by dredging (Fig. 6-2e.) Two ten- system is considered as relatively "rigid."
Sheet-Pile Bulkheads 555

6.1.2.4 Anchor Piles As illustrated in Figure 6-5, in hoth cases


the soil-sheet-pile wall interaction (deflec-
The hasic advantage of this type of an- tion) is rather different, a direct result of
chorage is its simplicity of construction (Fig. the different construction sequences. In the
6-2f.) When hulkhead height does not ex- case of the "hackfilled" construction proce-
ceed ahout 9 m, conventional steel or rein- dure, the construction sequence is typically
forced concrete piles can he conveniently as follows: dredging, sheet-pile-driv ing, first
used to serve the purpose. However, when stage of hackfilling to provide working area
the height of the wall exceeds 9 m, piles for installation of anchor system, construc-
with an increased ahility to carry pull-out tion of wall anchorage, and final hackfilling.
loads are used (Tsinker, 1977). During the construction process, after the
Extensive lahoratory and full-scale inves- hulkhead is hackfilled up to the level of
tigations hy Tsinker (1964) and Budin and anchor support, the middle part of the wall
Demina (1979) have indicated that anchor deflects forward and away from the fill; at
piles may have significant impact on the this stage, the still unloaded upper portion
active soil pressure distrihution and, there- deflects toward the fill. Upon completion of
fore, on sheet-pile wall performance. hackfilling, the upper part of the wall de-
flects away from the fill; naturally, at this
6.1.2.5 Soi/ and Rock Anchors stage, this part of the wall is loaded hy
active soil pressures. For details, consult
These usually consist of a stressed ten- Chapter 4 and Section 6.6 of this chapter.
don (rod or cahle) connecting a fixed anchor Construction of the "dredged" hulkhead
placed within the stahle soil or rock mass to is quite different. It usually proceeds in the
a hulkhead (see Fig. 6-2g.) These anchors following sequence: first, sheet piles are
are usually inclined downward, transmit- driven; then anchor system is constructed,
ting the vertical component of anchor force followed hy wall hackfilling, and, finally,
onto the sheet-pile wall. the soil on the harhor side of the wall is
When a tie-rod or cahle is connected to dredged. During this process the last stage
an anchor placed within the soil some yield naturally results in movement of the wall's
of the system is expected; when the anchor middle part toward the harhor and in an
tendon is fixed within the rock mass, the inward deflection of the part ahove the tie-
yield of the anchor system almost equals rod level. This type of wall deflection pro-
the elastic elongation of the tendon and is duces increased soil pressures on the wall
usually very small. Consequently, this type ahove the anchor level. Furthermore, the
of anchorage is usually considered as very soil may also start arching hetween the
"rigid." tie-rod level and the dredge line. This phe-
nomenon also affects the soil pressure dis-
trihution on a wall and is discussed in fur-
ther detail in this chapter.
6.1.3 sequence of construction
As pointed out hy Tschehotarioff (1973), the
6.2 SHEET·PILING- STRUCTURAL
construction sequence may have a pro-
nounced impact on the performance of the ANO DRIVINO ASPECTS
anchored sheet-pile wall. Essentially, it is
necessary to distinguish hetween "dredged" For detailed discussion on properties of the
or "partly dredged" hulkheads and the hasic structural materials (e.g., concrete,
"hackfilled" walls (Fig. 6-5). steel, and timher used for constructing in
556 Sheet-Pile Bulkheads

(a)

(b)

Figure 6-5. Effect of construction sequence on sheet-pile deflections. [From Tsinker (1983).]
(a) "backfilled" bulkhead and (b) "dredged" bulkhead. !-Original ground level before dredging;
2-dredged line; 3-anchor system; 4-backfill; 5a-sheet-pile deflection after the backfill is
completed to the anchor level; 5b-deflection of sheet pile before dredging; 6-final deflection.

the marine environment), the reader is re- pile must be cut off at an elevation that
ferred to Chapter 2. ensures that the sheeting will be continu-
In this section, sheet piles used for con- ously underwater to prevent decay or wood
struction of waterfront structures are dis- rot.
cussed mainly from structural and driving In tidal areas, sheet piles could be ex-
aspects point of view. tended up to the mean water level (e.g.,
halfway between "mean low" and "mean
high" water levels). In special cases, an ex-
otic kind of a very dense wood [e.g., Ekki
6.2.1 Timber Sheet Piles (for details see Chapter 2)] can be also con-
sidered. Where marine borers are present
As previously stated, timber sheeting is
usually considered for construction of low in the water, the piles used for wall con-
soil-retaining structures and where existing struction must be protected from these ma-
foundation soils are free of obstructions so rine animals one way or the other. For de-
as not to require hard driving. Also, timber tails on this subject, the reader is referred
sheet piles are used effectively for construc- to Chapter 2.
tion of waterfront structures which are used In N orth America, timber sheet piles are
for handling chemical materials that are usually made from pine, spruce, and Duglas
highly aggressive on steel or concrete. fir. Depending on structural requirements,
The obvious advantage of timber sheet the timber sheet piles are usually made
piles is their lightness and the consequent 6-30 cm thick. They can be up to 30 cm
lightness of the pile-driving equipment re- wide and up to 15 m long.
quired. If timber piles are used in perma- Depending on the thickness, timber piles
nent structures, then the top portion of the typically join together by V-joints and
Sheet-Pile Bulkheads 557

straight, or chamfer tongued joints (Fig. neous means (e.g., nails, bolts, glue, etc.),
6-6). The former is usually used when piles thus forming a laminated pile. Here, piles
are not thicker than 10 cm. Sheet piles are typically connected via a tongue-and-
thicker than 10 cm can be fabricated from groove joint by joining three planks to-
thinner planks joined together by miscella- gether in a staggered pattern.

(a) (c)

-~·f
(d)
3 Jlcm

(e)
3

(f)

5
7

4
Figure 6-6. Timber sheet piling: (a, b) typical joints; (1) v-joint, (2)
straight tongued jointing, (3) chamfer-jointed system; (c) sheet pile made
up from three planks Oaminated pile); (d) tip detail; (e) two sheet piles
jointed together bya common metal strip at the pile head; (O corner
details. 1-Timber plank; 2-bolt; 3-head metal strip 4-6 mm thick;
4-metal shoe; 5-metal dog; 6-corner pile; 7-double-tongued pile;
8-direction of pile-driving.
558 Sheet-Pile Bulkheads

Sometimes joints are not used on sheet sheet-pile wall breakwaters, and others). At
piles over 25 cm thick. In these cases, the present, steel sheet piles are produced by
piles must be installed with special care to steel mills worldwide. Major steel sheet-pile
ensure the absence of a substantial gap producers in North America are US Steel
between adjacent piles. Special piles are International Inc. and Bethlehem Steel Co.
normally installed at wall corners. These In Europe, steel sheet piles are rolled in the
piles are usually made from thick square United Kingdom (Frodingham, Larssen), in
timbers with grooves cut to accommodate Luxemburg (ARBED), and other countries.
the adjoining regular pile (Fig. 6-6). Standard sheet piles are fabricated by US
Timber sheet piles must be protected Steel International Inc.; the basic charac-
against brooming and splitting during the teristics are presented in Table 6-1.
driving process. Protection is usually ac- The strength and stiffness of basic steel
complished by reinforcing the cap or top of sheet piles can be greatly enhanced with
the pile with metal strips. These should weld-on plates. To prevent rust formation
tightly embrace the cap and any part of a on the inner contact surfaces, the reinforc-
pile exposed to the driving hammer. ing plates are welded to the sheet pile
Where required to accelerate the driving flanges by a continuous weld 5 mm (mini-
process, and foundation soils permit the mum) thick. The reinforcing plates are
technique, the timber piles may be driven welded to the pile flanges between inter-
in pairs held together by metal dogs. The locks. The material used for the reinforcing
driving is usually done with the tongue plates must be similar to that used in the
leading, for which purpose the end of each manufacture of the pile. Otherwise, corro-
pile is beveled on the tongue side so that sion will be expedited at the contacts be-
the pile is pressed against the previously tween the dissimilar material.
driven pile. In recent years, the required depth of
The tip "sharpness" depends on the type water at berthing facilities has been in-
of foundation soil; the cutting edge is usu- creased significantly due to dramatic
ally made blunter for heavier soil. Where changes in vessel size. Thus, the use of
piles are required to be driven into a hard sheet-piling with a very high section moduli
soil (e.g., gravei or a stratum of shale), the and moment of inertia is required. Piles
pile tip should be protected by a metal shoe with increased capacity are found in sec-
fabricated from steel sheets or of cast iron. tions similar to standard WF (I)-beams with
After installation, the timber sheet piles a rolled-on interlocking system. Also, mis-
become somewhat tight, a result of wood cellaneous combinations of regular strength
swell. Therefore, to provide for faster equal- and high-strength sheet piles are used. ·For
ization of water levels in front of and be- details, see Chapter 2. As stated earlier,
hind the wall, weep holes are normally where high-strength piles are required for
considered. bulkhead construction, steel pipe piles of
various diameters and wall thicknesses can
be used. These walls can be made tight by
the use of miscellaneous interlocking sys-
6.2.2 steel Sheet Piles tems, one of which is illustrated in Figure
6-7.
As previously stated, steel sheet piles are It should be pointed out that I-type and
most often used for construction of sheet- pipe steel piles are perfectly suitable for
pile bulkheads and other related marine driving into very hard soils. If further
structures (e.g., cellular-type bulkheads, strength is required for driving, these piles
Sheet-Pile Bulkheads 559

Table 6-1. Standard sheet piling

MASS SECTION MODULUS

PER
SQ m AREA DRIVINC PER m PER
PER m OF WALL A WIDTH WALL PILE

DESICNATION PROFILE ks ks mm

PZ38 84.8 186 10.8 457 2520 1150


x103 X 103 X103

PZ32 j~-tl"!
~"
""! .
(.t~s
z-·
~!hula)
83.3 156 10.6 533 2060 1100
x103 x103 x103

ft·
y

PZ27 742 1
60.3 132 1.1 457 1620
~ . W(lhua)
x101 x101 x103
y 1

PDA27 53.6 132 6.8 406 575 234


x103 x101 x101

3V4• Y ~)
(8311UD) 1 l\'
PMA22 53.6 107 6.8 498 290 144
"fPZ-+-~·
{ x103 x101 x101
TIJ2"(1SJIIIQ)

PSA28
·~·
1 y
55.5 137 7.1
x103
406 134
x103
54
x101
y318"(10mm)

PSA23
·-~·
1
y
45.7 112 5.8
x103
406 129
x101
52
x103
y

!fC
12St'8C(lt .... )
PSX32 z-:. x 65.5 156 8.4 419 129 54
y X 101 x103 x103
y

fe-s
1117'(1SIIllll)
PS32 z-a 1 59.5 156 7.6 381 102
1
y x101 x103
Tli/8"(10mm)
PS28 s-:. llc z 52.1 137 6.6 381 102 39
~ x103 x103 x103

can be reinforced with driving shoes or hard tially overstressed. This is basically at-
bites. tributed to pile fl.exibility.
There are a variety of advantages and The choice of sheet-pile profile (section),
disadvantages of this type of pile system, in addition to structural considerations and
details of which are given in Chapter 2. In economic aspects, depends on the founda-
terms of the steel sheet-pile-driving perfor- tion soil existing at the site (driving condi-
mance, an advantage of using steel sheet tions) and on acceptable wall defl.ections.
piles is that they are able to absorb local- Acceptable wall thicknesses and desired
ized overloading without being substan- chemical composition of the steel depends
560 Sheet-Pile Bulkheads

(a)

(b)

Type 1
Type 2

Type 3

Figure 6-7. Steel bulkhead constructed from steep pile piles: 1- steel pipe pile; 2-interlock-
ing system.
Sheet-Pile Bulkheads 561

on the marine environment, existing and soil-retention structures. These sheet piles,
expected contamination of both foundation used for marine application, are usually
and backfill soils by miscellaneous pollu- roade from a strong (fc1 = 30-40 MPa),
tants, available protection from corrosion as dense, high quality durable concrete. For a
well as design life span. Thus, the mini- marine pile, a fabrication rate of cement of
mum wall thickness is a function of its 360-380 kgjm 3 and a water-cement ratio
environment as well as its structural re- about 0.4 by weight are normally consid-
quirements. Furthermore, the proper selec- ered.
tion of the minimum wall thickness is espe- Further details on concrete used in a
cially important where steel sheet-piling is marine environment and the design re-
used for construction of docks speci~izing quirements as applied to the concrete ma-
in loadingjunloading of aggressive materi- rine structures, the reader is referred to
als that may degrade wall steel compo- Chapter 2 and Tsinker (1995).
nents, Freeze and thaw and the presence of sul-
The possibility of sheet-pile damage by fates, chlorides, miscellaneous organic com-
mechanical means due to berthing maneu- pounds, acids, and alkalies that are natural
vers of ships should also be taken into ac- in a marine environment are the most com-
count. Again, in some cases sheeting corn- mon agents of sheet-pile degradation. Typi-
posed of piles of different profiles may cally, the fundamental requirements for the
represent the most economica! solution to structure's longevity in a marine environ-
bulkhead construction. ment is the use ofsulfate-resisting Portland
Susceptibility to corrosion is a major cement (where required) and a concrete of
drawback of a steel sheet-piling system. low permeability.
Corrosion occurs mostly within the tidal Reinforcing steel, which is vulnerable to
range and depends on the properties of the corrosion, particularly in the zone where
water and the method of steel protection both mobile water and oxygen are present,
adopted. Apart from the corrosive qualities must be carefully designed and protected by
Of seawater, which vary widely, the pollu- any practica! means. This includes consid-
tion of the water, particularly by sewage eration of a concrete cover over the steel,
and industrial wastes, leads to widely vary- direct protection of the reinforcing bars by
ing degrees of attack. Protective measures applying protective coatings, use of corro-
used in a marine environment typically in- sion inhibitors, and last, but not least, use
clude protective coatings, reduction in al- of fiber reinforcement to control concrete
lowable stresses, use of heavier sections to cracking.
provide sacrificial thickness, and last, but For permanent structures, the cover over
not least, use of cathodic protection. Of ali the principal reinforcing bars in freshwater
of these, the use of a heavier section for and saltwater environments should not be
permanent structures is often the preferred less than 40 and 50 rom, respectively; it can
approach. Details on the corrosion phe- be reduced to 20 and 25 rom for structures
nomenon and protection of steel sheet- of limited life (e.g., temporary docks used
piling in a marine environment are given in for construction of hydroprojects and
Tsinker (1995). bridges, or docks used in military opera-
tions).

6.2.3 concrete Sheet Piles 6.2.3.1 Type of concrete Sheet


Piles
Reinforced concrete sheet piles have been
used extensively for the past 50 and more Three types of reinforced concrete sheet
years for construction of miscellaneous piles are used for construction of waterfront
562 Sheet-Pile Bulkheads

structures; they are rectangular or alterna- Concrete sheet piles are normally rein-
tively square, T-shaped, and tubular sheet forced by regular reinforcing bars or by pre-
piles. In some cases, piles of more compli- stressed reinforcement. The latter may be
cated shapes also have been used. Some comprised of high-strength deformed bars
characteristic cross sections of rectangular or high-strength wires. However, because of
concrete sheet piles used worldwide are de- their vulnerability to corrosion, prestressed
picted in Figure 6-8. wires are less desirable for use in marine

(a) (b) (c)

(d) (e)

(g) (h) (i)

Figure 6-8. Flat concrete sheet piles: (a) standard sheet pile with trapezoidal tongue and
t;
groove; (b) prestressed thin-walled sheet piles used in Hungary ( = 60 MPa); (c) double-grooved
sheet pile with timber tongue (guide) used in Poland; (d) typical double-grooved sheet pile with
semi-circular grooves designed to be filled with grout; (e) prestressed sheet pile (h = 150-300
mm) used in the U.S. coastal works; (t) prestressed sheet pile with special interlocking
arrangement provided for more accurate positioning of piles; used in Port-of-London, United
Kingdom for dock construction; (g) sheet pile with welded tongue developed in Russia for
simplification of formwork; (h) sheet pile with grooves of complicated geometry provided to
accommodate metal guide; (i) thin-walled sheet pile used in Poland for river bank protection.
1-Tongue; 2-groove; 3-regular or prestressed reinforcement; 4-stirrup; 5-timber tongue;
6-cement grout; 7-anchor bolt; 8-prestressed wire; 9-steel plate (tongue); 10-steel guide;
11-embedded steel plate; 12-steel tongue; 13-anchor bar.
Sheet-Pile Bulkheads 563

sheet piles. Naturally, sheet piles with pre- tion soil (e.g., compact gravei or similar ma-
stressed reinforcement are thinner than terial), a steel shoe is usually provided to
piles with regular reinforcement and are protect the pile tip from damage. The shoe
therefore lighter. is sloped similarly to that used in timber
An interlocking system is generally used piles. From a durability point of view, the
in rectangular concrete sheet piles and is concrete sheet piles used in permanent
traditionally accomplished by introducing a structures must have a minimum thickness
tongue-and-groove system or the installa- of 150 mm.
tion of a special interlocking member, as In general, concrete sheet piles should
shown in Figures 6-8f and 6-Sh. not be thicker than 400 mm, because thick
The tongue and groove can be of rectan- piles can meet a very hard soil resistance in
gular, trapezoidal, or semicircular con:figu- the driving process.
ration; the latter is usually provided in It must be noted, however~ that in the
lighter sheeting. Sometimes triangular past, concrete piles 100-600 mm thick have
tongues and grooves are used. The width of been used successfully for construction of
the groove is usually equal to one-third the retaining walls and breakwaters. The nor-
pile thickness, but normally not larger than mal width of the regular flat concrete sheet
700 mm. From a pile reinforcement point of pile is 500 mm. However, depending on
view, the groove is usually not deeper than driving conditions, the pile width can be
50 mm and runs continuously from the top reduced to 300-350 mm or increased up to
to the lower end on the leading side of the 1500 mm or more. Typically, flat concrete
pile. The tongue, which guides the pile piles are roade up to 15-20 m long.
against the previously driven pile, is usu- Large flat sheet piles have been used for
ally 1500-2500 mm long, and in some cases, construction of Port-of-Pensakola, Finland
where required for better tightness of the (Fig. 6-10). There, large flat sheet piles
wall, it may run from the top of the pile to a (1.22 X O. 71 m) were provided with ex-
point located about 1.0-1.5 m below the truded cylindrical voids that roade them
dredge line. Typical examples ofthe tongue- lighter and, therefore, more manageable.
and-groove arrangement and pile reinforce- After installation, the space between adja-
ment are illustrated in Figure 6-9. cent sheet piles was cement grouted. The
When installation is accomplished by grout was pumped into oversized plastic
hammer drive, the head of the flat concrete tubes installed in the interlocking system.
sheet pile (butt) is usually reduced to ac- Joints between adjacent concrete sheet
commodate a driving helmet. The driving piles must be adequately sealed to prevent
helmet must make a snug :tit with the butt granular back:fill material from being
of the pile, and it must enclose the butt as washed out. For this, in some cases, the
tightly as possible. For better protection of space between adjacent sheet piles is ce-
the pile head, timber is placed between the ment grouted; and in other instances, the
pile and driving helmet. In granular soils sheet piles are provided with an interlock-
and in some cohesive soils, better progress ing system similar to that used in steel
in pile-driving can be roade if a water jet is sheet piles (Lee, 1961). Both methods, how-
used (Tsinker, 1988). ever, are expensive and time-consuming. In
When piles are installed with help of a many cases, the sheet-pile wall back:fill is
water jet, the pile top is normally of stan- protected either by geotextile :filters which
dard width with the reinforcing bars or completely cover the interface between the
dowels extended above the concrete for an- sheeting and the back:fill or, altematively,
choring in the capping structure. lf the pile by the coarse granular material (e.g., gravei,
is expected to be driven into hard founda- rock:fill, etc.) installed behind the wall. The
564 Sheet-Pile Bulkheads

(a) (b)

Seciion
A-A
A-- -·-A

Seclion
B·B

Seciion
c-e
B-·- -·-B !•
"'
(c)

C-·-

Figure 6-9. Typical concrete flat sheet pile. [From EUA (1990).]1-Top of pile
shaped to accommodate driving helmet; 2-tongue; 3-normallongitudinal
reinforcement; 4-additional reinforcement at pile top; 5-space for sealing the
joints in the sheet-piling.

advantage of using the coarse granular As stated earlier, :flat concrete sheet piles
backfill as a filter material is a substantial due to driving resistance are typically fabri-
reduction in lateral thrust against the wall cated about 500 mm wide, which makes
due to the usually high value of interna! pile-driving quite a slow process due to
friction angle, l/J. In some cases, the granu- driving resistance. The pile installation pro-
lar :filter material in the form of a vertical cess is drastically expedited when wider
layer 0.5-1.0 ro thick is placed just behind sheet piles are used. One of the best types
the sheeting; in the case of a wide sheet of wide sheet piles used for sheet-pile bulk-
pile, the :filter roade of either granular ma- head construction is the T -shaped wide-
terial or from geotextile is installed directly :flange concrete pile. Piles of this type were
at joints between adjacent piles. occasionally used for waterfront construc-
Sheet-Pile Bulkheads 565

rience have confirmed the high degree of


constructability of bulkheads comprised of
wide-fiange T-shaped sheet piles. Their
durability and maintainability in a de-
manding busy port environment have also
been proven.
In modern construction practice, high-
strength concrete ( f~ = 30-40 MPa) and
prestressed reinforcing steel are normally
Figure 6-10. Concrete sheet pile used for dock con- used for the fabrication of relatively thin-
struction at Port-of-Pensakola, Finland. 1-Grout
injected in oversized plastic tube.
walled wide-fiange T-shaped sheet piles.
Because in T -shaped piles the thickness of
the fianges usually does not exceed 150-200
tion since the early 1930s. Examples of reg- mm, the, walls constructed from these piles
ular concrete, or prestressed T-shaped, · must be well protected from damage by
sheet piles used in waterfront construction berthing or departing ships by a carefully
are found in Europe, particularly in France, designed fender system. In some instances,
Germany, and Ukraine, where T-shaped better protection of the sheet-pile fianges
piles up to 1.5 m wide were successfully from ship impact can be obtained by in-
implemented (Gurevich, 1969). In the stalling the piles with their webs located on
United States, T-shaped prestressed con- the harbor side (Fig. 6-13).
crete piles were used in 1962 for construc- Wide-fiange T-shaped concrete sheet piles
tion of a sheet-pile bulkhead in Texas are usually joined to each other in one of
(McGregor, 1963). the following ways (Fig. 6-14)
In the past 30-35 years, T-shaped con-
crete piles have been extensively used in
(a) Mechanical connection made up from
the republics of the former Soviet U nion for structural angles [proposed by Matlin
construction of wharves in inland ports (1983)]
(Matlin, 1983). The height of the bulkheads
built there are typically between 9.0 and (b) Connection made up from the interlocking
parts of a regular flat steel sheet pile
14.5 m, with some variations where re-
welded to steel plates embedded into the
quired (Figs. 6-11 and 6-12). These 14.5- flanges ofT-shaped piles
m-high walls have been built as a double-
anchored bulkhead in which the bottom part (c) Regular joints covered by a gravei filter or
is comprised of T -shaped concrete sheet geotextile material similar to that used in
aflat concrete sheet-pile wall construction
piles and the upper part is erected from the
double-T-shaped prefabricated panels.
The extensive experience gained in Rus- The advantage of the type (c) joint is an
sia and Ukraine suggests that the wide- almost unobstructed fiow of water in and
fiange T-shaped concrete sheet piles offer a out the fill without washing out the fine soil
more efficient use of material in comparison particles while effectively relieving any un-
to the conventional fiat (rectangular) con- balanced hydrostatic water pressure behind
crete piles, help to expedite the construction the wall. However, underwater installation
process, and, in general, are more economi·· of this type of filter, as well as the lack of
cal systems, due to the smaller number of built-in guides provided by the mechanical
piles that must be handled at the fabrica- connections as are used in types (a) and (b),
tion and construction sites. Construction makes sheet-pile wall constructiori some-
experience and further dock operation expe- what difficult.
566 Sheet-Pile Bulkheads

u. •J.!

.. • ' '8-~~----~----------~~UL------~~----------~
.' ... . .. ..
.. ' .
.. . ·. : '. . :
l!. -U ... . ..

Figure 6-11. Cross section of sheet-pile bulkhead, H = 9.0 m, for construction on


inland waterways. 1-T-shaped wide-flange concrete sheet pile; 2-tie-rod with
turnbuckle; 3-prefabricated concrete anchor block; 4-concrete cope incorporating
service channel; 5-fendering.
Sheet-Pile Bulkheads 567

(a)

el . •o
. ·. .·.;
.... .
.. ·..
- 2 ·•

( a- a
a

b-b
Figure 6-12. Sheet-pile bulkhead, H = 14.5 m, for construction on inland water-
ways: (a) typical cross section and (b) prefabricated componenta. 1-T-shaped
wide-flange concrete sheet pile; 2-concrete superstructure; 3, 4-lower and upper
tie-rods, respectively, 5, 6-lower and upper concrete anchor blocks, respectively,
7, 8-lower and upper concrete copes, respectively, 9-gravel filter; 10, 11-upper
and lower sections offender system, respectively, 12-hinge.
568 Sheet-Pile Bulkheads

1----------·--

-
1 1
1
1
1
A 1 1
1 :--
1
! 1
1

1' J 1 • ~ 1·1 i 11 1 ' .


LĂ,q A,~~Ă .1 Ă 1.l
r . r
III ~~
1 1 11 1 .
IGI'd 10

llli·'
1
'11

Il
1,1
11.1
1.1
l1.1il
ŢWLJ
c~t-4 ~

A-A
Figure 6-13. Sheet-pile bulkhead anchored by raked concrete
piles: 1-T-shaped concrete sheet piles with limited penetration of
·-·
fianges into a foundation soil; 2-concrete piles 0.35 X 0.35 X 15.0 m
at 3.0 m center to center; 3-concrete cope; 4-geotextile filter
material; 5-dredge line; 6-original surface of the bank.

The concept of a mechanical connection ment and are normally installed with the
made from structural angles [type (a)] help of a water jet (Matlin, 1983; Tsinker,
proved to be effective and economica! 1988). This method was successfully used
(Matlin, 1983); it is "flexible" due to a rela- for pile installation in granular soils.
tively large gap between the angles, which Gurevich (1969) reported results of the
enables necessary adjustments to piles dur- successful use of powerful vibrators for in-
ing installation. Furthermore, after the stallation of concrete T -shaped sheet piles
backfill is placed behind the wall, a natural into soft and medium hard clays. As pointed
filter made up of coarse particles is formed out by Tsinker (1988), concrete piles can be
between angles that comprise the interlock successfully installed into the hard clay by
structure. using a high-pressure, low-water-content
Because of their large mass and exten- water jet. The method is discussed later in
sive cross-sectional area, the wide-flange this chapter (Section 6.5.2).
concrete sheet piles cannot typically be Special types ofthe wide-flange T-shaped
driven by conventional pile-driving equip- concrete and sheet piles of a composite de-
Sheet-Pile Bulkheads 569

piles 1.0-1.6 m in diameter can be effec-


(a) tively employed for construction of an-
chored bulkheads 12.5 m and higher
(Gurevich, 1969). The advantage of such
piles is that they can be installed practi-
(b) cally in any type offoundation soil. Depend-
ing on the subsoil conditions, these piles
can be installed by conventional impact-
driving equipment, by vibrators, water jet,
rock-boring equipment operating from in-
side of the pile, or a combination of some of
the above methods. In the installation pro-
cess, the soil can be removed from inside
Figure 6-14. Typical joints between T-shaped con- the pile by using the airlift technique or by
crete sheet piles: 1-T-shaped sheet pile; 2-steel special clamshells. ·
angle; 3-interlock from conventional steel sheet piles; An example of a large-diameter concrete
4-gravel filter installed at joint between adjacent
pipe pile used for construction of an an-
sheet piles. [From Matlin (1983).]
chored bulkhead at Port-of-Ventspils, Latvia
in the early sixties is depicted in Figure
sign have been developed for limited flange 6-16. This wall is 200 m long, 12.3 m high,
penetration in granular soils and hard clay and is comprised of prestressed concrete
(Figs. 6-13 and 6-15). These piles were pipe piles 1.6 m in diameter. Each 24 m
designed to penetrate the foundation soil to long pile was assembled from three sections
the depth needed to develop the required 8.0 m long. These sections were joined to-
passive soil resistance and at the same time gether by welding the metal parts previ-
safely retain the backfill material. For reli- ously embedded in each section. The piles
able retention of the backfill material, the have a wall thickness equal to 15 cm and
flanges of T-shaped piles are designed to were fabricated by the centrifuge process
penetrate the foundation soil to a nominal from high strength concrete ( f~ = 40 MPa)
depth just below the expected potential bot- in a specially designed spinning machine.
tom scour by natural or propeller-induced The centrifuge process used in pile fabri-
currents. cation achieves a very dense concrete of
Composite-type sheet pile systems com- good quality. Eventually, completed piles
prised of a steel H-pile and concrete flanges (weight 48 tonnes) were driven by the pow-
is seen in Figure 6-15. This pile was de- erful vibrators to a depth of 13.5 m into the
signed to be driven into the relatively hard foundation soil consisting of silty loam. The
clay by using conventional pile-driving space between adjacent piles were plugged
equipment. In order to prevent separation by tremie concrete.
of the concrete flanges from the H-pile dur- Similar piles were used for construction
ing the hard driving process, steel studs of a 350-m-long, 13.65-m-high, anchored
were welded to the H-pile web at 300 mm bulkhead on Volga River, Russia (Gurevich,
center to center along the pile length. The 1969). Special joints for the concrete pipe
bulkhead was designed to be completed by piles, used in this construction, were devel-
a cast-in-place concrete wall installed be- oped and successfully tested in Russia (Fig.
tween adjacent H-piles above the sheet-pile 6-17). This joint, although somewhat com-
flanges. plicated, allows for pile rotation during in-
Several projects completed in the former stallation and does not create substantial
Soviet Union have proved that concrete pipe resistance to pile-driving.
570 Sheet-Pile Bulkheads

D~
5
.....

-·.::-:----·----~.
_

,
_::::::..

j11 "'
11
11"
"'1 1 1
T 1 1 1 1 1 1
A 1 1 1 1 r- 1
1 1
1 1
'
1
-1 1
1
1
! i
1 1 11
1~
1 1 1
1 1
2 ~-

j
1
1 1 1 1
1 1 J
8
1 1 1 11 1 8

l
1
1 1 1 1 1 1
1 1 1 1 1
1 11-
-5 7 1 1 1 1
1
-~~~w--;:;·;. 1

~~ ~~~~~~~~~~i~~~
ţ- llii III ]i III MIII =tc
c/ay

c 1!1' !!!
III 1 !1 ~1!! 1!
UJ W Ul tu 1U L1l

D-D

c ,c 1.0
10

HINGE

Figure 6-15. Sheet-pile bulkhead comprised of composite-t ype sheet piles and
cast-in-situ concrete wall: 1-H-pile; 2-concrete flange; 3-concrete wall; 4-con-
crete cope; 5-timber curb; 6, 7-reinforce ment; 8-steel studs; 9-tie-rod hinge.

6.2.4 Selection of Sheet-Pile ally used for determini ng the strength of


Section the sheet-pili ng. When sheet-pili ng is used
in a permanen t structure , the potential loss
Normally , a sheet-pile section is selected on of strength due to material loss or deterio-
the hasis of strength and serviceab ility re- ration should be assumed. Protective mea-
quiremen ts in a marine environm ent. sures (e.g., allowance for corrosion, cathodic
Working strength design methods are usu- protection , use of coated reinforcem ent, etc.)
Sheet-Pile Bulkheads 571

Figure 6-16. Bulkhead constructed at Port-of-Ventspils, Latvia: 1-prestressed concrete pipe


piles 1.60 m in diameter; 2-prestressed concrete piles 40 X 48 cm; 3-tie-rod 80 mm in
diameter; 4-wooden piles; 5-longitudinal timbers to support tie-rods during construction;
6-concrete superstructure; 7-concrete plug; 8-concrete joint between pile and superstruc-
ture; 9-granular fill; 10-granular backfill; 11- original surface; 12- excavated surface.

duced due to only partial transfer of shear


slip at the pile interlocks.
The strength of concrete sheet piles is
usually governed by the permissible width
of cracks in cover over the reinforcement .
This is regulated by local and international
standards. For details on crack control in
concrete structures used in marine environ-
ment, the reader is referred to Chapter 2.
It must be pointed out that, in general,
in the selection of a sheet-pile section the
designer should always exercise a great deal
2 of judgment in order to account for some
Figure 6-17. Joint between concert pipe piles: specific andj or potentially adverse local
1-reinforcement; 2-metal protection; 3-pipe pile. conditions.

are normally considered. Loss of shear slip


at interlocks should also be considered. The 6.3 ANCHOR SYSTEMS
latter may occur where defective piles are
used or where the interlock is ruptured. The most common types of anchor systems
As pointed out by Williams and Little used to support sheet-pile walls are illus-
(1992), the bending strength of combined trated in Figure 6-2. These systems can be
sections of piling may be considerably re- divided into three basic groups, namely steel
572 Sheet-Pile Bulkheads

tie-rods secured to anchorages of miscella- during fahrication, transportation, and in-


neous designs, steel ground or rock anchors, stallation. The turnhuckles are also used for
and tension piles. tie-rod tensioning.
The :first group comprises steel anchor Round steel anchor rods are usually
rods secured to the anchoring structures; manufactured with upset threaded ends.
for example, continuous concrete wall, con- This type of construction provides for a
crete deadmen or prefahricated concrete lighter, more flexihle, and greater elonga-
plates, sheet-pile wall, piled supports of tion path anchorage than anchor rods with-
miscellaneous designs, or existing or new out upsetting.
structures adjacent to the sheet-pile wall. Although in most cases sheet-pile hulk-
Normally, the anchoring structure should heads are supported hy round tie-rods, other
he located at a certain safe precalculated systems [e.g., douhle channel (Fig. 6-3),
distance from the sheet-pile wall to ensure prestressed steel cahles, and prestressed
reliahle wall performance. concrete hars] are used. Technically, the
The second hasic group of anchors in- wall can he supported hy means of steel
clude the steeply racked tiehacks connected cahles. For this, the wall end of the cahle
to the soil or rock anchors. The first and should he fixed in a cahle jaw or regular
second types of anchors are used as single- socket and hinged at the wall structure.
level, or two-level, or multilevel systems. The other end should he furnished with a
To the third hasic group helongs raked threaded socket suitable for the cahle pre-
anchor tension piles connected directly to stressing equipment. Prestressing is re-
the sheet-pile wall via coping superstruc- quired to prevent movement that may occur
ture. due to significant elongation of the cahle
The following descrihes some applica- when exposed to the reaction force from the
tions design details, installation, and per- wall. The cahle is usually tensioned from
formance of the three hasic anchoring sys- the anchor side; the cahle ends can he re-
tem groups. versed, in which case the cahle will he ten-
sioned from the wall side. Owing to their
degree of flexihility, the cahles must he pro-
tected from exposure to hackfill hy placing
6.3.1 Anchor System them within a protective system, such as a
comprised of Tie-Rods plastic pipe or the like. In practice, hecause
and Anchorages of their sensitivity to the corrosion, steel
cahles are seldom used for construction of
Owing to its simplicity, this group of anchor permanent sheet-pile hulkheads operated in
systems is used more often than any other severe marine environments.
type of construction. The tie-rods are usu- Typically, a tie-rod is connected to the
ally placed horizontally; however, where re- sheet pile wall via a wale, which in most
quired they may he raked to a horizontal cases is installed at the wall interior (Figs.
position, sometimes quite steeply. The lat- 6-18 to 6-20); however, sometimes the wale
ter results in a vertical component of the is installed at the wall face (Fig. 6-20). In
anchor load transmitted to the sheeting that the former case, the wale is suitahly lo-
must he accounted for in the wall design. cated, as it is protected from physical dam-
In most cases, tie-rods are made from the age; however, in practice it is inaccessihle
pieces of round steel rods connected with for inspection and maintenance. In the lat-
each other hy turnhuckles or couplers. This ter case, the wale, although readily avail-
is done to make the anchor rods, especially ahle for inspection and maintenance, is vul-
those of large diameter, more manageahle nerahle to corrosion and physical damages.
Sheet-Pile Bulkheads 573

11 Sand

..--

UCfiOifll & · .&

Figure 6-18. Typical steel sheet-pile bulkhead: 1-steel sheet-piling; 2-anchor


tie-rod; 3- anchor steel sheet-pile wall; 4-turnbuckle; 5-double-channel wale at
sheeting; 6-double-channel wale at anchor sheet-pile wall; 7-cap (curb); 8-original
bank surface; 9-dredge line; 10-fasteners; 11-granular back:fill (sand).

The outside wale is typically used in tie-rod distress on the adjacent rods and the
sheet-pile wall composed from fl.at concrete wale system.
sheet piles. Here, it is usually embedded In most cases, wales are constructed from
into the concrete superstructure (Fig. 6-21). two steel channels placed symmetrically
The wale is a very important part of a back to back so that the anchor rod can
bulkhead. Its failure will inevitably cause rotate by an anticipated amount without
failure of at least part of a wall. Therefore, interference with the channels. The connec-
a wale design, fabrication, and installation tion between anchor rods and the wale must
must be given proper attention. The wale be carefully designed to prevent local over-
also stiffens and facilitates the alignment of stressing or buckling. Splices in the wale
the sheet-piling. structure are usually placed where the
For permanent wall construction, a mini- bending stress is at a minimum. Welded or
mum material thickness of 10 mm is nor- bolted splices must be roade in accordance
mally used for the wale. Furthermore, it is with the best steel construction and weld-
a good practice to consider some corrosion ing practices. Furthermore, due to the dan-
allowance as an addition to the conven- ger of corrosion, the stressed welds must be
tional safety factor, which is usually 2.0 for roade at least 2 mm thicker than would be
normal load combinations and 1.5 for the structurally required. Material used in the
extreme loading case. Additionally, it is also manufacture of the wales and wale fasten-
advisable to consider the potential effect of ers must not be dissimilar. Furthermore,
574 Sheet-Pile Bulkheads

(a) (b)

Figure 6-19. Examples of a steel sheet-pile wall construction. (a) Steel sheet-pile bulkhead in
process of construction. (Courtesy ARBED Inc.) (b) Double steel sheet pile wall pier.

the material must be similar to the sheet- 6-12). This ensures uniform tensioning of
pile material. Last, but not least, the wale all tie-rods during the backfilling process.
longevity can be extended by applying a The cope beam is usually cast-in-place after
coat of protective paint. For this, the wales the backfill is laid and densified.
must be thoroughly cleaned, primed with At sites with considerable tidal range or
red-lead (two coats are recommended), then substantial seasonal fl.uctuation in water
painted with one coat of coal-tar or epoxy levels, the bending moments in the sheet-
coating. If damaged during installation, the piling and wall penetration can be reduced
coating must be repaired. by the installation of anchor rods at two or
In the case of T-shaped wide-fl.ange rein- more levels (Figs. 6- 2b, 6- 2c, and 6- 2d).
forced concrete sheet-pile walls, the wale is Double or multilevel anchor systems are
not used; each pile is anchored by the dedi- also favored where wall embedment is lim-
cated tie-rod that is secured at hinges em- ited by the foundation condition; hence,
bedded into each sheet pile (Figs. 6-11 and sufficient passive pressure cannot be mobi-
Sheet-Pile Bulkheads 575

2 3

(8) DOUBLE INSIDE CHANNEL WALES

(b) DOUBLE OUTSIDE CHANNEL WALES

Figure 6-20. Wale systems: (a) inside wale system and (b) outside wale system.
1-Steel sheet-piling, 2-double-channel wale; 3-tie-rod; 4-spacer; 5-fasteners;
6-splice.

lized. In this case, one of the anchors can be wall (Fig. 6-19b), both sheet-pile walls pro-
placed underwater at a very low wall level, vide anchorage for each other.
as shown in Figure 6-22. As stated earlier, sometimes the tie-rods
The tie-rods transfer the wall reaction can also be secured at some structures
load to the anchorages. As stated earlier, located in close proximity to the wall (Fig.
these can be constructed in the fonn of a 6-3).
continuous wall comprised of a cast-in-place In the case of anchor plates, continuous
concrete wall or prefabricated concrete walls, and similar anchorages, the wall re-
plates (blocks) (Figs. 6-2, 6-11, 6-12, and action force is absorbed as passive earth
6-21), a continuous sheet pile wall that acts pressure developing in front of these struc-
as a cantilever (Figs. 6-18 and 6-19a), an tures. Hence, anchors of this type, along
individual steel or concrete pile (Fig. 6-16), with some elastic defonnation of the tie-rod
an A-frame-type arrangement fonned offor- itself, allow for some limited yield of the
ward and backward raking bearing, and structure at the top of the wall. The dis-
tension piles with a concrete cap placed at tance from the anchor plates or wall to the
the head of these piles (Figs. 6-2 and 6-3). sheet-pile wall is detennined as the mini-
In the case of a continuous double sheet-pile mum required distance for which the
576 Sheet-Pile Bulkheads

Figure 6-21. Port-of-St. Peterburg, Russia. Concrete sheet-pile bulkhead:


1-prestressed concrete cope sheet pile 450 X 500 mm; 2-concrete wall;
3-double-channel waler; 4-tie-rod, 75 mm in diameter with two turnbuckles;
5-prefabricated concrete anchor block; 6-riprap; 7-wooden pile to support
tie-rods during construction; 8-granular filter; 9-excavation line; 10-dredge
line; 11-granular backfill.

maximum available passive pressure is and the potentially very small yield of the
achievable and the overall stability of the pile system. This may affect the vertical soil
structure is obtained. Usually, the overall pressure distribution along the wall.
wall stability depends on the soil shear re- [Details are provided further in this chapter
sistance along a critical rapture line that (Section 6.6) and in Chapter 4.]
passes through the bottom of the anchor The A-frame-type anchor system is usu-
plate or wall and the toe of the wall. Fur- ally used in congested areas where it is
ther discussion on wall stability and opti- inconvenient or uneconomical to excavate
mum distance between the sheet-pile wall large areas of an existing paved apron, or
and the anchor system is offered later in where a new wall must be constructed in
this chapter. close proximity to existing structures sensi-
As a structural member, the tie-rod- tive to close excavation (e.g., existing walls,
anchor plate (wall) system is generally re- warehouses, storages, and others). In the
garded as rather flexible. In contrast, an presence of weak foundation soils the an-
anchor system made up from A-frame pile chor piles may enhance the overall wall
arrangement depends only on pile bearing stability.
and tension-carrying capacity; therefore, the
frame can be placed as close to the sheet-pile
wall as practical. However, from the safety
of installation viewpoint, a minimum dis- 6.3.2 Anchor svstem
tance of 1 m between end of the sheet piling comprised of Raked Piles
and the compressed pile should be consid-
ered (Fig. 6-23). A-frame-type anchorages Steel and concrete anchor piles are often
are generally regarded as rather stiff owing used in sheet-pile wall construction where
to the relatively short length of the tie-rods the wall height does not exceed 8.0-9.0 m.
Sheet-Pile Bulkheads 577

F
6
2 3·

11

Figure 6-22. Low penetration, double-anchored sheet-


pile wall: 1-steel sheet-piling; 2-wale; 3-tie-rod;
4-concrete anchor block; 5-hinge; 6-fender; 7-natural
surface; 8-dredge line; 9-bedrock surface; 10-granular
fill; 11-soft clay.

The main advantage of this type of con- For better wall performance, these joints
struction is its exceptional simplicity and are usually provided with shear keys.
minimum space requirements for wall con- Concrete pile dowels or dowels welded to
struction. It also does not require extensive steel sheet piles and steel anchor piles must
earthwork. The efficiency of sheet-pile bulk- have sufficient embedment into the cap to
heads anchored with raked piles is more ensure reliable interaction between both
pronounced where piles with local enlarge- sheeting and anchor piles. Where substan-
ments are used (Tsinker, 1977). Concrete tial wall movements are expected, in order
anchor piles are typically installed with the to avoid a large value of"fixed-end moment",
slope 3V: 1H. It has been found, however, pins or pinlike joints between sheeting and
that steel anchor piles, driven with a 1 : 1 anchor piles are provided. This can be ac-
batter, are practical and very economica!. complished by a heavy rebar reliably se-
This is especially valid when the wall is cured within both the concrete cap and an-
required to be built in close proximity to chor pile, by single- or double-pin joints, or
existing structures or where the depth of by other methods. Examples are illustrated
pile-driving is limited by underlying in Figure 6- 25. Where the structure is con-
bedrock. structed on compressible foundation soils of
Sheeting and anchor piles are normally substantial depth, or if nondensified backfill
connected with each other via concrete cap is to be placed to the full depth behind the
that embraces both sheet piles and anchor wall, the double-pin connection between
piles. Hence, both sheeting and anchor piles sheeting and anchor piles can be used as an
are fixed into the cap. An example of sheet- additional precaution against joint over-
pile wall anchored with concrete piles is stressing (Fig. 6-25d).
depicted in Figures 6-13 and 6-24. Regular single-pin connections (Fig.
Reinforced concrete caps generally have 6- 25e) are typically used where uncontrol-
expansion joints about every 15 to 20 m. lable and comparatively large stresses due
578 Sheet-Pile Bulkheads

Figure.6-23. Sheet-pile wall supported by A-frame pile anchorage


system: 1-steel sheet-piling; 2-wale; 3-tie-rod; 4-forward
(bearing) and backward (tension) concrete piles; 5-concrete cope
beam; 7-dredge line; 8-natural surface.

to residual settlement of the waii or defl.ec- of strain cracks to the required minimum,
tions of the sheet pile are expected. One they must be either prestressed or heavily
should be aware, however, that a hinged reinforced.
connection is much more costly than regu- A special type of concrete anchor pile
lar fixed joint. with an enhanced puii-out capacity has been
The connection between the sheeting and developed by Tsinker (1977). This type of
anchor piles is best accomplished when the pile was successfully used in several major
anchor piles are driven after the sheeting. marine projects built in Ukraine (Figs. 6- 26
Here, the sheet-piling can provide an effec- and 6-27). These piles have a substantial
tive guide for driving the anchor piles. enlargement at the lower part, which in-
Drawbacks of using anchor piles are in creases the pile's pull-out capacity dramati-
their required substantial length to resist cally. The puii-out load is taken by the con-
puii-out load and their exposure to the ventional high-strength tendon or by the
thrust from the backfill soil. Under certain assembly made up from regular rebars
conditions, the backfill soil may hang on the brought together in the required quantity
piles, thus inducing a substantial bending and installed at the centerline of the pile's
moment. shaft. The load-carrying tendon is provided
Furthermore, in the case of reinforced with an anticorrosion protection and is an-
concrete piles, in order to reduce the width chored by very simple means within both
Sheet-Pile Bulkheads 579

[ EL . 70M

M. W. L . EL . 6.8M .

N . W. L . EL. 4.2 M.

E L . 0.0 M .

..·.; .·:··

Figure 6-24. Cargo dock at Chernobil, Ukraine. Typical cross section and
pile installation process.
580 Sheet-Pile Bulkheads

(8) (b)

(O) (d)

Figure 6-25. Typical joints between sheet-piling and anchor


piles: (a) rigid joint; (b) hinged joint; (c) rigid joint between steel
piles; (d) double-pin joint; (e) single-pin joint. !-Concrete sheet-
ing; 2-concrete anchor pile; 3-concrete capping; 4-conven-
tional reinforcement; 5-steel rod with welded anchor dowels;
6-elastic material (e.g., neoprene); 7-steel sheet pile; 8-1-piles;
9-H-anchor piles; 10-burned opening; 11-pin; 12-joint
plates; 13-spacer; 14-steel plate welded to sheet pile;
15-dowels welded to sheeting and anchor piles.

the pile's enlarged part and the concrete component, a steel tendon, is insulated and
cap. therefore protected from corrosion, the pile's
The shaft itself is reinforced slightly, just application itself is not limited by the pres-
to be strong enough to protect the pile dur- ence of strain cracks. This pile was de-
ing the transportation and installation pro- signed for installation by the water jet
cesses. Because the pile's load-carrying method (Tsinker, 1988). In some cases,
Sheet-Pil e Bulkhead s 581

{a )

(b )

Figure 6-26. "Flexible" anchor pile with enlargem ent at the end: (a) typical
geometry and (b) reinforce ment.
582 Sheet-Pile Bulkheads

and anchor piles, and on the effects of an-


chor piles on soil pressure distribution along
sheet piles. The following basic conclusions
have been drawn from the above experi-
ments:

(i) The pile's ability to carry pull-out loads


is directly proportional to the size of the
widening(s) placed on its shaft. This is
largely attributed to the fact that for a
given displacement, the unit bearing
pressures does not depend on the size of
the pile widening(s).
(ii) The vibration of a pile jetted into the
sandy soil, or any other dynamic effects,
can substantially increase the pile's pull-
out capacity due to soil densification in a
zone previously loosened by the water jet.
(iii) The horizontal component of an anchor
force can considerably increase the carry-
ing capacity of the pile's shaft; however,
it has no practica! effect on the pile
widening's bearing capacity.
Note: Where the angle between the
resulting anchor force and pile axis is
equal to, or exceeds, about 21o-22°, the
pile's vertical displacement practically
ceases and only horizontal movement
Figure 6-27. "Flexible" anchor pile in the process of
installation using the water jet technique.
(shaft rotation) is observed.
(iv) Where soil conditions prohibit jetting of
piles with a single large enlargement,
however, it was simply placed with the two or more smaller widenings placed
lower end on a riverbed and with its other suf:ficiently apart from each other can be
end anchored to the sheet piling. used instead of a single large one, with a
The pile's performance was extensively similar effect.
tested (Tsink.er, 1972, 1977). Comprehen- (v) The mode of pile failure is mainly con-
sive physical scale models (scales 1 : 10 and trolled by the relative depth of the widen-
1 : 4) and full-scale tests were conducted on ing placed on a pile shaft and the relative
the piles jetted into predominantly medium density of the soil. In the case of shallow
sand. The investigation determined the anchors, the surface of soil rapture ex-
load carving capability of this type of pile tends from the widening to the soil sur-
(within limits of the study) and the load face; the body of displaced soil has the
distribution between the pile's shaft and form of the truncated cone. For deeply
located widenings, the surface of rapture
the enlargement at the pile end.
is of a local nature; it does not extend to
The field experiments not only included the soil surface. Factors affecting the
tests on an individual pile of different con- pile's movement to failure under sus-
figurations but also the performance of the tained loads are approximately constant
piles included into two real sheet-pile bulk- when widenings are located at a greater
heads. The latter provided an insight into depth. A pile's relative movement to fail-
the interaction between the sheet-pile wall ure increases sharply at shallow depths.
Sheet-Pile Bulkheads 583

Experimental load displacement curves trol the displacements of sheet-pile walls.


indicate that when widenings are located at Further, they can be technically and eco-
a greater depth, the pile's ultimate load can nomically feasible where a construction site
be achieved as the result of substantial is too narrow for installation of conven-
vertical displacements. As sheet-pile bulk- tional types of anchors. The disadvantages
heads are usually designed for limited hori- of these anchors might include their high
zontal displacements, the anchor pile dis- installation costs relative to conventional
placements and, therefore, loading should systems and, to some extent, the lack of a
be liniited accordingly. positive assurance as to their long-term
The acceptable value of pile vertical dis- performance.
placement can be increased by allowing for As a means of anchorage for sheet-pile
more movement in the sheet-pile wall. For bulkheads operated in severe marine envi-
this, the sheet-piling can be driven with an ronments, these anchors received relatively
inboard batter that generally should not little attention from civil engineers. Al-
exceed about 100 : 1. though there have been few problems re-
The other types of piles with enhanced ported, some engineers still lack the con-
pull-out capacity [e.g., screw piles (for de- fidence to accept ground anchors for
tails, consult Chapter 7), "spin-fin" piles permanent works. At present, ground an-
(Gilman and Nottingam, 1991; Pierce and chors in marine application are basically
Martinson, 1992), and the like] can also be used to support distressed existing bulk-
used as anchors for sheet-pile bulkheads. heads and other structures or where there
The discussion on the design of piles with is a need for dredging the sea bottom in
enhanced pull-out capacity is given later in front of the existing structure. In this case,
this chapter. ground anchors may be used as a second
Essentially, a sheet-pile bulkhead an- row in the wall anchor system. Multirow
chored by raked piles is a frame-type struc- anchorages of this type have also been used
ture. Therefore, its structural components in marine applications (e.g., for construction
(e.g., sheeting and anchor piles) must be of navigation locks and dry docks).
designed accordingly. Sheet-piling and an- By definition, the ground anchor is a sys-
chor piles, like ordinary piles, must support tem that is capable of transmitting an ap-
axial compressive and tensile loads and re- plied tensile load to a load-bearing stratum.
sist bending moments induced by soil thrust An ultimate function of the ground anchor
and wall displacements. Sheet piling must used to support sheet-pile bulkheads (as
have an adequate penetration into the foun- well as any other structures, e.g., dams,
dation soil to resist compressive loads of all navigation locks, dry docks, etc.) is to pro-
kinds while developing sufficient passive vide the required support to these struc-
pressures. tures by deriving its restraint from a suit-
able stratum (e.g., dense sand and gravei,
dense clay, or rock) located outside of a zone
where anchor stability may be affected if
6.3.3 cround <Rock) Anchors the structure were not adequately re-
strained.
This type of wall anchorage represents a Anchors are designed to have a useful
structural system that uses an anchor in life of no longer than the life of the struc-
the ground to secure a tendon. It is usually ture they are designed to support. Anchors
referred to as a ground or rock anchor. The of a temporary nature with a useful life of
basic advantage of ground (rock) anchors is no longer than about 2 years are usually
that they can effectively restrain and con- installed with no protection against deterio-
584 Sheet-Pile Bulkheads

ration. However, ifthe strata through which the anchor force and the strength of the
the anchor is installed contain materials, surrounding ground. The free, or service,
which could prove detrimental in a short length is part of the anchor that is not
period of time, some protection is required. bonded to the surrounding strata. Its func-
Permanent anchors are those intended for tion is to transfer the tensile load from the
use to ensure the stability of the structure wall to the bonded part of anchor.
throughout its designed service life. In this The head assembly incorporates the
instance, anchors must be protected against stressing length of the anchor bar (or ten-
deterioration of any kind. don). It is usually designed to retain the
In most cases, ground anchors used in tensile force applied to the anchor bar and
marine applications are constructed from to provide an adequate support to the struc-
high-tensile steel bars and occasionally from ture. This head assembly must be suitably
steel strands. Typically, an anchor system protected from physical damage and deteri-
comprises an opening in the ground into oration.
which the anchor rod or tendon is placed An advantage of the ground anchor sys-
and fixed. A standard ground anchor con- tem is that it may control movements of an
sists of three main components (Fig. 6-28): anchored structure by prestressing the an-
chors subsequent to construction. ·
(i) A fixed length which sometimes is re-
ferred to as the bond length (la) 6.3.3.1 Types of Ground Anchors
(ii) Free or service length (l b) Most ground anchors in use today are of
(iii) Head assembly soil-replacement type which basically in-
volves drilling a hole and grouting a steel
The fixed length is part of an anchor bonded bar or strand in place. In order to produce a
into the anchoring stratum from which the tieback of required capacity and perfor-
tensile strength is derived. The length of mance, it is often necessary to drill through
the fixed part is selected to ensure wall severa! strata that may include man-made
stability. It depends on the magnitude of granular back:fill as well as soft, weak, and

(8) (b) (C)

Figure 6-28. Typical ground anchor systems: (a) cylinder filled with grout; (b) cylinder
enlarged by pressure-controlled grout; (c) cylinder mechanically enlarged at one or more loca-
tions along anchor length. [Note: la = fixed anchor length, lb = stressed (free) anchor length.]
1-steel bar on strand; 2-secondary grout; 3-primary grout; 4-enlarged zone formed by
pressure (primary) grout; 5-primary grout :filled into mechanically enlarged void.
Sheet-Pile Bulkheads 585

compressible soils (e.g., compressible ma- done by air, water, and mud introduced to
rine clay, silty sand, etc.) and into compe- the drill hole. Because all cleaning out
tent soils (e.g., as gravels, dense sands, stiff methods leave small amounts of residue at
clays, or weathered or sound rocks). The the end of the anchor hole, it is a good
anchor fixed length, in general, must be practice to provide some extra length at the
formed in competent soils. end of the anchor hole (usually 0.3-0.5 m)
There are three basic types of ground to collect, or house, the fallen debris.
anchor system (Fig. 6-28). The first type Ai3 mentioned earlier, the anchor tendons
comprises a drill hole filled with cement for anchoring sheet-pile bulkheads operat-
grout or other fixing agents, depending on ing in marine environments are usually
the load to be mobilized. The second type is made from solid steel bars. To keep the
a predrilled cylinder that is expanded by anchor bars near the center of the hole,
the pressure grout to form an enlarged zone they are installed in the hole with spacers
filled with grout, whlch acts as the anchor- that are securely fixed along the length of
age. The third type is a cylinder mechani- the tendon. Spacers must be of strong and
cally enlarged at one or more locations along robust construction to withstand handling
its length; this is done by means of a special and installation stresses. There is little
cutting device. guidance in the literature on the distance
Because of the great diversity of ground between spacers. Hanna (1982) suggested
conditions to which the tiebacks are to be that they may be installed at 2 m centers or
anchored, a large number of systems, each less in the fixed anchor zone and about
one using different anchor rods or tendons, double that along the free anchor length.
drilling methods, and grout control proce- Prior to installation, the complete anchor
dures, have been developed and used. assembly must be carefully inspected: a
Ground anchor construction starts with check of dimensions, adequacy of spacers,
drilling of a hole into which the anchor is to damage to the protective system installed
be installed. The following factors affect the along both the free and fixed anchor lengths,
method of anchor hole drilling: physical damages, presence of corrosion,
and the grout pipes which may be located
(i) The type of ground to be drilled through within or outside of the anchor bar.
and type of anchor to be used An installation operation is relatively
(ii) The hole diameter, inclination, toler- simple and usually proceeds at a slow and
ances, and shape of the fixed-bond zone steady rate. If the tendons are too long,
(iii) The method of drilling then it is possible to assemble shorter pieces
joined together by couplings in the process
In general, the type of ground (e.g., soil or of installation. Furthermore, if the tendon
rock) controls .the choice of drilling system or its parts are very long and too heavy to
and the drilling and anchor installation be handled manually, then appropriate lift-
method to be used. Where the borehole can ing equipment may be required to maneu-
collapse during drilling or grouting, a cas- ver the bar into the hole. In general, man-
ing is used. For example, in an alluvial- handling the bars may result in severe
based stratum, it is common practice to bending. This may damage the bar protec-
case the hole over its full length and slowly tive system and must be avoided. It is good
withdraw the casing as the anchor hole is practice to use guides for bar installation.
grouted. The ground anchors should be laid out
Regardless of the ground type or drilling and installed in a way to ensure that the
method used, it is essential to efficiently depth of soil c<iver above them is equal or
remove the drilling spoil. This is typically greater than 5.0 m and is not less than 1.5
586 Sheet-Pile Bulkheads

m where anchors are installed in competent and final stressing, the annulus around the
sound rock. Where fractured or weathered free anchor length is grouted to provide
rock is encountered, the material is as- additional protection against corrosion (the
sumed to be a granular soil. second stage).
In a multiple-anchor system, the mini- In the first stage, the grout is normally
mum spacing between anchors in a line and extended about 1.5-2.0 m beyond the top of
vertically should be equal to 4b, where b is the fixed anchor zone to ensure there is no
the anchor diameter (Fig. 6-29). The inter- weak areas within the fixed anchor end.
action between anchors must also be con- However, single-stage grouting is also used.
sidered if they are more closely spaced, or if The disadvantage of this method is that in
the spacing is smaller than one-fifth of the the process of anchor loading, ali of the load
anchor length (Canadian Geotechnical may not be transmitted to the fixed anchor
Society, 1985). zone because of a certain amount of friction
Mter installation, the anchor bar is along the free anchor length.
grouted. The purpose of grouting is to fix In ali cases, however, it is good practice
the bar to the load-bearing ground. lts sec- to ensure that each grout injection stage is
ond purpose is to provide protection against a continuous operation. During installation
corrosion and to fill the voids in the ground. of pressure grout in the fixed anchor zone,
The grout is usually installed in two stages. it is technically possible to generate heavy
At stage one, the anchor bar is fixed in the pressure in the ground to form an enlarged
ground by grout, and after curing, testing zone. However, it is necessary to control
these pressures to avoid potential distress
to adjacent anchors, structures, or services.
Where quick holding power from the anchor
system is required, quick-set cementations
~s •
mixes or resins are used as the fixing mate-
H
rials. Resin is manufactured in the form of
two-component cartridges. These cartridges
~l.S b
are installed into the anchor hole and the
anchor bar is spun into the hole. In this
process, both components are mixed to-
gether. Typically, within a few minutes the
' STAtlliTY 10 11 CHECKlD resin sets to form a hard material.
The resin system is generally used where
anchor bars are relatively short. Manufac-
turers' specifications concerning resin in-
stallation and curing process must be
strictly followed. Manufacturers' advice is
also available and is advisable.
Protection of the anchor bar from corro-
sion is critica! for the bulkhead's longevity.
As in most cases of steel structures con-
structed in a marine environment, corro-
sion results from chemical, electrochemical,
PLAN and microbiological causes. In practice, de-
Figure 6-29. Requirements for minimum spacing
pending on the ground and groundwater
and depth for ground anchors. [From Canadian aggressiveness, specified protection mea-
Geotechnical Society (1985).] sures typically include a galvanized or
Sheet-Pile Bulkheads 587

stainless-steel cap over the anchor head, a At present, the acceptable limit for mini-
greased and plastic sheathed free-tendon mum crack width in cementation grouts in
length, and a tendon-bond length inside a an aggressive groundwater environment is
corrugated plastic duct. Three typical prac- 0.9mm.
tica! examples are illustrated in Figure In general, the cement grout and steel
6-30. are inherently incompatible under tension

Figure 6-30. Ground anchor details: (a) Anchor with-


out corrosion protection; (b) anchor with single corrosion
protection system; (c) anchor with double corrosion pro-
tection system. 1-Thread bar; 2-grout; 3-anchor
plate; 4-anchor nut; 5-drill hole; 6-smooth sheath-
ing; 7-tape; 8-sealing cap; 9-plastic nut; 10-mastic
corrosion inhibitor; 11-gasket; 12-grout sleeve;
13-corrugated sheathing; 14-end cap; 15-weld;
16-steel tube.
588 Sheet-Pile Bulkheads

because the steel is ductible, whereas ce- Where anchors are installed in soils such
ment grout is brittle and crack sensitive. as clays, known to be susceptible to creep, a
Littlejohn (1987, 1992) suggested that when 3-6-month anchor behavior monitoring pro-
smooth bars, wires, or strand tendons are gram should be considered. Through this
tensioned, cracks 1-2 mm wide, about program an engineer can observe the stabi-
50-100 mm apart, may occur. Therefore, lizing trend in a structure's behavior, then
under the above conditions, grout should conclude whether or not the structure is
not be considered a part of the corrosion safe. A conclusion can be drawn from the
protective system in an aggressive ground- monitoring program. The result may be to
water environment. Here, other protective extend the observation program, or the
means (e.g., resin grout) or plastic sheath- structure must be reinforced by the instal-
ing may be relied on. Furthermore, epoxy lation of additional anchors.
coated bars may also be a good, however not In general, the primary purpose of an-
perfect, answer to the question of corrosion chor stressing is to load each completed
protection to anchorages (Bruce et al., 1991). anchor to a known load level (in a stand-
At present, the double corrosion protec- ard manner) and to record the load-
tion system that includes both grouting and deformation behavior of that anchor. The
the plastic sheathing similar to one de- information obtained helps the engineer
picted in Figure 6-30 is generally regarded make the right decision about the anchor's
as the most reliable. reliability.
All of complete and stressed ground an- Where the anchored wall is constructed
in close proximity to existing structure(s)
chors are routinely load tested. The purpose
that is (are) sensitive to movements of the
of these tests is to demonstrate the anchor-
surrounding ground, a prestressing anchor
age's short-term ability to support a proof
is especially vital; it takes the slack out of
load greater than the design working load
the system and causes the system to behave
and the efficiency of load transfer to the
elastically. Prestressing can reduce the
fixed anchor zone. Proofloading is generally elastic deformation of the tendons to a
accepted within a range of 1.25-1.5 times minimum. For this reason, tiebacks are pre-
the design working load. The anchor testing stressed, preferably to a value approxi-
procedure usually specifies test load incre- mately equal to that expected from the de-
ments and duration of load applications. sign load. However, some ground movement
For example, if the ground is cohesive and may occur due to inadequately prestressed
liable to creep, the specified test duration tendons andjor from the "creep" of ground
may be increased from 2 to 24 hours. The anchors due to the rheological properties of
test time is usually extended as appropriate soil in which these anchors are embedded.
when anchorage's capacity fails to demon- In the latter case, the tendons should be
strate a stabilizing trend within tolerable retensioned to maintain movement of the
limits. ground within allowable boundaries. Budin
Similar to conventional tieback anchor- and Gurinsky (1985) suggested that the
ages for permanent works, safety factors of prestressing force applied to anchors can
2.0 for normalload condition and 1.5 for the be maintained with the help of a special
extreme loading are recommended. This, spring compensator. According to Budin and
however, may be reduced significantly at Gurinsky, this device was extensively tested
the engineer's discretion for temporary in- under laboratory and field conditions and
stallations. Safety factors applied to the proved tobe reliable for long-term use.
ground anchorages are discussed by The scope of this work is limited to a
Mitchell (1987). general discussion on the subject matter;
Sheet-Pile Bulkheads 589

for more details, the interested reader is


referred to works by Hanna (1982), Mc-
Mahon (1986), Juran and Elias (1991) and +5.10
many other related publications. Fiii
~
An example of high diaphragm wall built
Dlaphtegm·
in Rouen, France is illustrated in Figure w•ll
thlckneas
6-31. This wall, 16.8 m high, 1.33 m thick 1,33m

and 460 m long, is supported by ground Gt•v•lly ş• ~j:)


anchors placed within dense gravelly sand. , . ..-~ ~ dr•tJing

Holes for anchors are drilled at 1.75H: IV -


• --s.o;-10.o
L - ----------11: ." 7~·~~
1"l"J'l.-~
batter. Length of the anchored part is about "'t§!
~
12.5 m and stressing length is about 15.0 m. w.••- c/Nllt 'l':
The anchor design capacity is 150 tonnes. ---~----------:1!
s_ _,. ---- ,: ..L-
-18.0
Several characteristic examples illustrat-
ing the use of post-tensioned tiebacks used Figure 6-31. Quay wall constructed at Rouen on the
in marine wall construction, repair and Seine River, France; typical cross section.
modernization are given in McMahon (1986)
and Xanthakos (1979).
ingly. With a close space between the cap
and the wale, the sheeting alignment is
accomplished mostly with the wale.
6.4 WALL CAPPING When the cap is designed to meet a cer-
tain operational requirement (e.g., to pro-
vide support for vertical or horizontalloads),
Depending on functional requirements, the its strength must be revised accordingly.
wall capping is usually constructed in the In structural analysis, the capping beam,
form of a light steel cap, concrete beam, a particularly that made from reinforced con-
combination of both, or in a form of a con- crete, is · treated as a beam on an elastic
crete retaining structure extended above the foundation. In this analysis, the spring
top of the sheet-piling. The wall capping characteristics depend on the pile stiffness
usually serves either as an upper closure and length. In general, for heavy capping
for the sheet-piling and the base for the beams used on bulkheads serving seagoing
safety curb, or as a structural member that vessels, the spring characteristic in the hor-
is designed to distribute loads imposed by izontal plane is assumed equal to K h = 25
berthing vessels, mooring forces, andjor MNjm 3 , whereas in the vertical plane, it is
live loads. In a great many cases it is also assumed equal to about lOKh or Kv = 250
used as a foundation for bits and bollards MNjm 3 • Tobe considered as a beam onan
and as a load-distributing member between elastic foundation, the capping beam must
anchors. be properly reinforced and anchored to
Typically, the light steel cap is very flex- sheet-piling.
ible and, therefore, does not have sufficient When a capping beam is serving as a
strength to resist forces arising from wall foundation for bollards or other mooring
operations. However, it is strong enough to accessories, the wale is usually provided
absorb forces arising from the sheet-pile's with a dedicated anchor to absorb mooring
top alignment during the cap installation. loads without transmitting them directly to
The steel cap is usually installed after the sheeting. Figure 6-32 illustrates some
the back:fill is completed to the wale level standard capping structures that are typi-
and the top of the wall has deflected accord- cally used in bulkhead construction. The
590 Sheet-Pile Bulkheads

(a) (b) (c) (d)

. . : ·:
........Ji;{j '•.
[11''<...
-1

(e) (f)

10

Figure 6-32. Typical sheet-pile wall capping.

steel cap depicted in Figure 6-32a is nor- cap must be smoothed and protected by
mally used for the alignment of sheet-piling steel to prevent the hawsers, which run
and also as a base for the safety curb. It over the cap, from being damaged.
usually possesses no significant structural If granular backfill can be washed out
strength. Caps illustrated in Figures 6-32b from between the steel cap and the sheet-
and 6-32c, due to their considerable stiff- piling, then the gap between them should
ness, are able to distribute live loads along be filled with concrete. In doing so, the
a substantial length of bulkhead. They are vertical flange of the steel cap should be
also designed to absorb impact forces from embedded to an adequate depth into the
small service boats. Where required, the concrete, and secured at its location by
cap may be reinforced further by adding means of studs welded to the sheet piles.
additional stiffness to the cap structure. Alternatively, the top portion of the sheet-
If the capping beam is needed to support piling may be protected by a granular filter
an outboard raii of a portal crane, then it is material of well-balanced composition.
usually done as illustrated in Figure 6-32d. The concrete capping structure may serve
In this case, horizontal and vertical crane as the wall extension (Fig. 6-32e) or wale
wheel loads must be considered in the cap- (Fig. 6-320. In the former case, it is usually
ping and sheet-piling designs. It is good cast in place after the wall is backfilled up
practice to avoid sharp edges in a cap beam; to the bottom of the capping structure.
ali edges are to be chamfered when re- Therefore, most of the soil's lateral thrust is
quired. Furthermore, the top surface of the absorbed by the bulkhead via its wale-
Sheet-Pile Bulkheads 591

tieback system. In this case, the wall wale- If the concrete capping is designed as an
tieback system is usually designed for the extension of the steel sheet-piling, then its
expected fullloading associated with opera- reinforcement must be designed to continu-
tion of the structure; therefore, the concrete ously carry out the load from the sheeting.
capping is not designed as a structural ad- Very often reinforced concrete capping
dition to the wall's wale. structures are used as foundations for
Quite contrary to the system depicted in mooring accessories (i.e., bollards, lifts, etc.).
Figure 6-32e, the capping structure as In this case, if not provided with special
illustrated in Figure 6-32f performs as a auxiliary anchoring, the mooring force
wale and, therefore, must be designed ac- should be distributed within the capping
cordingly. In this case, the concrete capping structure, similar to that recommended for
is usually very stiff and cannot be as effec- superstructures used in gravity-type walls
tive in load equalizing between adjacent (Chapter 5).
anchors relative to a fiexible steel wale. In In general, expansion joints must be con-
the case of a stiff wale, an example of which structed in a way that changes in capping
is depicted in Figure 6-32f, design anchor length will not be hampered by the sheet-
forces should be increased at the engineer's piling. For this, the concrete capping must
discretion by a factor of 1.1-1.2. The latter be adequately reinforced longitudinally.
can be avoided, however, provided that all Joints are usually placed directly above the
tie-rods are uniformly tensioned before sheet-pile web and are normally coated with
back:filling. an elastic material that permits some
As illustrated in both examples shown in movement. Typically, expansion joints in
Figures 6-32e and 6-32f, to avoid local reinforced concrete capping are keyed for
stress concentrations in the tie-rod at the the transmission of horizontal forces along
concrete inner face due to rod defiection the bulkhead structure. Here, the keyed
joints must be appropriately reinforced by
during back:filling or in case of soil settle-
shear reinforcement.
ment, it is good practice to install a sleeve
Last, but not least, the chafe plates, or
pipe embedded into the concrete. This per-
rubbing strips, should always be provided
mits some rotation of the tie-rod. In the
where mooring lines pass over the capping
case shown in Figure 6-32f, for the sake of beam.
convenience of fabrication and installation,
the embedded part of the tie-rod should be
as short as practica!. 6.5 CONSTRUCTION METHODS
The capping structure and sheeting must
be securely joined to each other. In all cases
where the concrete capping is used, the re- 6.5.1 construction sequence
inforcement and the dowels welded to the
steel sheet piles or protruded from concrete As mentioned earlier in the discussion of
sheet piles must be designed in a way to the two basic methods for sheet-pile bulk-
prevent separation of the capping from the head construction, namely the "back:filled"
sheeting. The stirrups must be so designed or "dredged" design, the performance and
that the portions of the capping beam on stress-strain conditions are determined by
both sides of the embedded part of sheet- the construction sequence (Fig. 6-5). The
piling are securely connected. The sheet "back:filled" bulkhead construction method
piles are usually embedded into the capping is typically used when poor soils must be
beam to about 100-150 mm. excavated from the active (as well as pas-
592 Sheet-Pile Bulkheads

sive) soil pressures zone and substituted The rig must be sufficientiy robust to hold
with good quality granular soil. It is also the operating hammer firmly in leaders and
used when synthetic or other filter material in axial alignment with the sheet pile. Typi-
needs to be installed behind the reinforced cally, the leaders are fixed to the rig's boom,
concrete sheet piles to prevent granular which can usually move laterally and per-
backfill material from being washed out pendicular to the sheet-piling directions,
through the joints between the sheet piles. thus permitting the driving of batter piles.
The "dredged" bulkhead type of construc- Power sources (e.g., a boiler for steam or
tion is often used where there is no need to compressor for compressed air) that are
excavate existing soil from the active pres- used to operate pile-dr.iving hammers are
sure zone and where steel sheet piles with a located on the deck of the rig-supporting
standard interlocking system are used. The barge or pontoon, or on a traveling carriage
construction sequences for both methods are where the rig is operating from the bottom-
outlined in Section 6.1. fixed trestie.

6.5.2.1.2 Hammers
6.5.2 Sheet-Pile Driving There are various types of pile-driving
hammers that are used for pile and sheet-
The purpose of this section is to provide the pile installation. Typically, the drop ham-
basics on sheet-pile installation require- mers, single-acting hammers, double-acting
ments. Detailed information on pile-driving hammers, differential hammers, diesel
equipment and methods of pile installation (single or double acting) hammers, and vi-
are found in practically any standard text bratory pile drivers are the principal types
on foundation engineering. of equipment used for pile-driving and
Depending on material, weight, and sheet-pile-driving.
foundation soil conditions, the sheet piles The drop or gravity hammer is raised
can be installed in place either by im- and then dropped by its free fall on the pile
pact hammers of miscellaneous designs, head. The drop hammer is usually raised
(e.g., drop hammer, single-acting hammer, manually. A single-acting hammer is essen-
double-acting hammer, diesel hammer, etc.), tially a drop hammer except that the ham-
by the water jet method, or by a vibratory mer is raised either by steam or air pres-
pile driver. sures. In double-acting hammers, the
hammer is raised by steam or compressed
6.5.2.1 Pile-Driving Equipment air and then accelerated down by the same
source of power. This delivers more impact
6.5.2.1.1 Pile-driving rig
energy per blow than a drop hammer or
The pile-driving hammer is operated on a single-acting hammer of equal weight. A
rig which is usually mounted on a barge or similar effect is achieved in differential act-
pontoon. An example is seen in Figures ing hammers, but with a lower volume but
6-24, and 7-36 and 7-37 (Chapter 7). Fur- higher pressure of air or steam (Compton,
thermore, sheet piles as well as batter an- 1981). The impact hammer delivers the blow
chor piles can also be driven from a tempo- on a pile head via an elastic cushion placed
rary bottom-fixed trestie. The pile-driving within a pile helmet. The cushion (mini-
industry offers different types of rigs. The mum thickness 150 mm), which is usually
basic purpose of the rig is to ensure that the made from hardwood, softwood, or plywood,
pile is firmly held in position and to main- is placed between the top of a concrete pile
tain the axial alignment of the sheet-piling. and the drive head.
Sheet-Pile Bulkheads 593

Pile-driving hammers of all k.inds are of both the pile and vibrator; thus, the piles
rated by the maximum impact energy they are vibrated to the predetermined eleva-
are able to deliver to the pile head. How- tion. A comprehensive list of impact and
ever, the actual amount of energy delivered vibratory hammer data is given in Prakash
to the pile is influenced by the type of and Sharma (1990).
hammer (e.g., single- and double-acting
airjsteam hammers, diesel hammer), the 6.5.2.2 Hammer Selection
k.ind of soil the pile is driven in, the cushion
properties, the pile material, the pile- As stated earlier, the selection ofthe par-
length-toe configuration, and other con- ticular hammer for the specific project de-
ditions. AII the above factors should be pends on the type of piling (sheet-piling) to
considered when selecting pile-driving ham- be driven, foundation soil, and hammer
mers. U seful information is obtained by re- availability.
lating the energy ratio to the actual poten- In general, because of the slow rate of
tial energy delivered by the hammer, either operation, drop hammers are used on small
measured directly (Lik.ins and Ransche, projects or in remote areas. Single-acting
1988; Hannigan and Wesbster, 1988) or, for hammers deliver the impact energy to the
single-acting diesel hammers, determined pile by the fall of the ram due to gravity.
from the blow count rate (Fellenius et al., Naturally, the amount of impact energy de-
1978). For more information consult Chap- pends on the ram's weight. Thus, heavy
ter 7. rams can be used to deliver large amounts
of impact energy to the pile. Double-acting
6.5.2.1.3 Vibratory drivers hammers have lighter rams; however, they
operate at much higher speeds than single-
This type of pile-driving equipment em- acting hammers. These hammers are gener-
ploys a pair of rotating weights that are set ally used to drive relatively small and light
eccentrically from the center of rotation to piles. Differential hammers combine the ad-
produce an oscillation that drives the pile vantages of single-acting and double-acting
through the soil. These drivers are most hammers and they have heavy rams operat-
effective in granular soils. The vibratory ing ata high speed. As stated earlier, vibra-
driver capacity depends on the actual en- tors are most effective in pile installation in
ergy it can deliver to the pile during the granular soils; in some cases, vibrators can
pile-driving operation. drive the pile into its final set in a matter of
In general, there are two types of vibra- minutes.
tory drivers: low-frequency vibrators which In general, the effectiveness of vibration
can produce up to 30 cycles per second oper- depends directly on the power transmitted
ating frequency, and high-frequency vibra- to the sheet pile and, thus, on the soil-pile
tors which have an operating frequency friction coefficient. The coefficient reduces to
range of 80-120 or more cycles per second. about one-third of that obtained in the ab-
Low-frequency vibrators are usually pow- sence ofvibration (Bilett and Sieffert, 1989).
ered by either electric or hydraulic motors, A major disadvantage of using vibrators
whereas the latter type of vibrators are for piles that bear vertical load is that the
powered by an internal combustion engine. last 1.0-1.5 m of the pile's final set must be
The vibration generated by vibrators liq- driven by an impact hammer in order to
uefy the surrounding soil, thus reducing the take full advantage of the foundation soil's
skin friction along the pile's surface during bearing capacity. However, because in most
driving. Furthermore, the pile is pushing practica! cases sheet-piling carries little or
down into the soil under a combined weight no vertical load and they are usually de-
594 Sheet-Pile Bulkheads

signed to resist lateral loads only, there is ensure that the piles are vertically and hor-
no need in using impact hammers to drive izontally aligned. In the latter case, the pile
the vibrated sheet piles to their final set. leading interlock must be well guided. When
driving concrete sheet piles with standard
6.5.2.3 Driving Methods tongue-and-groove joints the pile must be
kept tight to the one previously installed.
To avoid pile damage during the driving This can be accomplished by a special
process, it is important that the driving mechanism (Fig. 6-33). The first sheet pile
energy be delivered centrally on the pile must be installed with special care; the in-
head. The effect of interlock friction, which
terlock engagement of the subsequent piles
acts only on one side, can be countered, if
with previously installed piles must be peri-
necessary, by a suitable adjustment of the
odically checked. Where required, a better
point of impact. For better results in driv-
ing sheet piles longer than 15 m, guides interlock engagement can be achieved by
placed vertically apart as far as practica! at dredging a deep (as practica!) trench before
a minimum of two levels should be used. the piles are driven. The latter also results
This is particularly important when a heavy in a reduced depth of sheet-pile-driving. The
current is present at the construction site. subsequent soil replacement, particularly in
When a fl.oating pile driver is employed, the passive zone, must be given proper at-
it must be securely moored at the location tention; it must be well compacted before
to effectively eliminate movement. In some placement of the backfill, to ensure ade-
cases the fl.oating pile driver's huli is used quate sheet-pile resistance to lateral forces.
as a guide for pile-driving. When piles are Soil also may be excavated and replaced
driven without leads, which can be the case by the noncohesive material when difficult
on small projects, care must be taken to subsoil conditions do not allow for pile pen-

~ . ~ ·.. '/.

1•::: :• •' ).f :• ,,


~dl :1: !:
W4-4LJl]
1 .... 11
lJI:J
500 i 110 i
A-A
Figure 6-33. Sheet-pile-d.riving: 1-sheet pile; 2-steel guides (alternatively, a floating tem-
plate); 3-guide-supporting piles; 4-pile-holding mechanism; 5-tension wire rope; 6-rollers;
7-clamping bars.
Sheet-Pile Bulkheads 595

etratîon. To detect pile deviations from the


design position (e.g., rake, warping, defor-
mation at the top, etc.) the driving process
must be continuously monitored. Frequent
and careful observations are especially im-
portant when piles are driven into difficult
subsoils that house potential obstacles to
penetration. lf the sheet pile ceases to pene-
trate the soil, driving must stop immedi-
ately and the next sheet pile in line must be
driven. Driving a sheet pile which pene-
trates the soil only with great difficulty and
shortly before reaching the final set, there-
by inviting assumption of a damaged pile
tip or interlock, should not be driven any
further.
In the case when several adjacent piles
cannot be installed to the design depth, an
attempt should be made to remove the ob-
stacle. This can be achieved by partial exca-
vation or introducing a water jet to the area
in an attempt to shift the obstacle to the
side or move it below the design depth of
penetration. The water jet technique is dis-
cussed later in this section. In general, it is
preferable to limit the number of sheet piles
that are not installed to the design depth. Figure 6-34. Installation of steel piles in a stag-
The acceptable number ofunderdriven sheet gered mode.
piles in a row should be determined by an
appropriate analysis. the interlocking components. Steel sheet
Tongued-and-grooved timber and con- piles are typically driven in sets of two to
crete sheet piles should always be driven three sheet piles joined together by tuck
with the tongue leading. This rule also ap- welds.
plies to steel sheet piles. Sheet piles of any In the case where difficti.lt subsoil condi-
material tend to drive outward and also to tions exist andjor deep sheet-pile penetra-
creep in the direction in which the wall is tion is required, in order to reduce resis-
being driven. Lee (1961) suggested that tance to sheet-pile-driving the male part of
"this action may be prevented by pitching a sheet-pile joint can be cut short, 1.0- 1.5
and driving partly a few piles at the begin- m below the dredge line.
ning of the wall and then completing the As noted earlier, in some cases sheet piles
driving of these few in the reverse order, tend to lean sideways. This usually occurs
that is, from the last pile in the row back because ali types of sheet pile inevitably
towards the original pile." Sometimes, to deviate from a vertical plane as they are
prevent sheet piles from leaning in the di- driven, encountering obstacles within the
rection of driving, they are driven in a stag- foundation soil which deflect them from
gered pattern (Fig. 6-34). Before driving, their original direction. The result is that
each pile should be thoroughly inspected for while the top of sheeting is kept in a straight
straightness, lack of cracks, and integrity of line (AB) by its guide system, its lower
596 Sheet-Pile Bulkheads

portion follows a somewhat undulated line ple, on a typical job a 35 X 35-cm concrete
(A'B'). The result is that the sheet piles pile, 15 m long, can be jetted into dense
lean sideways in the driving direction, as sand in less than half an hour. The energy
depicted in Figure 6-35. If this cannot be savings are also considerable. In addition
prevented, then a compensating taper pile pile jetting is a "quiet" driving process in'
should be installed to permit further that it does not shake the ground. Further-
driving. more, it allows for quick reinstallation of
incorrectly driven piles.
Pile jetting is typically recommended
6.5.3 Pile Jetting when piles or sheet piles are expected to be
driven into sand or loose gravei. This
Water jets in various forms have been used method could also be very effect ive for driv-
for years in marine construction. This in- ing piles into hard clay, provided that some
cludes placement of subsea cables and special techniques are used. It should be
pipelines, installation of anchors for off- noted that jetting is seldom required for
shore moorings, and, of course, installation driving steel sheet piles, but is more often
of piles and sheet piles. During the past used for driving concrete or heavy timber
30-40 years, many thousands of vertical piles and sheet piles. It is especially effec-
and batter concrete piles of regular con- tive, and probably the only means, for in-
struction, anchor piles with a large widen- stallation ofwide-flange, T-shaped, or simi-
ing at their lower end, and sheet piles have lar concrete sheet piles (Gurevich, 1969;
been installed by the water jet method. McGregor, 1963; Matlin, 1983).
In general, pile jetting proved to be a Jetting techniques can be effectively uti-
very effective method for the installation of lized for driving vertical and batter piles. It
concrete piles and sheet piles. Contractors is almost indispensable for driving anchor
who used the water jet method for pile piles with a local widening at the end
driving have found that the time savings, (Tsinker, 1972, 1977), similar to that de-
as compared to standard concrete piles with picted in Figure 6- 24. Furthermore, jetting
a h ammer, are very significant. For exam- minimizes damage to wood or concrete piles
designed for driving into dense sand, grav-

- elish soils, or hard clay. However, the final


set ofbearing piles installed by jetting tech-
niques is usually obtained by driving the
last meter or so without jetting. This is
usually required for densification of the
granular soils loosened by the water jet,
thus increasing skin friction and the pile's
bearing capacity. For more information on
dynamic effects on granular soil densifica-
tion, the reader is referred to Chapter 4.
The use of the water jet (sometimes in
II f&..EVATIOI'I combination with compressed air) in a pile-
driving operation differs radically in princi-
A ~I~A~---~--------~------~1B'8
pal from driving with a h ammer or vibrator.
1' LA H In this simple procedure, the water jet is
Figure 6-35. Sheet-pile leaning in driving direction.
discharged near the pile tip or along its
(Note: pile guiding syst em is not shown.) 1-Regular shaft, loosening the surrounding soil so that
sheet pile; 2- taper pile.- the pile drops through the soil into the
Sheet-Pile Bulkheads 597

underlying material. Typical pile jetting that the adjacent piles are not adversely
conditions are illustrated by Figure 6-36. affected, particularly wooden piles.
When piles are driven by water jet into Pile jetting into silty sands usually en-
sand, the flow rate of water is a more im- hances geotechnica l properties of these soils
portant factor than the jet velocity. On the by removing some of the fine particles; the
other hand, when piles are jetted into gravei latter increases the pile's frictional skin re-
or hard clay, the jet velocity is a more im- sistance. Thick layers of clay that may exist
portant factor than its flow rate. For either within a predominan tly sandy foundation
procedure to be effective, the velocity of can block the return flow of water, thus
the water jet must be suflicient to loosen preventing it from escaping to the surface
the soil and make it quick. Furthermor e, (Fig. 6-36b). In this case, the column of
the volume of water must be sufficient to water, and its upward velocity, must be
force water to escape by rising to the sur- suflicient to wash the loosened clay parti-
face along the pile's shaft (Fig. 6-36a). lf des surroundin g the pile to the surface.
the pump capacity is insuflicient to provide This will provide a clear annular space for
the required volume of water, then com- effective water return. This effect is partic-
pressed air may be added to the water jet to ularly important when piles are jetted
help the water travel upward. Sometimes it through a soft clay stratum.
may be necessary to limit the horizontal If piles are to be jetted into hard clay,
radius of the escaping column of water so then a sufliciently thick layer of sand should

(a) (b) ( c)

Figure 6-36. Pile jetting in different foundation soils: (a) sand; (b) sand with clay stratum;
(c) sand stratum underlaid with clay. 1-Pile; 2-jet pipe; 3-water jet; 4-sand; 5-clay;
6-loosened sand; 7-jet hole; 8-fine partide deposition.
598 Sheet-Pile Bulkheads

be placed on top of the clay so that the piles bles and coarse gravei, could produce simi-
are driven into the clay foundation through lar effects. For successful jetting through a
the layer of sand (Fig. 6-36c). The sand soil containing large cobbles or a lot of
particles will then fiii the slot cut in the gravei, it is essential that a sufficient quan-
clay stratum by the water jet. Subsequent tity of water be used. This phenomenon,
dynamic pile-driving to its final set is nor- however, can be successfully used to in-
mally required to consolidate the sand in crease the pile's bearing capacity by feeding
the void around the embedded portion of the Iarge-sized material (e.g., gravei or
the pile. crushed stone) into the jet hole to allow its
Sometimes pile jetting could be pre- concentration at the designated locations,
vented by the presence of large boulders in such as the pile's tip or the bearing surface
the foundation soil. During pile jetting, of an anchor pile with widening, as illus-
these boulders can move down. along with trated in Figure 6-37 (Shestopal, 1959;
the water jet and toward the pile's tip. Tsinker, 1988).
Eventually, these boulders can interfere It is obvious that the amount of water
with the pile's movement to the extent that required for pile driving does not present
movement of both the jet pipes and the pile any problem if the piles are jetted offshore.
are effectively blocked. The presence oflarge It could, however, create some problems
obstacles in sand foundations, such as cob- when piles are driven on shore, that may

(a ) ( b)

• t

Figure 6-37. Increase in pile capacity: (a) pile in compression and (b) anchor pile (in
tension). 1-Pile; 2-jet pipe; 3-crushed stone andjor large gravei; 4-loose sand in jet hole;
5-water jet.
Sheet-Pile Bulkheads 599

require to dispose of large quantities of one centrally located pipe embedded into
water during jetting. In the latter case, the pile, During the pile-driving process,
sufficient means of water escape are re- the jet pipes are usually pushed into the
quired. ground, then moved up and down along the
pile to obtain a "lubricating" effect. It usu-
6.5.3.1 General Procedure ally requires only two jet pipes for installa-
tion of a regular concrete bearing pile or
Conventional pile jetting procedures re- regular concrete fl.at sheet pile. Accordingly,
quire very simple equipment, as detailed in at least four symmetrically placed jet pipes
Figures 6-38 and 6-39. Typically, it in- are used for installation of wide-fl.ange T-
cludes a centrifugal pump equipped with a shaped concrete sheet piles.
fl.ow meter and pressure gauge, a minimum During pile jetting, special attention
of two steel jet pipes connected to the pump must be given to the symmetry of the jet-
by the same-diameter fl.exible pressure ting streams and their intensity; in case
hoses, and a winch for handling the jet additional jets become necessary, care must
pipes. The required jet pipe diameter is be taken for their proper positioning. On
usually varied between 50 and 100 mm. To the other hand, by changing the position of
increase the velocity of discharged water, the jet pipes, as well as intensity of the
the lower end of the pipe is usually pro- water jets, an operator can change the posi-
vided with a nozzle as illustrated in Figure tion and inclination of the jetted pile.
6-39. The internal diameter of the nozzle is For improved performance, the oper-
normally taken as one-third or more of the ator(s) push the jet pipe down first, haul it
pipe's internal diameter. Side holes 6-10 up and down several times, then while
mm in diameter help to loosen the soil keeping the jet pipe down, they drop the
around the nozzle. In the case of concrete pile into the jet hole. This sequence of steps
piles, the water can be discharged through are repeated until the pile reaches the de-

Figure 6-38. Typical pile jetting setup; 1-barge;


2-pump; 3-pile handling winch; 4-jet pipe; 5-fiexible
hose; 6-T-joint; 7-manometer; 8-fiow meter; 9-valve;
10-pile.
600 Sheet-Pile Bulkheads

(a) (b)

2 4

•1 .,o
-t-
1
IlO
.,o
5 N

Figure 6-39. Typical water jet assemblage: 1-water supply fiexible


hose; 2-fiange; 3-bent pipe; 4-straight pipe; 5-coupling;
6-nozzle.

sired depth. As previously stated, the bear- sand stratum, three distinctive zones in
ing piles are driven to their final set by the jet hole structure are observed, as
dynamic methods. This, however, may not illustrated in Figure 6-40. In the imme-
necessarily apply to sheet piles designed to diate vicinity of the water jet (Zone 1),
resist thrust only. The aforementioned tech- the sand's natural structure is destroyed,
niques can be applied to driving of both resulting in a heayy sand-water liquid
mixture. Also, the excess water pumped
vertical and batter piles.
into Zone 1 escapes to the surface along
Few reports on the investigation of pile
the pile's shaft (Zone 2). The upward fl.ow
jetting have been published to date. The of water in Zone 2 "lubricates" the pile
most comprehensive test results are dis- and assists its downward movement.
cussed by Shestopal (1959) and Tsinker Some water infiltrates into the adjacent
(1972, 1977, 1988). Shestopal basically in- Zone 3 and maintains a high pore pres-
vestigated the jetting of steel pipes, whereas sure there. However, in sand, this excess
Tsinker reported results of tests carried out pressure dissipates almost instanta-
on concrete piles of regular construction and neously after termination of pile jetting.
anchor piles with widening at the lower The latter phenomenon is characteristic
end, jetted into sand. The most significant for piles driven into both dry and water-
results of the investigations can be summa- bearing sands.
rized as follows: (ii) Concrete piles jetted into dry sands have
six to nine times more carrying capacity
(i) In the process of pile jetting into a clean than identica! piles jetted into water-
Sheet-Pile Bulkheads 601

water jet is withdrawn, the loose sand


settles around the pile in the jet hale,
being compacted by the movement of wa-
ter filtrated down and out the jet hale.
Further dynamic actions, such as use of a
hammer impact or vibration for the pile
final set, do not increase the pile's capac-
ity significantly.
(iii) In most laboratory and field tests, it has
been found that there is some increase in
soil density outside of the jet hole. How-
ever, pull-out tests on the piles indicated
a reduction in pile skin friction resis-
tance of up to 10% in comparison to
hammer-driven piles (Tsinker, 1972). For
more details, consult Chapter 7.
(iv) Despite the fact that some soil particles
are washed out, the pile jetting process
does not change the sandy soil's granulo-
metric composition significantly. Accord-
ingly, inclusions such as cobbles or large
gravei tend to settle near the deeper zone
of the jet hale.

6.5.3.2 water Rectuirements


On the hasis of numerous :field tests,
Figure 6-40. Typical structure of a jet hole: 1-jet Shestopal (1959) suggested that the follow-
area2 (soilliquified and suspended) (Zone 1); 2-rising
fiow (Zone 2); 3-filtration zone (Zone 3). ing empirica} equations may be used to de-
termine the quantity of water required for
pile jetting into a sandy soil:
bearing sands. Dynamic effects, however,
such as hammer impacts or pile vibra- With a dry sand stratum (water table located
tion, sharply increases the capacity of below pile tip):
piles jetted into the water-bearing sands.
This happens because the loosened sand
grain structure in the jet hale collapses (6- 1)
and consolidates as a result of the dy-
namic impact. Basically, ali types ofloose
With a saturated sand stratum
sand are susceptible to liquefaction.
However, a few seconds after sand lique-
faction occurs, the entire column of loose ; = 530(d50 )1.3 l 0 ·5 + 0.0177Tlk (6- 2)
sand in the jet hole sinks, causing densi-
fication of the sand. The repeated dy-
namic impact on jetted piles does not where
increase their capacity significantly. This Q = flow rate of water required for pile jetting
phenomenon is discussed in detail in (m3 jh)
Chapter 4.
D =pile diameter, or width (m)
The post-pile-driving sand liquefaction
phenomenon was not observed in cases d 50 = average size of sand particles (mm)
where concrete piles or steel pipes are l = embedded length of pile (m)
jetted into a dry sand stratum. When the k = filtration coefficient (mj day)
602 Sheet-Pile Bulkheads

When pile jetting through several soil Table 6-3. Value of factor kr in Eq. (6-4) for
strata, the coefficient of filtration based on a hoses made from different materials
weighted average can be determined from Jet Pipe Hose Material
Internal Diameter,
knln (mm) Rubberized Rubber
k=E-1- (6-3)
33 33 50
50 133 200
where kn and ln are the filtration coeffi- 65 567 850
cient and length, respectively, for soillayer 76 1333 2000
n.
Source: From Marine Structures Handbook, Nickolayev
On the hasis of extensive experience ed. (1972).
gained in jetting different kinds of piles in
the East European countries, and particu-
larly in the former Soviet U nion, the flow exceeds 5 mjs, then it is necessary to in-
rates indicated in Table 6-2 are recom- crease the number andjor diameter of the
mended for pile installation. jet pipes.
When selecting a pump(s) for pile jetting The selection of jet pipe diameter and
the head loss (H) in the water supply hoses pump capacity should be made in the fol-
must be considered. The following equation lowing sequence:
can be used to calculate the head losses:
1. Using data from Table 6-2, select the
desirable value of fiow rate and water
(6-4) head required for pile-driving.
2. Determine the pressure losses in the hoses
where Q is the same as in Eqs. (6-1) and and pipes.
(6-2) and 3. Determine the required pump capacity
lh = total length of the water supply hoses (m) and number of pumps.
Kr = empirica! coefficient dependent on hose
material obtained from Table 6-3
6.5.4 Earthwork
The pressure loss in the jet pipes and in
the manifold can be determined using the Earthworks, in association with sheet-pile
data given in Table 6-4. If the pressure loss wall construction, basically includes place-
in the jet pipes is excessive and the velocity ment of the backfi.ll material and dredging

Table 6-2. Volume ofwater pressures

Pile Section (Diameter) (cm)


30-50 50-70
Depth Headatthe Jet Pipe Jet Pipe
of Pile- Tip ofthe Interna! FlowRate Internal FlowRate
Driving Nozzle Diameter ofWater Diameter ofWater
Soil (m) (MPa) (mm) (m 3 jmin) (mm) (m 3 /min)

Silt, silty sand, 5-15 0.4-0.8 37 0.4-1 50 1-1.5


fine sand, soft 15-25 0.8-1.0 68 1-1.5 80 1.5-2
clay, sand
Sand and hard 5-12 0.6-1.0 50 1-1.5 68 1.5-2.0
sand loam, sand 15-25 1.0-1.5 80 1.5-2.5 106 2-3
with gravei
Sheet-Pile Bulkheads 603

Table 6-4. Flow rate ofwater (in m 3 /h) (numerator) and head loss (in m) (denominator) per 100 m
of steel pipe

Velocity
Interna! Pipe Diameter (mm)
ofWater
(mjs) 25 40 50 60 70 80 100 125 250
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

1.0 1.80 4.50 7.10 10.2 13.9 18.1 28.3 44.2 63.6
6.40 3.70 2.90 2.3 2.0 1.7 1.3 1.0 0.84
1.1 1.95 5.0 7.8 11.2 15.2 20 31.1 48.6 70
7.60 4.4 3.4 2.8 2.3 2.0 1.6 1.2 1.0
1.25 2.20 5.70 8.8 12.8 17.3 22.6 35.4 55.2 79.5
9.70 5.6 4.3 3.5 3.0 2.6 2.0 1.6 1.3
1.50 2.70 6.8 10.0 15.3 20.8 27.1 42.4 66.3 95.4
13.5 7.8 6.1 5.0 4.2 3.6 2.8 2.20 1.8
1.75 3.10 7.9 12.4 17.8 24.3 31.7 49.5 77.3 111.0
18.1 10.5 8.1 6.6 5.7 4.8 3.8 3.0 2.40
2.0 3.5 9.10 14.1 20.4 27.7 36.2 56.6 88.4 127
23.0 13.4 10.5 8.6 7.2 6.2 4.9 3.8 3.2
2.5 4.40 11.3 17.7 25.5 34.6 45.2 70.7 110.0 159.0
34.7 20.5 16.0 1.1 11.1 9.6 7.5 5.9 4.9
3.0 5.30 13.6 21.2 30.5 41.6 54.3 84.3 133.0 191.0
47.9 28.8 24.7 18.6 15.8 13.6 10.7 8.4 6.9
3.5 6.3 15.9 25.2 35.2 48 63 100 154 222
69 39.4 35 34.8 21 18.2 16.2 11.3 9.3
4.0 7.2 18.1 28.8 40.4 54.7 72 114 176 253
90 51 45.5 32.5 27.2 24 21.2 14.7 12.1
5.0 9.0 22.6 36 50.5 68.5 90 142 220 316
140 76 71 51 41.5 37 32.7 23 18.9

Note: For used rusted pipes, the head loss should be increased by a factor of 1.15-1.3. Also, head loss in valves,
connecting curved parts, and so forth could be taken as equal to the loss of a straight pipe 5 m long of relevant diameter.
Source: From Marine Structures Handbook, Nickolayev, ed. (1972).

of material in front of the wall. Dredging the help of conventional equipment such as
may also be required when the "backfill" dump trucks, conveyors, clamshells, bull-
type of construction is planned (Fig. 6-5) ordozers, and so forth. The fiii material should
where a poor foundation soil exists and must be clean and contain as little silt andjor
be replaced by good quality granular soil. clay particles as practica!.
Soil obtained from borrow pits must be
6.5.4.1 Backfilling carefully investigated for its acceptability.
In general, backfill includes aU kinds of If the available sand contains a large per-
free-draining granular materials ranging centage of silt and clay components, then a
from clean sand and gravei to rockfi.U of flushing process may be used to wash these
miscellaneous composition. Depending on fine-grain particles from the fill material. In
the sources from which this material is this respect, when fiU is placed hydrauli-
coming and available means of transporta- cally, dredging the fiU material at its source
tion and placement, backfill is usually and placing it behind the waU are of essen-
placed behind the waU either hydraulicaUy tial significance. For details on this subject,.
or by dumping at the designated areas with the reader is referred to Recommendations
604 Sheet-Pile Bulkheads

of the Committee for Waterfront Structures To enable the backfill to settle uniformly
(EAU, 1990). and to ensure that the existing subsoil can
In most cases of marine wall construc- adapt itself to the increased surcharge,
tion, the fill is placed underwater, and ifnot there should be a suitably long interval
compacted in one way or another, may re- between the start and completion of back-
main in a loose state for an extended period filling. Planning the backfill placements
oftime. should be consistent with the entire process
To reduce the risk of soil liquefaction, as of wall construction.
well as to avoid excessive maintenance of An example of a "backfilled bulkhead" is
pavement, railway and crane tracks, and illustrated in Figure 6-41. In this case, the
similar works, the loose granular backfill sheet piles are driven first (Stage 1), then
should have a minimum relative density of the first stage of backfill (2) is placed and
D = (e - emin)/(emax - emin) = 0.5, where compacted. This portion of backfill provides
emax and emin are maximum and minimum a construction area for installation of the
soil void ratios, respectively. For details, see lower tieback system (3). Prior to the first
Chapter 4. stage of backfill placement, ali unsuitable
Granular backfill is usually effectively soils must be removed. In some cases, soft
densified to the required value of relative mud stiH remains at the site. Here, the fiU
density either by vibrofiotation or vibrocom- placement should proceed from the wall to-
paction methods. Both methods are dis- ward the land, thus effectively displacing
cussed in detail in Chapter 4. Sometimes the soft mud away from the wall. For de-
the acceptable value of relative density, D, tails, see Chapter 4.
can be achieved by steady pumping of water After construction and careful inspection
in large volumes onto the top of the fill. of the lower anchorages, the second stage of
However, to prevent wall overloading by the backfilling (4) is carried out. At this
substantial hydrostatic pressures, an effec- stage, material placement proceeds from the
tive drainage system must be in place. land toward the harbor. The latter ensures

..
. ..
..
Figure 6-41. Construction sequence of double-anchored sheet-pile
bulkhead: 1-T-shaped concrete sheet piles; 2-backfill (first stage);
3-low-level anchor system; 4-backfill (second stage); 5-precast
concrete panel; 6-upper-level anchor system; 7, 8, 9-third, fourth,
and fifth stages, respectively, of wall rock fill. [From Matlin (1983).]
Sheet-Pile Bulkheads 605

that the anchorages will have some of their in the upper anchorages, each prefabricated
design capacity developed before any lateral panel is usually anchored by two tiebacks
soil thrust on the wall is produced. In this only.
case, the back:fill soil is usually placed in Finally, the third (7), fourth (8), and fifth
the dry. The soil placement proceeds in lay- (9) stages of back:filling are carried out in
ers 0.4-0.6 m thick, and the soil is com- the dry. The final stage of wall construction
pacted in one way or another; vibrocom- is casting of the upper concrete capping
paction is used most often. At this stage, fill beam.
placement is terminated at a certain eleva- The example discussed demonstrates the
tion, to provide a suitable construction area general approach to wall backfilling. This,
for installation of the upper row of anchors. of course, can vary on a case-by-case hasis.
Completion of the second stage of fill For example, where a sheet pile wall is
placement usually results in a uniform ten- anchored by raked piles, which are actually
sile force transmitted to the lower anchor "stiff' anchors, in most cases rigidly joined
system. This is attributed to the ability of with the sheet piling via capping structure,
individual sheet piles to deflect indepen- the back:filling is usually accomplished by
dently of each other. Note that in the case placing the backfill material in layers to
in question, each sheet pile is anchored in- prevent overloading of the anchor piles by
dividually. When deflection movements soil hanging on them. If this occurs, it may
stop, the stiff lower capping beam is con- induce excessive bending moments in the
structed, the upper prefabricated panels (5) piles, resulting in pile overstressing and
are erected, and the upper anchor system cracking. An example is illustrated in Fig-
(6) is installed. To ensure uniform tension ure 6-42.

Figure 6-42. Cracks in overloaded concrete anchor pile: 1-sheeting with synthetic filter
material; 2-anchor pile; 3- capping beam; 4- cracks.
606 Sheet-Pile Bulkheads

6.5.4.2 Dredging model tests and field investigations carried


out in the past 40-50 years worldwide.
Dredging in front of the sheet-pile wall is A great deal of important investigations
conducted in cases where a "dredged bulk- have been carried out in the former Soviet
head" type of construction is used. Union. Results of these investigations have
The soil dredging in front of the wall is been published in Russian in a great num-
usually accomplished by a standard dredg- ber of local papers and books. However,
ing equipment suitable to remove the exist- because of tens of years of virtual isolation
ing soil to acceptable tolerances. It must be ofRussian science from the free world, these
carefully supervised to ensure that substan- publications have been largely unknown to
tial overdredging does not occur. During the the scientist working on the other side of
dredging operation, the wall's deflection the iron curtain. To fill the gap, some of the
must be accurately surveyed after each most important works accomplished in the
dredge cut, so that the beginning of the former Soviet Union are discussed in this
potentially excessive wall movements away section.
from the backfill are detected in its early Efforts to establish a realistic soil pres-
stage. sure value against a sheet-pile bulkhead
Completion of the dredging operation goes back a long time in the history of
must be followed up by wall inspection con- engineering arts and science. Yet, in 1906
ducted by divers to confirm that no dam- an interesting experiment has been under-
ages to the sheet piles have occurred. taken by Danish engineers Christian and
The specified dredging tolerances basi- Nielsen. They built a pier comprised of rein-
cally depend on the type of equipment to be forced concrete sheet piles that could not
used, type of soil to be dredged, and site have stood up according to the conventional
environmental conditions (e.g., current, application of Coulomb's theory to retaining
wind, waves, and sometimes ice). Obvi- wall design. Although obviously underde-
ously, the dredging tolerances must be signed according to even presently used
taken into account for wall design. theories of soil-structure interaction, the
pier was for several decades operational and
remained in excellent condition. The stabil-
ity of this structure was largely attributed
6.6 EARTH PRESSURES ON to a reduction in active lateral pressures in
FLEXIBLE WALLS: the mid-span zone between the anchor level
STATE-OF-THE-ART REVIEW and the dredge line. In 1910 Ehlers pub-
lished results of some observations on tim-
ber sheet-pile walls used in existing mas-
Sheet-pile bulkheads are generally re- sive relieving platforms. He found that the
garded as flexible structures with a varying stresses in the sheet piles, if computed on
degree of flexibility. the hasis of the Coulomb distribution of
The state of knowledge regarding the lateral earth pressures, may reach values
sheet-pile wall-soil interaction in general that are incompatible with sheet-piling ac-
terms is discusşed in Chapter 4. Further tual strength. Although Ehlers did not at-
details follow. tempt to evaluate the change oflateral earth
To better understand the nature of the pressure exerted on the sheeting, he sug-
phenomenon on earth pressure exerted on gested that the reduction in this pressure is
sheet-pile walls and its response to this attributed to sand arches, which may be
pressure, this section gives the state-of-the- formed between the pier structure and the
art review of the most important scale- failure plane in the backfill material.
Sheet-Pile Bulkheads 607

In 1933 Russian scientist Yaropolski ments in the wall that are much smaller
(1935) tested an anchored sheet-pile wall than those predicted by classical theory,
14.0 m long and 4.59 m high. The wall was and in reaction forces in anchor system that
comprised of timber sheet piles, 19 cm thick, are somewhat greater than those obtained
driven into sand stratum to a depth of 2. 73 by classical soil pressure theory.
m and anchored by timber tiebacks secured Bjerrum et al. (1972) defined soil-
at timber A-framed anchor piles. The wall retention "flexible structures" as those
was designed in accordance with Coulomb's " ... which will show distortional deforma-
theory of soil pressure distribution. The in- tions under the stresses they are designed
tention was to fail it by an application of to withstand, of such an order of magnitude
surcharge load of approximately 4.0 tjm 2 • that they will have a substantial influence
The wall, however, withstood all applied on the earth pressures or their distribution."
loads with no substantial overstress; the Additionally, the anchored sheet-pile
sheet piles, however, defl.ected by 22 mm at bulkheads have a tendency to yield away
mid-span. from the fill at the anchor point. The sheet-
Y aropolski concluded that wall defl.ection pile wall deflections may result in soil arch-
and the friction between the wall and back- ing behind the sheeting, which, in turn,
fill material (a :::::: 30°) contributed to a sub- may be followed by an increase in the earth
stantial reduction in lateral earth soil pres- pressure on a relatively unyielding portion
sutes acting on the sheeting. It should be ofthe wall.
noted that the beneficia! effects of wall fric- Ohde (1938) was one of the first to estab-
tion on soil-structure interaction was
lish a relationship between wall deflections
pointed out by Franzius (1924) prior to these
and the nature of the soil pressure on it. He
tests. There are more examples of similar
found that the total value of active soil
nature found in technical publications.
pressure on the fl.exible wall was equal to
Since then, a great deal of theoretical, as
that determined by the Coulomb theory.
well as experimental studies, on sheet-pile
Accordingly, this pressure is redistributed
bulkheads have been conducted. A large
vertically along the sheeting and concen-
number of speciality conferences have been
held in the past 50 years at which the trates at the anchor zone and in the wall's
problem of sheet-pile wall-soil interaction passive pressure zone. Ohde concluded that
has been discussed. Literally hundreds of this form of soil pressure distribution oc-
papers on the subject matter have been curs due to the effects of soil arching in a
published in proceedings of conferences and zone between the dredge line and the an-
other special publications. chor. On this hasis, Ohde proposed a calcu-
Persuing the available literature for re- lation method that ultimately leads to a
sults on scale-model tests, full-scale obser- reduction in the sheet-pile bending moment
vation, and finite element analyses pub- and to an increase in anchor pull-out values
lished to date, it is possible to conclude up to 60%.
that the earth pressures on sheet-piling, It should be rţoted that before Ohde's
that is, flexible soil retention structures, are thesis, the effects produced by an assumed
very much different from those acting on soil arching effect had been discussed quali-
gravity-type walls and also from those pre- tatively by Baumann (1934) and Stroyer
dicted by the classical earth pressure theo- (1935). The concept of soil arching, devel-
. ries. Furthermore, the sheet-piling fl.exibil- oped by Stroyer, is included in the Danish
ity has a substantial infl.uence on the earth Code of Practice for design of sheet-pile
pressure's vertical redistribution along the walls, known as Danish Rules. According to
sheet-pile wall. This results in bending mo- Danish Rules the soil pressure distribution
608 Sheet-Pile Bulkheads

empirically simulates the soil arching effect much greater value of passive pressure than
(Fig. 6-43a). that obtained by Coulomb's theory, and to
Similar recommendations for sheet-pile the concentration of this pressure much
walls supported by unyielding anchorages closer than expected to the dredge line.
has been roade by Goncharov (1962). The Thus, the effective span of the sheeting is
recommendations are derived from a large- smaller than that obtained from a calcula-
scale-model test with real steel sheet piles tion using the classical method and, hence,
(Fig. 6-43b). Goncharov found that the total so is the bending moment; the bending mo-
value of soil pressure exerted on the wall in ment reduction in tests by Tschebotarioff
the active zone is 10-20% less than the was not as great as that obtained by Danish
value obtained by classical theories. At the Rules.
same time, the pull-out anchor force is about First and foremost, Tschebotarioff has
20% greater than that obtained by classical pointed out that differentiation between the
theory. conditions of soil-structure interaction cre-
Well-instrumented large-scale model ated by backfilling behind a flexible an-
tests (1: 10) were carried out from 1943 to chored sheeting and by dredging in front of
1948 by Tschebotarioff (1948, 1973). The the anchored bulkhead is essential. Basic
results of these tests are also discussed in results of Tschebotariotrs observations are
papers by Tschebotarioff and Brown (1948) illustrated in Figure 6-44. Tschebotarioff
and Tschebotarioff and Welch (1948). found no evidence of vertical arching when
Similar to the earlier discussed investi- the model bulkheads driven into sand were
gations, Tschebotarioff obtained an appre- back.:fill with sand (Figs. 6-44a and 6-44b).
ciable reduction of bending moments in the As stated by Tschebotarioff, the absence
sheet-pile wall. However, unlike the previ- of arches during most of the back.:filling is
ous investigators, Tschebotarioff concluded due to the actual absence of the abutments
that this phenomenon is attributed not to for the arch. Furthermore, both the progres-
the soil arching effects but rather to the sive elastic elongation ofthe tie-rod and the
yield of its support as back.:filling goes on,
induce the breakdown of any arch in the
(a)
granular back.:fill. However, definite evi-
dence of soil arching was obtained when the
sand was dredged in front of the wall, and
all the back.:fill was in place behind it (Fig.
6-44e). In this case, the magnitude of the
deflection observed in a zone between the
dredge line and anchor level indicates that
a considerable reduction in earth pressure
in this zone has occurred. The latter is a
result of a considerable increase in earth
Figure 6-43. Basic assumptions for soil pressures pressure at the bottom of the wall and above
distribution in (a) Danish Rules and (b) as proposed by
Goncharov (1962). Note: In Danish Rules q = k[(lOh the anchor level. In other words, where the
+ 2L)j(l0h + 3L)]pA in which PA is the average anchor support is nonyielding, the deflec-
active pressure over the height of the bulkhead, h is tion of the bulkhead will press the upper
the height of submerged soil equivalent to the load portion of the wall (above the supports)
above the anchor level, and k is the empirica! coeffi- against the backfill, thus inducing a high
cient ("" 0.8). In Goncharov's method q 1 = q 2 is the
active soil pressure as determined by the classical
soil pressure in this zone.
(Coulomb) theory, a = (0.3 to 0.4)q, and the active It should be pointed out that similar ef-
pressure diagram has a parabolic shape. fects can be expected in back.:filled bulk-
Sheet-Pile Bulkheads 609

(a) (b) (c) (d) (e)

Figure 6-44. Basic types of lateral earth pressure distribution obtained in


large-scale-model tests carried out by Tschebotarioff: (a, b)-backfilled bulkheads;
{c) pressure distribution after strong vibration of sand in front of the model; (d)
pressure distribution under normal yield of the anchor; (e) effect of dredging in
front of the bulkhead. [From Tschebotarioff and Brom (1948).]

heads when live loads (both concentrated In the aforementioned tests carried out
andj or uniformly distributed) are placed at by Tschebotarioff, the resultant, from the
a certain distance away from the wall's edge. residual passive pressures, was typically lo-
In this case, a live load produces a lateral cated much closer to the dredge line than
earth pressure, mainly on a part of the wall that which would usually be obtained by
below the anchor level; the latter causes the the Coulomb's theory; the magnitude of
wall span between the dredge line and an- these pressures corresponded to certain
chor level to defiect outward, and the upper bulkhead displacements (Fig. 6-45a). Ac-
part above the anchor to defiect inward, cording to Tschebotarioff, as bulkhead
thus inducing an extra soil pressure on this backfilling proceeds, the active pressure
part of the wall. This phenomenon has been gradually build up, and the wall defiections
observed and reported by Koslov (1963) increase froni the original position (1). The
during field tests on 12.05-m-high sheet-pile
wall's successive positions (2), (3), and (4)
bulkheads.
result in corresponding passive pressure di-
In Tschebotariofi's experiments the se-
agrams (2), (3), and (4) extended further
vere vibration of the sand în front of the
completed wall had an effect similar of and further down along the passive pres-
dredging (Fig. 6-44c). However, the slight- sure maximum limiting line, determined as
est yield of the anchor strongly reduced the KPydi, where KP is the coefficient of pas-
e:ffect of the pressure redistribution along sive pressure, 'Y is the soil unit weight, and
the sheeting. This is caused by reducing the di is the depth of soil. In Tschebotariofi's
load concentration in the anchor zone and experiments, this phenomenon was ob-
increasing the value of the passive pressure served by varying the depths of the embed-
(Fig. 6-44d). ded part of the sheeting, during which the
Exceptionally severe vibrations of the bulkhead's relative fiexibility remained un-
sand behind the wall substantially in- changed. Furthermore, the fixation at the
creased the lateral pressures on the wall, tip of the sheeting was found to be produced
and resulted in a bending moment increase by neutral or active pressures only, not ex-
by about 60% and more, compared with ceeding the K 0 line (pressure "at rest")
values obtained after normal backfilling. values.
610 Sheet-Pile Bulkheads

·:·Ai,
..:'J i,,
.:: ~-.

fJ

(a) (b)

Figure 6-45. Relationship between deflection and soil pressures distribution on a


flexible bulkhead below the dredge line: (a) according to Tschebotarioff (1973); (b)
obtained by Lasebnik and Chernishova (1966) for granular soils with a different
modulus of elasticity (E 0 ); 5 through 8-passive pressure diagrams for sands with E 0
equal to 0.6, 1.2, 2.5, and 3.5 MPa, respectively.

The conclusions made by Tschebotarioff arching), are the cause of smaller bending
were further confirmed by Lasebnik and moments in the sheet-piling, compared with
Chernishova (1966) who carried out large- values obtained by conventional classical
scale-model tests on steel sheet-piling methods.
driven into granular material and back- It is noteworthy to indicate that the close
filled with sand. These latter investigators proximity of the passive pressure resultant
confirmed test results obtained by Tschebo- to the dredge line have also been found in
tarioff and also concluded that the shape of field measurements carried out during con-
the pasive pressure diagram is highly de- struction of a new pier in Brooklyn, New
pendent on the soil density in the passive York (Kyle and Kapp, 1957) and the well-
zone. For example, in a relatively loose sand instrumented and well-monitored anchored
(E0 = 6.0 MPa) the shape of the passive bulkhead constructed at Long Beach, Cali-
pressure diagram is very close to parabolic, fornia (Duke, 1953).
and in a dense sand (E 0 = 35 MPa) it is The next milestone in the sheet-pile
close to that obtained by Tschebotarioff bulkhead investigation were tests carried
(here, E 0 is the soil modulus of deforma- out by Rowe (1952). This investigator car-
tion). Naturally, in densier soils the resul- ried out a series of a very comprehensive
tant force from the soil passive pressure is tests on a rather small-scale model of vary-
located closer to the dredge line (Fig. 6-45b). ing fiexibilities. This simulated conditions
Both Tschebotarioff, and Lasebnik and in which the sheet-pile tie-rod had sufficient
Chernishova concluded that changes in yielding to destroy any soil arches built up
shape and magnitude of the passive pres- behind the sheeting. The wall stiffness was
sure diagram, along with a decrease in the defined by the coefficient of fiexibility ( p)
active pressure just above the dredge line determined as p = (H + D) 4 jEI, where H
by shearing stresses (and not by vertical is the wall height; D is the pile embedment,
Sheet-Pile Bulkheads 611

and E and I are the modulus of elasticity In subsequent works (see list of refer-
and the moment of inertia of the sheet pile, ences), Rowe added that for elastic tie-rod
respectively. yield during dredging, an additional mo-
In each test, the tie-rods were allowed to ment reduction occurred as a result of
yield after dredging, until the bending mo- sheet-pile inward defl.ections. This also re-
ment in the sheeting reached a maximum. sulted in high soil pressures above the
During his tests, Rowe observed that the anchor.
maximum bending moment in the sheet- Essentially, Rowe's tests were in general
piling has decreased substantially with in- agreement with the large-scale-model tests
creased wall fl.exibility. This effect was carried out by Tschebotarioff in 1944 and
found to be greater in dense sand than in 1948 as discussed earlier.
loose sand. As a result of his investigation, Rowe
First and foremost, Rowe gave a quanti- (1952) developed the bending moment re-
tative correlation between the fl.exibility of duction curves that allow for reduction of
the sheet pile and the maximum bending bending moments obtained by the classical
moment, an effect which had previously method. An example is illustrated in Figure
been discussed qualitatively by Stroyer 6-46. It must be remembered, however, that
(1935) and Bauman (1934). As Rowe let the these curves are obtained from tests of
anchors yield, he found no arching in the rather small models and refl.ect purely labo-
sand located between the anchor and dredge ratory (stable) conditions; therefore, they
level where the sheet pile shows its maxi- must be used with caution in the design of
mum defl.ection. Rowe concluded that the real fl.exible walls.
moment reduction is a result of sheet-pile Since 1952, Rowe has published a num-
defl.ection below the dredge line. This de- ber of papers in which he proposed a method
fl.ection resulted in a distribution of the soil for analysis of fl.exible sheet-pile walls.
pressures below the dredge level that was These theoretical analysis show good agree-
much different than the distribution envis- ment with his 1952 model tests (Rowe,
aged in the classical soil pressure theories. 1955a, 1955b, 1955c, 1956, 1957a, 1957b).
Because of the large outward movement Dismuke (1991), in his comments on the
near the dredge level, the soil pressure on results of Rowe investigations, pointed out
the embedded part of the wall just below that for a given tie-rod level, the reduction
dredge level was higher than predicted by in bending moment in the sheet-piling is
the classical theories, and, due to the due chiefl.y to the raising of the point of
smaller outward movement at the sheeting contrafl.exure in the sheeting that occurs as
toe, the soil pressure on the back of the wall the result of the sheet pile's fl.exibility and
at the bottom was also higher than pre- relative soil compressibility. The latter is
dicted by the classical theories. Hence, in- illustrated in Figure 6-48; the sheeting is
creased pressure on the front of the wall crudely modeled as a beam supported at the
just below dredge level and on the back of top and bottom. As is seen in this figure, the
the sheeting at its bottom together pro- fl.exible beam carries load W by rotating
duced a moment which tends to consider- about a curved rigid support B, effectively
ably reduce the bending moment in the reducing the span relative to a stiff beam
sheet-pile zone between the tie-rod and the loaded by an identicalload W, and rotating
dredge level. The magnitude of the moment about a similar rigid support B. Naturally,
reduction increased with the defl.ection of a smaller effective span of the fl.exible beam
the sheet piles and was therefore greater results in a smaller bending moment. Ac-
for fl.exible walls than for stiff walls. cording to Dismuke, changes in the effec-
612 Sheet-Pile Bulkheads

1.0

.9 \. \
~~
1"
" "
15PENETRATION INMEOIUM(~f CTANO iOMPACT
COARSE GRAINEO SOILS
~ ~ 1-'
~

~ r<: t'to..
t' r--. 1'........
""" ............
:?_: ~ ._ r--..... r--.
!""""

PE ET ltf'Tk!! I~E ~R .~M,!.E.


~
- r-- """"'r- 1- ~
~~EG~AIN P ~ILSI ~~
.3
-
6 7 8 10 12 14 16 18 20 30 40 ro 60 ro eo 100 1!50 200

p = l~ţ0 l 4 IN t~
2
8 j PER RUNNING FOOT OF WALL

Figure 6-46. Bending moment reduction with increase in sheet-pile fiexibility based on Rowe's
(1952) experiments. [From NAVFAC DM-7.2, 1982).] MDES =design bending moment; Mmax =
maximum bending moment as determined by classical theory of soil pressure distribution.

EL.IS.S
R ~~:!:.:..!!:L_ EL.I4.S

R4 •12.4 ~~/~m
R1•11.2 ~o/dm
R1 • 8.55 ~t/d•
R1•8.4 ~t/dm

y •1.84
~
EL.O.O EL.O.O

o...._ _.....,••• 10 10 10 40 10 kQ.dlft


liiil
(a) (b) (c)

Figure 6-47. Typical relationship among defiections, bending moments, and soil pressure
obtained from large-scale-model tests by Lasebnik (1961); 1, 2, 3-defiection of sheet piles with
fiexibility coefficient ( p) equal to 0.55, 2.12, and 8.2, respectively; 4, 5, 6-bending moment
diagrams in sheet piles with p, correspondingly equal to 0.55, 2.2, and 8.2, respectively;
7-bending moment diagram obtained by the classical method; 8, 9-soil pressures on sheet
piles with p equal to 2.12 and 8.2, respectively; 10, 11, 12, 13, 14-active and passive soil
Pressure diagrams obtained by classical theory.

tive span length by only 5% would result in hensive large-scale-model tests carried
bending moment change of about 16%. out by Lasebnik (1961). Reviewing Rmye's
The next important contributions to bet- findings and the test results carried out by
ter understanding the phenomenon of Tschebotarioff and other investigators,
soil-sheet-pile wall interaction were the Terzaghi proposed a design procedure in
paper by Terzaghi (1954) and the compre- which he recommended the use of Rowe's
Sheet-Pile Bulkheads 613

bending moment reduction factors when the


sheet piles are to be driven into fairly ho-
mogeneous clean sand; smaller reduction
factors were proposed for use in silty sand,
and in loose silty sand, silt, and clay no 1
Span CA 1 ..,
moment reduction should be used. Span DA
Terzaghi pointed out that "no bulkhead
theory can possibly anticipate all the vari- c
eties of subsoil and hydraulic conditions o
that may be encountered in practice, and
every case requires a certain amount of CA<OA
independent judgment." He also noted that
(a) (b)
the designer should not follow the rigid de-
sign rules but rather base his or her judg- Figure 6-48. Analogy to show effect of fiexibility
ments on the best assumptions on subsoil under working conditions: (a) fiexible beam and (b)
properties. When subsoil properties are in stiffbeam. [From Dismuke (1991).]
question or uncertain, Terzaghi concluded
that the sheet-pile embedment should be
designed with generous allowances. reaction force. Reduction of bending mo-
ments is especially drastic in the range of
Lasebnik (1961) reported results of a
p = 0.2 to 6. A further increase in bulk-
comprehensive large-scale-model test on a head flexibility has no practical effect on
sheet-pile bulkhead. A total of 500 tests the reduction in bending moments. An-
have been carried out on sheet piles of dif- chor forces in rigid walls ( p = 0.55 to 1.0),
ferent flexibilities. with nonyielding anchorages, could be
A flexibility criteria established by 30-40% greater than those determined
Lasebnik was defined as p = H 4 jEI, where by Coulomb's theory.
H is the length ofthe sheet-piling and EI is 2. The total active pressure exerted on a
the wall stiffness. He found that in dense flexible wall is 25-30% smaller than that
sand, the sheet-pile embedment does not acting against a rigid nonyielding wall.
significantly effect the sheet-pile wall flexi- 3. The shape of the passive pressure dia-
bility criteria, and, therefore, that in cases gram is similar to that obtained by
when a dense foundation material exists, Tschebotarioff (1948) and Rowe (1952).
the wall's height should only be considered The triangular passive pressure diagram,
for establishing the wall flexibility. similar to that obtained by Coulomb's
Typical deflections, bending moments, theory, was found only in the case of rigid
and soil pressure curves for sheet piles with sheet piles driven into loose sand.
the flexibility coefficients ( p) equal to 0.55, 4. In an anchored sheet-pile wall the values
2.12, and 8.2, respectively, obtained by of bending moments and anchor forces
Lasebnik are illustrated in Figure 6-47. depend greatly on the yield capabilities of
Not surprisingly, results of Lasebnik's the anchor system, especially in bulk-
investigations are in good agreement with heads driven into dense soil; the anchor
yield decreases the anchor force and the
tests carried out earlier by Tschebotarioff
bending moment at bulkhead mid-span,
(1948) and Rowe (1952). On the hasis of his
and increases the bending moment at the
observations, Lasebnik concluded the fol- point of the sheet piles fixity into the soil.
lowing: In Lasebnik's experiments, the wall de-
flection, !::., varied from 0.01Hw to
1. The bulkhead flexibility ( p) has a dra- 0.0005Hw, where Hw is the wall height.
matic effect on the bending moment and This was based on Schekhiev formula
614 Sheet-Pile Bulkheads

(Gurevich, 1969) t1 = t1 0 a, where 6. 0 is pending on the sheet pile's flexibility ( p),


the elastic elongation of the tie-rod, and yield of anchor system (ti), and density of
a = 2 to 5 depending on the type of foundation soil. Added roughness has no
anchor. practical effect on value and redistribu-
5. The difference in bending moments be- tion of the active soil pressure on a sheet
tween "backfilled" and "dredged" bulk- pile.
heads (d 1 jd 2 = 2.3 to 2.5) could reach
25-30% (d 1 and d 2 are defined in Figure Recommendations made by Lasebnik are
6-5). Typically, the anchor force in a summarized in Table 6~5. AII structural
"dredged" bulkhead is 10-20% larger schemes depicted in Table 6-5 are classi-
than in a "backfilled" one. :fied in accordance with their respective
6. In an anchored sheet-pile wall system the fiexibilities: "rigid" ( p = 0.55 to 1.0),
anchor was placed at different heights "medium rigid" ( p = 2 to 4), and "fiexible"
from the top of wall that varied from O to ( p ;;:::: 5.5), and they are related to the "free
0.25Hw. In all ofthese tests, the soil pres- earth support" (F.E.S.) and ":fixed earth
sures against the cantilever portion never support" (Fx.E.S.) conditions. De:finitions
reached values of passive pressures. and discussions on F.E.S. and Fx.E.8. are
7. The wall friction (8) value changes de- given later in this chapter.

Table 6-5. Summary oflarge-scale-model tests carried out by Lasebnik (1961)

BULKHEAD
MEDIUM RIDGIDITY FLEXIBLE
2<P=~<4 P=rr;;.:.
H4
55
F. E.S.
Sheet-Pile Bulkheads 615

A comprehensive model study of sheet- Results of comprehensive field testing of


pile wall performance has been carried out anchored sheet-pile bulkheads have been
by Dubrova (1963). Dubrova's observations reported by Gurevich (1969). This study was
basically confirmed the results of the earlier carried out during construction of the river
discussed observations by other workers. port of Ust Donetsk, Russia, in 1958-1961
Furthermore, Dubrova pointed out that ver- on two test sites. The anchored bulkhead on
tical soil pressure distribution is sensitive Site 1 (length 56.2 m) was comprised of
to very small wall displacements [e.g., T-shaped, prestressed reinforced concrete
(1j7000)Hwl· He also concluded that where sheet piles, 15 m long and 155 cm wide,
some wall displacements occur at anchor with a web 65 cm deep. Sheet piles were
levels, the total active soil pressures on jetted to a depth of 5.5 m into a sand foun-
flexible walls is 10-20% smaller than that dation. According to Lasebnik's criteria,
obtained by classical theory. Further obser- these piles are defined as rigid ( p = 1.1).
vations indicated that any wall displace- Each sheet pile was anchored by a dedi-
ments result in an increased angle of wall cated anchor system consisting of 18-m-long,
friction, o, which in turn increases the ver- 75-mm-diameter steel tie-rods secured at a
tical component of soillateral pressure. prefabricated 3.5-m-high reinforced con-
The importance of model tests is well crete wall. Tie-rods were placed horizon-
recognized. However, measurements that tally with a negative sag equal to 100 mm.
are carried out during model tests are typi- The height of the test wall was 9.5 m. The
structure was tested twice, in 1959 and
cally discontinued soon after the backfill
1960. The basic results of these tests are
and surcharge load have been placed. Fur-
shown in Figure 6-49.
thermore, in model tests, long-term effects
The backfilling of this structure has been
such as soil creep, tidal water fluctuations,
accomplished in the following stages. lni-
repeated loading, dynamic effects of waves
tially, still unanchored sheeting was back-
and cargo handling and hauling equipment,
filled with medium sand to a depth of about
earthquakes, chemical and physical changes
1.8-2 m. This uncompacted and, therefore,
in soil, and so on are difficult to evaluate. quite loose fill pushed the sheet piles for-
Therefore, only long-term full-scale testing ward up to 10-15 mm at the sheet-pile cope
can produce a real quantitative picture of beam level. At this point, the anchor system
sheet-pile wall performance. Unfortunately, was installed and the tie-rods were postten-
the full-scale tests cannot be as comprehen- sioned by a force of about 9-10 tonnes. This
sive as model tests and, of course, they are force pulled the top of the wall backward by
very costly. That is probably why the num- about 15-30 mm; when the compacted
ber of full-scale investigations reported to backfill reached the tie-rod level, an addi-
date is much smaller than the number of tional backward movement of about 3-5
model tests. mm occurred at the cope beam level. When
Besides the already aforementioned in the backfill was completed, the absolute
this section field investigations carried out movements of the bulkhead top was about
by Duke (1953) and Kyle and Kapp (1957), nil. This happened because the anchor sys-
results of field observations have been re- tem yielded up to 10-15 mm toward the
ported by Wiegman (1953), Casagrande and basin. After a surcharge of 5 tjm 2 was
Wilson (1953), Hakman and Buser (1961), applied, additional forward movement of
Matich et al. (1964), Gurevich (1969), about 20 mm at the top of the bulkhead
Tsinker (1964), Budin (1969), and others. occurred. At the same time, the deflection
Some of these investigations are discussed at the bulkhead mid-span has reached about
in this section. 7 mm and the sheet-pile toe moved forward
616 Sheet-Pile Bulkheads

IL .•. I
IL.U

IL.!-5

--
IL.O.O

(a) (b) (c) (d)

Figure 6-49. Port-of-Ust Donetsk, Russia. Results offield investigation of sheet-pile


bulkhead conducted at Site 1; (a) typical cross section of test wall; (b) wall defiection;
(c) measurement of soillateral pressures taken after backfill is completed up to an
elevation (EL) of 8.7 m; (d) measurement of soil pressures after backfill is completed.
la, 2a, 3a-correspondingly, concrete sheet pile, concrete anchor wall, and steel
tie-rod; lb-defiection of sheet pile after it was pulled back by posttensioning of the
tie-rod that preceded the backfilling; 2b-defiection of sheet pile after completion of
backfill to an elevation of 5.5 m; 3b-defiection of sheet pile after completion of
backfill to an elevation of 8. 7 m; 4b-defiection of sheet pile after application of
5-tjm2 surcharge load; le, 3c-measured soil pressure after completion of backfill to
EL= 8.7 m, with corresponding surcharge loads q =O and q = 5 tjm 2 ; 2c, 4c-
active pressures as determined by Coulomb theory, corresponding to q = O and q = 5
tjm 2 ; 5c-passive pressure after Sokolovski (1965) theory; ld, 2d-actual active and
passive pressures after completion of backfill to EL = 9.5 m. [From Gurevich (1969).]

by about 10 mn1. Ali these sheet-pile move- along the sheet-pile wall that is obviously
ments occurred when the water was at its quite different than that obtained by classi-
minimum level (El = 3.5 m). A year later, cal theory. The soil pressure in the wall
the bulkhead had experienced a spring flood active pressure zone at the tie-back level is
and the top of the structure deflected about substantially (2-2.5 times) greater than
5 mm toward the fiii. Sometime later in the that determined by Coulomb's theory. Rigid
same year, when portal cranes were in- sheet piles deflected relatively little at the
stalled and created a dynamic loading on mid-span. However, this was sufficient to
the granular backfill material, the top of obtain values of active pressures smaller
the bulkhead moved an additional 3-5 mm than that given by the Coulomb theory. An
toward the basin. At the same time, the increase in active pressure between eleva-
initial tie-rod sag, equal to 100 mm, was tions of O and 3.5 m is explained by initial
reduced to 40-70 mm and the anchor sys- prestressing of anchors and subsequent
tem movement toward the basin reached movement of the sheeting against loosely
15-20 mm. The wall deflections at different placed backfill. Movements of the buried
stages of back:fill placements are illustrated part of the bulkhead contributed to a de-
in Figure 6-49b. crease in active soil pressure in this area.
The aforementioned sequence of backfill Total active soil pressure was measured at
placements and associated deflections of the this stage of the test and was found to be
wall resulted in a soil pressure distribution about 90% of that determined by the
Sheet-Pile Bulkheads 617

Coulomb theory. Movement of the embed- uniform surcharge load and to fail when
ded portion of the wall appeared to be in- loaded by a surcharge load of 6 tjm 2 .
sufficient to develop the ultimate value of In accordance with these objectives, the
passive pressure against the wall. 11.15-m-long, T-shaped, prestressed rein-
Measurements of soil pressures in both forced concrete sheet piles 155 cm wide with
passive and active zones, made one year a 45-cm-deep web were jetted into the
later (1960), showed a very minor decrease medium sand to a depth of only 3.25 m.
in active pressure at mid-span, and some According to Lasebnik's rigidity criteria,
increase in soil total pressures in both ac- these sheet piles are defined as rigid ( p =
tive and passive zones. The fi.rst phe- 1.5). Each sheet pile was supported by a
nomenon is explained as a result of concrete dedicated tieback system in which the
creep. The second was explained as a possi- length ofanchor tie-rods was cut to 14.5 m
ble result of backfi.ll of soil consolidation. (i.e., 3.5 m shorter than that determined by
As pointed out by Polshin and the conventional method). The height of the
Sinelschikov (1953), the change in soil den- anchor wall was reduced from the standard
sity alone cannot signifi.cantly change the of 3.5 m to only 1.5 m. When the backfill of
total soil pressure. For example, the change medium sand was completed, the surcharge
in sand void ratio from 0.8 to 0.6 automati- of 8 tjm 2 in increments of 2.0 tjm 2 was
cally leads to an increase in the angle of applied. Furthermore, to create the most
interna! friction of up to 22%. Therefore, an unfavorable stability condition of the an-
increase in soil unit weight would be neu- chor system, a 3.6-m-high pile of sand was
tralized by a decrease in the coefficient of placed behind the anchor wall.
lateral soil pressure; thus, the total in- As it turned out, in spite of heavy over-
loading, no damage occurred to the struc-
crease in soil pressure would not exceed
ture. A thoroughly conducted inspection of
about 7%.
the wall above and below the water level
At Site 1, soil pressure redistribution
has not revealed even the smallest cracks
along sheet piles resulted in much greater
in the sheet piles. The heavy load and pres-
reaction forces transmitted to the anchor
ence of a rather loose soil in front of the
system than would be determined by
anchor wall resulted in a 40-mm anchor
Coulomb's theory; the measured tie-rod
system yield toward the basin. At the same
elongation reached about 20 mm, and the time, the sheet-pile toe moved 15 mm in the
average anchor force without a surcharge same direction, and the bulkhead deflection
load applied was about 48 tonnes. The max- at the mid-span reached 17 mm. The basic
imum measured anchor force attributed to results of this test are shown in Figure
the surcharge load of 5 tjm 2 was 56 tonnes. 6-50.
Had it been determined by the Coulomb As is seen in Figure 6-50, the wall move-
theory, the anchor force would be equal to ments resulted in active soil pressures that
about 30 tonnes, or about 1.87 times smaller significantly differed from those determined
than that actually measured. This was the by Coulomb's theory; in the wall's upper
result of both soil pressure concentration in zone, the actual soil pressure exceeds that
the wall anchor zone and the initial tension determined by classical theory by about
applied to the tie-rods as discussed earlier. 20%, and in its lower part the active pres-
A daring experiment had been carried sure tends to be almost nil. At the same
out at Site 2. The sheet-pile bulkhead (7.9 time, the passive pressure reached its al-
m high) constructed there (Fig. 6-50) was most ultimate value. Wall movements, com-
designed with no factor of safety applied. bined with wall friction effects, resulted in
This wall was designed to survive a 4-tjm2 a total reduction in lateral soil pressure by
618 Sheet-Pile Bulkheads

EL.I3.0

EL. 7.9

EL.5.75

W.L. EL. 4.5

11' lt------::;;r----'-':.:.=.._----1
2
EL.O.O 1
~= 32°
Y=I.ITtm•

o 1 2 3
T/m 2
(a) (b) (c)

Figure 6-50. Port-of-Ust Donetsk, Russia. Results of field investigation of sheet-pile bulk-
head conducted at Site 2: (a) typical cross section of test wall; (b) soil pressure diagrams; (c)
surcharge load q = 8 tjm 2 diagrams. la through 5a-correspondingly, concrete sheet pile,
concrete anchor wall, steel tie-rod, pile of sand and surcharge load q = 8 tjm 2 (timber cells
filled with sand); lb-measured active and passive pressures after completion of backfill to
EL= 7.9 m (q =O); 2b, 4b-correspondingly active pressure as determined by Coulomb's
theory for surcharge load q =O and q = 8 tjm 2 ; 3b-measured active and passive pressures
(q = 8 tjm 2 ); 5b-passive pressure after Sokolovski (1965); lc-measured pressure produced
by surcharge load and (q = 8 tjm 2 ); 2c-surcharge load (q = 8 tjm2 ) pressure diagram
determined by Coulomb's theory. [From Gurevich (1969).]

about 20% compare with that determined which they are composed and the type of
by classical theory. anchorage used.
The increase in active pressures devel- (b) Total active pressures on conventional
oped by the surcharge load was about 30% bulkheads usually do not exceed that de-
less than that determined by Coulomb's termined by Coulomb's theory.
theory, and the shape of this pressure dia- (c) The active pressure vertical distribution
gram was triangular, with its maximum along sheet-piling depends on the charac-
ordinate located at the anchor point. The ter of bulkhead defl.ection; concentrating
anchor force was about 11% larger than it at the anchor zone and in the passive
that determined by Coulomb's theory, and pressure area.
the factor of nonuniformity of measured an-
chor forces (Rmax/Raverage) was about 1.25. Tsinker (1964) carried out the :field investi-
Finally, it was concluded that the sur- gation of a sheet-pile bulkhead built at
charge load placed behind the anchor wall Port-of-Kiev, Ukraine. The structure com-
did not affect the wall's stability. prised rigid T -shaped concrete sheet piles
The basic conclusions that can be drawn anchored by raked "flexible" anchor piles
from the aforementioned observations are with a widening at their ends. Anchor
as follows: piles were driven an average distance of 2.5
m along the sheet-pile wall. The sheet-
(a) Conventional anchored sheet-pile bulk- pilejanchor pile stiffness ratio EI8 pjEIAP
heads should be considered as fl.exible was equal to about 12, where E i~ the
structures regardless of the material of modulus of elasticity of concrete, and I 8 p
Sheet-Pile Bulkheads 619

and JAP are the moments of inertia of the were monitored by special devices attached
sheet pile and anchor pile per unit length of to two sheet piles and two anchor piles. The
wall, respectively. wall was backfilled in two stages with
All piles were jetted into a medium sand. medium sand as indicated in Figure 6-51.
To the depth of about 2 m below the dredge The readings from the installed instru-
line, this sand initially was quite loose; ments were taken at two loading stages; at
however, in the process of pile-driving, it Stage 1, when the wall was backfilled to the
was densified. Sheet and anchor piles were design grade, and at Stage 2, when the
joined together by a concrete cope beam. 2.5-m-high pile of sand was placed to simu-
The test wall, 5.5 m high and 20.0 m long, late a surcharge load on the wall equal to
included three instrumented sheet piles and about 3.8-4.0 tjm 2 •
two instrumented anchor piles. The strain The wall defiections (movements) oc-
gauges were installed and calibrated on all curred almost immediately after it was
piles at the fabrication yard. About 90% of backfilled to the design grade (Fig. 6-51a).
the strain gauges installed in sheet piles At this stage, the top of the wall moved
and only about 65% of gauges installed in forward by about 9.2 mm, and at the dredge
anchored piles have survived transporta- line the wall movement away from the fiU
tion from the fabrication yard to the project was equal to about 31.5 mm. At the same
site and hard driving, and were functional time, no significant defiection in wall mid-
throughout the test. The wall deformations span was measured. The bending moment

(a)

-
o
t·m
5

(b)

Figure 6-51. Field tests carried out at Port-of-Kiev, Ukraine: (a) sequence ofwall
backfilling; (b, c) bending moments and deflections measured in sheet and anchor piles.
1-Sheet pile; 2-anchor pile; 3, 4-bending moments (in t m) and corresponding deflections
(in cm) during the first stage of wall backfilling and after application of a surcharge load of
- 4.0 tjm 2 , respectively.
620 Sheet-Pile Bulkheads

diagram for the sheet piles was similar to raked piles. The wall, 7.3 m high, was
that obtained by the "fixed earth" method constructed in 1962-1963 in Port-of-St.
(Fig. 6-51b). Despite the rigidjoint between Peterburgh, Russia. It was comprised of
sheet-piling and anchor piles, the bending standard flat concrete sheet piles and con-
moment at the cope beam level in both the crete anchor piles. The sheet pile 81 anchor
sheet and anchor piles was insignificant. pile per unit length of wall stiffness ratio
Axial forces measured in both sheet piles Elsp/EIAP = 6. Sheet piles and anchor piles
and anchor piles correlated very well. The were driven into the stratum of soft varved
axial load in anchor piles measured at five clay [ <P = 16°, c = 0.1 kgjcm 2 ] 4-5 m thick
sections along the pile indicated consistent underlaid by hard loam (</J = 27°, c = 0.2
load reduction from the pile top (fixity at kgjcm 2 ). The wall was backfilled with
cope beam) down to the pile end. For exam- medium sand ( <P = 30°). The top 3-4 m of
ple, at the pile head the measured axial the backfill was heavily densified by the
force was equal to about 38.0 tonnes, at 3.0 vibrowing method. [For details on the
m down from the pile head (zone of loose vibrowing method of soil densification, see
fiii) it was equal to about 34.0 tonnes, and Broms (1991).]
at 6.0 m from the pile head it was equal to Measurements of wall deformations,
about 27.75 tonnes. The basic findings from bending moments, and soil pressures have
the above tests are as follows: been carried out on two well-instrumented ·
sheet piles and two anchor piles. The meas-
1. The actual bending moment in the sheet- urement of the aforementioned parameters
pile wall anchored by raked piles is about were carried out after the wall was back-
10-12% smaller than that obtained by
filled up to the bottom of the cope structure,
conventional methods with an assumed
wall friction a = icf>.
then at the top of the wall, and finally after
2. The values of bending moments measured
the application of a surcharge load 4.0 tjm 2 •
in anchor piles, as well as the shape of The test results are illustrated in Figure
bending moment diagrams, indicate that 6-52.
they are the result of both the weight of The field investigations carried out by
the backfill material that ş.rches on piles Tsinker and Budin have confirmed the
and the pile rotation about the fixity point basic results of comprehensive scale-model
within foundation soil. In the sand foun- tests carried out earlier by Zarkhi (1961).
dation the fixity point is located at about The latter investigator has arrived at the
the dredge line level. following conclusions:
3. Where sheet-pile walls are anchored by
raked piles, the soil between the sheeting
and the anchor piles serves as a medium 1. The vertical redistribution of the active
to link all wall components together. The soil pressure in a sheet-pile wall anchored
total lateral soil pressure that acts on the by raked piles is less pronounced than a
wall is basically distributed between the regular anchored bulkhead. This is basi-
sheet-piling and the anchor piles accord- cally attributed to wall rotation about its
ing to their stiffnesses. point offixity in foundation soil away from
4. The granular backfill material hangs on the fill and the shielding effect of anchor
the anchor piles due to the arching effect. piles.
This phenomenon is responsible for the 2. The earth pressure on the coping struc-
reduction in axialload in anchor pile along ture is close to that obtained by the
its shaft from the pile head down to the Coulomb's theory; between the dredge line
lower end. and anchor point the active pressure is
somewhat less than that obtained by
Budin (1969) reported results of field Coulomb's theory, and down below, it is
tests of sheet-pile bulkhead anchored by somewhat higher than obtained by the
Sheet-Pile Bulkheads 621

J,O

(a) (b)

Figure 6-52. Field test carried out at Port-of-St. Peter-


burg, Russia: (a, b) bending moments (in t m) and soil
pressures (in tjm2 ) measured in sheet and anchor piles.
1-Bending moments measured after the wall was back-
filled up to the top of the sheet-pile level; 2-bending
moments and active and passive pressures measured after
the wall was completely backfilled and loaded with a
surcharge of - 4.0 tjm 2 • [From Budin (1969).]

Coulomb's theory. This phenomenon is a It follows from the preceding discussion


result of a load shielding effect produced that the crucial issue in the soil-anchored
by the anchor piles. The latter reduces sheet-pile wall interaction is to what extent
the soil pressure in the wall span area the maximum bending moment in a sheet-
and transfers the extra load to the em-
ing and the wall reaction force are affected
bedded wall part.
by .soil "arching."
3. The change in anchor pile batter from The definition and explanation of the soil
1 : 1 to 1 : 3 naturally results in increase mechanism is given by Terzaghi (1943) in
in axial loads in both sheet piles and
his classic work Theoretical Soil Mechanics
anchor piles; however, the bending mo-
ments in both piles become smaller.
as follows:
Zarkhi concluded that the wall with an
anchor pile driven at 1 : 3 slope needs If one part of the support of a mass of soil
about 8-10% less material than one with yields while the remainder stays in place the
anchor piles having an inclination 1 : 1. soil adjoining the yielding part moves out of
4. The large shear force that is acting in the its original position between adjacent station-
anchor pile increase the pile's carrying ary masses of soil. The relative movement
capacity by 20-50%. This conclusion was within the soil is opposed by a shearing resis-
confirmed by tests carried out by Tsinker tance within the zone of contact between the
(1972). yielding and the stationary masses. Since the
shearing resistance tends to keep the yielding
mass in its original position it reduces the
For more discussions on results of sheet- pressure on the yielding part of the support
pile wall investigations, the reader is re- and increases the pressure on the adjoining
ferred to Gurevich (1969), Tschebotarioff stationary part. This transfer of pressure from
(1973), Casagrande (1973), and other inves- a yielding mass of soil onto adjoining station-
tigators. ary parts is commonly called the arching ef-
622 Sheet-Pile Bulkheads

fect and the soil is said to arch over the Rowe's tests provided quantitive informa-
yielding part of the support. Arching also tion on the amount of moment reduction due
takes place if one part of a yielding support to each of these effects, and on the conditions
moves out more than the adjoining parts. under which these effects may be relied on.
His model tests indicated that flexure below
From this statement and from the detailed the dredge level is the most stable and reli-
discussion on soil arching phenomenon able source of moment reduction. The results
given in Chapter 4 it follows that arching showed, however, that flexure above the an-
basically involves two parts: a reduction of chor level could be reduced or destroyed by
additional anchor yield after dredging, and
the earth pressure on a yielding portion of a
thus cannot be counted on for all conditions.
structure, and an increase of the earth The tests also showed that the moment reduc-
pressure on the adjoining portions. tion due to pressure reduction between the
Basically, these two effects are responsi- anchor and dredge levels was unstable and
ble for the soil pressure redistribution which could be destroyed by anchor yield and back-
occurs around a sheet-pile wall when it de- fill settlement, except in the case of piles
f:lects. In conjunction with aforementioned, encastre at the anchorage.
Bjerrum et al. (1972) made the following Rowe's model tests thus indicate that the
observations: type of behavior visualized in the Danish
Rules and discussed at the Rotterdam Confer-
... , the earth pressure on both the back and ence in 1948-reduced earth pressure be-
the front of the wall are changed from the tween the dredge and anchor levels due to
classical free earth support distribution as a arching-is in most cases a relatively unim-
result of the wall deflections. On the back of portant factor in the behavior of sheet-pile
the wall, the pressure at the top and the walls. Far more important in most cases are
bottom are increased due to the smaller out- the effects of flexure below the dredge level
ward deflections at these points, while the and flexure above the anchor level, which
pressure between the dredge and anchor lev- together accounted for most of the moment
els tends to be reduced due to the relatively recaption in Rowe's tests.
large outward deflection there. It can thus be appreciated that while the
These changes in earth pressure are re- type of arching considered in 1948 1 has been
sponsible for the three sources of moment found by Rowe to be not so important, the
reduction which Rowe found in his model more important causes of moment reduction
tests: The increased pressures on the back of identified in his tests are in fact also due to
the wall at the bottom and on the front of the arching. They arise from the increases in
wall just below dredge level together produce earth pressure on relatively. unyielding por-
a clockwise moment which reduces the maxi- tions of the wall. Thus, while the considera-
mum moment in the wall, the effect which tion exclusively to the reduction in earth
Rowe called "flexure below the dredge level". pressure on the more yielding part of the
The increased pressure acting on the part of wall, in actual fact the most important effects
the wall above the anchor produces a counter- of arching are due to the increase in earth
clockwise moment which also reduces the pressure on the less yielding parts of the wall.
maximum moment in the wall, the effect
which Rowe called "flexure above the anchor
level". Finally, the reduced pressure on the Furthermore, with reference made to a
back of the wall between dredge and anchor number of field observations Bjerrum et al.
levels also reduces the maximum moment in (1972) pointed out that the amount of soil
the wall, the effect which Rowe called simply arching is controlled by the def:lected shape
"arching''. of the wall defined by the def:lection ratio

1 European Conference on Soil Mechanics and Foundation Engineering in Rotterdam, 1948.


Sheet-Pile Bulkheads 623

(relative deflection). Typical values of de- rest" condition. The aforementioned phe-
flection ratios were found to be on the order nomenon can be partly explained by the
of 0.2-1.0%. For a comparison, the ratio of presence of the so-called "heredity effect."
the wall anchor yield to the height of the The nature of this effect is that live loads
wall is typically 0.1-0.2%. From the previ- that may act on a structure can develop
ous discussion it is obviou~ that the effects residual stresses in the bulkhead compo-
of soil arching on the soil lateral pressure nents. This happens because the forces and
depend on the wall construction sequence deformations developed in the sheet-pile
(e.g., "dredged" and "back:field" bulkheads). bulkhead, due to active pressure produced
In the former, the considerable reduction in by the live loads, are considerably smaller
soil pressure in the zone between the dredge than the forces required to restore the wall
line and the anchor line can be expected. to its original position against the passive
Subsequently, a considerable increase in soil resistance of the retained material. The
pressure can occur at the bottom of the wall recognition of the "heredity" effect may
and in the zone above the anchor. In the sometimes be a key to the proper under-
case of the "back:filled" wall the arching standing of the results of some :field obser-
phenomenon is less pronounced and can be vations and help to avoid misjudgement of
effective only when :fill is placed above the results of some important tests. For a de-
anchor. tailed discussion on the "heredity effect,"
It should be noted that since the begin- consult Chapter 4. Last, but not least, it
ning of this century, the phenomenon of soil must be understood that the soil parame-
arching has been the subject of many ters and, subsequently, the soillateral pres-
theoretical studies. In recent years severa! sures may be greatly influenced by the
attempts have been made to use finite ele- time-dependent factors such as consolida-
ment techniques to develop a mathematical tion changes in ambient temperature, rheo-
model of soil arching. logical properties of foundation soil, chemi-
In conclusion of this section, it must be cal changes in back:fill and foundation soil,
pointed out that any reported results on and others.
sheet pile-wall investigation must be treated
with caution. For example, test results ob-
tained by means of scale model, and partic- 6.7 DESIGN OF SHEET-PILE
ularly by small models, can be subject to
misinterpretation in terms of their applica-
WALLS
bility to the full-scale bulkheads.
Furthermore, the results of measure-
ments made on real structures several years 6.7.1 Design Criteria
after the wall construction can be equally
misleading. For example, Casagrande Typical design criteria for sheet-pile bulk-
(1973), making reference to several :field head design will routinely include informa-
tests, pointed out that in sever al of the :field tion on site geotechnical conditions, para-
tests the measured soil pressure on flexible meters of the proposed back:fill material,
anchored walls after several years in ser- details on hydraulic conditions (e.g., tides,
vice was appreciably greater than the waves, currents, ground water), and infor-
active pressure and corresponded approxi- mation on live and dead loads.
mately to pressure in the "at rest" condi- The aforementioned information should
tion. The latter fact, however, does not nec- be supplemented with a survey of the con-
essarily mean that flexible sheet-pile walls struction site with an indication of the loca-
must be designed for the pressure in the "at tion of the boreholes drilled at the site.
624 Sheet-Pile Bulkheads

Before any design calculations can be initi- the active earth pressure behind the wall
ated, controlling parameters such as eleva- and decreases passive earth pressure in
tions of the dredge line and top of the wall, front of the wall. The latter effects may
maximum, minimum, and mean tide levels result in a quick-sand condition developing
should be established. At this stage, poten- in front of the sheeting resulting in a tem-
tial overdredging, scour, or both in front of porary loss of a part or all soil passive
the wall must be considered. The latter is resistance. In such a case, both the wall toe
dependent on the construction tolerances kickout and a vertical dropping of the
that are normally related to the particular sheeting may occur.
dredging equipment, the presence of strong Three basic scenarios of wall loading are
currents, as well as effects attributed to depicted in Figure 6-53. The wall loading
ship operation (e.g., water jets produced by as indicated in Figure 6-53a at the mini-
ship propellers andjor side thrusters). mum water level will result in maximum
N ext, the normal and extreme design
reaction force in the tieback system and, at
loads and load combinations are to be iden-
the same time, will result in a higher resis-
tified and formulated. The design loads must
tance of the anchor wall (plate). The stabil-
be placed in the most adverse, but practica!,
ity of the anchor system must also be
combinations. This should include consider-
ation of a thorough evaluation of the poten- reviewed at the maximum water level be-
tial for development of unbalanced hydro- cause of substantial reduction in soil pas-
static pressure behind the wall. It should be sive pressure on the anchor wall. The wall
pointed out that sometimes the rapidly de- loading as indicated in Figure 6-53b should
veloped and unusually low water level in be used to determine the maximum value of
front of a sheeting due to tides, wind effect, bending moment in sheet-piling. Where
or severe storm waves while the ground- waves are the factor that may substantially
water level behind the sheeting remains affect the value of the hydrostatic pressure
substantially higher may product signifi- behind the wall, the design scheme indi-
cant unbalanced hydrostatic pressures on cated in Figure 6-53c may govem the wall
the sheeting. design.
If foundation soils are granular, the high It must be noted that each project is
hydrostatic pressure will coincide with site-specific, and, therefore, potential load
seepage under the sheeting that increases combinations must be given proper atten-

(a) (b) (c)

Figure 6-53. Anchored sheet-pile wall: Typical load combination: (a) for computing maximum
reaction force, Ra, maximum bending moment in upper part ofthe wall (above the anchor point),
and analysis of anchor plate stability; (b) for computing of the maximum bending moment in the
wall; (c) for determining the maximum hydrostatic load on the sheet-piling and its effect on
bulkhead strength and stability.
Sheet-Pile Bulkheads 625

tion and not be limited to those indicated in The nonhomogeneity of soils results in
Figure 6-53. uncertainly in the properties of the backfill
For information on hydraulic resistance and foundation soils, as well as uncertainly
of steel sheet-pile joints, the reader is re- in the mechanism of the soil-structure in-
ferred to the work of Sellmeijer et al. (1995). teraction. This presents problems in com-
The design criteria must also include the puting the soil pressure not only against a
basic foundation and backfill soil parame- flexible wall but especially against an an-
ters such as angle of interna! friction, </>, chored sheet-pile wall. This is probably why
cohesion, c, wall friction, 8, soil unit weight the aforementioned problem cannot be
in dry, saturated and submerged condi- solved with mathematical precision and why
tions, 'Y, as well as coefficients of active one should only expect to obtain results of
pressure K A, passive pressure, K P, and, approximate accuracy from computation.
where required, the "at rest," soil pressure, Sometimes attempts to make "precise"
K 0 . All of the above data are required for mathematical evaluations of an anchored
computing the soil pressure on a wall. For flexible wall can result in misleading and
basic information on soil pressure calcula- dangerous conclusions. Despite various
tions, the reader is referred to Chapter 4. theories that have been suggested to derive
Last, but not least, the wall design crite- analytical solutions for the bending mo-
ria should include information on proposed ments, reaction force, and the sheet-pile
structural materials, safety factors, and al- penetration, most practitioners still rely
lowable stresses for both normal and ex- heavily on good judgment and experience.
treme load conditions.
The evaluation and formulation of the
design loads and load combinations are fol-
lowed by computation of the required 6.7.2 Design of cantilever
sheet-pile penetration, the maximum bend- Walls
ing moments, shear and anchor forces in
the wall structural components, computa- A cantilever wall is one which must be
tion of the required section modulus and driven to a sufficient depth into the founda-
selection of the appropriate section of sheet tion soil to become fixed as a vertical can-
piles, and the design of the anchor system. tilever to resist the lateral active soil
In computing the soil pressures on flexi- pressure. Depending on the sheet-piling
ble walls, one must always realize that even material (e.g., wood, steel, or concrete), cross
though all theories of soil pressures on a section, and quality of backfill soil, this type
sheet-pile wall developed to date sometimes of the wall may be suitable and economica!
differ from each other in the fundamentals for heights up to 6-7 m. According to Head
involved, they usually result in more or less and Wynne (1985) 75% of all temporary
credible estimates of the loads exerted on retaining walls built in the United Kingdom
flexible walls by soils, if soil engineering are the cantilever-type bulkheads.
parameters are well defined. It must be The lateral support of a cantilever wall
remembered, however, that natural soil de- depends solely on the passive pressure ex-
posits are usually quite complex; therefore, erted on its embedded part. To avoid the
the design geological profiles at the wall excessive stresses and severe yield due to
location and the soil engineering properties potential scour, erosion, or overdredge in
must be based on reliable information ob- front of the wall, a sheet-piling penetration
tained from an adequate number of ex- depth into the foundation soil should be
ploratory borings and laboratory tests of sufficient to account for the increase in the
representative samples. freestanding part ofthe wall (e.g., the height
626 Sheet-Pile Bulkheads

of the piling above the dredge line) and the H, and a preliminary estimation of the
decrease in its penetration. depth of penetration, D. In a very loose
An early design method for cantilever sand, the wall penetration may be as-
walls based on experimental observations sumed to be about 2H, and in a very
was introduced by Engels (1903), who as- dense sand or clay, tobe about 0.75H
sumed a parabolic distribution of passive 2. Compute the active and passive soil lat-
earth pressure along the depth of penetra- eral pressures. The earth active pressure
tion in equilibrium with active earth pres- is usually computed assuming the angle
of wall friction 8 = (0.5 to 0.75)cp. With
sure due to the retained soil. In this method,
reference roade to Sokolovski's (1965)
the depth of penetration was evaluated by theory, Budin and Demina (1979) stated
requiring an equilibrium ofhorizontal forces that the angle of wall friction in passive
and moments; the failure of the wall under zone can be taken as cp, but not greater
the applied bending moments was recog- than 30°, and in the active zone, 8 = c/J/2
nized as a limited state. can be considered. In practice, however,
Subsequently, several of the limit equi- the value of 8 for calculating the coeffi-
librium design methods to analyze can- cient of passive pressure K p greater than
tilever sheet-pile walls have been developed ~ cp is very seldom used.
with respect to the shape of soil pressure
distribution along the sheeting and the The typical earth pressure diagrams . for
method of computing soil pressures (Krey, sheet-piling driven into the granular and
1932; Blum, 1931; Rowe, 1951; Hansen, cohesive soils are depicted in Figures 6-54a
1953a; Clayton and Milititsky, 1986; and and 6-54b. The hasis of these soil pressure
others). Most of these methods are identica! diagrams is that under the action of active
and are differentiated from each other only soil pressure the wall tends to rotate about
by assumed de:finitions for the safety factors point O, located close to the sheet-pile lower
and the values of wall friction, 8, which are end. This results in the development of soil
used for determination of the earth lateral passive pressures on both sides o.f the em-
pressures. bedded portion of the wall, as indicated in
Bica and Clayton (1989 and 1992) offered Figures 6-54a and 6-54b. One must real-
a comprehensive state-of-the-art review of ize, however, that in reality the soil pres-
design methods for free embedded can- sure diagrams have a more complicated
tilever walls installed in granular soils. The shape, and straight lines used to outline the
latter investigators pointed out that this soil pressure intensities are used only to
type of wall lacks suf:ficient experimental simplify the design process.
data; and therefore, more studies, particu- The areas of soil pressure diagrams must
larly large-scale-model tests and :field obser- comply with the requirements of static
vations, are required to better understand equilibrium; the sum of forces in horizontal
the soil-cantilever wall interaction. and vertical directions and the sum of bend-
Usually, the cantilever wall is designed ing moments about any point in the wall
in accordance with principles discussed ear- must be equal to zero. As stated earlier, in
lier in this chapter and the recommenda- most practica! design cases, simpli:fied soil
tions given in Chapter 4. The conventional pressure diagrams are used, and, normally,
design procedure include the following the passive pressure diagram on the back
steps: side of the piling is substituted by a concen-
trated equivalent force E~ (Fig. 6-55). This
1. Establish the preliminary wall geometry. approach, although usually resulting in
This should include a determination of some theoretical inaccuracies, simpli:fies the
the required freestanding part of the wall, calculations. The wall penetration, D 0 , must
Sheet-Pile Bulkheads 627

(a) (b)

Figure 6-54. Cantilever sheet-pile wall: soil pressure diagrams: (a) in granular soils and (b) in
cohesive soils (undrained condition) with granular backfill material. a and p are the ordinates of
the total active and passive pressures, respectively, and p' is the ordinate ofpassive pressure on
the back site ofthe sheeting; p' = a(KpjKA)- yDKA and S = [KpD 2 - KiH + D) 2 ] /
(Kp - KAXH + 2D).

P-•
p •

w ~

Figure 6-55. Cantilever sheet-pile wall: simplified resultant soil pressures diagrams: (a) in
granular soils and (b) in cohesive soils. Notation is indicated in Figure 6-54. ·

satisfy the aforementioned equilibrium re- 3. Compute the maximum bending moment
quirements. at the point of shear force equal to zero,
The computed penetration depth, D 0 , is prior to increasing the depth of embed-
usually increased by 20-40% [e.g., D = ment.
D 0 (1.2 to 1.4) that provides a safety factor
of approximately 1.5-2.0. Alternatively, the The design of a sheet-pile wall embedded in
reduced value of coefficient of the passive cohesive soils is complicated by the fact
earth pressure can be used. that the strength of these soils may change
628 Sheet-Pile Bulkheads

with time due to their rheological proper- driven with a negative slope to account for
ties; accordingly, the lateral soil pressure future movements toward the basin.
may also change. Therefore, the depth of
embedded part and the strength of the pil- 6.7.2.1 Graphical Method
ing must satisfy the pressure conditions
that exist immediately after sheet-pile in- The advantage of this method, particu-
stallation (undrained strength condition; larly in cases of complex or irregular load-
cp =O) as well as the long-term conditions ing, is its simplicity and fairly rapid cal-
after the strength (parameters cp and c) of culations. Furthermore, the results are
the cohesive material has changed. In the presented in the form of simple and clear
former case, it is assumed that the soil sketches from which the necessary design
derives all its strength from cohesion, and parameters (e.g., bending moments and
no strength from internal friction is consid- depth of penetration) can be determined by
ered. At this stage, the analysis is usually simple measurements. It should be noted
carried out in terms of total stress using a that the accuracy of results obtained by
cohesive value, c, equal to 50% of the con- graphical analysis is usually of greater pre-
fined compressive strength, qu = 2c (cp = O) cision than soils engineering -parameters
(Teng, 1962). used for the computation; therefore, this
Where relevant, the long-term conditions method is regarded as sufficiently accurate
for sheet-piling in cohesive soils must also to serve the purpose.
be considered. In this case, the analysis The soil pressure on the wall is computed
should be carried out using effective stress as discussed earlier in this chapter and in
parameters c 1 and cp 1 obtained from the Chapter 4. Again, because the theoretical
consolidated drained tests, or from consoli- depth of sheet-pile penetration at the begin-
dated undrained tests in which the pore ning of calculations is unknown, an arbi-
pressure measurements are made. Under trary initial depth must be chosen, as dis-
the drained condition, the value of c is very cussed earlier.
small and, in practice, is assumed to be The maximum bending moment and the
equal to zero; for cohesive soils, the value of anchor force are then determined by the
cp usually varies between 20° and 30°. following steps (Fig. 6-56): ·
Because the lateral pressure in cohesive
soils may approach that in granular soils (a) The soil pressure diagram is divided into
where c = O over a long period of time, the a number of approximately equal panels
long-term conditions may be analyzed as that are replaced by corresponding equiv-
prescribed for granular soils. It must be alent force vectors, acting at the gravity
understood, however, that in some cases centers of these panels. For further sim-
(e.g., soft clay or the like) the wall deflection plification, the force vectors may be as-
due to soil creep can be too large and there- sumed as acting through the center of
trapezoidal sections (Figs. 6-56a and
fore unacceptable. 6-56b). In general, the resulting accuracy
A detailed discussion on performance of associated with such an approximation is
sheet-pile walls driven into cohesive soils is compatible with the accuracy of graphical
given in Section 6.8. A rough estimate ofthe design method.
wall lateral deflection may be obtained by
(b) Once the equivalent forces are deter-
considering it as a simple cantilever beam mined, then the vector diagram and string
to be held rigidly at the point with maxi- polygon (defiected line) are constructed
mum bending moment and subjected to the (Figs. 6-56c and 6-56f) as follows. On a
design soil pressures on both sides of the horizontal base line commencing at the
wall. When required, the sheet piles can be right, the successive force vectors repre-
Sheet-Pile Bulkheads 629

(a) (b) (c) (d) (e)


p

,s
E'p
.J
(f)

Figure 6-56. Cantilever sheet-pile wall: graphic design method: (a) soil pressures
diagram; (b) load vectors; (c) string polygon; (d) moment vectors; (e) wall defiection;
(D vector diagram.

senting soil active pressure are laid. Cor- diagram is drawn. Both lines intersect at
respondingly, the successive force vectors point 1. From this point, the line 1-2 is
representing soil passive pressure are laid, drawn parallel to line p1 in the vector
commencing from the end of the line rep- diagram; the latter line is intersecting ac-
resenting the active soil pressure, but in tion line of load vector "2" at point 2. The
the opposite direction. process is continued through all sections,
To avoid confusion between base lines including both active and passive pres-
representing active and passive pres- sure lines. This results in the elastic line
sures, they may be conveniently sepa- that is none other than the funicular poly-
rated as indicated in Figure 6-56f. gon of moments of the beam that repre-
The pole, p, is then selected at a dis- sents the sheeting.
tance from the base line; it should be such
Finally, the base line po is drawn
to give a convenient curvature of the elas-
through point 1 in the string polygon. This
tic line included in the string polygon.
line, which is parallel to line po in the
From pole, p', lines are drawn to the ends
of all the active and passive load vectors. vector diagram, intersects the elastic line
at point 9. The position of point 9 deter-
(c) Next, the wall elastic (bending moment) mines line of action of force E~ and the
line is drawn as follows (Fig. 6-56c): theoretical value of wall penetration, D 0 •
Starting at the top line that represents
action line of the load vector "1," the line
that is parallel to line po in the vector The value of the maximum bending mo-
630 Sheet-Pile Bulkheads

ment is then computed as follows: The discussed design procedure was com-
puterized by DeNatale and Ibarra-Encinas
(6-5) (1992), who proposed a computer program
that solves the generalized equation of a
where wall stability for the required penetration
and maximum bending moment.
p = the pole distance Most recently, King (1995) proposed a
Zmax = maximum ordinate obtained from the revised method for the analysis of a can-
string polygon
tilever sheet-pile wall driven into cohesion-
less soil. The proposed method is based on
The required sheet-pile penetration is de- the current limit-state design method; how-
termined as D = D 0 + llD, where the addi- ever, it takes into account soil pressure dis-
tional penetration depth, llD, is computed tribution in the passive zero that occurs due
from to plastic deformations in soil mass. Accord-
E' . ing to King, the proposed method produces
t::..D = P (6-6) results that are close to those observed in
2q(Kp- KA) the centrifuge model tests.

where
E~ = passive pressure acting at the
6.7.3 Design of Anchored
sheeting lower part (obtained from Bulkheads
Fig. 6-560
q = vertical stress on the back side of By definition, anchored sheet-pile bulk-
the wall at the level of force E~; heads are walls that derive their support by
q = ajKA means of soil passive pressure on the front
K A and K P = coefficients of soil active and pas- of the embedded portion of the sheeting and
sive lateral pressures, respec- anchor(s) near the top of the sheeting.
tively Figure 6-57 illustrates the general rela-
tionship between the sheet-piling penetra-
As stated earlier, the required penetration, tion and the wall defl.ected shape as a func-
D, with sufficient accuracy and reliability tion of soil compressibility and strength as
may be determined as equal to D = (1.2 to well as lateral soil pressure distribution. In
1.4)D 0 • case (a), the passive pressure that the earth
The wall def:lection can be obtained by exerts on the embedded part of a sheeting is
the graphic statics procedure as illustrated insufficient to prevent the wall from lateral
in Figures 6-56d and 6-56e. It is similar to displacement and rotation about the point
that used for computing bending moments. of anchorage. This results in a displacement
For this, the string polygon converted into of the wall toe by the value /1. Cases (b), (c),
the bending moment diagram is used as the and (d) illustrate the effect of increased
load diagram for constructing the moment depth of penetration. In cases (b) and (c),
vector diagram (Fig. 6-56d) and the string the passive pressure has increased enough
polygon (not shown in Figure 6-56). Fi- to prevent the wall displacement at point
nally, the wall def:lected line is drawn as 'c', however, wall rotation about the anchor-
depicted in Figure 6-56e. A more detailed age point and point c continue. In case (d)
description of the procedure is given in passive pressures have sufficiently devel-
Terzaghi (1943). It should be noted, how- oped on both sides of the wall to prevent
ever, that this procedure is quite laborious both lateral displacement and rotation at
and, in practice, not often used. point C.
Sheet-Pile Bulkheads 631

c
(a) (b) (c) (d)

Figure 6-57. Anchored sheet-pile wall: effects of sheet-pile penetration on soil pressures
distribution and deflected shape of the wall.

From the aforementioned follows that the penetration, but rmmmum bending mo-
minimum penetration of a sheet-pile wall ments and minimum anchor force.
conforms to the condition outlined in case The following is a discussion of both
(b), and the maximum penetration corre- methods.
sponds to condition (d).
Two basic design methods used for de- 6.7.3.1 conventional Design
sign of anchored sheet-pile walls are "free Procedures
earth support" and "fixed earth support"
procedures. The first one is based on the 6.7.3.1.1 Free earth support
assumption that the soil into which the method fF.E.S.J
lower end of the sheet pile is driven is
incapable of producing an effective restraint As stated earlier, under the free earth sup-
from passive pressure to the extent neces- port condition the piling is driven deep enough
sary to induce a negative bending moment; just to assume wall stability, assuming that
in this case, the sheet pile is driven just the maximum passive pressure exerted on the
deep enough to assure the wall stability, front portion of the embedded part is fully
mobilized and no passive pressure develops
assuming that the maximum possible pas-
on the back side of the sheet piles. Under
sive pressure is fully mobilized. The second these conditions, the piling is assumed to act
procedure is based on the assumption that as a beam supported at the anchor level and
the sheet pile is driven deep enough so by soil passive pressure below the dredge line.
that the soil beneath the dredge line pro- The subsequent design is based on the mo-
vides the defl.ected shape of the pile which ment equilibrium condition of all forces about
reverses curvature at the point of con- the wall anchor point.
trafl.exure and eventually becomes vertical. In the conventional design procedure, the
Consequently, the bulkhead acts as a built- · lateral earth pressure distribution is usually
in beam subjected to positive and negative computed by the Coulomb's theory. For com-
bending moments. puting KA and Kp, the wall friction, B, in the
active zero is conservatively assumed equal to
As will be seen from the following discus-
zero, and in the passive zone, it is taken as
sion, the former design procedure results in i<P to </>, where KA, Kp and <P are the coef-
minimum penetration, but maximum bend- ficients of soil active and passive pressures
ing moment in sheet-piling and maximum and the angle of soil internal friction, respec-
pull-out anchor load. The latter design tively, KA and Kp can be calculated as rec-
method results in sheet-piling maximum ommended in Chapter 4. As stated in Chapter
632 Sheet-Pile Bulkheads

4, the coefficients of lateral passive pressure Rankin's theories (Fig. 6-58a). For details
K P can also be calculated by Sokolovski's consult Chapter 4.
(1965) formula which assumes the log-spiral 2. Establish the distance from the dredge
failure surface in the passive zone: line to the point of zero pressure (y) as
indicated in Figure 6-58a, and compute
l+sincf>cos~ • .~. the value of active pressure (EA) and the
Kp = e• tan"' (6-7) distance between EA and RA, that is HA"
1-sincf>
Here R A is the wall reaction force.
3. To satisfy the equilibrium, the wall must
where ~ = 8 + arcsin(sin ojsin cf>). be embedded deep enough so that the
Conventional soil pressure diagrams as are moment due to the net passive pressure
used in F.E.S. method are shown in Figure will balance the moment due to the re-
6-58. sultant active load EA" Therefore, the sum
Ai; stated earlier the model of wall failure is of moments about the anchor point is ob-
assumed to be the rotation about the anchor tained from
point. For postulated collapse behavior, the
required depth of embedment of sheet piles
below the dredge line is determined by taking
the moments about the anchor point. The
anchor force is then determined by consider- =0
ing equilibrium of horizontal forces. Cornput-
ing is usually proceeded in the following where Ep = 0.5(p - a)D 1 , in which D 1 =
sequence. D- y. Then the above equation, MA= O,
is solved for D 1 , and the required pene-
Sheet-Pile Wall in Granular Soil tration D = D 1 + y is established.
This penetration is usually extended by
1. Compute the active and passive soil pres- about 20% to ensure safe performance of
sures. These are determined by the classi- the wall.
cal method, namely by Coulomb's or Subsequently, the anchor force is corn-

a b

Figure 6-58. Anchored sheet-pile wall: free earth support method soil pressure diagram: (a) in
granular soils and (b) in clay andjor sand.
Sheet-Pile Bulkheads 633

puted from Table 6-6. Empirica! values of KM and KR

jDl + y + HB htfH ~ 0.25 htfH > 0.25


RA =Ep HA ha!h2 KM KR KM KR
4. The maximum bending moment, Mmax• is 1 0.60 1.65 0.56 1.70
computed at the point of the zero shear 2
0.65 1.60 0.60 1.65
3
force in the wall, below the anchor point 1
0.70 1.50 0.65 1.55
3
level. o 0.74 1.40 0.70 1.45
5. The design values of the bending mo-
ment, MD, and the anchor pull out force Note: Parameters h 1 , h 2 , h 3 , and H are as depicted in
Figure 6-5.
Rn, are computed as follows:
Source: From Budin and Demina (1979).
MD = MmaxKM (6-8)
RD = RAKR (6-9) (d) As recommended by Budin and
Demina (1979) and presented in Table
where K M and K R are factors that con- 6-6
sider reduction in bending moment and
the corresponding increase in value of an- According to the latter investigators, the
chor tension force, respectively. The hasis values of KM and KR as summarized in
for the use of K M and K R are discussed Table 6-6 are based on data obtained from
in the previous section. As stated earlier, numerous large-scale model and field obser-
the values of K M and K R depend on wall vations of anchored sheet-pile bulkheads
flexibility and engineering properties of
that actually account for the sequence of
the foundation soil and the backfill mate-
wall construction (e.g., "dredged" or "back-
rial. In practice, the values of K M and K R
are determined by one of the following filled" bulkhead construction). On the
methods: other hand, as is seen from Table 6-6, the
(a) By Rowe's (1952) recommendations wall flex:ibility is not accounted for directly.
summarized in Figure 6-46 Coefficients K R as are given in Table 6-6
(b) As recommended by Janbu et al. account for both the load concentration at
(1981) (Fig. 6-59) the anchor location and for the potential
(c) On the hasis of Lasebnik's (1961) increase in the anchor loading due to the
large-scale-model test as summarized unequal yield (or initial tension) of adjacent
in Table 6-5 tie-rods. As pointed out earlier, depending

1.0
~Il '" ' \ 1\
1\
For steel and wood

' '
11 l''
E L\. \ p =
30 Mm VMm
...o
.le
.......... .~ r-1o~ L in meter •nd
t 0.5 ~ ""' ro..
.2 .......... r-... .... ~
.. ~l L=H+D
M,Jntm/m
c
o 1 ~ """ ~
.,~

u Dense IJ 1\
::;,
-o \. ·Medi 1m den~e
QJ
1
0:::
o
0.5 2 5 ~10 20 50
Flexibility p CT m 3/t

Figure 6-59. Bending moment reduction curves KN = f( p). [From Janbu et al. (1981).]
Note: p = L4 fEL, where Lis the totallength ofthe sheet pile and EI is the wall stiffness.
634 Sheet-Pile Bulkheads

on the importance of the structure, the val- difficult and time consuming task. In prac-
ues of KR = 1.2 to 1.4 are usually consid- tice this problem is usually solved by the
ered in conventional practice. graphic method; most recently a number of
the computer programs have been devel-
Sheet-Pile Wa/1 Embedded into oped for analysis of an anchored sheet pile
Cohesive Soils and Backfilled bulkhead (consult section 6.7.3.7).
with Granular Material Although in the case of a sheet-pile wall
designed in accordance with Fx.E.S. method
Normally, similar to cantilever wall both
the longer sheet piles are needed, the de-
the short term (undrained condition, cp = O)
sign bending moment in these piles and
and long term (drained condition, c = O)
anchor force are smaller than in the bulk-
should be examined. The typical resultant
head designed according to the free earth
soil pressure diagram for the undrained soil
support condition. Experience indicates that
condition is illustrated in Figure 6-58b.
in the most practica! cases the anchored
Furthermore, for given soil conditions the
sheet-pile walls designed according to the
permissible height of the wall (Hcr) must be
F x .E.S. method are more economica! than
more than required by design (H)
those designed as F.E.S. walls. Further-
(6-10)
more, the longer sheet piles provide better
assurance against wall outward movement
due to any kind of erosion that may occur at
where qu = 2c is as defined in section 6.7.2.
the dredge line.
The design sequence in general is the same
The basic philosophy and sequence of wall
as is used for design walls driven into gran-
design, according to F x .E.S. method, is es-
ular soils.
sentially the same as discussed earlier in
the case of the free earth support condition.
6.7.3.1.2 Fixed Earth support
Method (Fx.E.S.J
6.7.5.2 Graphical Method
If the sheet-pile penetration is not lim-
ited by the closely located bedrock andjor
the sheet-pile-driving is not obstructed by (aJ Free earth support method
the difficult soil condition, then the sheet- (F.E.SJ
pile bulkhead could be designed according
to the fixed earth support method. This Theexample is illustrated in Figure 6-60.
method is based on the assumption that the In this case the vector diagram and string
wall elastic def:lections are such that the polygon are drawn as in the case of a can-
deflected shape reverses its curvature at tilever wall.
the point of contraflexure and becomes ver- N ext, from point A' which is an intersection
tical at point c (Figure 6~57d). Usually, this of line OA' with the action line of the an-
type of design best assures the wall's safe chor force, R A, the tangential line A'D is
performance. drawn through point D on the string poly-
To produce the aforementioned elastic gon. The position of point D determined the
deflection the wall must be driven deep required (theoretical) sheet-pile penetra-
enough so that the soil beneath the dredge tion, D 0, and line pA' in vector diagram
line provides the required restraint on the that is parallel to line A'D in string polygon
wall deformations. The elastic line method determines the theoretical value of anchor
of assuming a depth of wall penetration and force RA_.
analysis of the resulting deflected shape to Again similar to the cantilever wall maxi-
see that it agrees with the assumption is a mum bending moment Mmax' in sheet pil-
Sheet-Pile Bulkheads 635

(a) (b) (c) (d) (e)

Figure 6-60. Fixed earth support method graphic method of wall design: (a) resultant soil
pressure diagram; (b) load vectors; (c) string polygon; (d) moment vectors; (e) wall deflection; (t)
vector diagram.

ing is obtained from expression Mmax = The theoretical value of the anchor load
pz:Uax and the design bending moment is R A is determined by the line pA in the
computed from Mv = KMMmax; correspond- vector diagram that is parallel to line A'C
ingly, the design anchor force R'v = K R RA. in the string polygon. The design penetra-
and the design penetration, D', can be con- tion, D, may be obtained as equal to (1.2 to
sidered equal to (1.2 to 1.4)D0. 1.4)D 0 as previously discussed.
The wall penetration, D, may also be
lbJ Fixed Earth support Method computed from D = D 0 + llD, where the
lFx.E.S) additional penetration, !lD, is obtained
from the following formulation:
The graphical design method, that is
based on the F x .E.S. condition is illustrated
in Figure ·6-60. The design procedure is
very much the same as is used in the F.E.S.
method. The only difference is that the
baseline A'C in the string polygon is drawn where q is the surcharge load and ali other
in a way to obtain, although negative and parameters are the same as in the preced-
positive, equal values of bending moments ing discussions.
in the upper and lower parts of the wall The design values of the bending mo-
(Fig. 6-60c). In this case, Mmax = ±PZmax· ment (Mv) and the reaction load Rn are
636 Sheet-Pile Bulkheads

obtained by expressions (6-8) and (6-9), The basic ideas included in Danish Rules
respectively. If sheet-pile penetration, D", in terms of assumed soil pressures distribu-
is limited to D 0 < D 0 , then the baseline in tion along the wall has been discussed ear-
the string polygon is drawn somewhere be- lier in this chapter. In this section, some
tween the lines A'D and A'C. The latter practical details, as depicted in Figure 6-61,
obviously will result in a decrease in the are provided. In the case in question, the
negative bending moment in the lower part sheet-pile wall is assumed as simply sup-
of a sheeting, and an increase in the posi- ported at points A and B, where B is located
tive bending moment in the wall upper part. at the gravity center of the passive pressure
The anchor load R A will also be increased diagram. The active and passive pressures
accordingly. are obtained by classical theory similar to
Finally, it should be pointed out that the that as in previously discussed conventional
short and rigid anchored walls (e.g., walls design procedures. The active pressure is
comprised from large-diameter tubular steel modified by a parabola that decreases the
or concrete piles or very rigid reinforced soil lateral pressure in the mid-span region
concrete sheet piles of miscellaneous cross of AB by the amount q, and increases the
sections) are usually designed by the F.E.S. pressure by 1.5q at anchor level (point A).
method; relatively fl.exible walls comprised The factors q and 1.5q account for soil
from steel sheet piles are usually more eco- arching resulting in earth pressure concen-
nomica! if designed as fixed in foundation tration at points A and B with subsequent
soil by the Fx.E.S. method. reaction in earth pressure at wall mid-span
(point 0). The shape of curve A 10B is
parabolic and the magnitude of the factor q
6.7.3.3 Danish Ru/es is determined by
As discussed earlier, this method is based
on the observation of numerous existing 10L' + 4Lpm
q - k -.,....---= (6-12)
sheet-pile bulkheads and is essentially em- 10L' + 5L
pirica!. The design procedure according to
Danish Rules is based on a concept of soil
vertical arching that affects the soil pres- where
sure distribution. The method proved to be L' = height of equivalent layer of soil above point
economica! and was successfully used for A, including the equivalent height of sur-
charge load q 0 L' = yh + q 0
years for the design of miscellaneous soil-
retaining structures comprised of fl.exible L = effective span (AB)
single anchored sheet-pile walls primarily Pm = equivalent uniformly distributed pressure
in cohesionless soils. Many of these struc- on the wall between simple supports A and
B that will produce the bending moment,
tures, which were constructed many years ML, equal to one, produced by trapezoidal
ago, are still operational. Although subject pressure diagram AVV1 ZB obtained by
to considerable criticism, Danish Rules rep- Coulomb's active pressure distribution;
resent the least conservative approach to Pm = 8MLfL2
fl.exible wall design. Some authorities in the
field with reference to the good design record The factor k is computed from
of Danish Rules suggested that if design
results obtained by any other design method
are more conservative than those obtained
by use of Danish Rules, then the latter
_ [
k- 1 + - -
0.01 ( (1 + n)Ea )o.sl-l
sin cf> Lfa
should be preferred.
Sheet-Pile Bulkheads 637

A,

....CII

RL~
J __________________________
B

~
a
z

Figure 6-61. Danish Rules design method.

where where
cf> = average angle of internal friction of backfill RA(L) = reaction force at point A that corre-
materiallocated between points A and B sponds to the earth pressure diagram
n = ratio of the negative bending moment at the obtained by classical method (AVV1 ZB)
anchor level to the maximum positive bend- EA{o) = resultant of soil pressure above the an-
ing moment of the span, L, below the an- chor point A
chor line M0 = cantilever moment at point A due to the
E = Young's modulus pressure EA(O) above the anchor line
a = distance between extreme fibers of sheet-
piling (wall thickness) The wall reaction at point B is determined
fa = permissible compressive bending stress in as follows:
the material ofwhich a sheet-piling is roade.
(6-14)
The value of k typically varies from about
0.8 to 0.9 for steel, and for design purposes
0.9 is used. where R B(L) is the reaction force at point B
The bending moments and anchor reac- that corresponds to the earth pressure dia-
tion can be determined by graphical method, gram AVV1 ZB.
as discussed earlier, or analytically. In the The maximum design bending moment,
latter case the reaction force R A at point A Mv, is determined from the following ex-
can be obtained from the following approxi- pression:
mate relationship:

RA = RA(L) + EA(O) + M 0 jL- f<iqL (6-13) Mn = ML- M 0 j2- 0.09qL2 (6-15)


638 Sheet-Pile Bulkheads

where ML is the maximum bending mo- Tschebotarioff, the pile penetration D


ment corresponding to the earth pressure should be roade equal to 1.7D.
diagram AVV1 ZB.
Because the design soil pressure diagram 1_7.3.4 Tschebotarioff's Method
already accounts for the soil pressure con-
centration at the anchor support and for the On the hasis of his large-scale-model
reduction at the midspan, the values of the tests, Tschebotarioff (1973) has developed a
design anchor pull, RA, and bending mo- simplified "equivalent beams hinge-at-the-
ment, Mv, obtained from Eqs. (6-13) and dredge-line" procedure for the design of an-
(6-15), are not required to be modified by chored sheet-pile walls (Fig. 6-62). This
the respective factors K R and K M typically procedure is based on the assumption that
used in conventional design methods. the location of the zero-moment point in the
The theoretical sheet-pile penetration, D, sheet-pile wall is at the dredge line, where
is determined by the equilibrium condition the upper portion of the sheeting is as-
that assumes the total earth passive pres- sumed to be hinged to the lower one that is
sure computed according to Coulomb's the- embedded into the foundation soil.
ory (8 = 0.5cf>) to be equal to the reaction Where a layer of weak or very loose soil
force R B. This is usually achieved by a is present at the dredge line, the location of
trial-and-error method. The design penetra- the hinge should be moved from the dredge
tion is increased to D(2) 0·5 or 1.42D. Ac- line down below, where the sound soil con-
cording to Tschebotarioff (1973), this pro- ditions are established. Once the elevation
vides a safety factor of approximately 1. 7, of the hinge is established, the bending mo-
not 2.0 as originally assumed; if a safety ments in the sheeting above the hinge and
factor of 2.0 is derived, then, according to the reaction forces at the hinge and anchor

.C:'

.C:'

SAND

(a) (b)

Figure 6-62. Simplified design method by Tshbotarioff (1973); (a) sheet-pile wall
driven into natural sand deposit and (b) computation of safety factors for sheet piles
fixed in clay.
Sheet-Pile Bulkheads 639

level can be computed by the application of obtained as follows:


elementary statics to a beam on two sup-
ports with an overhanging end. In natural 3.15
sand deposits, the penetration D is deter- (6-18)
P1 = RBID
mined to be D = 0.43H, where H is the
height of the wall above the dredge line. (6-19)
Sheeting driven into soft plastic clays
may not have a distinct fixation point; The safety factor, F5 , against shear is
therefore, the design of all types of
sheeting penetrating into the medium to (6-20)
soft plastic clay should follow the conserv-
ative free earth support method as dis-
The above approach usually results in a
cussed earlier; on the other hand, according
smaller penetration than that obtained by
to Tschebotarioff and Ward (1957), steel
conventional methods.
sheeting driven into stiff overconsolidated
Although Tschebotarioffs design method
and nonsensitive clay does get an effective
is generally regarded as reliable, caution
fixation in such material.
must be exercised, especially when dealing
An approximate estimate of the magni-
with soft and sensitive clays. Furthermore,
tude of such a fixation can be made by a
in determining the sheet-pile penetration,
trial-and-error procedure as illustrated in
the possibility of scour or inadvertent over-
Figure 6-62b.
dredging should always be kept in mind.
The pile fixation is dependent on the soil
pressure ordinates p 1 and p 2 ; this pressure
6. 7.3.5 Russian Method
is to be resisted without shearing or sub-
stantial consolidation deformations of the This method, which is part of the
clay if the fixation is to take place as as- Russian standard on the design of the
sumed. This is checked as follows: R 8 , com- waterfront structures constructed on inland
puted earlier, is applied at the hinge loca- waterways~ is based on a great deal of
tion; the active pressures of the clay below large-scale-model tests and field investiga-
the dredge line are ignored if the vertical tions carried out in Russia and summarized
pressure at the dredge line level behind the by Gurevich (1969).
wall is smaller than 2c, which will normally For decades, this method was success-
be the case for water depth not exceeding fully used for the design of numerous
about 9.0 m when the shearing strength of sheet-pile bulkheads comprised of rigid con-
the clay is about c = 70 to 75 MPa. crete sheet piles basically of T -shaped cross
Then sections. The method considers soil vertical
redistribution that is the result of wall de-
(6-16) flection (e.g., wall rotation and displace-
ment at the anchor level and within the
foundation soil) as well as elastic bending
and
about supports at the anchor level and in
the foundation soil. As discussed in detail in
(6-17)
Chapter 4 and earlier in this chapter, each
of the aforementioned wall movements that
Assuming that passive pressures vary are functions of wall flexibility and soil
parabolically and the sign of passive pres- properties affect changes vertical distribu-
sures changes at -fD, then from Eqs. (6-16) tion of soil pressure along the sheet pile.
and (6-17), the value of p 1 and p 2 are The latter along with elastic elongation of
640 Sheet-Pile Bulkheads

the tie-rod are considered in the design where KA is calculated for 8 = 0.14>. The
method in question. soil pressure ordinated between points B
In this method, the horizontal compo- and C are computed from the following ex-
nents of soil active pressures are deter- pression recommended by Dubrova (1963):
mined for a particular part of the wall, as
indicated in Figure 6-63. Part AB (h) is ( 'YY + qB)(tan 2 u + KA)
the distance from top of the wall to the ay =
2
anchor level; normally it is not smaller than
0.2H. However, if hc < 0.2H, then point B 0.5["/f/>'y 2 + 2('Yhc + q)f/>'y]
is assumed to be located a distance of 0.2H 2h 8
from the wall top. Typically, hc < 0.35H.
X tan u(1 + tan2 u) (6-23)
The distance between points B and C, and
C and D (h 8 ) is determined as follows: where
1.15H- hc 'Y = unit weight of soil
hs = 2 (6-21) qB = vertical stress at level of point B; qB =
'Yhc +q
Obviously, part DE is equal to D - 0.15H.
The horizontal component of the active (6-24)
soil pressure at point A is computed by
Coulomb's theory:
KA = coefficient of active pressure calculated for
(6-22) 8=0

4>' = 0.017514>1 (6-25)

From formula (6-23) the stress at point B


(y = O) will be equal to

(6-26)

and at point C (y = h 8 )

0.5["/f/>'h~ + 2('Yhc + q)f/>'h 8 ]


2h 8

/ X tan u(1 + tan 2 u) (6-27)

The stress line between points B and C


K!J- / obtained from formula (6-23) is essentially
/ curved. However, with suf:ficient accuracy,
L - - - -1-------E it is usually assumed to be straight. The
Figure 6-63. Russian design method: soil pressure soil pressures at points D and E are com-
diagram: 1-stiff concrete sheet piles; 2-flexible steel puted by classical theory using the wall
piles. [From Gurevich (1969).] friction 8 = O.
Sheet-Pile Bulkheads 641

Expression (6-23) is basically applicable The active soil pressure diagram for a fl.ex-
for the design of rigid concrete sheet piles ible wall is depicted in Figure 6-63 by
with relative defl.ection !:!.. = !:!..' / h s = broken line.
1/1000 or less, where !:!..' is the wall defl.ec- The passive pressure is determined by
tion at point C. the classical method with the wall friction
For relatively fl.exible piles (e.g., steel 5 = cp, but not exceeding 30°. It is recog-
sheet piles), the results obtains using Eqs. nized that the classical method of comput-
(6-23), (6-26), and (6-27), should be modi- ing the passive pressure is conservative.
fied by the factor K provided in Table 6-7. The design bending moments, the reaction
The maximum value of the sheet-pile rela- force of anchor tie-rod level, the sheet-pile
tive defl.ection, !:!.., can be obtained from the penetration, and the defl.ected line are de-
following expression (Dubrova, 1963): termined by the graphical method. In this
method, the factors K M and K R are not
il' = 2Mh 8 j5EI (6-28)
used for computing the design values of the
where bending moment and anchor reaction force.
The sheet-pile penetration is computed sim-
M = bending moment
ilarly as is done in the conventional method.
E = sheet pile's material Young's modulus
It may be obtained also by increasing the
I = moment of inertia of the sheet-pile cross theoretical embedment D 0 by about 20%.
section

The value of the bending moment M is 6.7.3.6 Lasebnik's Method


computed from the following expression: (in Tsinker, 1983)

Lasebnik recommends the following


(6-29)
approach and sequence of sheet-pile wall
design.
where
{3 =factor that depends on the ratia hc(2h 8 +
hc); values of f3 are given in Table 6-8 1. The density of the foundation soil is es-
tablished. The soil is considered as "dense"
qavr = average active pressure between points A
when D z 0.1, and "loose" when D < 0.1.
and D; qavr = w(2h 8 + hc), where w is the
Here, D is the soil relative density.
area of the soil active pressure diagram
obtained by classical theory 2. The preliminary calculation using the free
earth support or fixed earth support

Table 6-7. Active Pressure Reduction Factor (K) as a function ofsheet-pile


relative defl.ection

Relative defl.ection
 = Sjh 8 0.0010 0.0025 0.0050 0.0075 0.0100
Stress reduction
factor K 1.0 0.8 0.65 0.55 0.5

Table 6-8. Factor f3 in Eq. (6-29)

hcf(2h 8 + hc) o 0.1 0.15 0.2 0.25 0.3 0.35


f3 0.125 0.107 0.094 0.082 0.069 0.056 0.043
642 Sheet-Pile Bulkheads

method is made. At this stage, the values 1990; Pearlman and Wolosick, 1990; Lenzi
of Ba = O and BP = te!> are considered. The and Rosetti, 1992; and others). Although
bending moment obtained as a result of many of these programs are stiH not suffi-
preliminary calculations is used for pre- ciently developed to be used for practica!
liminary selection of the sheet-pile design purposes, their use in the study
section.
of various phenomena involving soil-
3. The preliminary designed bulkhead is structure interaction has proved to be very
then identified with one of 24 schemes useful.
depicted in Table 6-5. The identification Bowles (1988) has developed a finite ele-
is done according to the preliminary val-
ues of d and p and the assumed values of ment program that includes the stiffness
D and ~- Here, d is the sheet-pile pene- matrix commonly used for the analysis of
tration, p is the wall flexibility, and ~ is beams on an elastic foundation with soil
the wall displacement at the top. The springs removed above the dredge line. Ac-
value of ~ depends heavily on the anchor cording to Bowles, his program directly
yield. Anchors such as rock bolts or piled gives the reduced bending moment in
structures are considered as relatively sheet-piling similar to that which would be
unyielding. Others, like tie-rods secured obtained using Rowe's moment reduction
at different types of anchor walls or procedure.
ground anchors, are assumed as yielding. The program, developed by Nicholson
The value of ~ also depends on the den- Construction (Pearlman and Wolosick,
sity of a backfill material placed in front
of the anchor wall. 1990), is also based on beams on the elastic
foundation theory with further extension to
4. The final calculation is then carried out the structural spring model for the embed-
using the values of Ba and BP as recom- ded portion of the wall. It generates bend-
mended in Table 6-5, but the total area of
the passive pressure diagram has to be ing moments and the shear and deflection
equal to that obtained by the conven- in the sheet-pile wall and is independent of
tional method, but with KP determined the number of anchors. The program inputs
from Eq. (6-7). The design value of the include pile stiffness and stiffness of the
bending moment is obtained using the springs (which are input as an elastic se-
-relevant value of KM. For bulkheads with cant modulus), tributary width of the verti-
p ~ 4 and designed according to the fixed cal elements, and allowable soil capacity to
earth support method, the obtained value avoid overloading of the spring elements. In
of d could be reduced by 20-25% (de- the passive zone, the spring forces are cal-
pending on the soil foundation density). culated and compared with passive soil ca-
For yielding anchors, K R = 1, and for
pacities. Any springs that exhibit tension
unyielding anchors (e.g., rock anchors,
piled systems, and the like), the value above grade are set to zero.
K R = 1.4 to 1.5 is recommended. As with any geotechnical analysis, and
an analysis that involves the finite element
method in particular, the input of credible
6. 7.'3. 7 Finite Element Method soil parameters is important. Therefore, the
(computerized ana/ysisJ
designer must realize that the success of
In recent years, several computer pro- finite element methods used for analyzing
grams based on finite element analysis have sheet-pile bulkheads is dependent heavily
been developed to analyze stresses and on the accuracy of the estimation of the
strains in anchored sheet-pile walls. Most of modulus of subgrade reaction (spring char-
these programs treat the sheeting as a :ftex- acteristics) that are dependent on the char-
ible beam on an elastic foundation (Bowles, acter of the soil-structure interaction [e.g.,
1988; Duncan, 1988; Sundaravadivelu et al., potentials for soil arching (wall yield), wall
Sheet-Pile Bulkheads 643

stiffness, impact of soil dredging in front of sheet piles and anchor piles are perform-
the wall, aud, finally, time-dependent ing independently of each other.
changes in soil]. 2. Compute the earth pressures on each
Hence, if the designer selected the finite structural component.
element analysis method for the design of 3. Assume sheet and anchor piles as rigidly
the anchored sheet-pile wall, great care joined at a capping superstructure and
should be exercised when choosing and fixed in the foundation soil and determine
evaluating soil data and the construction the design length of both piles between
procedures involved. fixity points.
More information on installation effects 4. Analyze structure as a two-dimensional
and their importance in the design of frame loaded by relevant soil pressures. If
earth-retaining structures by finite element the calculation results in a substanta!
methods are given in Gunn and Clayton difference between the assumed and com-
(1992). A new constitutive finite element puted sheet pile and anchor pile stiff-
model of a braced wall performing in stiff nesses, then the wall analysis should be
repeated in the same sequence.
clay is described by Simpson (1992).
5. Review the overall wall stability.

6.7.4 Design of Sheet-Pile Note that where (EI). per linear meter of
Bulkheads Anchored wall is more than 10(EI)P per linear meter
of wall, the wall structural components may
by Raked Piles be analyzed independently (e.g., regular
sheet-pile wall and independently perform-
This type of construction essentially repre-
ing anchor piles). Here, (EI). and (EI)P are
sents a space frame comprised from sheet-
the stiffnesses of sheet piling and anchor
piling and anchor piles joined together by a
piles, respectively.
rigid caping superstructure; the backfill
The following is a discussion of two con-
materiallocated between the sheet and an-
cepts of the sheet-pile wall anchored by
chor piles links them together and effec-
raked piles analyses proposed by Budin
tively turns sheet pile-anchor pile-backfill (Budin and Demina, 1972) and Tsinker
material system into a hybrid-type struc- (1972).
ture. The presence of the backfill material The design procedure proposed by Budin
affects the soil pressure distribution be-
was successfully used for the design of sev-
tween the wall structural components (e.g., eral sheet-pile walls comprised of concrete
sheet and anchor piles) that is basically sheet piles and anchor raked piles with
distributed proportionally to the stiffness of
(EI).j(EI)P ~ 5 to 6. The concept of the
these components. Budin's design procedure is illustrated in
The sheet-pile wall anchored by raked Figure 6-64. According to this method, the
piles is usually designed in the following
sheet-piling is loaded by the soil active
sequence: pressure a. which is comprised of two com-
ponents a'. and a';, where a's represents the
1. Establish the wall structural scheme and
pressure of soil that is confined between the
assume preliminary stiffness (EI) of both
sheet and anchor piles, where E is
sheeting and anchor piles included in zone
Young's modulus of elasticity and I is the abc;d (Fig. 6-64b) and a'; is part of a full
moment of inertia. This is usually done on active pressure exerted on the wall at con-
the hasis of the designer experience, ex- ditionally assumed plane o-o. The latter is
isting precedents, or by preliminary wall related to (EI)J(EI)P. Note that the plane
analysis based on an assumption that o'-o' is related to soil arching on anchor
644 Sheet-Pile Bulkheads

(b)

(c)

Figure 6-64. Design of sheet-pile bulkhead anchored by raked piles: computation of design
length of sheet pile (H,). (a) Design scheme; (b) graphic computation of EA(max); (c) soil pressure
on sheeting; (d) resultant soil pressure diagram; (e) load vectors; (f) string polygon; (g) vector
diagram. Note: In (f) line "1" is used when the sheet pile is fixed at the coping superstructure,
and line "2" when sheet piles are hinged there.

piles: where
Emax = total maximum lateral thrust produced
a5 = a's +a'~ (6-30) by soil included in zone abc;d; EA(max) is
computed by the conventional grapho-
statics method as is illustrated in Figure
For simplification of the design process, the 6-64b
wall friction in the active zone, 8a, is as- y = unit weight of soil
sumed to be zero. Under this condition, h = depth from top of a sheeting to the level,
the values of a~ and a'~ are computed as where the value of a~ is computed
follows:
cf> = angle of internal friction

2EA(max) - yzh tan(45° - cf>/2) If soil in zone abc; d is cornposed of layers


a' - ---'---'------:-:-----:-:--- (6-31)
s - h - z cot(45o- cf>/2) with different geotechnical parameters (e.g.,
Sheet-Pile Bulkheads 645

B and cf>), then for each layer of soil, n, a~(n) piles. The component normal to the pile of
is to be computed as follows: this load, a'v, is determined from the follow-
ing expression:

where
(6-32)
q0 = surcharge load
n = soil arching factor that depends on the b1d
where h 0 , h 1 , h 2 , •.. , hn are the distances
ratio; b is the distance between adjacent
from the top of the sheeting to the bottom of piles and d is the pile width (diameter)
the respective layer of soil.
Furthermore, where the values of 4> in The arching factor, n, is obtained from the
adjacent layers of soil differ from each other,
following:
the values of the corresponding a's(n) ob-
tained from formula (6-32) should be cor- bjd < 6, n = 1.02- O.OBbjd
rected by a factor k = KA(n+l/KA(n)• where bjd > 6, n = 0.81- 0.045bjd
KA(n) and KA(n+ l) respectively are coeffi-
cients of the active pressures in the upper The total soil pressure normal to the anchor
and underlying soil layers. pile, aP, is equal to
The soil pressure at a depth h 0 = z X
cot (45° - 4>/2) is computed as a~(O) = (6-36)
yh 0 KA"
The ordinates of the full active pressure Note that the axial component of a soil
an on the assumed plane o-o are deter- weight that is arching on the anchor piles
mined by the conventional method from (a~) along with weight of superstructure
an = (q 0 + L yhn)KA, and the balance be- and live load (q 0 ) must be added to the
tween the full pressure and that con- design scheme:
tributed to the area abc;d (an -a~) is dis-
tributed between sheet-piling and anchor
piles according to their stiffness ratio.
Hence, The diagrams of the soil pressure exerted
on anchor piles are illustrated in Figure
(EI). 6-65. Because the whole structure is treated
a'; = (an - a~) (EI).+ (EI)p (6-33) as a frame, it is necessary to know the
effective freestanding lengths of both sheet
Similarly, the active soil pressure acting on piles (H.) and anchor piles (HP) fixed in the
anchor piles, a~, is computed from the fol- foundation soil.
lowing expression: Essentially, at fixity points the piles' de-
fl.ections are equal to zero. Hence, depend-
, , (EI)p 2 ing on the joint system at the cope level
aP = (an -a.) (EI)s + (EI)p cos a (6-34) (e.g., stiff or hinged), the sheet and anchor
piles are treated as fixed or hinged at the
where a is the angle between the anchor top and fixed in the foundation soil. Natu-
pile and the vertical. rally, when piles are joined rigidly with a
In addition to the active horizontal soil capping superstructure the piles' design
thrust, the anchor pile is also exposed to lengths are shorter than when joints are
the weight of the soil that is arching on the designed in form of hinges.
646 Sheet-Pile Bulkheads

(d) (e)

(f)

Figure 6-65. Design of sheet-pile bulkheads anchored by raked piles: computations of design
length (min) or anchor pile (HP). (a) Design scheme; (b) soil pressure on anchor pile; (c) resultant
soil pressure diagram; (d) load vectors; (e) string polygon; (0 vector diagram.

The passive soil pressures on both sheet chor piles, may transmit some additional
and anchor piles are determined by conven- pressure on the sheet piles. However, this
tional methods, and the resulting soil pres- pressure in most cases is insignificant, and
sure diagrams are used for the determina- usually neglected.
tion of the theoretical fixity planes of these The effective freestanding lengths H 8 and
piles in foundation soils. HP are determined by a graphical method,
The coefficient of passive pressure acting as illustrated in Figures 6-64 and 6-65.
on anchor piles is determined with due con- Finally, the structure is designed as a frame
sideration given to the angle (a) between with given H 8 and HP, loaded by externa!
piles and the vertical and the "effective" forces such as Ra(s) and Ra(p) (Figs. 6-64g
surface cd in front of these piles that is and 6-65f); the upper part of the soil active
loaded by weight of soil, q ', confined in pressure, related to hc (Ee) should be added
zone abcd (Fig. 6-65a). Budin (Budin and to the design scheme. The weight of the
Demina, 1979) pointed out that passive superstructure (W) must also be included
pressure, which represents the ultimate soil as a design load. Then frame may be ana-
shear resistance developed in front of an- lyzed by any suitable method.
Sheet-Pile Bulkheads 647

It is apparent from the preceding discus- pressure exerted on the sheeting. He also
sion that Budin's method is quite compli- concluded that in most practica! cases the
cated and time-consuming. The method is joint between both the sheeting and anchor
most effective for the design of anchored piles and the capping structure with suffi-
walls with (E/) 5 j(EI)P ~ 6, in which com- cient accuracy for design purposes can be
plete fixity of both piles into the capping considered as hinged. Hence, the sheet-pile
superstructure is ensured. However, in most wall can be considered as a regular an-
practica! cases, (E/) 5 j(EI)P is much greater chored sheet-pile bulkhead with the addi-
than 6. Furthermore, in practice it is dif- tion of the axial component of the anchor
ficult to avoid some rotation of the pile head force produced by anchor piles. The latter
embedded into the capping superstructure. can be designed as a beam fixed in the
Complete fixity of regular pile into the con- foundation soil and free supported at the
crete superstructure can be achieved when top of the sheet-piling. Furthermore, this
the pile is embedded to about O. 75-1.0 m. beam can be assumed as loaded by lateral
On the hasis of his observations, Tsinker and vertical soil pressures, the puii-out
(1964) concluded that the design of sheet- force, and the bending moment that is at-
pile bulkheads anchored by raked piles, tributed to wall displacement toward the
specifically where (E/) 5 j(EI)P ::=::: 10, may be basin with subsequent rotation about the
simplified by ignoring the shielding effects fixity point (Fig. 6-66). Regular concrete or
produced by anchor piles on the active soil steel anchor piles driven into noncohesive,

(a)

-·-~

Figure 6-66. Design of anchor piles: (a) design scheme; (b) soil pressure diagram; (c) effect
of passive pressure; (d) system displacement.
648 Sheet-Pile Bulkheads

medium density soils or hard clay are usu- Accordingly,


ally considered as fixed at point B if they
are extended beyond this point to at least (6-39)
5.0-7.5 m. In practice, however, depending
on the local subsoil condition, the height of where q 1 and q 3 are the vertical and hori-
the wall, the quality of back:fill and live zontal components of the load intensity at
loading, the pile length (BE) extended be- the pile head (point A), respectively, and
yond point B varies between 7.0 and 15.0 m
or more. (6-41)
Note that the lateral top of the wall dis-
placement is basically a result of the settle-
ment of the sheet piles and the upward
displacement of the anchor piles due to the
action of the respective compressive and where
pull-out forces. Hence, the total design
YAhc = weight of soil at the pile head level
bending moment at point B (MB) can be
q = uniform distributed surcharge load
expressed as follows:
All other components of Eq. (6-42) are simi-
(6-38)
lar to those used in earlier discussions. For
submergedgranularsoil, n = 0.85jb, where
where b is the distance between adjacent piles.
Ma= bending moment in the pile, attributed to Subsequently,
the action of soil lateral and vertical pres-
sures (e.g., q 1 through q 4 as indicated in (6-43)
Figure 6-66b)
MP = bending moment due to the soil passive where q 2 and q 4 are the vertical and hori-
pressure (Figs. 6-66b and 6-66c) zontal components of the load intensity at
Ma= bending moment attributed to the dis- point B, respectively, and
placement of the pile head (Fig. 6-66d)
(6-44)
As indicated by Tsinker (1964), of ali three
components included in Eq. (6-38) the pile (EI)p
displacement, d, resulting in Ma, is the q4 = byBl(cos a)KA (EI)p +(EI). (6-45)
major contributor to the value of MB.
In accordance with the load diagram de- Therefore,
picted in Figure 6-66b,
q~ = byBl(cos a)n
(6-39)

where
l = pile length between the capping
superstructure and the fixity point
in foundation soil Assuming
q~ and q~ = ordinates of the soil lateral pres-
sure and the weight of the soil that
hangs on pile due to arching effect, · 2 2 (EI)p ]
K = n [ sm a+ (cos a)KA (EI)p + (EI).
respectively
Sheet-Pile Bulkheads 649

we obtain Kp(a) = coefficient of passive pressure deter-


mined by the classical theory, taking into
q~ = K(q + 'YAhc) consideration the pile angle with verti-
(6-4 7) cal, a, and soil friction, a, which is equal;
q~ = K[l(cos a)yB] for a concrete pile 8 = %cp and steel piles
8 = icfl
Hence, Ma can be computed from d = pile width (diameter)
K' =a factor that takes into consideration the
M = -Klz(q + 'YAhc + 'YBlcos a) (6-48) three.dimensional effect of the passive
a 8 15 pressure distribution in front of the an-
chor pile; with a safety factor of 2, K'
If the backfill soil parameters change along can be determined from the following
the pile length, l, then Ma is computed e:xpression (Liakhnitsky et al., 1956):
from
1 ( 8lf- (2l 1 + d- b) 3 )
K' = 1+-
24 lfd

(6-52)
(6-49)
where b is the distance between adja-
If the value of f/J changes, then the corre- cent piles.
sponding ordinates of the horizontal soil
pressure are multiplied by the factor AM and C are determined by basic static
KA(n+ 1/KA(n)• principles and are computed from the fol-
The bending moment resulting from the lowing expressions:
passive pressure (MP) is obtained according
to the design scheme depicted in Figure qplt
6-66c: AM= 24(2l + ll) (6-53)

20l 2 _:_ 15lll + 3lf


(6-50) c= ---,---,...---
15(2[ - ll)
(6-54)

where
AM= area of moment diagram, produced by pas- Hence, MP is expressed as follows
sive pressure, considering the beam AB as
free supported at A and B
C = distance from the gravity center of the
moment diagram produced by the passive
pressure to point A The moment MI!. at point B, which depends
on the pile displacement, !!. , at point A, is
The intensity of the passive pressure at computed from
fixity point B is computed from
ll(El)p
(6-51) MI!.= -3 zz (6-56)

where The horizontal displacement of the pile


'YB = unit weight of soil in the passive zone head, ll, depends on both the pile displace-
l 1 = part of the anchor pile exposed to pas- ment, llP, which includes both the anchor
sive pressure as shown in Figures 6-66b pile puii-out displacement and the elastic
and 6-66c elongation of the pile shaft, and the dis-
650 Sheet-Pile Bulkheads

placement of the sheet pile, 11 s, which in- If the wall comprises a superstructure that
cludes both pile settlement and elastic is extended significantly high above point
shortening. Both 11 8 and !1P depend on the A, then the system's displacement at point
effects of the corresponding axial compo- A, 11u(A) is computed from
nents of the anchor force R a acting at point (6-58)
Llu(A) =du - hc tan(Llf3)
A. Of all the causes of system displacement,
11, the pile pull-out movement, 11 , is the where
most significant; in general, all oth;r effects Llu = horizontal displacement on the top of the
(e.g., sheet-pile settlement and both sheet wall
and anchor pile elastic elongation or hc = wall height above point A
shrinkage) can be neglected (Tsinker, 1972). il f3 = angle that is actually equal to f3 - 90°; for
The value of 11' (Figs. 6-66b and 6-66d), a preliminary calculation, the following
which is very small, can also be ignored. values of tan(Ll f3) can be used; for metal
The permissible value of pile displace- sheeting tan(Llf3) = 0.008, and for con-
ment, 11 P, is controlled by the permissible crete sheet piles tan(Ll f3) = 0.0045.
value of the wall displacement, !1u; the lat-
ter is usually assumed tobe equal to (0.004 Subsequently, the value of !1u(A) is included
to 0.005)H for sheet-pile walls comprised of in formula (6-57) instead of !1u for comput-
stiff concrete sheet piles and (0.008 to ing 11 p(A)· The value of 11 p(A) used in final
0.010)H for bulkheads including steel design should be based on the results of pile
sheet-piling, where H is the height of the field tests.
wall from the dredge line. Finally, the designer must be aware that
The pile pull-out capacity is determined in the process of construction before sheet
by standard procedures that are given in and anchor piles are joined together by cap-
any standard text on foundation engineer- ping superstructure, the anchor pile may be
ing. It is also discussed in detail in Chapter exposed to substantial bending stresses re-
7. Normally, permissible tensile load on the sulting from the effects of the weight of the
anchor pile should not exceed 50% of the inclined freestanding portion of the pile.
ultimate load capacity of the pile (safety These stresses, if significant, must be in-
factor F = 2). cluded in the total balance of bending and
It should be pointed out that the pres- tensile stresses, resulting from M D and the
ence of substantial shear forces acting on axial tensile force.
the anchor pile can significantly increase its As discussed earlier, the "flexible" anchor
pull-out capacity. As stated earlier, the an- piles with widenings at the end have been
chor pile pull-out displacement, 11 p(A)' extensively used for anchoring of sheet-pile
walls (Fig. 6-26 and 6-27). The comprehen-
should correspond to the permissible value
of the horizontal displacement of the pile sive load tests carried out by Tsinker (1972,
1977) have indicated that the ultimate ca-
head, !1u. Here, !1p(A) is the pile upward
pacity of these piles correspond to substan-
movement at point A. Thus, in accordance
tial pull-out displacements. Therefore, if
with wall geometry, as illustrated in Figure
piles with enhanced pull-out capacity are
6-66d, 11 p(A) is computed from the follow-
used for anchoring of sheet-pile walls, their
ing expression:
design capacity must be related not to the
ultimate value of pile capacity but to the
2 permissible horizontal displacement of the
Llp(A) = [ (l cos a) + (Llu + l sin a) 2] 0.5 wall top. With reference to the pile pull-out
2 2 0.5
displacement, !1p(A)' and the effective bear-
-[(l cos a) + (l sin a)] (6-57) ing area of the widening, Fe, the pile pull-
Sheet-Pile Bulkheads 651

out capacity, Rd, can be detennined from The pile strength at point A must be
the following expression recommended by adequate to resist the axial component of
Tsinker (1972, 1977): the design value of the anchor force, R p(A)'
which is equal to Ra/sin a, and the re-
(6-59) quired pile capacity, R P, must be equal to
R p(A) modified by the axial components
where
(both positive and negative) of loads q 1
Ru =pile ultimate pull-out capacity through q 4 that are computed in a manner
R P = design pull-out load similar to that used for q~ and q~. Where
r = empirical coefficient obtained from large- the foundation and back:fill soils are com-
scale model and field tests; r = f(!lp(A)IFe), prised of unifonn granular materials, the
where Fe is the effective bearing area of approximate value of Rd can be computed
the pile widening. The values of r are from
obtained from the diagrams given in
Figure 6-67. (6-61)

(6-60) where
R p(u) = ultimate shear resistance of soil on the
where pile lateral surface below point A'
A. = effective area of the skin friction R w = bearing capacity of the pile widening that
r. = ultimate unit skin friction resistance corresponds to a certain pile pull-out dis-
fb = ultimate bearing pressure at pile widening placement
effective bearing area Ke = empirical coefficient; Ke = 1.25 to 1.50
Q = submerged weight of the pile
Skin friction resistance is usually ignored
in the area within sliding wedge behind the
wall (AA'), 50% of the design value of the
skin friction is used in the A'B' area and
100% ofthe design value ofthe skin friction
is considered below point B (Fig. 6-68).

Figure 6-68. Numerica! example: 1-sheeting;


Figure 6-67. Determination of the coefficient r = 2-concrete anchor pile 0.35 X 0.35 m; 3-medium
f(!lp(AJIFJ. Note: F. = 2ab (Fig. 6-26). sand; Note: m = 2 (a= 26.56°).
652 Sheet-Pile Bulkheads

By definition, the soil shear resistance on f~ f(x 1 ) dx = function that approximates value
the pile shaft is equal to the soil lateral of pile skin friction from zero em-
pressure on a pile multiplied by the appro- bedment to a depth of 3.0 m
priate friction coefficient. In turn, the soil J/; f(x 2 ) dx 2 = the same as above, but from a
lateral pressure is a function of the soil depth of 3 m to the pile lower end
vertical stress multiplied by the coefficient z. = effective length of pile; obviously
of lateral pressure. In the case of an anchor le = 3 + h in meters
pile with a large enlargement placed on the c/J = angle of intemal friction
pile shaft, the uplift load produced by this
y = unit weight of soil
enlargement reduces the vertical stresses in
the soil above the enlargement's bearing K0 = coefficient of soillateral pressure
area that, in turn, reduces the shear re- in the "at rest" condition.
sistance on a pile surface. For vertical piles,
Tsinker (1977) proposed the following for- For preliminary calculations, the values of
mulation: both M f(x 1 ) dx and Jt' f(x 2 ) dx 2 can be
obtained from Table 6-9.9

Example
(6-62)
Determine preliminary value of Rp(u) for the
conventional anchor pile (no widenings) depicted
in Figure 6-68. The pile is installed in medium,
where
and (c/J = 32°, y = 10 kN/m 3 , and K 0 = 0.47) by
Kj = coefficient that takes into account means of a water jet (Kj = 0.9). The pile angle
the method of pile installation; with the vertical is a = 26.56°.
for piles installed by the water Assume that the pile portion BC is located in
jet method or by vibrator, Kj = the stable backfilljfoundation zone with effec-
0.9; for piles driven by hammer tive length, z•. Hence, the conservative value of
Kj = 1.0 R p(u) which does not take into account the ef-
u = parameter of pile shaft fects of the soil mass located above point B is

Table 6-9. Numerical value of Jt f(x 1) dx and ff f(x 2 ) dx 2 in Eq. (6-62) (kN) per linear meter
of concrete pile

ff f(x2) dx

Type From Depth = 3.0 m to h (m)


ofSoil Jt f(xr)dx 4 5 6 10 15 20 25
Coarse
sand 91 50 105 160 415 760 1120 1550
Medium
sand 73 37 76 117 295 535 800 1100
Fine
sand 49 26 54 84 215 390 670 815
Note: For steel piles the values ofboth /J f(x 1 )dx and ff f(x 2 )dx should be reduced by a factor of0.9.
Sheet-Pile Bulkheads 653

computed from the following: constant stress or load. The creep stress
may ultimately lead to structural failure ..
Rp(u) = KjU [ f 18 f(x2) dx- ( 6 f(x';) dx The typical creep rate behavior of soils
cos a 13 13 and deformation-dependent models have
been discussed by Murayama et al. (1984),
-(tan cp)yh 2 2( 1Kt ]
2 Lacerda (1976), Pusch and Feltham (1981),
_ Ko)
and others. The most recent discussion on
soil creep is given in Kuhn and Mitchell
0.9 X 1.4 [ · (1993).
= 976 - 160 - 0.623
cos 26.56° The practice of using sheet-pile bulk-
heads built on clayey soils shows pro-
. o 47 2 ]
nounced specific features of their behavior,
X 10 X 15 2 2( 1 ~ 0.4 7)
such as an increase of stresses in the
= 738kN sheet-pile wall over time under sustained
loads. Sheet-pile bulkheads constructed on
The allowable value of Rw that is subsequently
soils with pronounced rheological properties
used for the determination of the required effec-
tive bearing area at the widening F. = 2ab is are very sensitive to postconstruction defor-
obtained from Eq. (6-61) as follows: mations of the foundation soil which, in
some cases, may lead to the failure of this
type of structure. The interaction between
(6-63)
the sheet-pile wall and the creeping founda-
tion soil depends on the composition of the
The value of F. as a function of R w can be soillayers into which the wall is embedded.
determined by the conventional approach In practice, there are two basic scenarios of
(load within elastic semispace) developed by sheet-pile bulkhead construction (Fig.
Tsinker (1977) or by more accurate finite 6-69):
element method. In both cases, however,
knowledge of accurate value of soil engi- 1. The embedded part of the wall is com-
neering properties and, particularly, the soil pletely situated in the creep foundation
deformation modulus, E 0 , is required. soil (continuous creep foundation, Fig.
The best result, of course, is obtained 6-69a).
from large-scale-model tests or field tests.
2. The embedded part of the wall extends
Pile axial capacity is discussed in
through the top layer of a creep soil of
Fellenius (1991) and in any standard text limited depth and terminates embedded
on foundation engineering. The comprehen- in a bottom layer of a noncreep soil (inter-
sive discussion on a subject matter is also bedded creep foundation, Fig. 6-69b).
given in Chapter 7.
In practice, continuous creep foundations
are not suitable for sheet-pile wall construc-
6.8 SHEET-PILE BULKHEADS tion because usually the strength of soft
BUILT ON CREEP SOILS clayey soils does not provide sufficient pas-
<BY M. GURINSKY) sive pressure for wall stability. Therefore,
sheet-piling is usually driven into a lower
Soils, specifically cohesive soils, exhibit vis- layer of noncreep soil. This practice, how-
cous creep behavior in which deformation ever, sometimes leads to a radical change in
and movement proceed under a state of the time-dependent, stress-strain condi-
654 Sheet-Pile Bulkheads

(b)

..J.:' H*(O)
~ Q*(O)

01]
(a) (b)

Figure 6-69. Practical cases of sheet-pile wall in- 0


stalled in creep soil: 1-creep soil; 2-underlaying
noncreep soil; 3-granular backfill. Figure 6-70. Cantilever sheet-pile bulkhead in-
stalled in creep soil. Design scheme: (a) typical cross
section and (b) embedded part :fixed in creep soil.
tions in wall components, such as the sheet- 1-Creep soil; 2-underlaying noncreep soil; 3-back-
fill.
ing and anchor system resulting in heavy
overstress and overload that may lead to
structural failure (Budin, 1969, 1982).
E = Young's modulus of sheet-pile material
Therefore, under the aforementioned foun-
dation soil conditions, the long-term perfor- I = moment of inertia of the sheet pile
mance ofthe sheet-pile wall is of paramount
importance and should not be overlooked by Laboratory and theoretical studies of the
the designer. general behavior of creep soils under stress
On the hasis of his laboratory and full- indicate the importance of their viscous pa-
scale observations, Budin (1969, 1982) de- rameters on the rate of shearing deforma-
veloped a conceptual theoretical approach tions of the material that follows the ap-
to the design process of sheet-pile bulk- plied stress. If the direct proportionality
heads constructed on creep foundation soil. between the deformation velocity gradient
He has established that in the case of an normal to the sharing surfaces and applied
interbedded creeping base (e.g., varved shear stress is assumed as a quantitive de-
clays) the process of relaxation of contact scription of the quality of the creep soil,
reactive pressures of the soil, p(x, t), that then the material is said to be proportion-
is applied onto a flexible sheet-pile wall ally constant.
with rigidity EI could be described by the The viscosity coefficient, 7J, is determined
following differential equation (Fig. 6-70). from tests of undisturbed soil samples us-
ing a special apparatus. Descriptions of this
a 2 p(x, t) HTJ ap(x, t) apparatus as well as soil testing methods
=- (6-64) and procedures are found in most relevant
ax2 EI at
texts on soil mechanics. For preliminary
calculations, the approximate values of 7J as
where
provided in Table 6-10 can be considered.
X = vertical coordinate
For 7J ~ 10 6 kN day jm2 , the foundation is
t = time usually not considered as creep.
H = depth of the layer of creeping soil For a detailed discussion on the rheologi-
TI = viscosity coefficient of creeping soil during cal properties of creeping soils, the reader is
the period of stabilized creep (in kN day jm 2 ) referred to the fundamental work by Scott
Sheet-Pile Bulkheads 655

Table 6-10. Approximate values ofviscosity (t = 0), respectively, and


coefficient of creep soil, Tf, during the
stabilized creep (in kN day jm2 ) f[ p(x, O)] = Ky(O, O)
Soil f[ p(x, t)] = Ky(O, t)
(6-66)
Clay ofhigh plasticity 2.3 X 10 4 F[ p(x, O)] = Kli(O, O)
Clay ofmedium plasticity 2.5 X 10 4 F[ p(x, t)] = Kli(O, t)
Clay of low plasticity 3.5 X 10 4
where K is the horizontal spring character-
istic of the underlying noncreeping soil, and
(1963) or to any other relevant text on the y(O, t), y(O, 0), 0(0, t), and 0(0, O) are the
subject of soil mechanics. displacement and angle of rotation of the
The basic assumptions for formula (6-64) wall at x = O at any given moment of time,
are as follows (Fig. 6-70): t and for the initial time (t = 0), respec-
'
tively.
The following is a discussion of the con-
(a) The passive pressure diagram p(x, O) is
ceptual approach to the design of sheet-pile
obtained from the conventional static
analysis for the initial moment of the
walls tobe constructed on creeping founda-
soil-structure interaction (t = O) tion soils. The discussion is based on the
soil-sheet-pile wall interaction as described
(b) The load in active zone N(x) is constant by Eq. (6-65).
and is not time dependent.

Provided that all required data are avail- 6.8.1 cantilever Sheet-Pile
able, Eq. (6-64) can be used to compute the Bulkhead
distribution of bending moments M(x, t),
shear forces, and defiections in a sheet-pile Consider the conventional approach to de-
wall. However, Eq. (6-64) does not take into termining the passive pressure distribution
consideration the fiexibility ofthe lower part on the wall structure. For cohesive soils,
of the wall embedded into noncreeping soil. this will be associated with the trapezoidal
This situation can be accounted for by using soil pressure diagram p(x, O) (drained con-
the following expression: dition) that can be represented by its rec-
tangular and triangular components. For

1 !H dx },o p(x, t) dt
the triangular component,
EI x [t
- dx
HTJ o X (6-67)

= f dx f[M*(x,t) -M*(x,O)]dx where


o o y = specific gravity of soil
EI K p = coefficient of reactive (passive) pressure
+-{x{F[p(x,t)] -F[p(x,O)]} H = height of the layer of the creep soil
K

+f[p(x,t)]- f[p(x,O)]} (6-65) The initial and boundary conditions can be


described as follows:
where M*(x, t) and M*(x, O) are the sum- p(x, t)lt~o = p(x, O)= yK/H- x)
marized bending moments in a certain sec-
p(x, t)lx~o =O (6-68)
tion of the wall at a certain postconstruc-
tion period of time, t, and at the initial time p(x, t)ix=H =O
656 Sheet-Pile Bulkheads

For the rectangular distrihution of the ini- and


tiai reactive load, p(x, O)= p(O) = constant,
the houndary and initial conditions can he
descrihed as follows:

p(O, t) = O The summarized hending moments in Eq.


(6-65) can he expressed as follows:
dp(H, t)
---=0 (6-69)
dx
M*(x, t) = [M(x, t)]a- M(x, t)
p(O, O) = p(O) M*(x,O) = [M(x,O)]a -M(x,O) ( 6 - 72 )

The initial relationship hetween the reac-


tive soil pressure and the hending moment, where [M(x, t)]a and [M(x, O)]a are the
M*(O, 0), and shear force, Q*(O, 0), acting in components of the hending moments result-
a certain section of the wall with the coordi- ing from the action ofthe load N(x); M(x, t)
nate x = O at the initial moment of the and M(x, O) are the the components of
soil-wall interaction (t = O) and at a given hending moments resulting from reactive
time t are ohtained on the hasis of work hy loads p(x, t) and p(x, 0). The resulting
Snitko (1963) as follows (Fig. 6-70): hending moments and shear forces in sec-
tion x = O can he expressed as follows
24
f[p(x,O)] = ["2(M*(O,O) + tQ*(O,O)xl)
M*(O, t) = [M(O, t)]a- M(O, t)
24 M*(O, O) = [ M(O, O)]a - M(O, O)
f[p(x,t)] ["2(M*(O,t) + tQ*(O,t)xl) (6-73)
= Q*(O, t) = [ Q(O, t)Ja- Q(O, t)
Q*(O, O) = [ Q(O, O)Ja - Q(O, O)
12 ( 3M*(O, O) )
F[p(x,0)]=["2 12 +2Q*(O,O)
For N(x) =constant,
12 ( 3M*(O, t) )
F[p(x, t)] = ["2 12 + 2Q*(O, t)
[M(x, t)]a = [M(x, O)]a,
(6-70)
[M(O, t)]a = [M(O, O)] a
In the case in question, the right-hand side
of the initial equation (6-65) can he rewrit- and [ Q(O, t)]a = [ Q(O, O)]a, and on the hasis
ten as follows: of the formulas (6-71), (6-72), and (6-73),
the hasic equation (6-65) can he expressed
as follows:
[ dx [[M*(x,t) -M*(x,O)]dx
o o
EI
+ Kl 2 {[M*(O, t)- M*(O, O)]ii(x)
EI
-H
TJ
1 dx J dx j
0
x

X
H t

0
p(x,t)dt

+ [ Q*(O, t) - Q*(O, O)] f 2 (x)} (6-71)


= [dxjx[M(x,O)-M(x,t)]dx
o o
where
EI
+ Kl 2 {[M(O,O) -M(O,t)]ii(x)

+ [ Q(O, O) - Q(O, t)] fix)} (6-74)


Sheet-Pile Bulkheads 657

Subsequently, the following solution for be obtained from the following equations:
p(x, t) can be derived from Eq. (6-74).
M*(x, t) = [ M(x, t)]a- M(x, t)- R~l(x)
p(x,t) =p(x,O) M*(x,O) = [M(x, O)]a- M(x, O)- R!l(x)
(6-80)
Xexp(--H-~-[-F-(-x)_~_IR1~2-:-t-~-(x_)_E_I_]) (6-75)
where
R!(t) and R~(O) = anchor reactions at the mo-
The functions F(x) and <l>(x) depend on the ment of time t and at the
geometry of the initial reactive load dia- initial moment of time, re-
gram p(x, 0). spectively
For the triangular initial reactive load, l(x) = arm of the force R~ relative
these functions are expressed as follows: to a certain specific section of
the wall; l(x) = L - hk - x
(for definitions, see Figure
(3ajf3 + (9/2)/3 + 6a + 2) 6-71).
~( x) = --------=--- (6-76)
0.25a(l - a- a 2 j3)
The expressions for bending moments and
shear forces in section with coordinates x =
O and y =O are as follows:

M*(O, t) = [M(O, t)]a- M(O, t)


where a= xjH and f3 = ljH (Fig. 6-70) - Ra(t)(L - hk)
and for the rectangular diagram of the ini-
tialload M*(O,O) = [M(O,O)]a- M(O,O)
- Ra(O)(l - hk)

Q*(O, t) = [ Q(O, t)]a- Q(O, t)- Ra(t)


(6-78)
Q*(O, O) = [ Q(O, O)]a - Q(O, O) - Ra(O)
6(3ajf3 + 4a + 3f3 + 2) (6-81)
~(x) = (6-79)
a(l- aj2)

Thus, by using the appropriate value of the


functions F(x) and <l>(x) as described by
Eqs. (6-76) through (6-79) and assuming
the initial load distribution p(x, O) =
p(O) = constant, the reactive soil pressures
p(x, t) at any time in question, t, can be
determined by Eq. (6-75).

6.8.2 Single-Anchor Sheet-Pile y


Bulkhead o..___.._@
Figure 6-71. Design of single-anchored sheet-pile
For the single-anchored sheet-pile wall (Fig. bulkhead on creep foundation: 1-creep soil; 2-un-
6-71) the resulting bending moments can derlying non-creep soil; 3-granular backfill.
658 Sheet-Pile Bulkheads

By solving Eq. (6-80) for Eqs. (6-81) and In Eq. (6-84), the function F(x) is obtained
also using formula (6-71) and assuming from Eq. (6-78). The function B(x) depends
[ M(O, t)]a = [ M(O, O)]a and [ Q(O, t)]a = on the initial geometry of the reactive pres-
Q(O, O)]a, the following expression is ob- sure diagram, described by p(x, 0). lf the
tained: distribution of the load p(x, O) is triangular
[Eq. (6-67)], then
EI
-H Jx dx !H dx it p(x, t) dt
7] 0 X 0
mt'a[(L- hk)jH- aj3]
B(x) = 2 (6-85)
+ { dx {[M(x,O) -M(x,t)]dx 1- a- a /3
o o

- { dx { llRa(t)l(x) dx where
o o
a=xjH
EI m tr = coefficient for determining the support re-
+ Kl 2 {[ M(O, O) - M(O, t) actions of the statically undetermined
beams; in the case of a triangular dia-
-llRa(t)(L- hk)]fi(x) gram,
+[Q(O,O)- Q(O,t)- t1Ra(t)]f2(x)}
(6-82)
(6-86)
where ilRa(t) is the increment of anchor
reaction for the specific period of time t,
ilRa(t) = R~(t)- R~(O) and
In the case of a rectangular initial load
f 1(x) = 12( 3lx + 2) p(x, O) diagram,

fix) = 12( 2x + ~) B(x) =


mrecaH[(L- hk)jH- (dj3)]
2 ( 1 _ta) (6-87)

The initial anchor reaction R~ (O) and the


reaction at a specific time, R~(t), are ob- where
tained from the following formulations:
(6-88)
R~(t) = [Ra(t)]a- Ra(t)
(6-83)
R~(O) = [Ra(O)]a- Ra(O) The function <l>*(x) also depends on the
geometry of the initial reactive load p(x, O)
Solving Eqs. (6-81) through (6-83), the ex- diagram. In the case of a triangular reactive
pression for the distribution of the reactive load distribution, <l>*(x) is obtained from
(passive) pressure, p(x, t), is as follows: the following:

p(x,t) =p(x,O)
24{[H 2j6- mtr(L- hk)](3xjl + 2)
EIKZ 2 t ) +(H 2 j2- mtr)(2x + ~l)}
( <1>*( x) = ------:::-----:=-----
X exp - H7J{[F(x)- B(x)]Kl 2 + <l>*(x)EI} x(H2 - Hx + x 2j3)
(6-84) (6-89)
Sheet-Pile Bulkheads 659

In the case of a rectangular initial reactive The design depth of water is 4.6 m. The sheeting
load diagram, the formula for calculating is driven into the stratum comprised of the creep
<l>*(x) is as follows: clay (1) and noncreep loam (2) with the following
engineering parameters.
12{[H 2 j2- mr•c(L- hk)](3xjl + 2)
+ (H + mr•c)(2x + ~l)} Clay: submerged unit weight 'Yc = 8 kNjm 3 ,
~*(x)= ----------~----~---------­ angle of internal friction 4>c = 16°; cohesion
x(H- xj2)
c = 0.01 MPa; TJ = 2.3 x 10 4 kN · day jm 2
(6-90)
Loam: submerged unit weight y 1 = 10
kNjm 3 ; angle of internal friction 4>1 = 27°;
Thus, using the appropriate values of the
cohesion c = 0.02 MPa; spring characteristic
functions F(x), B(x), and <l>*(x) as de- K= 0.32 MPa
scribed by Eqs. (6-78) and (6-85) through
(6-90) and assuming the initial distribution The wall is backfilled with granular material
of the reactive load p(x, O) = p(O) = (sand) with the following parameters: unit
constant, the reactive soil pressure p(x, t) weight in dry condition 'Ys(d) = 16 kNjm 3 ; sub-
at any time t can be determined by Eq. merged unit weight 'Ys(s) = 10 kNjm 3 ; angle of
(6-84). internal friction 4>. = 29.5. The wall is loaded by
a uniformly distributed surcharge load q = 40
kNjm 2 ; the wall rigidity EI= 63400 kN m 2 .
Problem (From Budin. 1982) Determine the bending moment in the wall
structure after 40 years in service.
Consider the following parameters as indicated
in Figure 6-72:
L = 12.6 m
Solution
hk = 2.35 m First, determine the active a, and passive (reac-
H = 5.3 m tive), p(x, 0), pressures for the initial period of
l = 1.95 m the sheet-piling-soil interaction (t = 0). The re-

(a) (b)

Figure 6-72. Design of single-anchored sheet-pile bulkhead on a creep foundation; Numerica!


example. (a) Design scheme and (b) soi! pressures diagram (MPa). 1-Creep soil; 2-underlying
noncreep loam; 3-granular back:fill (sand).
660 Sheet-Pile Bulkheads

Table 6-11. Analysis of a single-anchored sheet-pile bulkhead; numerica! example

p(x, O) p(x, t = 40 years)


Component Component
xjH Triangular Rectangular Sum Triangular Rectangular Sum

0.25 0.0715 0;0436 0.1151 0.0196 0.0131 0.0327


0.5 0.0476 0.0436 0.0912 0.0125 0.0116 0.0241
0.75 0.0238 0.0436 0.0674 0.0076 0.0120 0.0196
1.0 o 0.0436 0.0436 o 0.0153 0.0153

sults are indicated in Figure 6-72b and pre- Table 6-12. Analysis of a single-anchor sheet-pile
sented in Table 6-11. The initial bending mo- bulkhead; numerica! example
ment in the wall obtained from static analysis
xjH F(x) B(x) <l>*(x)
for t = O is M(O) = 256 kN · m.
Next, the wall is analyzed for the long-term 0.25 1.2/1.7 0.5/1.1 112.5/120.5
performance (t = 40 years). To calculate 0.5 2.5/3.3 1.3/2.2 108.5/110
p(x, t = 40) from Eq. (6-84), determine func- 0.75 3.9/5.2 2.5/3.7 128/112
tions F(x), B(x), and <P*(x) by using Eqs. 1.00 5.7;7.0 4.2/5.3 156.5/128
(6-85)-(6-90), where the coefficients mtr and
Note: Values indicated in the numerator and denominator
mrec are determined from Eqs. (6-86)-(6-88) are related to the triangular and rectangular components,
and are as follows respectively, of the generally trapezoidal diagram p(x, 0).

mtr = 0.844 and mrec = 0.56


6-13. The values of Kl 2 and t in this case are
The values of the functions F(x), B(x), and equal to 1215 kN and 14,600 days, respectively.
<P*(x) that correspond to different values of xjH The corresponding passive pressure diagram
are shown in Table 6-12. Other values to be p(x, t = 40 years) is shown in Figure 6-72b.
used in the design are Finally, compute the new bending moment in
the wall structure by using a new reactive pres-
L- hk l sure diagram, p(x, t = 40 years). In this particu-
- -H= 1 9.4 and H = 0.368
lar example, the bending moment in the bulk-
head structure after 40 years in service obtained
The coordinates of the diagram p(x, t = 40) are by a graphic method is equal to 310 kN m, as
determined from formula (6-84). The results of compared to the initial bending moment of
these calculations are shown in Tables 6-11 and 256 kNm.

Table 6-13. Analysis of a single-anchor sheet pile bulkhead; numerica! example

X
[F(x)- B(x)]Kl 2 + <l>*(x)El
R Hry{[F(x)- B(x)]
xKI 2 + <l>*(x)El}

0.25 770,000/825,00 1.29/1.2 0.275/0.301


0.5 743,000/752,000 1.34/1.32 0.262/0.267
0.75 875,000/766,000 1.14/1.29 0.32 /0.275
1.0 1,070,000/946,000 0.93/1.05 0.395/0.35

Note: Values indicated in the numerator and denominator are related to the triangular and rectangular components,
respectively, of the generally trapezoidal diagram p(x, 0).
Sheet-Pile Bulkheads 661

6.8.3 Multianchor Sheet-Pile


Bulkhead
The design diagram is shown in Figure
+ [ Q(O, O) - Q(O, t)
6-73.
The nature of interaction of the multian-

jţO t::.R/t)]f2(x))
chored sheet-pile wall with a clayey creep
(6-91)
base is described by the following equation
(Gurinsky, 1986):

fH dx it p(x, t) dt
where
EI
-H
TJ
1
x dx
0 0
t::. R /t) = incremental change of the force at an-
X
chor j for the period of time t
Z/x) = arm of the anchor force R/t) for the
= {dx{[M(x,O)-M(x,t)]dx specific section of the wall
o o
n = number of anchors
n
-j j L
X X

dx t::.R/t)l/x) dx
O O j~O From Eq. (6-91) the following solution for
p(x, t) is obtained:

+ ~~ ([M(O,O) -M(O,t) p(x,t) =p(x,O)

X
f
X '~ x- :~-~-Kl=;1: -:~
( - -=s=-TJ-:-:{(-::::F:-:-( _-o-::::B::-/::-x-::-::-)) )
(6-92)

X Kl 2 + f/J(x)EI}
Ra.. ~ ..:
" c; Some components included in Eqs. (6-91)
Rt. ,. r== ® and (6-92) are similar to those previously
Rz._ ~
~
~
used in the cases of cantilever and single-

s
f:;: anchored bulkheads.
The functions B/x) and ~x are deter-
J/IL-1 ..
..- mined from the following formulations. In
. RIL ~ ..... N(z) cJ" ..:a
the case of triangular distribution of load
p(x, 0),
{3
-~
=t::
...!.;,
....
L- EJ-- 0 C1 a)
'Z~
B.(x) = mtra ( - -
3
J J H
:a:: (J)
(6-93)
o

-
" - (!)
y ~(x) = 24([~ 3 - .t mj'(L- .t C1)]
J=O 1~0

Figure 6-73. Multianchored sheet-pile bulkhead in-


stalled on creep soil; design scheme: 1-Creep soil;
2-Underlying noncreep soil; 3-granular backfill.
B62 Sheet-Pile Bulkheads

+( ~ 2 jţo m}r) (2x + ~))


- +(H+ jţO myc)(2x+ ~))
x[x(H 2 -Hx+ ~2 )r 1 (6-94) X[ x( H- i) r 1
(6-96)

In the case of rectangular load distribution, The coefficient m 1 that correlates the an-
chor force R1 and the geometrica! parame-
ters of the structure is determined from the
·
B.(x) = mrec aH ( L- I;n_ C.
;-O J - ~3
)
following relationship:
J J H

(6-95) Ri
m.= (6-97)
J p(x,O)

and
The coefficients m 1 are obtained by solving
the following system of equations:
4J(x)=12([~ 2 - 1ţ0 myc(L- 1ţ0 c1)]
[A]{R} = {P} (6-98)

X ( 3lx + 2) where the matrix A is given by

aii= 4RJ( IL- t ţz) [2( Eţz) E ţz],


l=O
2
IL-
l=O
+
l=j+l
i >j

[A]= aiJ=BR1 (1L- t


l=O
g1 )
3
, i=j (6-99)

aii= 4RJ( IL- t gz) [2( t ţz) E ţz],


l=O
2
IL-
l=O
+
l=i+l
i <j

with

L- C0 {R} = (6-100)
and IL= RJ
H

{R} = the column of unknown anchor forces {ji} = the column offree members
Sheet-Pile Bulkheads 663

In case of rectangular diagram of the reac- The solution for the sheet pile wall with two
tive pressure anchors is obtained by assuming n = 1
and j = O, 1. Accordingly, the intensity of
p(O)H(1 - 4ţL)
the reactive pressure of creep soil on the
IL- g
p(O)H[1- 4( 1 )] double-anchored wall during postconstruc-
tion time interval, t, is expressed by modi-
{p} = p(O)H[ 1- 4( IL- 1 ţ0 gz)] (6-101) fied Eq. (6-92) as follows:
••••••••••••• o. o •••••••

p(O)H [ 1- 4( IL - lţo g 1)]


p(x, t) = p(x, O)

In the case of a triangular diagram of the


reactive pressure
X exp {- -H-TJ- :-{[: -F-(.,. . x-:-)-~-I-:0- :...,.)-=-E-
2
(.,... 1(:-x-:-::-)] }

xKl 2 + cp(x)EI}
H (6-103)
p(0)5(1 - 5ţL)
H
p(0)5[1- 5(ţL- gl)] where function F(x) is determined from
Eq. (6-77) and (6-78).
{p} = (6-102)
p(O) ~ [1-5( IL- lţO gz)] The functions B 0 (x), B 1(x), and -;j)(x) are
computed by Formulas (6-93) through
•• o o •• o •••••••••••••••• (6-96), where in the case of a triangular
p(O) ~ [1-5( ţL- lţO gl)] geometry of the soil initial reactive pressure

diagram
mt"a(L- hk)/H- a/3
E 0 (x) = -------;;--- (6-104)
1- a+ a 2 j3
mira(L- hk- c)jH- aj3
E 1( x) = --------:::------ (6-105)
1- a+ a 2 j3
24{[H 3 /6- mt"(L- hk)- mir(L- hk- c)](3xjl + 2) + (H 2 /2- mt"- mir)(2x + ~l)}
"J>(x) =
x(H 2 - Hx + x 2 j3)
(6-106)

In the case of a rectangular geometry of the soil initial reactive pressure diagram,
m 0ecaH[(L- hk)jH- aj3]
Eo(x) = 2(1- aj2) (6-107)

miecaH[(L- hk - c)jH- aj3]


E 1( x) = --------,-------- (6-108)
2(1- a/2)
12{[H 2 /2- m 0ec(L- hk)- miec(L- hk- c)](3xjl + 2) + (H- m 0ec- miec)(2x + ~l)}
"J>(x)=---------------------------:~-----:----------~----~----~
x(H- xj2)
(6-109)
664 Sheet-Pile Bulkheads

Subsequently, the coefficients mj' and m?c Sand: submerged unit weight 'Ys = 10
are obtained by solving eqs. (6-97) through kNjm 3 ; angle of internal friction <Ps = 25°;
(6-102) for j = 1 and n = 1. Hence, for the spring characteristic K = 0.28 MPa
upper anchor, m~ and m({c will be ob-
tained from the following: The wall is backfilled by granular material (sand)
with the following parameters: unit weight in
dry conditions 'Yf(d) = 17.0 kN m 3, submerged
unit weight 'Yt(s) = 10.0 kNjm 3 ; angle of inter-
na! friction <Pr = 30°. The wall is loaded with a
uniformly distributed load q = 40 kNjm 2 and
where JL = LjH and t= (L- hk - c)jH its rigidity EI= 83,200 kN m 2. Determine the
and bending moment in the wall structure after 40
years in service.

Solution
For the lower anchor, mir and miec will be
First, by using the conventional "fixed earth
obtained from the following:
support method," determine active, a, and pas-
sive (reactive), p(x, t = 0), pressures for the ini-
H 2 2p,3 (5ţ- 1)- ţ 2 (3p,- 0(5p,- 1)
mtr = - - - - - - - - - - - - - - , = - - - - - - tial period of sheet piling-soil interaction. The
1 20ga 4p,3 - ţ(3p, - 0 2 results are indicated in Figure 6-74b and pre-
(6-112) sented in Table 6-14. The initial bending mo-
ment in the wall obtained from static analysis
H 3p, 3 (4ţ- 1)- ţ 2 (3p,- 0(4p,- 1) for t = O is M(O) = 107 kN m.
mrec= - - ----------~---- Next, the wall is analyzed for long-term per-
1 4p,3 4p,3 - ţ(3p, - 02
formance (t = 40 years). To calculate p(x, t =
(6-113) 40) by Eq. (6-103), determine the functions
B 0 (x), B 1(x), and ~(x) by using Eqs. (6-104)
through (6-109) where the coefficients mtr and
mrec for the diagram p(x, t = O) are determined
Problem from Eqs. (6-110) through (6-113) and are as
follows:
Consider the following input data for design of
the double-anchored steel sheet-pile bulkhead m~ = 1.72, m~ec = 0.91,

parameters as indicated in Figure 6-74a.


mi'"= 3.91, miec = 2.12
L = 16.9 m
hk = 1.6 m The values of the functions F(x), B 0 (x), Blx),
C = 3.4 m and ~(x) that correspond to different values of
xjH are shown in Table 6-15. Other values to
H= 6.6m
be used in further analyses are
l = 1.6 m
L- hk L- hk- c
The designed depth of water is 4.0 m. The sheet- H = 2.32 and H = 1.8
ing is driven into the stratum comprised of the
creep clay and loose sand with the following The coordinates of the diagram p(x, t = 40
engineering parameters: years) are determined from Formula (6-103).
The results of these calculations are shown in
Clay: submerged unit weight 'Yc = 10 kNjm 3 ; Tables 6-14 through 6-16. The values of Kl 2
angle of internal friction <Pc = 20°; cohesion and t in this case are equal to 717 kN and
c = 0.03 MPa; TJ = 3.54 X 10 4 kN day jm 2 14,600 days, respectively. The corresponding
Sheet-Pile Bulkheads 665

(b)

.--
10,J

c-8,7
O.OfJJ

R-

SAND
Figure 6-74. Design of double-anchor sheet-pile bulkhead installed on creep soil; Numerica!
example. (a) Design scheme and (b) soil pressures diagram (MPa).

Table 6-14. Analysis of a double-anchored sheet-pile bulkhead; numerica! example

p(x,O) p(x, t = 40 years)


Component Component
xjH Triangular Rectangular Sum Triangular Rectangular Sum

0.25 0.162 0.0857 0.2477 0.108 0.0297 0.1377


0.5 0.108 0.0857 0.1937 0.0735 0.0272 0.1007
0.75 0.0538 0.0857 0.1395 0.0420 0.0302 0.0723
1.0 o 0.0857 0.0857 o 0.0364 0.0364

Table 6-15. Analysis of a double-anchor sheet-pile after 40 years in service will be equal to 150 kN
bulkhead; numencal example m, compared to the initial bending moment equal
of107 kN m.
xjH F(x) <l>(x)

0.25 1.74/2.7 1.25/0.29 2.2/0.52 100/41.8 It should be mentioned that ali calcula-
0.5 3.7/5.36 3.17/0.65 5.49/1.16 116/38.8 tions discussed in this section can be easily
0.75 6j 7.85 6.1/1.13 10.41/ 1.91 141/42.6
1.0 8.72/10.4 10.2/1.82 17.4/3.22 243/51.6
computerized.

Note: Values indicated in the numerator and denominator


are related to the triangular and rectangular components, 6 g ANCHQRAGE DESIGN
respectively, of the generally trapezoidal diagram p(x, 0). ..-";•.;.......o...-.,;".;..;..;....,.-.--.-..;...-..;.....,,__-.,_ _ __

passive pressure diagram p(x, t = 40 years) is


As indicated earlier, the sheet-pile walilat-
shown in Figure 6-74b. eral restraint can be achieved by a variety
Finally, calculate the new bending moment in of means which basically fali into three cat-
the wall structure by using the new reactive egories; (a) piled anchorages whose capacity
pressure diagram p(x, t = 40 years). In this ex- to resist the wall reaction force depends on
ample, the moment in the bulkhead structure the capacity of tension and bearing piles;
666 Sheet-Pile Bulkheads

Table 6-16. Analysis of a double-anchor sheet pile bulkhead; numerica! example

xKl 2 + Cl>(x)EI
H1){[F(x)- B 0 (x)- B 1(x)]
X Kl 2 + Cl>(x)El}
0.25 914,900/348,100 0.407/1.07 0.666/0.643
0.5 964,600/322,200 0.386/1.16 0.681/0.314
0.75 1,169,200/354,300 0.318/1.05 0.728/0.38
1 2,018,6001 429,400 0.182/0.87 0.834/0.42

Note: Values indicated in the numerator and denominator are related to the triangular and rectangular components,
respectively, of the generally trapezoidal diagram p(x, 0).

(b) systems that include soil and rock an- 6.9.1.2 A-Frame Piled
chors; and (c) systems that rely on passive Anchorages
soil resistance.
As discussed earlier in this chapter, this
type of anchorage is used effectively where
the short distance between the sheet-pile
6.9.1 Piled Anchorages wall and the adjacent structures prevents
the use of a conventional anchor system
The piled anchorages include the use of such as tie-rods secured at anchorages
raked piles and A-frame systems. whose capacity depends on the passive soil
reaction. The A-frame system formed by
6.9.1.1 Raked Piles piles is connected to the sheeting via tie-rods
These anchor piles develop their resist- (Fig. 6-23).
ance to pull-out force through friction be- Naturally, when loaded by a wall reac-
tween the pile and soil in the stable zone tion force, the piles that are angled toward
behind the wall; only the effective length of the wall are in compression while the piles
a pile, z. (as discussed earlier), should be angled away from the wall are in tension.
considered as effective in mobilizing the pile In conventional design practice, the com-
friction resistance. If piles with local widen- pression and tensile forces are determined
ings (enlargements) are used, then in addi- from a force polygon (Fig. 6-75). The com-
tion to the pile friction resistance, the bear- pression (R) and tensile (R 1) loads in piles
ing capacity attributed to the effective are computed as follows:
bearing area of the widening(s) is added to
the pile capacity. In both cases, however, (6-114)
the pile pull-out displacement compatible
with permissible horizontal displacement of
(6-115)
top of the wall should be considered. Detail
design of such a type of anchor system
should be based on the results of field tests. where
Particular attention should be given to Ra =design value of anchor load; Ra =
the sheet-anchor pile connecting details R~a, where R~ is the anchor force
that may be subject to a combination of per unit length of wall and a is the
axial loads and bending moments in both distance between adjacent bends of
sheet and anchor piles. A-frames
Sheet-Pile Bulkheads 667

(a)
(b)

~
a
1"' 1
t
'.:!"
Rt R' tRt
RJ
·'
a
1'
\R,
Figure 6-75. Design of A-frame piled anchor system: (a) design scheme; (b) load
polygon; (c) design scheme of the load distributing cope beam.

ac and a 1 = angles of compressed and tensioned ments. This can be solved by a suitable
piles, respectively, with the computer program such as GROUP1 devel-
vertical. oped by Reese et al. (1990), or a similar
program. The latter assumes that in addi-
The latter approach, however, results in a tion to compression and tensile axial forces,
conservative solution. Similar, to the sheet- the piles are exposed to lateral displace-
pile wall anchored by raked piles axialloads ments, resulting in shear forces and bend-
in piles that are included in the A-frame ing moments. In the pile design, it is cus-
anchor system can be substantially altered tomary to consider the shear friction values
due to the fact that these piles are joined above line AB (Fig. 6-75) equal to 50% of
together by a stiff capping beam. To resist the normal design values; to account for
an imposed anchor (lateral) load by means shear forces attributed to system displace-
of soil shear requires both inclined piles to ments, the skin friction in the zone below
move against the soil. Obviously, under ax- line AB can be increased up to 15%. A
ial load, the piles in compression should common design consideration is that at least
move down and those in tension should 75% of the design loads, Re and Tt> are
move out of the soil. Both these movements, transferred to the zone below line AB. The
however, are restricted by the capping coping (load distributing) beam is designed
beam; hence, the pile-beam system will be as a continuous beam, loaded by a uni-
displaced laterally. Such movement will formly distributed load R~, an anchor force
cause piles to rotate, placing the lateral per unit length of wall.
load on the surrounding soil. In the process,
piles will either push against or pull away
from the capping beam. Both of the afore- 6.9.2 Sheet-Pile Anchor Wall
mentioned factors substantially change the
axialloads in the piles. Hence, a more accu- A sheet-pile anchor wall is typically com-
rate solution would be based on an analysis prised of short steel of concrete sheet piles
that accounts for system lateral displace- driven ata certain distance from the bulk-
668 Sheet-Pile Bulkheads

head (Figs. 6-18 and 6-19). The tie-rods of the anchor wall do not intersect. Soil
are normally connected to the anchor wall pressures on the anchor wall are usually
via the walling system that is similar to the computed assuming the angle of wall fric-
"parent" one, place behind the wall. The tion in the passive zone is 8 = ţep, and in
anchor wall derives its resistance from pas- the active zone, it is 8 = O. The wall is
sive pressure developed as the tie-rods pull analyzed in the following sequence (Fig.
the wall against the soil. 6-76):
The anchor wall is normally analyzed by
methods similar to that used for the analy- 1. Assume preliminary values of D 1 and D.
sis of a "parent" wall. In this analysis, it is 2. Compute values of passive and active soil
assumed that the full passive pressure in pressures acting on anchor wall and de-
front of anchor wall is developed only if the termine the resulting soil pressure dia-
active pressure zone behind the "parent" gram (p -a). In the process, the sur-
wall and the passive pressure zone in front charge load, q, is placed in such a way as

(a)
q

(d)

p _ _ _ _---r

(e)

Figure 6-76. Design of anchor sheet-pile wall: (a) soil pressure diagram; (b) load vectors; (c)
string polygon; (d) vector diagram; (e) determination of the optimal distance between the
sheet-piling and the anchor wall.
Sheet-Pile Bulkheads 669

to obtain the maximum active pressure front of the anchor wall. Consequently,
acting on the anchor wall and does not the anchor wall resistance is reduced by
increase the passive pressure at the same the amount Eh = 0.5yh 2 (Kp - KA).
time (Fig. 7-76a). Furthermore, to ensure Hence, in the case in question, the anchor
the wall safe performance, the value of wall should be designed to resist anchor
p - a is commonly reduced by a factor force, Ra(h)• equal to
of0.8.
3. The resulting load ofthe 0.8(p- a) pres-
(6-117)
sure diagram is used for the graphical
computation of the wall design parame-
ters [e.g., penetration (D) and design
bending moment Mn = pzmaxl· The solu-
tion is usually found in a trial-and-error 6.9.3 Individual Vertical Anchor
process: components of the pressure dia-
gram 0.8(p -a) must balance the load Piles
(E~ + kRa) as indicated in the vector dia-
gram (Fig. 6-76d), where k = 1.3 to 1.5 is This type of anchorage is illustrated in Fig-
the safety factor. Line AO in the string ure 6-16. It is analyzed in the same way as
polygon, which is drawn parallel to line the continuous anchor sheet-pile wall, but
p'C in the vector diagram determines the with consideration given to the space be-
wall theoretical penetration, D 0 ; full pen- tween the individual piles (Fig. 6-77). The
etration is computed as D = D 0 + f:!.D, coefficient, K', which accounts for the in-
where f:!.D is determined by Eq. (6-6). crease in soil pressures on a pile due to the
In order to be effective, the anchor wall three-dimensional effect of soil pressure
must be located outside the potential ac-
distribution is determined from K' =Vjv,
tive failure plane (BD) developed behind
a sheet-pile wall (Fig. 6-76e). Therefore, where v is the volume of a prism ABCC'B' A'
the minimum distance between the sheet- and V represents the volume of soil in-
ing and the anchor wall (Lmin) is deter- cluded in prism ABGFLL'F'G'B'A'. From
mined from this ratio, the following formulation is
derived:
Lmin = HM tan(45°- cpj2)

+ (D 0 + D 1 ) tan(45° + cpj2) K' = 1


(6-116)
+ 2_ ( 8(D 0 + D 1 ) 3 [2(D 0 + D 1 ) + d- b ] 3 )
Note, that when the wall is designed in 24 (D 0 + D 1 ) 2 d
accordance with the "fixed earth support (6-118)
method" (Fx.E.S.), HM is considered as
the distance from the wall top to the point
of maximum bending moment in the All components included in Eq. (6-118) are
sheeting lower part; when the wall is de- as indicated in Figures 6-76 and 6-77. Note
signed in accordance with "free earth sup- that in this formulation, as in Eq. (6-52),
port method" (F.E.S.), HM is assumed as the factor of safety of 2 is included.
the distance from the top of the wall to
According to Smirnov et al. (1979), Eq.
the sheet-pile tip.
Sometimes site conditions do not per-
(6-118) produces an adequately accurate
mit the installation of an anchor wall at estimate of the pile load capacity when <fJ .:=:::
Lmin· When the wall is placed closer than 30°. lf <fJ > 30°, then </J = 30° should be con-
Lmin• the close proximity of the active sidered in soil pressure analysis.
failure zone behind the sheeting prevents In the case of individual vertical anchor
development of full passive pressure in piles, !::..D is computed from a modification
670 Sheet-Pile Bulkheads

(a)
b b (b)

_1
:1
cS
........
O+

~
.!:.1

~ 1 \
-1
8=45'!_0.54>

1 \
J_
D' L' c• c L D

Figure 6-77. Resistance wedge in front of vertical anchor piles.

ofEq. (6-6) face. This condition usually exists when the


anchor plate embedment ratio D 2 jD 0 is
E' equal to or less than 2 (Fig. 6-78). If the
t!..D = P (6-119)
2q(Kp- KA) dK' n height of the anchor plate is small com-
pared with the embedment depth, then the
where n is the added safety factor; n = 0.8. anchor would fail by ploughing through the
The Canadian Foundation Engineering ground without producing failure zones ex-
Manual (Canadian Geotechnical Society, tending to the ground surface. In the case of
1985) suggests that in the design of vertical a continuous shallow anchor wall (plate), its
piles installed in a homogeneous soil and stability should satisfy the following condi-
subjected to horizontal loads, it is possible tion:
to assume that each pile can sustain a hori-
zontal load equal to the passive earth pres- (6-120)
sure acting on an equivalent wall with a
where
depth of 6b and a width of 3b, where b is
the pile width or diameter, or face-to-face KF = safety factor, KF = 1.6 to 2.0 de-
distance b. pending on the importance of the
structure
R a = anchor force per unit length of the
wall
6.9.4 Deadman <Plate) Anchor
EP and EA= passive and active pressures, re-
spectively, on the anchor wall that
These can be constructed in a form of a are usually computed considering
continuous wall made from plane or rein- the wall friction l>a = O in the ac-
forced concrete, or from prefabricated con- tive zone and 8 = ~1> in the pas-
crete plates (anchor blocks) placed a certain sive zone
distance from each other. According to theo-
retical analysis of the anchor plate resist- For concrete anchor walls and plates, this
ance in soil, plates are usually defined as was confirmed by an extensive series of
shallow or deep anchors. In this respect, a large-scale model tests and field observa-
shallow anchor might be defined as one for tions (Gurevich, 1969). If anchorages are
which failure zones extend to the soil sur- comprised of individual anchor plates placed
Sheet-Pile Bulkheads 671

a certain distance from each other, then the to be extended up to the ground surface.
individual anchor plate stability must sat- Accordingly, the values of EP and EA are
isfy the following condition: computed for the depth of soil extended
from the bottom of the anchor plate to the
ground surface.
In most practica! cases, anchorages such
where as deadman or plates are constructed from
KF, Ra, EP' plane or reinforced concrete. Sometimes,
and EA= same as in Eq. (6-119) however, the anchor plates are fabricated
b = distance between adjacent anchor from steel sheet piles and are, therefore,
rods not as stiff as concrete plates. According to
d = width of anchor plate Othman and Edil (1993), the capacity of a
K' = coefficient which accounts for the fiexible anchor plate at the same displace-
increase in soil pressure on a plate ment is much lower than that of a rigid one.
due to the three-dimensional effect Because in case of an anchored sheet-pile
of soil pressure distribution; K' is bulkhead the wall lateral displacement
determined from Eq. (6-118). must not exceed a certain critical limit, the
adequate stiffness of the anchor plate must
Dismuke (1991) suggested that an iso- always be ensured.
lated anchor plate has an effective width of The optimal (minimum) distance from the
about 1.6 times its actual width; the latter sheet-piling to the anchor plate (Lmin) is
applies, provided that the centers of adja- computed from Eq. (6-116).
cent isolated anchorages are not less than The height of the anchor plate should be
1.6 times their width. At this spacing, sufficient to prevent it from excessive move-
according to Dismuke, the sum of the ment due to the effects of the soil compress-
resistance of an isolated anchor plate is ibility; normally, it should satisfY the fol-
approximately equal to the resistance of a lowing conditions:
continuous wall of the same height.
Equations (6-120) and (6-121) are valid
where the anchor plate (deadman) does not (6-122)
rotate about its support at the tie-rod. To
prevent rotation, the anchor rod should be where
placed at gravity center of the resultant soil 6. = permissible displacement ofthe anchor plate
pressure diagram represented by EP -EA- that corresponds to the permissible dis-
This condition is practically achieved where placement of the sheet-pile wall
the tie-rod is placed 0.4D 0 from the bottom g= soil spring characteristic; for medium dense
of the anchor plate (Fig. 6-78). Hustad and sand, it can be set to about 0.8; the latter is
Langford (1989) described the case of an- usually related to D 2 jD 0 = 2 to 3.
chor wall rotation, among other things, due
to an incorrectly located attachment point Again, the anchor plate overall design
of the tie-rod at anchor wall. should be based on the least favorable load
The height of the anchor plate, D 0 , is combinations, as indicated in Figures 6-53
determined from the condition stipulated in and 6-78. The anchor pull-out force may
Eqs. (6-120) and (6-121). The most rational have a maximum value at low water level;
solution is usually achieved where condi- however, under this condition, the anchor
tion D 0 ~ 0.5D 2 is satis:fied. lf the overall plate resistance may be at its maximum.
height of the anchor plate, D 0 , is equal or On the other hand, at a maximum water
exGeeds 0.6D 2 , then it is usually considered level, the anchor pull may be the smallest;
672 Sheet-Pile Bulkheads

where
N-y = (Nq - l)tan(1.4tjl)
Nq = e7T tan 4> tan 2 (45° + t/1/2)
.
Q
Regardless of the concern about the theoret-
ical correctness of formulation (6-123),
there is an overwhelming acceptance of
Terzaghi's approach. To enhance the accu-
racy of the bearing capacity calculation
without altering the basic equation, Das
Figure 6-78. Anchor plate (deadman); design (1990) and Kumbhojkar (1993) presented
scheme.
the numerica! evaluation of Terzaghi's N'Y.
The other proposed methods and state-
of-the-art review are given in works by
however, the anchor plate resistance may Biarez et al. (1965), Meyerhof (1973),
also be minimal. Hence, both conditions Ovesen (1964), Andreadis et al. (1981),
should be carefully examined. Dickin and Leung (1985), Dickin (1988),
It must be noted that Eqs. (6-120) and Shaheen et al. (1987), Sutherland (1988),
(6-121) used for computing anchor plate and others.
stability have proved to be reliable for the
stability analysis of shallow anchors. The
aforementioned approach, however, repre-
sents an oversimplified solution to the prob- 6.10 WALINO ANO TIE-ROD
lem and leads to conservative results be- DESIGN
cause it fails to address the vertical
equilibrium condition in soil passive and As pointed out earlier, the walings (or wales)
active zones around the anchor plate that that are used in anchored sheet-pile bulk-
arises from friction forces between the soil heads commonly consist of a pair of hori-
and the anchor plate. The complex failure zontal steel channels bolted to the sheeting
mechanism around the anchor plate and its with webs placed back-to-back. Waling is
effects on anchor plate stability was pointed typically designed as a continuous beam
out by Buchholz (1930) and others. supported at locations oftie-rods and loaded
As stated earlier, the deep anchor (D 0 < by the design reaction force that is equal to
0.5D 2 ) failure is characterized by the soil K R R a, where K R and R a are as discussed
local failure mechanism, and the analyses earlier (section 6.7.3.1). Normally for per-
of a deep located anchor plate are usually manent structures, the selected sections of
based on bearing capacity theories. the wale should have minimum thickness of
Terzaghi (1943) proposed that the force, Ra, the web equal to at least 10 mm. The tie-rod
required to puii out a deep anchor is ap- is designed for a pull-out force of KRRab,
proximately equal to the bearing capacity of where b is the distance between adjacent
a continuous footing with a width D 0 whose rods.
base is located ata depth D 2 = D 0 /2 below The wale-tie-rod system forms a particu-
the ground surface and can be obtained larly important part of the anchored sheet-
from the following formulation: pile wall, as failure of any one of these
could lead to progressive failure of the oth-
ers, followed by extensive collapse of the
(6-123) wall. In recognition of this, the safety factor
Sheet-Pile Bulkheads 673

ranging from 1.5 to 2.0, which is dependent Cohesive soils with a plastic index (liquid
on the importance of the structure, is usu- limit minus plastic limit) greater than 20
ally considered for normal load combina- must be carefully assessed for creep.
tion; it can be reduced to 1.2-1.5 for ex- As pointed out earlier, soil creep is a
treme loading. If the tie-rods are postten- time-dependent deformation of the soil
sioned and, therefore, the possibility of structure under a sustained loading. It de-
nonuniform transfer of anchor load is less pends on the composition and structure of
likely to occur, then the safety factor can be soil minerals, preconsolidation history, and
reduced to 1.3-1.5. The working stress in natural moisture content (Murayama and
the tie-rod for the normal load condition Shibata, 1958; Singh and Mitchell, 1968,
should not exceed 0.6FY, and for the ex- Bustamante, 1980). On the hasis of a theory
treme load combination, the working stress given by Singh and Mitchell (1968),
should not exceed 0.8FY, where FY is the Bustamante (1980) proposed that the creep
yield stress of steel. displacement of the ground anchor under a
Use of a high-yield steel for fabrication of sustained load condition can be estimated
tie-rods should be avoided for two basic rea- by using the following equation:
sons: first, anchor rods of smaller diameter AeaT
are more sensitive to strength loss due to !:ll = !:ll 0 + - - ( t l - m - 1) (6-124)
steel corrosion, and second, the higher elon- 1- m
gation associated with steels of less yield where
strength provides for better adaptability to R = applied sustained puii-out force
overstress due to the uneven elongation of
M 0 = initial displacement prior to
the adjacent tie-rods. Anchor tie-rods are creep
normally placed in such a way as to be a, and m = interface creep parameters ob-
situated in troughs formed by sheet piles, A, tained from the experimental
as depicted in Figure 6-20. Joints in the log !:ll-log t and log !:ll - T
walings are normally situated opposite the curves, in which !:ll is the dis-
troughs formed by sheet piles and also as placement rate.
close as possible to the point where the
bending moment is minimal; the joints In practice, the critica! creep load oftiebacks
in the upper and lower channels may be is obtained from pull-out field tests.
staggered. Where relevant, the tieback anchorages
must be tested for cyclic loading. Docu-
mented technical data on the long-term per-
formance of ground anchors subjected to
6.11 GROUND (ROCK) ANCHORS repetitive loadings are very limited. Tests of
seawall anchors carried out in France
For long-term performance, these anchors (Pfister et al., 1982) showed that for peak
are installed only in good quality soils that cyclic load levels smaller than 63% of the
do not exhibit creep. However, in all cases ultimate static pull-out capacity, Pu, anchor
of wall construction, the anchor bond length displacement became negligible after five
should never be located in organic soils or load cycles. However, for larger cyclic loads,
in cohesive soils with an Atterberg liquid the anchor displacement continues to in-
limit greater than 50. The bond length may crease at a constant or increasing rate.
be founded in cohesive soils having a liquid Begemann (1973) reported that repetitive
limit less than 50 if its liquidity index is uplift loads on steel H-piles in sand under a
less than 0.2 (i.e., the natural moisture con- cyclic load amplitude as low as 35% of Pu
tent is near the plastic limit). generated progressive pull-out of the piles.
674 Sheet-Pile Bulkheads

Most recently, Benmokrane and Ballivy The ground anchor design procedure is
(1981) reported results of a 5-year monitor- typically carried out in the following steps:
ing of load losses on prestressed cement-
grouted rock anchors. The tested anchors 1. The design working load is determined
were grouted to various lengths within a from sheet-pile wall analysis and, subse-
sound rock. The study resulted in the fol- quently, wall displacement criteria are es-
tablished. Note that the anchor must be
lowing observations:
located beyond the potential sliding
(failure) surface behind the wall.
1. The instantaneous loss ofload depends on 2. Select the type of anchor(s), inclination,
the load applied by the jack to the anchor length, and spacing, as well as the corro-
and on the free length of the bar. The sion protection system.
anchored length does not effect the in- 3. Verify that the anchor capacity is suffi-
stantaneous loss of load. For a fixed free cient to withstand the design working load
length, the instantaneous loss of load in- without unacceptable yield.
creases as the load applied by the jack to 4. Verify that the anchor performance does
the anchor increases, and vise versa. For not produce detrimental effects on the
constant applied loads, the instantaneous sheet-pile wall (e.g., unacceptable settle-
losses diminish when the free length in- ment due to effects of vertical component
creases. For applied loads approaching the of the anchor force).
elastic limit of the anchor bar, an average 5. Verify that the wall system global stabil-
instantaneous loss of load varying be- ity with respect to general sliding along a
tween 6.2% and 11.2% respectively was potential failure plane is maintained with
noted for the anchors tested for both free an acceptable factor of safety.
lengths, namely 2900 and 1450 rom. 6. Complete the wall-anchor system struc-
2. There were two distinct phases of pre- tural design with respect to the applied
stress losses. Phase I is characteristic by bending moments and axial and shear
a fairly rapid loss of load with time, oc- forces.
curring within about a half-year period.
Thereafter (Phase II), a slower and more In practice, the objective of the designer is
uniform rate -of load loss is observed. This to specify anchor parameters that are re-
period is comparable to the one pointed quired for long- (short-) term stability ofthe
out by Littlejohn and Bruce (1979). wall-anchor system; the final design is usu-
3. The average final loss of load recorded is ally carried out by the contractor. Any de-
8.2% for initialload. signs, however, are usually treated as pre-
liminary, and usually, the anchor system
parameters are finalized through field
For more information on this subject, the te sting.
reader is referred to Hanna (1982). As stated In determining the anchor system pa-
earlier, the basic concept of the ground an- rameters, the designer must consider vari-
chor design relies on the transfer of the ous potential failure mechanisms that are
tensile force of the sheet-pile wall reaction basically related to failures of steel tendons,
to the ground (rock) through friction (or soil-grout bond, and grout-tendon bond,
adhesion) mobilized along potential sliding and to the ground mass global stability.
surfaces (or rock joints). In the process of With respect to the tendon failure, the de-
design, the main engineering concern is to sign load should not exceed 60% of the
restrain anchor displacements in order to ultimate tensile strength for the final de-
secure the stability of the structure with an sign of permanent anchorages; this may be
appropriate factor of safety. increased to up to 80% of steel ultimate
Sheet-Pile Bulkheads 675

strength for the design of temporary an- gate, resulting in bond mobilization of
chorages. deeper zones. Once the stress is transferred
From the tendon corrosion viewpoint, the to the end of the bond zone and the uiti-
choice of solid bars as opposed to wires or roate bond is exceeded, anchor failure by
strands is preferable. In general, however, puii-out occurs.
strands and wires have advantages with Because during load transfer, the bond
respect to tensile strength. As stated ear- zone elongation is decreasing from the up-
lier, the anchors should be extended far per bond length down to its end (and so
enough beyond the potential failure plane does it relative to the soil movement), mobi-
to prevent failure of the ground mass in lization of skin friction resistance in bond
which they are embedded. The failure zone decreases accordingly.
mechanism is characterized by the instabil- Experience has indicated that if the an-
ity of a mass of soil in front ofthe bond area chor is grouted properly, an increase in the
followed by puii-out of this zone. The mini- bond length beyond 10-12 m usually does
mum recommended distance of the anchor not result in much of an increase in anchor
bond area from the failure plane is usually capacity.
assumed tobe 0.15H, where H is the free- Finally, the bond between steel tendon
standing portion ofthe wall (Fig. 6-29). The and grout must not exceed the value of
likely plane of failure in sound bedrock is anchor design capacity; the bond between
along a cone generated at about a 45° angle tendon and the grout can be enhanced by
from the anchorage. In fractured or bedded added embedment.
rock, the cone shape and size varies with During the design process, the angle of
the distribution of bedding and cleavage anchor inclination should be carefully eval-
planes, and the grout takes in fissures. Rock uated. It is usually considered between 10°
mass failure seldom occurs in anchors em- and 45° to the horizontal. Smaller angles
bedded more than 5.0 m below the bedrock are usually preferred in order to minimize
surface, because the bond strength between the vertical component of anchor load on
the rock and grout or the grout and tendon sheeting, and steeper angles are normally
is significantly less than the strength of the considered in order to reach deep bearing
sound rock. strata (e.g., rock) or to avoid interference
An anchor system installed sufficiently with existing structures. The effects of slop-
deep into the ground beyond the failure ing anchors on sheet-piling are discussed in
zone rnobilizes skin friction between the in- detail by Browzin (1981). The minimum
terfaces. The bond between the anchor and spacing between anchors is limited by the
the ground depends on the normal stress permissible distance between the bond zone
acting on the bond zone grout and the adhe- of adjacent anchors to avoid a group effect
sion andjor friction mobilized between the which reduces individual anchor capacity.
ground and the grout. Underreamed or By rule of thumb, the lesser of either three
belled anchors may also develop base and a half diameters of the bond zone or 1.5
resistance of the increased annular area m is commonly used as a minimum spacing
(Fig. 6-29). requirement (Fig. 6-29). A horizontal spac-
In general, in uniform soil or rock the ing between adjacent anchors of 2-3 m is
ground-grout bond is mobilized progres- commonly used.
sively as the stress is transferred along the The required bond length, and bond
bond length. The stress mobilization or load length parameters with sufficient accuracy
transfer to the bond length is dependent on could be approximated by the method given
bond length elongation; in the process of in the Canadian Foundation Engineering
load transfer, the anchor continues t<? elon- Manual (Canadian Geotechnical Society,
676 Sheet-Pile Bulkheads

1985) as follows:

(a) Cohesionless Soils o. 75

In these soils the anchor estimated pull-


out resistance, R, is obtained from the "' o. 50
following formulation
0.25

(6-125)
oL---L-~---L--~--~~~~--~--~
o 20 40 60 80 100 120 140 160 180
UNORAINEO SHEAR STRENGTH, kPo
where
a; = effective vertical stress at the mid- Figure 6-79. Reduction factor, a, in Eq. (6-126).
point ofthe load-carrying bond length [From Canadian Geotechnical Society (1985).]
As = effective unit surface area ofthe bond
length
As, which is function of the
Ls = effective load-carrying (bond zone) anchor diameter, is usually
length of anchor difficult to estimate; it is
K f = coefficient which depends on the type highly dependent on ground
of soil and condition as given in Table porosity, grout permeability,
6-17 and grout pressure during
installation.
(b) Cohesive Soils
These are basically considered as being (c) Rock
stiti to very hard clays. The pull-out ca- Anchorage in rock is based on the allow-
pacity, R, can be estimated from the fol- able grout-rock bond (shear) stress acting
lowing: over the fixed-in-rock length. It is usually
assumed that the allowable bond stress
(6-126) should be smaller than 1/30 times the
unconfined compressive strength of the
where rock andjor 1/30 times the unconfined
As and Ls = the same as in Eq. (6-125) compressive stress of the grout. In gen-
eral, it should not exceed 1.3 MPa.
a = reduction factor related to
undrained shear strength;
obtained from Figure 6-79 The pull-out capacity of the tremie-grouted
ru = average undrained shear straight-shaft rock anchors, or those used
strength of the clay over the in very stiff to hard clays can be estimated
anchor (bond zone) length from the following formulation:
It should be noted that the
effective unit surface area, (6-127)

Table 6-17. Anchorage coefficient Kr in Eq. (6-12)


where
Compactness Condition D = anchor diameter
Soil Type Loose Compact Dense L = effective bond length
Silt 0.1 0.4 1.0 ru = ultimate shear stress at the grout-ground
Fine sand 0.2 0.6 1.5 interface
Medium sand 0.5 1.2 2.0
Coarse sand, gravei 1.0 2.0 3.0 Littlejohn and Bruce (1975) recommend us-
Source: From Canadian Geotechnical Society (1985). ing a Tu value of about 10% of the rock
Sheet-Pile Bulkheads 677

uniaxial compressive strength, or Tu= aSu by classical theory to the passive soil reac-
in cohesive soils. In the latter, a is the tion required to maintain the limit equilib-
cohesive factor and Su is the average . rium of the anchored wall-soil wedge
undrained shear strength of soil. The factor system.
a generally varies within the range 0.3-0. 75 Brom's method provides a safety factor
(Tomlinson, 1957; Peck, 1958; Woodward et with respect to failure along a potential
al,. 1961). sliding surface passing through the an-
Again, the final anchor engineering pa- chors. However, the wall global stability
rameters in any type of soil or rock must be with respect to general sliding along poten-
established by means of the field pull-out tiai failure surfaces that do not cross the
test. anchors should be also investigated using
conventional methods of slope stability
analysis.
6.9.11.1 Global stability of Results of the large-scale (field) tests re-
Sheet-Pile Wall ported by Anderson et al. (1983) confirmed
supported by Ground that the aforementioned methods used for
Anchors review of the wall global stability are reli-
The single-level ground anchor system able. However, in some cases when heavy
that is the most appropriate for the sheet- surcharge loads are applied to the backfill,
pile bulkhead construction is analyzed for the wall movements may be considerable
global stability with respect to sliding along and, therefore, may not be acceptable.
a critica! failure plane, DF, as indicated in
Figures 6-80 and 6-81. This failure plane
extends from the base of the sheeting to the 6.12 OVERALL STABILITY
midpoint of the anchor. OF ANCHORED
As illustrated in these figures, the poly- SHEET-PILE WALL
gon of forces acting on the soil wedge at
limit equilibrium involves the active soil
The conventional ship circle method is usu-
pressure acting on the wall, soil shearing
ally used to check the overall stability of
resistance, and wall reaction. The safety sheet-pile walls of any design and particu-
factor of this system is defined as a ratio of larly those designed according to the free
the sum of the available resisting forces earth support method (minimum embed-
to the total force required to maintain ment of sheet-piling). This essentially trial-
stability. and-error method is discussed in Chapter 4
Kranz (1953) recommended the limit and is found in most standard texts on soil
equilibrium of the soil wedge neglecting the mechanics and foundation engineering
passive pressure on a wall. This method is [e.g., comprehensive work by Lamb and
adopted in the French code of practice Whitman (1969) and others]. This method
(Bureau Securitas, 1977) to allow for the can be applied to soil having both interna!
design ofmultianchored walls. Broms (1968) friction (</>) and cohesive (c). In this method,
extended Kranz's method, taking into ac- the factor of safety is defined as the ratio of
count the passive pressure and sheet-piling resisting moments to driving moments.
basal reaction force in the equilibrium of For the single-anchor level walls de-
forces acting on the wall-wedge system. signed by the free earth support method
According to Broms, the safety factor is (F.E.S.) (Fig. 6-82a) German Standard EAU
defined as the ratio of the limit passive 1990 recommends Kranz's (1953) method in
resistance of the foundation soil estimated which a composite failure surface BEF made
678 Sheet-Pile Bulkheads

LIVE LOAD- TOBE INCLUDED


IN ANALYSIS ONLY IF LOWER
ACTIVE FAILURE WEDGES FAILURE PLANE 15 INCLINED
/ T O HORIZONTAL AT ANGLE > ~

~ • ANGLE OF SHEARING
RESISTANCE
AREQ'D • CALCULATED ANCHOR
PULL FOR WALL STABILITY
PA= ACTIVE FORCE (FROM H TOF)'
P 1 • ACTIVE FORCE (FROM C TOD)'
c1
= cu x l
WATER PRESSURES NOT INCLUDED
BASE OF IN THESE CALCULATIONS
RETAINING
WALL

(o) FCRCES ACTING ON ANCHORING BODY

R1 DIRECTION

~ANCHOR
DIRECTION

Aposs. • POSSIBLE MAGNITUDE


OF A
NOTE, MAGNITUDE OF R1 MUST
BE CHECKED TO ENSURE
COMPATABILITY WITH
ANCHOR DIRECTION

(b) VECTOR DIAGRAM

Figure 6-80. Graphlc analysis of sheet-pile wall, anchored by single-level ground anchors.
[From Canadian Geotechnical Society (1985).]
Sheet-Pile Bulkheads 679

EQUIVALENT
IMAGINARY
ANCHOR
SEPARAT ION
WALL
JOINT
H

SOIL MASS
2
ACTIVE FAILURE
WEDGE

SOIL WITH
rut<fut' 0 )

1 SOIL WITH
f f2 (C2 • 0)

f = ANGLE OF SHEARING RESISTANCE

AREQ'D = ;~~C~!~~ES~:::~~~~R PULL

PA= ACTIVE FORCE (FROM H TO F) •


(o) FORCES ACTING ON ANCHORING pl =ACTIVE FORCE (FROM C TO O) •
BODY ( SOIL MASSES 1 ANO 2) C 1 =eul x l
• WATER PRESSURE NOT INCLUDED
IN THESE CALCULATIONS

R1 DIRECTION

A POSS. • POSSIBLE MAGNITUDE OF A


NOTE, MAGNITUDES OF R1 ANO R2
MUST BE CHECKED TO ENSURf
COMPATABILITY WITH ANCHOR
DIRECTION

(b) VECTOR DJAGRAM

Figure 6-81. Graphic analysis of anchored wall in two soillayers. [From Canadian
Geotechnical Society (1985).]

up of an active failure wedge behind the supports itself on anchoring soil section
anchor plate, a passive resistance wedge in BCDE with a shear force that is created at
front of anchor plate, and a connecting plane interface BC. The anchoring section BCDE
between these wedges and the pile base is lies on the lower failure plane BE which
assumed as practica! for a review of the extends from the lower edge of the anchor
wall global stability. In this scheme, the soil plate to the sheet-piling tip. This section is
wedge ABC with dead load W and slip sur- held stable by shear forces which exist at
face BC loads the sheet-piling. The wedge interface BE inclined to the horizontal by
680 Sheet-Pile Bulkheads

( a) (b )

Figure 6-82. Overall stability of sheet-pile wall supported by conventional anchor system: (a)
anchored wall stability according to Kranz (1953) and (b) anchored wall stability according to
Bogdanov (Guvevich, 1969). !-Active pressure failure plane; 2-lower (sliding) failure plane;
3-passive pressure failure plane; 4-logarithmic spiral sliding surface. HM and Hw = moor-
ing force per unit, length ofwall and hydrostatic pressure, respectively; Ra = anchor force per
unit length ofwall; Ea and EP = earth active and passive pressures, respectively; W = weight of
soil within failure wedge, W1 = weight of soil in anchoring body of soil; 8 = angle of wall
friction.

the friction angle cp. The anchoring section mum free anchor length should be ensured
BCDE is loaded by the wall reaction Ra before carrying out stability calculations.
transmitted to it via the anchor plate, by its This minimum free anchor length is either
own weight W1 , and by the active pressure the length required to fix the anchorage in
Ea, exerted by the failure wedge DFE be- a firm stratum or the length required to
hind the anchor plate. The anchoring sec- have the front of the fixed anchorage length
tion is stable when it provides resistance beyond a line drawn through the wall base
(shear) force at interface BE greater than at angle of 45°, whichever is greater. To
Ra + Ea. The system is considered as stable ensure global stability, it is recommended
when the safety factor [the ratio of resisting that individual equilibrium analysis for
forces versus driving forces (normal load each layer should be carried out and should
combination)] F :2=: 1.5. Ifthis is not the case, satisfy the general condition of stability.
then the anchor plate must be placed fur- The French Code also strongly emphasized
ther back from the sheeting. Alternatively, that in the multianchored systems the in-
the anchor plate can be installed deeper. fluence of one anchor on the stability of the
It should be noted that EAU 1990 recom- adjacent one should also be taken into con-
mends the Kranz's method for analyzing sideration.
the stability of single anchored walls, Another modification to Kranz's method
whereas the original work of Kranz (1953) applied to a single-anchor wall system was
considered multianchor systems. proposed by Littlejohn (1972). The latter
An evaluation of Kranz's method used for author proposed the stability analysis
walls installed in layered soils is given in method using a logarithmic spiral sliding
the original French Code of Practice (Rec- surface, as indicated in Figure 6- 82. The
ommendation Concernant la Conception, le radius from the center of the spiral to any
Calcul, l'Execution et le Control des Tirants point on the curve forms a constant angle of
d'Ancrage, 1972). This document suggests interna! friction, cp, with the normal to the
that in the case of ground anchors the mini- curve and the line of action of the resultant
Sheet-Pile Bulkheads 681

forces on each part of the sliding surface lizing a slope in decomposed schist rock
passing through the spiral center. There- (DeBeer and Wallays, 1970).
fore, none of the forces along the sliding The resisting force per unit length ofwall
line will develop a moment around the spi- provided by anchor piles, p (Fig. 6-83), is
ral center; hence, moments can be neglected estimated as p = p' jb, where p' is the
when considering the equilibrium of mo- total permissible shear force acting on the
ments around the center. Littlejohn's modi- pile and b is the distance between adjacent
fication developed for a single-anchor wall piles. The pile length, D, below slip surface
was further extended to multianchored 0-0, is considered as a dowel and its shear
walls by Schulz (1976) and Ostermayer resistance is analyzed in a similar manner
(1977). as that used in conventional anchor pile
Anderson et al. (1983) carried out a se- design that is fixed at the top. The graphi-
ries of tests of a sheet-pile wall anchored by cal method as discussed earlier can be con-
means of ground anchors. The latter inves- veniently used for pile-resistance analysis.
tigators concluded that all the aforemen- The design is carried out in the following
tioned methods basically lead to stable sys- sequence (Fig. 6-83).
tems. However, considerable movements
occurred in surcharge loaded walls which 1. Compute the soil pressure on the pile
were designed according to the aforemen- length, D, below the slip surface 0-0. The
intensity of the resulting soil pressure, p,
tioned French Code of Practice. Anderson et
similar to the conventional anchor pile
al. concluded that movements of this mag- design, is determined from p = ( p -
nitude may not be acceptable in some prac- a)dK', where p and a are intensity of
tica! cases. passive and active pressures, respec-
In the latter case, the most critical fail- tively, d is the width of pile, and K' is
ure angles ai and {3i are found by the the coefficient determined from Eq.
trial-and-error approach. As noted in the (6-118).
previous section, Broms (1968) modified 2. Draw the vector diagram and the string
Kranz's method to allow for the axial force polygon, and determine the permissible
in the wall. In both the Kranz's and Brom's value of zmax that is function of the per-
methods, the essential requirement is that missible bending moment, acting on the
pile in combination with the design ten-
the part of the fixed anchorage should be
sile load; zmax = 0.85M;fp 0 , where p 0 is
located within the active pressure zone obtained from the vector diagram and M;
which affects the earth pressure on the is the bending moment which in combina-
sheeting. This zone is determined by a line tion with the axial tensile load R P can be
drawn from point B at 45° + cf>/2 to the resisted by the anchor pile at a permissi-
horizontal (Fig. 6-82a). Ifthe sheet-pile wall ble stress level in pile material(s); 0.85 is
is anchored by means of racked piles inter- a coefficient that accounts for the poten-
sected by a slip surface, the soil reinforcing tial presence of loose soil, pile damage
effect (dowel action) provided by these piles during installation, and so forth.
is usually considered. 3. The baseline AB is drawn in the string
Experiments conducted by Hansen polygon, as indicated in Figure 6-83d.
This line determines the minimum re-
(1953b), Mackenzie (1955) and Dastidar
quirements for the pile embedment, D,
(1956) have indicated that piles which in- into the ground below the slip surface
tersect a slip surface increase the slope's 0-0. Then line pa, which is parallel to line
global stability. Short rigid dowels consist- AB, is drawn in the vector diagram (Fig.
ing of steel beams grouted in predrilled 6-83e) to obtain the shear force, p, that
holes have been successfully used in stabi- acts at the slip surface on the pile. Fi-
682 Sheet-Pile Bulkheads

-
a -
e 9

o
-
Figure 6-83. Dowel effect on stability of sheet-pile wall anchored by raked
piles: (a) structural scheme; (b) soil pressure on pile below failure plane 0-0;
(c) load vectors; (d) string polygon; (e) vector diagram.

nally, the shear force per unit length of 6.13 SEISMIC DESIGN OF
wall p, that is taken into account as an
ANCHORED SHEET-PILE
additional shear resisting force is com-
puted as p = p jb. In most practica! cases, WALLS (BY W. S. DUNBAR>
the pile length, D, is extended deep
enough below the slip surface 0-0 to sat-
isfy equal moments condition as depicted 6.13.1 Observed Failure Modes
in Figure 6-83d.
However, if the pile penetration is in- Gazetas et al. (1990) documented the field
sufficient to develop the equal moment performance of 75 anchored sheet-pile walls
condition and the latter is required from
in harbors during earthquakes. Their obser-
the wall overall stability viewpoint, then
vations can be summarized as follows:
longer anchor piles should be used. Other-
wise, a smaller bending moment in the
lower part of the pile can be considered. 1. Most failures resulted from liquefaction of
the loose, saturated, cohesionless soils
used for the backfill andjor existing in
Naturally, in ali design cases the wall the supporting base. The most dramatic
overall stability must be reviewed for the examples of this type of failure are the
worst combination of live and dead loads. sheet-pile wall failures which occurred in
the Niigata, Japan harbor during the 1964
The former includes surcharge and point
earthquake. A typical example of soil liq-
loads acting on the wall, the hydrostatic uefaction and its impact on a sheet-pile
load, and mooring forces. A safety factor of bulkhead is discussed in Chapter 4 (Fig.
1.5 is usually considered as sufficient for 4-2).
normal load combination, and 1.2 for ex- 2. Anchor failure due to either inadequate
treme loading is generally regarded as passive soil resistance against the an-
adequate. chor or degradation of soil resistance
Sheet-Pile Bulkheads 683

due to a buildup of pore water pressure chor system is sized and located so that the
during ground shaking. Anchor failure passive resistance at the toe of the wall
is often accompanied by permanent sea- equals the driving rotational moment about
ward tilting of the wall. the anchor.
3. Less frequent types of failure are as
follows:
• Seaward displacement of the wall, ac-
companied by backfill subsidence
6.13.3 Assumption
• Seaward movement to the toe due to
inadequate embedment of the wall or In the equations to follow, the soil used for
to degradation of soil strength during backfill is assumed to be dry and cohesion-
ground shaking, both resulting in in- less. In a system composed of saturated soil,
adequate passive soil resistance the specific unit weight of the soil would be
• localized seaward tilting of the upper used instead of a dry unit weight. In addi-
part of the wall due to yielding of the tion, the acceleration ratio, tan 1/J, of the
sheet pile Mononobe-Okabe formulation would have
• flexural failure of the wall to be modi:fied in accordance with Eqs.
(4-37) and (4-38) of Chapter 4. Note that,
These observations resulted in numerous soils having cohesion are usually not used
investigations of the design procedures used for backfill of sheet pile bulkheads and par-
for anchored sheet-pile walls. Particular ticularly those in a marine environment. It
attention was given to the design of the is furthermore assumed that the backfill is
anchor system. sufficiently compacted so that any liquefac-
tion failure occurs following a failure by
large displacement.
6.13.2 Seismic Design
Procedure 6.13.4 Factor of Safety Against
The standard method of evaluation of the
Failure by Rotation
dynamic loads is by means of pseudo-static
The force resultants due to an earthquake
procedures involving the Mononobe-Okabe
acting on the wall are shown in Figure
formulation described in Chapter 4.
6-84. The dynamic component of the active
Neelakantan et al. (1992) proposed a de-
pressure EA_E, is given by
sign procedure in which failure of a single-
anchor sheet-pile wall, if it occurs, is by
rotation about the anchor tie-rod attach- (6-128)
ment point such that the toe of the wall
moves seaward. This is a more preferable where E AE and EA are the active seismic
type of failure from the point of view of and static pressures, respectively, on the
continued use ofthe facility. For this type of wall; E AE and EA are computed in accor-
failure to occur, the driving moment about dance with recommendations provided in
the anchor tie-rod attachment point must Chapter 4.
exceed the restraining moment provided by According to Neelakantan et al. (1992)
the passive resistance at the toe of the wall. (Fig. 6-84), the dynamic components of the
This implies that the tie-rod and anchor active pressure EA.E are assumed to act at
block must be proportioned to withstand the midheight (HA/2) of the wall, and both
the total horizontal thrust on the wall. A the dynamic and static components of the
"balanced" design is achieved when the an- passive pressure E'pE and Ep are assumed
684 Sheet-Pile Bulkheads

that gives F = 1.0. Mathematically, this is


equivalent to solving the nonlinear equa-
tion 1.0 - MRjM0 = O for tan 1/J = tan 1/J* .

... 6.13.5 Size and Location of


Anchor Block
Consideration of the equilibrium of forces
on the wall leads to the following expres-
sion for the required anchor force RE which
can be resisted by the anchor system:

Figure 6-84. Forces acting on sheet-pile wall during


earthquake.
RE= h<1- k)[(H + D) 2 KAE- D 2KPE]
(6-131)

to act at a height D j3 from the bottom of


the wall. In order for failure by rotation about the
Considering a free-body diagram of the anchor tie-rod to occur, the maximum an-
forces on the wall shown in Figure 6-84, chor force should occur when the critica!
the driving moment about the tie-rod acceleration ratio reaches tan 1/J *. For an
attachment point M 0 is given by acceleration ratio tan 1/J*, the coefficients of
seismic active and passive earth pressures,
KAE and KPE> may be computed from Eqs.
(4-31) and (4-33) given in Chapter 4. The
ultimate load, Rult> that can be sustained
by an anchor plate of height D 0 is then
given by

The restraining moment M R is given by

Thus, for failure by rotation about the


anchor,

where EPE is the passive seismic earth


pressure; EPE = Ep- E'PE· Ep is computed
in accordance with recommendations given This results in the following expression for
in Chapter 4. Accordingly, the factor of Do:
safety of the wall against failure by rotation
about the anchor is then given by F =
MR!Mo. (6-133)
The wall will begin to rotate as assumed
when F = 1.0. The value of the acceleration
ratio, tan 1/J = k h/(1 - k) [see Chapter 4, lf the anchorage system were designed for
Eq. (4-31)] that gives F = 1.0 is defined as an acceleration ratio less than tan 1/J *, an-
the critica! acceleration ratio, tan 1/J*.It may chor failure would precede any other kind
be determined by finding the value of 1/J of failure mode.
Sheet-Pile Bulkheads 685

Experiments on models oftied-back walls Perfonn static analysis


performed by Neelakantan et al. (1990)
did not confirm the existence of a definite Obtain D/H ratio

active slip surface consistent with the


Mononobe-Okabe formulation. Instead, a
diffused slip surface was observed.
Solve
Neelakantan et al. (1992) therefore sug-
1-MR/Mo=O
gested that placing the anchor far from the
for lan v'
active slip surface would be conservative
and recommended that the anchor be placed
such that the active wedge of the wall does
not interfere with the passive wedge in front
ofthe anchor plate. Using this criterion, the
minimum distance of the anchor block from
the wall is given by Eq. (6-116) in which
D 2 is substituted for D0 + D 1 . Determine size and
location of anchor

Figure 6-85. Sheet-pile wall seismic design; bal-


6.13.6 Balanced Design anced design procedure.

Procedure
The objective of the balanced design proce- is usually assumed that the seismic waves
dure described by Neelakantan et al. (1992) giving rise to the additional dynamic forces
is to size and locate the anchor such that are vertically propagating shear waves.
the driving rotational moment about the However, the step geometry of the wall re-
anchor tie-rod attachment point equals the sults in complex diffracted wave patterns as
restraining moment provided by the passive well as reflections from the wall which
resistance at the toe of the wall. This bal- propagate in a more horizontal direction.
ance is achieved by solving the equation for The backfill material willlikely exhibit non-
the safety factor with F = 1.0 for the criti- linear, hysteretic behavior and pore pres-
ca} acceleration ratio tan f/1 *. However, there sure buildup during strong shaking, which
is another constraint, namely the design may result in the amplification of ground
acceleration ratio. The critica} ratio tan f/1* motion within the backfill. Also, hydro-
must equal or exceed the design ratio. If it dynamic forces would be exerted on the
does not, then the embedment depth D seaward side of the wall and possibly on the
must be increased. This will result in a backfill side, depending on the saturation of
larger value of the critica} acceleration the backfill (Steedman and Zeng, 1990).
ratio. At present, it is not possible to quantify
The steps of the balanced design proce- the effect of these factors in any reliable or
dure are shown in Figure 6-85. practica} manner. Most of the above factors,
In conclusion, it should be noted that particularly any nonlinear soil behavior,
although the above design procedure has a would become important in the event of
firm empirica} and experimental hasis, it severe, long-duration shaking (i.e., in the
ignores several complicating factors which event of a large near-field earthquake). In
can affect the actual behavior of a sheet-pile general, one should be aware of such com-
wall during an earthquake. For example, it plications and exercise judgement and uti-
686 Sheet-Pile Bulkheads

lize experience to evaluate their effect on the sheeting, and overstress and failure of
the design and performance of the wall. the tie-rod or its components such as cou-
plings or turnbuckles due to inadequate
design or excessive surcharge loading
and/ or backfill settlement. The anchor-
6.14 SHEET-PILE WALL FAILURE age may also fail because of a location too
close to the anchoring block to the sheet-
It is very important to learn about struc- ing or an excessive surcharge loading
placed behind the anchoring block.
tural failures that occur in order to prevent
them from occurring in the future. Field 2. Soil Failure at Dredge Line and Be-
measurements carried out in the past 35-40 low Resulting in Movement of the Toe
years, in general, con:firmed that the cur- of the Piling Outward at and below
rently available design procedures used for Dredge Level (Fig. 6-86b). Failure of
this type occur due to inadequate pene-
sheet-pile wall design, in principle, lead to
tration of sheet-piling that may be the
satisfactory results. Failures of sheet-pile
result of an inadequate design, scour of
bulkheads, however, are stiU occurring and miscellaneous nature, or overdredging; in
some interesting case histories of wall fail- clayey soils, it may occur due to excessive
ures have been recently reported. soil creep.
The basic causes of wall failure are illus-
3. Failure of the Sheet-Piling in Bend-
trated in Fig. 6-86 and can be summarized
ing (Fig. 6-86c). This type of failure may
as follows: occur basically due to incorrectly evaluat-
ing the nature of the soil-structure inter-
1. Failure of the Wall Anchorage (Fig. action that may result in the use of a
6-86a). This occurs because overstress wrong section of piling, or as a result of a
and failure ofthe anchor components such substantial increase in the sheet-piling
as wale and bolts attaching the wale to effective span due to soil scour or over-

(a) (b)

\ ·=·=·=·=·=·=·=·=·=·=·=·=·=·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:
l<t---.......----+-
1
1
___...,,
··:·:·:·:·:·:·:·:·:-:-:-:-:-:-:·.·.·=·=-:.:-: ··"'i=·=·=·=·=·==·=·······=·.·.:::.:.....
.... 1
~~ 1
-..:::- ..... ./...

(C) (d)

1
1
l

Figure 6-86. Mode of failure of anchored sheet-pile wall; (a) failure of anchorage
system; (b) failure of soil in front of the wall; (c) failure of piling in bending; (d) global
failure along the slip circle.
Sheet-Pile Bulkheads 687

dredging, or due to long-term effects such and the failures occurred as a result of
as soil creep or adverse changes in the overloading the clay.
backfill due to miscellaneous dynamic ef- In the other interesting case described by
fects (e.g., earthquake, heavy vibration, Sowers and Sowers the anchored bulkhead
etc.)
failed because the batter anchor piles de-
4. Overall Failure Along Circular Slip flected too much from the weight of sand
Surface (Fig. 6-86d). This is discussed in backfill as the clay below these piles consol-
the preceding section. idated.
Among other causes of failure was poorly
It should be noted that of all the afore- designed and performed backfill drainage
mentioned types of wall failures which have that has caused overloading of the bulk-
occurred from time to time most are at- heads by the hydrostatic pressure.
tributed to failure of components of the an- Several case histories of anchored sheet-
chor system, to inadequate penetration of a pile bulkhead failures that occurred in
sheet-piling below the nominal dredge line, Sweden are discussed by Broms and Stille
or to overall failure of the structure. Very (1976). There, most of failed bulkheads
few cases of sheet-pile failure due to mate- were anchored by ground anchors inclined
rial overstress have been reported. at 45°. The basic failure mechanisms were
Sometimes bulkhead failure occurred due as follows:
to heavy overloading. It should be pointed
out too often old wharves are overloaded 1. Insufficient sheet-pile penetration depth
because their original design loads have resulted in soil failure in front of the wall.
been forgotten during their years of service 2. Pile embedment was insufficient to resist
with changes in management and owner- the axial component of the anchor force.
ship or due to other reasons. Furthermore, The latter resulted in pile settlement and
sometimes the excessive soil pressure on substantial outward movement of the top
sheeting may be developed during the wharf ofthe wall.
modification by piles driven behind the ex- 3. Wall overall failure.
isting wall. The latter results in a heavy 4. Failures caused by stress corrosion of the
soil consolidation and backfill displacement anchor bars or by slippage of the anchor
toward the sheeting. bolts or the wedges which lock the anchor
An interesting observation of a series of bars or strands.
anchored (braced) sheet-pile wall failures
have been reported by Sowers and Sowers Most of observed failures were attributed to
(1967). The analysis of the failed walls led frost action that caused substantial soillat-
the latter authors to the conclusion that the eral expansion. The latter was the result of
failures were not the result of the inade- poorly performed wall drainage system.
quacy of modern earth pressure theories. The failures reported by Broms and Stille
Instead, they were caused by the neglect of (1976) indicated the importance of a review
the backfill loads, construction operations of the combined effects of axial load and
that produce excessive earth pressure, bending moment on sheet-pile wall perfor-
poorly designed support systems, inade- mance. They also suggested that the afore-
quate allowance for deflection, deterioration mentioned effects can decrease the sheet
and corrosion of sheet piles and elements of pile's bearing capacity.
anchor system, and poor design and con- The most recent report on bulkhead fail-
struction details. Several of the reported ure was presented by Daniel and Olson
failures are of special interest, as the bulk- (1982). In this case, the wall driven into the
heads were constructed on clay foundations highly plastic clay was designed in accor-
688 Sheet-Pile Bulkheads

dance with accepted design methods. How- author concluded that the overall stability
ever, less than a month after construction, ofbulkheads built in Beaumont clay forma-
the 240-m-long sheet-pile wall anchored by tion, and possibly other expensive fissured
steel tie-rods secured at anchor panels overconsolidated clays, should be analyzed
moved outward nearly 3.0 m. using residual strength parameters which
The latter authors explained the cause of are less than peak value. Fluctuation of
the failure by the designers lack of under- water levels should be considered in the
standing of soil behavior under project cir- stability analysis.
cumstances. Furthermore, Browzin (1985), Interesting case of failure of a sheet-pile
Pilecki (1985), Steenfelt (1985), Surendra wall built on a creep clay also is described
and Mundell (1985), and Tinoco (1985) in in an article by Budin (1969).
their discussion of Daniel and Olson's (1982)
article pointed out that the most likely REFERENCES
causes ofbulkhead failure are design errors
that are as follows: ANDERSO~, W. F., HARMA, T. H., and
ABDEL-MALEK, M. N., 1983. "Overall Stability
of Anchored Retaining Walls." ASCE Journal
1. The depth of penetration was too small, of Geotechnical Engineering, Voi. 109, No. 11.
which resulted in insufficient soil passive ANDREADIS, A., liARVEY, R. C., and BURLEY, E.,
resistance. 1981. "Embedded Anchor Response to Uplift
2. The anchor tie-rod was too short; There- Loading." ASCE Journal of the Geotechnical
fore, the anchor plate was located too clase Engineering Division, Voi. 107, No. GTl.
to the failure wedge behind the wall. BAUMANN, P., 1934. "Analysis of Sheetpile Bulk-
3. The size of the anchor plate was inade- heads." ASCE Proceedings, Voi. 60; also pub-
quate and, therefore, was not able to de- lished with discussions in ASCE Transla-
velop sufficientresistance to puii-out force. tions, Voi. 100 (1935).
Furthermore, in an attempt to stabilize BEGEMANN, H., 1973. Alternating Loads and
the moving bulkhead, approximately 3 m Pulling Tests on Steel I-Beam Piles. Proceed-
of backfill between the wall and anchor ings 8th International conference on Soil Me-
plate was excavated. The latter resulted chanics and Foundation Engineering, 2.1.
in a reduction in anchor capacity and in BENMOKRANE, B and BALLIVY, G., 1991. "Five-
increased potential for continued horizon- Year Monitoring of Load Losses on Pre-
tal displacements. stressed Cement-Grouted Rock Anchors."
4. The structure global stability was insuf- Canadian Geotechnical Journal, Voi. 28.
:ficient. BIAREZ, l., BOUERANT, L. M., and NEGRE, R.,
5. The soil parameters were chosen incor- 1965. "Limiting Equilibrium of Vertical Barri-
rectly. ers Subjected to Translation and Rotation
Forces." Proceedings of the 6th International
Conference on Soil Mechanics and Founda-
Another interesting case of sheet-pile bulk-
tion Engineering, Voi. II.
head failure is discussed by Sullivan (1972).
BICA, A. V. D. and CLAYTON, C. R. l., 1989.
In this case, the sheet-pile bulkhead was
"Limit Equilibrium Design Method for Free
constructed on the Houston ship channel in Embedded Cantilever Walls in Granular
the strong overconsolidated Beaumont clay Materials." Proceedings Institution of Civil
formation. The bulkhead experienced con- Engineers, Part 1.
siderable movements after deepening the BICA, A. V. D. and CLAYTON, C. R. I., 1992. "The
channel. The movement occurred when an Preliminary Design of Free Embedded Can-
extreme low water level created sufficient tilever Walls in Granular Soil." Retaining
overstress for a long enough period to cause Structures, Clayton, C. R. I., ed. Thomas
clay rapid progressive failure. The latter Telford, London.
Sheet-Pile Bulkheads 689

BILETI, P. and SIEFFERT, T. G., 1989. "Soil-Sheet Walls for Northern Regions. Stroyizdat Pub-
Pile Interaction in Vibro-Piling." ASCE lisher, Leningrad (in Russian).
Journal ofGeotechnical Engineering, Vol. 115, BUDIN, A. YA. and DEMINA, G. A., 1979. Quays
No. 8. Handbook. Stroyizdat Publisher. Moscow (in
BJERRUM, L., and CLAUSEN FIRMANN, C. J., and Russian).
DUNCAN, J. M., 1972. "Earth Pressures on BUDIN, A. Y. and GURINSKY, M. A., 1985. "Com-
Flexible Structures-A State-of-the-Art Re- putation of Spring Compensators U sed for
port." Proceedings of the 5th European Con- Ground Anchors." Osnovaniya, Fundamenty i
ference on Soil Mechanics and Foundation Mechanica Gruntov, (Soils, Foundations and
Engineering, Voi. 2, Madrid. Soil Mechanics) No. 6, November-December
BLUM, H., 1931. Einspannungsverhaltnisse bei (in Russian, translated into English), Moscow,
Bohlwerken. W. Ernst and Sohn, Berlin. Rus sia.
BoRTHWICK, M. A., 1836. "Memoir on the U se of BUREAU SECURITAS, 1977. Recommendations
Cast Iron in Piling, Particularly at Brunswick Concerning the Concepts, the Calculations,
Wharf, Blackwall." Transaction Institute Civil the Execution, and the Control of Ground
Engineers, Vol. 1. Anchors. Paris, France.
BOWLES, J. E., 1988. Foundation Analysis and BUSTAMANTE, M., 1980. Capacite s'anctage et
Design, 4th ed. New York. comportement des triants injectes, scellas
BROMS, B. B., 1968. "Swedish Tieback System for dans une argile plastique. These docteur-
Sheet Pile Walls." Proceedings ofthe 3rd Con- ingenieur ENPC, Paris.
ference on Soil Mechanics and Foundation CANADIAN GEOTECHNICAL SOCIETY, 1985.
Engineering, Budapest, Hungary. Canadian Foundation Engineering Manual,
BROMS, B. B., 1991. "Deep Compaction ofGranu- 2nd ed. prepared by the Canadian Foundation
lar Soils." Foundation Engineering Hand- Committee, Vancouver, Canada.
book, 2nd ed. Fang, H-Y. (ed.). Van Nostrand
CASAGRANDE, A. and WILSON, S. D., 1953.
Reinhold, New York. "Stability of Bulkhead at Boston Army Base."
BROMS, B. B. and STILLE, H., 1976. "Failure of Report to U.S. Corps of Engineers, August.
Anchored Sheetpile Walls." ASCE Journal of
CASAGRANDE, L., 1973. "Comments on Conven-
Geotechnical Division, Vol. 102, No. GT3.
tional Design of Retaining Structures." ASCE
BROWZIN, B., 1981. "Anchored Bulkheads: Hori- Journal ofthe Soil Mechanics and Foundation
zontal and Sloping Anchors." ASCE Journal Division, Vol. 99. No. SM2.
of the Geotechnical Engineering Division, Vol.
CLAYTON, C. R. 1. and MILITITSKY, J. 1986. "Earth
107, No. GT5.
Pressure and Earth Retaining Structures."
BROWZIN, B. S., 1985. "Analysis of Stability Surrey University Press, London, U.K.
Against Overturning About Point O (Toe Fail-
ure)." Discussion on paper by Daniel and COMPTON, G. R., Jr., 1981. "Selecting Pile In-
Olson, 1982. ASCE Journal of Geotechnical stallation Equipment." MKT Geotechnical
Engineering Division, Vol. III, No. 3. Systems.
BRUCE, D. A., FIEDLER, W. R., and TRIPLETI, R. CRAIG, R. F., 1983. Soil Mechanics, (3rd ed.) Van
E., 1991. "Anchors in the Desert." Civil Engi- Nostrand Reinhold, New York
neering, December. DANIEL, D. E. and OLSON, R. E., 1982. "Failure of
BUCHHOLZ, W., 1930. "Erdwiderstand auf Anker- an Anchored Bulkhead." ASCE Journal of
platten." Jahrbuch Hafenbantechnishen Ges- Geotechnical Engineering Division, Vol. 108,
sellshaft, Vol. 12. No. GT10.
BumN, A. YA., 1969. "Behaviour of Sheetpile Re- DAS, B. M., 1990. Principles of Geotechnical En-
taining Walls in Soils Susceptible to Creep." gineering, 2nd ed. PWS-Kent Publishing Co.,
Soil Mechanics and Foundation Engineering, Boston.
No. 6, (translated from Russian). DASTIDAR, A. G., 1956. Pilot Tests to Determine
BUDIN, A. YA., 1982. Thin Sheetpile Retaining the Effect of Piles in Resisting Shear Failures
690 Sheet-Pile Bulkheads

in Clay. Report on research performed at Report, Department of Supply and Services,


Princeton University, March, Princeton, NJ. Canada. Research Project, Vols. I and II.
DEBEER, E. E. and WALLAYS, M., 1970. "Stabili- FRANZIUS, 0., 1924. Versuche mit Passivem Druck
zation of a Slope in Schists by Means of Bored Bauingenieur, Berlin.
Piles Reinforced with Steel Beams." Proceed- GAZETAS, G., DAKOULAS, P., and DENNEHY, K.,
ings 2nd Congress on Rock Mechanics, 1990. "Empirical Seismic Design Method for
Yugoslavia. Waterfront Anchored Sheet Pile Walls." ASCE
DENATALE, J. S. and IBARRA-ENCINAS, G. A., 1992. Design and Performance of Earth Retaining
"Total Stress Analysis of Cantilever Sheet Structures. Special Publications, New York.
Piling in Layered Clay." ASCE Journal of GILMAN, J. F. and NOTTINGHAM, D., 1991. "En-
Geotechnical Engineering, Voi. 118, No. 7. vironmental Wave Barriers in Small Craft
DICKIN, E. A., 1988. "Uplift Behavior of Horizon- Harbors." ACSE Proceedings Specialty Con-
tal Anchor Plates in Sand." ASCE Journal of ference. "World Marina," Long Bearch, CA.
Geotechnical Engineering, Voi. 114, No. 11. GONCHAROV, Y. M., 1962. "Design of Thin Walls
DICKIN, E. A. and LEUNG, C. F., 1985. "Evalua- and Consideration due to Soil Redistribution
tion of Design Methods for Vertical Anchor Along Wall Height." Foundations and Soil
Plates." ASCE Journal of Geotechnical Engi- Mechanics, No. 5, Moscow (in Russian, trans-
neering, Voi III, No. 4. lated into English).
DISMUKE, T. D., 1991. "Retaining Structures and GUNN, M. J. and CLAYTON, C. R. I., 1992. "Instal-
Excavations." Foundation Engineering Hand- • lation Effects and their Importance in the
book, Fang, H-Y. (ed.). Van Nostrand Design of Earth-Retaining Structures."
Reinhold, New York. Geotechnique, Voi. 42, No. 1.
DUBROVA, G. A., 1963. Soil-Structure Interac- GUREVICH, V. B., 1969. The River Port Marine
tion: Computation of Earth Pressures {rom Structures. Transport Publisher, Moscow (in
Granular Soils on Marine Structures with Due Russian).
Consideration of Structure Movements. River GURINSKY, M. A., 1986. "Investigation of An-
Transport Publisher, Moscow (in Russian). chored Sheet Pile Bulkheads on Clayey Creep
DUKE, C. M., 1953. "Field Study of a Sheet-Pile Foundations." thesis presented to Leningrad
Politechnical Institute in partial fulfullment
Bulkhead." Translation ASCE, Voi. 118,
of the requirements for the Ph.D. degree (in
Paper No. 2572.
Russian).
DUNCAN, J. M., 1988. "FES Program for Bulk-
IIAKMAN, P. A. and BUSER, W. M., 1961. "Deflec-
head Analysis." Civil Engineering Depart-
tion and Stress Measurements on New Sheet
ment, Virginia Polytechnic Institute and State
Pile Bulkhead at Port Toledo, Ohio." Pre-
University, Blacksburg, Virginia.
sented at the October 19, ASCE Convention,
EAU, 1990. Recommendations of the Committee New York.
for Waterfront Structures, 5th ed. Ernst and
HANNA, T. N., 1982. Foundations in Tension-
Sohn, Berlin. Ground Anchors. Trans Tech Publication and
EHLERS, H., 1910. "Beitrag sur Statischen McGraw-Hill Book Co., New York.
Berechnung Von Spundwanden." Hamburg. HANNIGAN, P. J. and WEBSTER, S. D., 1988.
ENGELS, H., 1903. "Zur Berechnung der "Evaluation ofDrive System Performance and
Bohlwerke." Zentbl. Bauverw., Part 1, 44 Hammer Cushion Parameters." Proceedings
June. of the 3rd International Conference on the
FELLENIUS, B. H., 1991. "Pile Foundations." Application of Stress-Wave Theory on Piles.
Foundation Engineering Handbook, Fang, BiTech Publishers, Vancouver, Canada.
H-Y. (ed.). Van Nostrand Reinhold, New York. HANSEN, J. B., 1953a. Earth Pressure Calcula-
FELLENIUS, B. H., SAMSON, L., THOMPSON, D. E., tion. Danish Technical Press, Copenhagen.
and TRow, W., 1978. "Dynamic Behavior of HANSEN, J. B., 1953b. The Stabilizing Effect of
Foundation Piles and Driving Equipment." Piles in Clay. CN-Post, No. 3. Christianie and
Terratech Ltd. and the Trow Group Ltd. Final Neilsen, Copenhagen.
Sh~et-Pile Bulkheads 691

HEAD, J. M. and WYNNE, C. P., 1985. "Designing Kiev, Ukraine, in partial fulfillment of the
Retaining Walls Embedded in Shift Clay." requirements for the degree of Doctor of
Ground Engineering, Vol. 18, No. 3, London. Philosophy (in Russian).
HUSTAD, P. A and LANGFORD, J. D., 1989. LAsEBNIK, G. Y. and CHERNISHOVA, E. I., 1966.
"Evaluation of a 50-year-old Barge Dock on "Effects of Geometry of Soil Pressure Dia-
the Missouri River." ASCE Proceedings of gram on Stress Distribution in Anchored
Congress on Foundation Engineering: Cur- Sheet Pile Walls." Gidrotechnicheskos
rent Practice and Principles, Vol. 2, Denver, Sfroitelstvo, No. 5 (in Russian).
CO.
LEE, D. H., 1961. Deep Foundation and Sheet
JANBU, N., BJERRUM, L., and KJAERNSLI, B., 1981. Piling. Concrete Publication Ltd., London.
"Anchor Bulkheads." Port Engineering,
Brunn, P. (ed.). Gulf Publishing Company, LEIMDORFER, P., 1979. "Sheet Piling. A Compara-
Houston, TX. tive Study of Steel Sheet Piling Structures."
The Dock and Harbour Authority, February.
JURAN, J. and ELIAS, V., 1991. "Ground Anchors
and Soil Nails in Retaining Structures." LEIMDORFER, P., 1979. "La Saga des Palplanches."
Foundation Engineering Handbook, 2nd ed. PIANC Bulletin No. 34, Vol. III.
Fang, H-Y, (ed.). Van Nostrand Reinhold, New LENZI, M. and ROSETTI, R., 1992. "Evaluation of
York. Ground-Structure Interaction in the Con-
KING, G. J. W., 1995. "Analysis of Cantilever struction of a Wharf Inside the Agip Base
Sheet-Pile Walls in Cohesionless Soil." ASCE North Operating Sector of Port of Ravena."
Journal ofGeotechnical Engineering, Vol. 121, Proceedings 10th International Harbour
No. 9. Congress, Antwerp. Netherlands.
KosLOv, S., 1963. "Field Test of the Anchored LIAKHINTSKY, V. E. et al., 1956. Port Related
Sheet Pile Wall." Morskoy Flot (Seagoing Marine Structures. Maritime Transportation
Fleet), No. 9 (in Russian). Publishing House, Moscow.
KRANz, E., 1953. Uber die Verankerung von LIKINS, G. and RAscHE, R., 1988. "Hammer In-
Spundwiinden. W. Ernst and Sohns, Berlin. spection Tools." Proceedings of the 3rd Inter-
KREY, H., 1932. Erddruck, Erdwiderstand und national Conference on the Application of
Tragfah~gkeit des Baugrundes, 4th ed. W. Stress-Wave Theory on Piles. Bi-Tech Pub-
Ernst and Sohn, Berlin. lishers, Vancouver, Canada.
KUHN, M. R. and MITCHELL, J. K., 1993. "New LITTLEJOHN, G. S., 1972. "Anchored Diaphram
Perspectives on Soil Creep." ASCE Journal of Walls, in Sand-Anchor Design." Ground Engi-
Geotechnical Engineering, Vol. 119, No. 3. neering, Vol. 5, No. 1.
KUMBHOJKAR, A. S., 1993. "Numerica} Evalua- LITTLEJOHN, G. S., 1987. "Ground Anchorages:
tion of Terzaghi's N./' ASCE Journal of Corrosion Performance." Proceedings Institu-
Geotechnical Engineering, Vol. 119, No. 3. tion Civil Engineers, Part 1, No. 82, June.
KYLE, J. M. and KAPP, M. S., 1957. "A Flexible
LITTLEJOHN, G. S., 1992. "Advancing Anchorage
Bulkhead for New York Harbour." Proceed-
Technology." Civil Engineering, July.
ings of the 4th International Conference on
Soil Mechanics and Foundation Engineering, LITTLEJOHN, G. S. and BRUCE, D. A, 1975. "Rock
London, Vol. 2. Anchors State-of-the-Art Part 1: Design and
LACERDA, W. A., 1976. "Stress-Relaxation and Part II: Construction." Ground Engineering,
Creep Effects on Soil Deformation." Ph.D. May.
thesis, University of California, Berkeley. LITTLEJOHN, G. S. and BRUCE, D. A, 1979.
LAMB, T. W., and WHITMAN, R. V., 1969. Soil "Long-Term Performance of High-Capacity
Mechanics. John Wiley and Sons Inc., New Rock Anchors at Davenport." Ground Engi-
York. neering, Vol. 12, No. 7.
LASEBNIK, G. Y., 1961. "Investigation of An- MAcKENZIE, R. R., 1955. "Strength of Deadman
chored Sheet-Pile Bulkheads." Thesis pre- Anchors in Clay." Masters thesis, Princeton
sented to the Kiev Politechnical Institute, University.
692 Sheet-Pile Bulkheads

MACKLEY, F. R., 1977 (reported by Buckley P. J. OHDE, J., 1938. "Sur Theorie des Erddruckes
C). "The History and Development of Sheet Unter Besonderer Berucksichtingung der
Piling." Proceedings Institution of Civil Engi- Erddruckvereiling." Boutechnick, Nos. 10/11,
neers, Part 1, No. 62, London. 13, 19, 25, 37,42, 53/54.
Marine Structures Handbook, 1972. Nickolayev, OSTERMAYER, H., 1977. Practice in the Detail
G. N., ed. Transport Publisher, Moscow (in Design Application of Anchorages. A Reuiew
Russian). of Diaphragm Walls. Institute of Civil Engi-
MATICH, M. A. J., HENDERSON, R. P., and 0ATES, neers, London.
D. B., 1964. "Performance Measurements on OTHMAN, M. A. and EDIL, T. B., 1993. "Force-
Two New Anchored Bulkheads." Canadian Displacement Behavior of Flexible Plate
Geotechnical Journal, Vol. 1, No. 3. Anchors." ASCE Journal of Geotechnical
MATLIN, A., 1983. "Wide Flange Concrete Sheet Engineering, Vol. 119, No. 3.
Pile Wharves." ASCE Proceedings Specialty OVESEN, N. K., 1964. "Anchor Slabs, Calculation
Conference PORTS '83, New Orleans, LA. Methods and Model Tests." Bulletin No. 16,
McGREGOR, T. H., 1963. "T-Shaped Concrete Danish Geotechnical Institute.
Sheet Piles. Renew Deteriorated Bulkhead." PEARLMAN, S. L. and WoLOSICK, J. R., 1990.
Engineering News Record, January 3. "Anchored Earth Retaining Structures." De-
McMAHoN, D. R. (ed.), 1986. "Tiebacks for Bulk- sign Procedures and Case Studies." ASCE
Specialty Conference on Design and Perfor-
head ASCE Special Technical Publication
No. 4. mance of Earth Retaining Structures. Cornell
University, Ithaca, NY.
MEYERHOF, G. G., 1973. "Uplifi Resistance of
PECK, R. B. 1958. "A Study on the Comparative
Inclined Anchors and Piles." Proceedings of
Behaviour of Friction Piles." Highway Re-
the 8th International Conference of Soil
search Board, Special Report No. 36.
Mechanics and Foundation Engineering, Vol.
2.1, Moscow. PFISTER, P., EVERS, G., GUILLAND, M., and
DAVIDSON, R., 1982. "Permanent Ground An-
MITCHELL, J. M., 1987. "Ground Anchorages:
chors: Design Criteria." U.S. Federal Highway
Safety Facto Selection." Proceedings Institu- Administration Report No. FHWA-RD-81-150.
tion Civil Engineers, Part 1, No. 82.
PIERS, D. and MARTINSON, R. E., 1992. "Seismic
MARAYAMA, S. and SHIBATA, T., 1958. "On the Rehabilitation of Seattle's Pier 69." ASCE
Rheological Characteristics of Clays." Part 1, Proceedings Specialty Conference PORTS '92,
Bulletin No. 26, Disaster Prevention Re- Seattle, WA.
search Institute, Kyoto.
PILECKI, T. J., 1985. Discussion ofpaper "Failure
MURAYAMA, S., MICHIHIRO, K. and SAKAGAMI, T., of Anchored Bulkhead" by Daniel and Olson,
1984. "Creep Characteristics of Sands." Soils 1982. ASCE Journal of Geotechnical Engi-
Foundations, Vol. 24, No. 2. neering Division, Vol. III, No. 3.
NAYFAC DM-7.1, 1982. "Foundation and Earth POLSHIN, D. E. and SINELSCIDKOV, S. J., 1953.
Structures." Department of the U.S. Navy, "Discussion of Angle of Internal Friction in
Naval Facilities Engineering Command, Sand." Proceedings of the Research Institute
Alexandria, VA. of Soil Mechanics and Foundation, Moscow
NEELAKANTAN, G., BUDHU, M., and RICHARDS, R., (in Russian).
1990. "Mechanics and Performance of a Tied- PRAKASH, S. and SHARMA, H. D., 1990. "Pile
Back Wall Under Seismic Loads." Earthquake Foundations in Engineering Practice," John
Engineering and Structural Dynamics, Wiley and Sons, New York.
Vol. 19. PuSCH, R. and FELTHAM, P., 1981. "Computer
NEELAKANTAN, G., BUDHU, M., and RICHARDS, R., Simulations of Creep of Clay." ASCE Journal
1992. "Balanced Seismic Design of Anchored of Geotechnical Engineering Division, Vol.
Retaining Walls." ASCE Journal of Geotech- 107, No. 1.
nical Engineering, Vol. 118, No. 6. Recommendation Concernant la Conception, le
NICKOLAYEV, G. N., 1972. Marine Structural Caleul, l'Execution et el control des Tirants
Handbook, Transport Publisher, Moskow, s'Ancrage. Recommendation T.A.72, Bureau
Russia (in Russian). Securitas, 1st ed. Paris.
Sheet-Pile Bulkheads 693

REESE, L. C., AwosHIKA, L., LAM, P. H. F., and SINGH, A. and MITCHELL, J. K., 1969. "General
WANG, S. T., 1990. Documentation of Com- Stress-Strain-Time Function for Soils." ASCE
puter Program GROUP1, Ensoft, Inc., Austin, Journal ofthe Soil Mechanics and Foundation
TX. Division, Vol. 94, No. SMl.
RoWE, P. W., 1951. "Cantilever Sheet Piling in SMIRNOV, G. N. et al., 1979. Ports and Port
Cohesionless Soil." Structural Engineering, Related Structures. Stroyisdat Publisher,
September. Moscow (in Russian).
RoWE, P. W., 1952. "Anchored Sheet-Pile Walls." SNITKO, N. K., 1963. "Static and Dynamic Pres-
Proceedings of the Institution of Civil Engi- sures on Soils and Design of Retaining Walls.
neering, London, Vol. 1. Stroyizdat Publisher, Leningrad (in Russian).
RowE, P. W., 1955a. "A Theoretical and Experi- SOKOLOVSKI, J., 1965. Statics of Granular Media.
mental Analysis of Sheet-Pile Walls." Pro- Pergamon, London.
ceedings of the Institution of Civil Engineer- SOWERS, G. B. and SoWERS, G. F., 1967. "Failures
ing, Part 1, Vol. 4. of Bulkhead and Excavation Bracing." Civil
RoWE, P. W., 1955b. "The Flexibility Character- Engineering, Vol. 37, No. 1.
istics of Sheet-Pile Walls." Structural Engi- STEEDMAN, R. S. and ZENG, X., 1990. "The Seis-
neering, May. mic Response ofWaterfront Retaining Walls."
RoWE, P. W., 1955c. "Sheet-Pile Walls Encastre ASCE, Design and Performance of Earth Re-
at the Anchorage." Proceedings Institution of taining Structures, New York.
Civil Engineers, Part 1, Vol. 4. STEENFELT, J. S., 1985. "Discussion of paper
RoWE, P. W., 1956. "Sheet-Pile Walls at Failure." 'Failure of Bulkhead', by Daniel and Olson,
Proceedings Institution of Civil Engineers, 1982." ASCE Journal of Geotechnical Engi-
Part 1, Vol. 5. neering Division, Vol III, No. 3.
RoWE, P. W., 1957a. "Sheet-Pile Walls in Clay." STROYER, R. N., 1935. "Earth Pressure on Flexi-
Proceedings Institution of Civil Engineers, ble Walls." Proceedings of the Institution of
Vol. 7. Civil Engineers, Vol. 1, London.
RowE, P. W., 1957b. "Limit Design on Flexible SULLIVAN, R. A., 1972. "Behavior of Warf M-
Walls." Proceedings Midland Soil Mechanics fected by River Fluctuation." ASCE Journal of
and Foundation Engineering Society, Vol. 1, the Soil Mechanics and Fo-undation Division,
Paper No. 2. Vol. 98, No. SM9.
SCHULTZ, H., 1976. "The Definition of Factor of SUNDARAVADIVELU, R., IDICHANDY, V. G., GANDHI,
Safety ofMulti-tied Back Walls." Proceedings, S. R., and RAJu, V. S., 1990. "The Tie-Rod
6th European Conference on Soil Mechanics Force Measurements in Cargo Berth." ASCE
and Foundations Engineering, Vol. 2. Vienna. Journal ofWaterway, Port, Coastal and Ocean
ScoTI, R. F., 1963. Principles of Soil Mechanics. Engineering, Vol. 116, No. 1.
Addison-Wesley, Reading, MA. SURENDRA, M. and MUNDELL, J. A., 1985. "Dis-
SELLMEIJER, J. B., CooLS, J. P. A. E., DECKER, J., cussion of paper 'Failure of Anchored Bulk-
and POST, W. J., 1995. "Hydraulic Resistance head' by Daniel and Olson, 1982." ASCE
of Steel Sheet Pile Joints." ASCE Journal of Journal of Geotechnical Engineering Division,
Geotechnical Engineering, Vol. 121, No. 2. Vol. III, No. 3.
SHAHEEN, W. A., CHANG, C. S., and DAMARs, K. SUTHERLAND, H. B., 1988. "Uplift Resistance of
Soils." Geotechnique, Vol. 38, No. 4.
R., 1987. "Field Evaluation of Plate Anchor
Theory in Sand." Proceedings Offshore Tech- TENG, W. C., 1962. Foundation Design,
nology Conference, Paper 5419. Prentice-Hall, Inc. Englewood Cliffs, N.J.
SHESTOPAL, A. 0., 1959. Jetting of Pipes, Piles, TERZAGHI, K., 1943. Theoretical Soil Mechanics.
and Sheet Piles. Hydro Project Institute, John Wiley and Sons, New York.
Moscow (in Russian). TERZAGHI, K., 1954. "Anchored Bulkheads."
SIMPSON, B., 1992. "Retaining Structures: Dis- Transactions, ASCE, Vol. 119, Paper No. 2720.
placement and Design." Geotechnique, Vol. Trnoco, F. H., 1985. "Discussion ofpaper 'Failure
42, No. 4. of Anchored Bulkhead', by Daniel and Olson,
694 Sheet-Pile Bulkheads

1982." ASCE Journal of Geotechnical Engi- Odessa Institute of Civil Engineers, Odessa,
neering Division, Voi. III, No. 3. Ukraine (in Russian).
TOMLINSON, M. J., 1957. "The Adhesion of Piles TSINKER, G. P., 1977. "Performance of Jetted
Driven in Clay Soils." Proceedings 4th Inter- Anchor Piles with Widening." Proceedings
national Conference on Soil Mechanics and ASCE, GT3, March.
Foundation Engineering, Voi. 2. TSINKER, G. P., 1983. "Anchored Sheet Pile Bulk-
TsCHEBOTARIOFF, G. P., 1948. "Large-Scale Model heads: Design Practice." ASCE Journal of
Earth Pressure Tests of Flexible Bulkheads." Geotechnical Engineering, Voi. 109, No. 8.
ASCE Proceedings, Voi. 74, No. 1: also pub- TSINKER, G. P., 1988. "Pile Jetting." ASCE
lished in ASCE Transactions, Voi. 114 (1949). Journal ofGeotechnical Engineering, Voi. 114,
TSCHEBOTARIOFF, G. P., 1979. Foundations, No. 3.
Retaining and Earth Structures, 2nd ed. TSINKER, G. P., 1995. Marine Structures Engi-
McGraw-Hill Book Co., New York. neering: Specialized Applications. Chapman
TscHEBOTARIOFF, G. P., 1948. "Large Scale Model and Hali, New York.
Earth Pressure Tests on Flexible Bulkheads." WIEGMAN, D., 1953. "Messungen an fertigen
Proceedings ASCE, January. Spundwandbauwerken." Vortrage der Bau-
TSCHEBOTARIOFF, G. P. and BROWN, P., 1948. grundtagung, 1953, Deutsche Ges. F. Erd und
"Lateral Earth Pressure as a Problem of De- Grundbau, Hamburg.
formation or of Rapture." Proceedings Inter- WILLIAMS, S. G. O. and LITTLE, J. A., 1992.
national Conference on Soil Mechanics and "Structural Behavior of Sheet Piles Inter-
Foundation Engineering, 2, Rotterdam. locked at the Center of Gravity of the Com-
TSCHEBOTARIOFF, G. P. and WELCH, J. D., 1948. bined Section." Proceedings Institution of Civil
Effect of Boundary Conditions on Lateral Engineers, Structures and Buildings, London.
Earth Pressures." Proceedings International WoonwARD, R. J., LUNDREN, R., and BoiTONO, J.
Conference on Soil Mechanics and Founda- D., 1961. "Pile Loading Test in Stiff Clays."
tion Engineering, 2, Rotterdam. Proceedings 5th International Conference on
TscHEBOTARIOFF, G. P. and WARD, E. R., 1957. Soil Mechanics and Foundation Engineering,
"Measurements with Wiegmann Inclinometer 2.
of Five Sheet Pile Bulkheads." Proceedings XANTHAKOS, P. P., 1979. Slurry Walls. McGraw-
4th International Conference on Soil Mechan- Hill Book Co., New York.
ics and Foundation Engineering, Voi. 2, YAROPOLSKI, 1. V., 1935. Field and Laboratory
London. Investigations of Strength and Stability of
TSINKER, G. P., 1964. "Field Investigation of Per- Sheet Piles Included in Sheet Pile Bulkhead.
formance of the Sheet-Pile Bulkhead Proceedings (TS) IIVT, Voi. 155. Gosstroyis-
Anchored by Means of Piles." The Water dat Publisher, Moscow.
Transport, No. 11 (in Russian). ZARKID, A. Z., 1961. Laboratory Tests ofthe Sheet
TSINKER, G. P., 1972. "Rational Design of Anchor Pile Wall Anchored by Raked Piles. Proceed-
Piles with Widenings at the Ends for An- ings LIVT, Voi. 83, Transport Publishing
chored Sheet Pile Bulkheads." Ph.D. thesis, House, Moscow.
7
Piled waterfront Structures

7.1 INTRODUCTION installation applied to offshore structures


are given in Heavy Construction News
As discussed elsewhere in this text, the (May, 1983), in paper by Shrivastava and
design of a dock structure is influenced by Youdale (1986), Gerwick (1986), and in
many factors; primarily those such as prop- many other publications.
erties of the founding soils at the project In general, piled structures are less sen-
site, required depth of water at the wharf sitive to local scour than other types of
face, intensity of surface loads resulting bottom-fixed waterfront structures. Rela-
from cargo handling and hauling equip- tively flexible and light piled structures are
ment, cargo storage requirements, and mis- suitable for construction in seismically ac-
cellaneous environmentalloads. tive areas where construction of a massive
Where the load on the quay apron is very gravity-type structure cannot be feasible
heavy (e.g., load created by storage of iron due to substantial ground motion. In order
ore or steel products), the foundation soil is to absorb the ground motion in seismically
weak, and the depth of water at the wharfs active areas, modern piled structures are
face is substantial, the light piled struc- usually designed in the form of ductile mo-
tures of miscellaneous designs can present ment-resistant frames with no batter piles
a viable and economica! solution to dock included. Although these structures are
construction. In this type of construction, more expensive than conventional ones,
the load transferred to the piles through they are able to survive earthquakes of sub-
various structural components is carried out stantial magnitude due to their ability to
by piles directly to a lower firm strata. It absorb seismically induced energy by sub-
must be pointed out that piled-type struc- stantial elastic deformation (Fotinos, 1986;
tures have also been successfully con- Cooper, 1991). Additionally, the potential
structed on a bedrock foundation, where the for soilliquefaction during earthquakes also
piles were socketed into the bedrock. Exam- favors the use of piled wharf structures
ples and technical details of this type of pile over gravity-type installations or sheet-pile

695
696 Piled Waterfront Structures

bulkheads. In general, construction of a tion) in combination with a concrete super-


piled structure can reduce the potential for structure. Both vertical and batter piles are
liquefaction because of densification of normally used to resist the design forces
granular soils due to dynamic effects of and loads. Sometimes, where appropriate
pile-driving. Also, overall wharf structure and economically justified, regular timber
stability (the underdeck slope included) is piles and sheet piles are also employed.
improved due to the so-called "dowel effect" In some instances, depending on the pur-
(for details, consult Chapters 3 and 6). The pose as well as on the design conditions, the
latter effect is due to shear strength of pil- piled structure comprises just vertical piles
ing driven through critica! failure circles. able to resist lateral forces of a miscella-
The open piled structures are usually fa- neous nature. Conversely, batter piles are
vored for construction at locations exposed normally used where the structure must be
to heavy waves because they enable practi- prevented from significant lateral displace-
cally free passage of waves. Normally, con- ments. The batter piles are installed in such
ventional open piled structures are not a manner that the structure is not dis-
feasible in heavily-ice-infested waters. For placed appreciably under the effects of the
details on this subject, the reader is re- lateralloads. In this case, both vertical and
ferred to Tsinker (1995). batter piles will not be subjected to sub-
In the past, the piled waterfront struc- stantial bending stresses. In structural
tures have been constructed where the analysis, it is usually assumed that piles
depth of water varies between 10 and 15 m. behave as elastic springs in both vertical
However, the growing trend in the use of and horizontal directions. As regular piles
large and very large vessels for transport- are relatively flexible, their buckling stabil-
ing bulk cargo, as well as liquid and nonliq- ity is of paramount importance. The latter
uid commodities, requires the construction necessitates a limitation of the transmis-
of deep water terminals providing for drafts sion oflarge bending moments to these piles.
of 15 m to more than 30 m. The most recent As stated earlier, this is achieved either by
example of a deep water piled waterfront installation of the batter piles or by increas-
structure, built at Port-of-Rotterdam (water ing the number of vertical piles.
depth 24.5 m) is given by De Gijt et al. The dock superstructure placed on piles
(1993). Earlier, a similar structure was is typically constructed in the form of a
constructed for the Westshore Terminal, relatively thing and, therefore, flexible slab,
British Columbia, Canada to accommodate or in the form of the L-shaped (cantilever or
super bulk carriers with a minimum draft counterfort) wall sitting on piles. The con-
of 29.0 m (Anonymous, PCI Journal, crete superstructure is either cast in place,
N ovember jDecember 1989). erected from prefabricated components, or
Sometimes the cost of piled structures constructed as a combination of both tech-
can be clase to the cost of other types of niques. Where timber piles or sheet piles
waterfront structures. Hence, normal eco- are employed in wharf construction, the
nomic considerations may be decisive in se- deck structure should be placed at least 0.5
lecting the desired type of waterfront con- m below the mean water level to prevent
struction. In most practica! cases, pile- them from decay.
founded waterfront structures comprise It should be pointed out that piled water-
steel or reinforced concrete sheet piles in front structures, particularly those of open-
combination with steel or reinforced con- type construction, are very sensitive to un-
crete regular piles (relieving platforms), or derwater impact by ships with bulbous
just regular steel or concrete piles of miscel- bows. If one or several piles get damaged,
laneous cross sections (open pile construc- then the deck's strength may be affected
Piled W aterfront Structures 697

immediately; a consequence may be deck the case of the relieving platforms, an effi-
damage over a length longer than just the ciently designed fender system can guaran-
impact area. Damage to the structure can tee the structure safe operation. The
range from collapse of the front pile(s) and latter is also true for the open type of
subsequent failure of the quay-side crane construction.
track, to piercing holes through the front It must be realized, however, that too
sheeting. The latter may result in a loss of large an overhang of the deck over piles can
back:fill material. Consequently, the broken result in a heavy, cost-prohibitive deck
piles have to be replaced, and the hole in structure. Too large an overhang may also
the front sheeting piling must be properly necessitate moving the heavy cargo han-
sealed to prevent the back:fill material (usu- dling equipment further away from the
ally sand) from flowing away. Naturally, berthing line at the expense of equipment
the lost back:fill material must be replaced outreach. The most economica} solution is
with a new material. In all instances of usually found in a compromise among the
wharf damage, the underwater repair may selection of an appropriate type of fender
take a long period of time and may be very system, the projection of the deck structure
difficult and expensive. beyond the front row of piles, and the posi-
Usually, damage to piled structures may tioning of cargo handling equipment.
occur where the ship approaches the berth In general, the structure must be de-
at an angle in excess of8-10 degrees andjor signed in a way that damage to one or
where the dock superstructure is not ex- several piles will not lead to complete de-
tended far enough to prevent the contact struction of the dock structure in a chain
between the ship's bow and piles. Ideally, reaction. The designer must also be aware
the harbor mastersjpilots should know the of the risk of sheet-pile damage and conse-
critica} approach angle for all ships calling quence of the outflow of back:fill material in
at port and prevent them in one way or the the event the sheet piling get pierced by a
other from approaching the berth at both ship's bulbous bow.
unacceptable angle and approach speed.
This is normally achieved by better control
of ship movements while approaching the 7.2 General
dock andjor by using tugs to assist ship
berthing maneuvers. As discussed in Chap-
ter 2, better control of ship berthing maneu-
vers can be achieved by the installation of a 7.2.1 Structural Schemes and
sonar system for measurements of the ship Structural components
distance from the dock and her approach
speed and angle. Normally, this type of Basic structural schemes of piled water-
equipment should be standard on quays that front structures are illustrated in Figure
are designed to accommodate large vessels 7-1. These structures can be broadly classi-
with bulbous bows. The economy of installa- fied as the open pile construction with sus-
tion ofthis equipment is based on risk anal- pended deck (Figs. 7-la-7-ld, 7-lf, and
ysis of potential damages that may be in- 7 -lh) relieving platforms of miscellaneous
flicted on the dock by a ship. construction (Figs. 7-le and 7-lg). The for-
In the case of an open structure type, the mer type of structure can be further subdi-
ideal solution to the problem would be pro- vided into those constructed in the form of
jection of the deck (platform) far enough to piled piers, dolphins, and others (Figs. 7-la
prevent the ship's bow from making contact and 7-lb), and those constructed as an ex-
with the front row of supporting piles; in tension of the terminal apron where the
698 Piled Waterfront Structures

{a) {b) {c)

,, l ,"

{g)

_5_ _ 3

Figure 7-1. Basic types ofpiled waterfront structures: (a) pier on vertical piles; (b) pier on
vertical piles with batter piles included to resist lateralloads; (c) marginal piled wharf on
vertical piles only; (d) marginal wharf on vertical piles with a soil-retaining structure placed on
the land side ofthe deck; (e) relieving platform; (O marginal wharfwith a sheet-piling located on
the land side; (g) anchored relieving platform on vertical piles only; (h) piled platform on vertical
piles with anchor slab (Scandinavian type); (i) piled platform resting on closely located bedrock;
(j) anchored piled platform placed on closely located bedrock with land-side located sheeting; (k)
anchored relieving platform placed on closely located bedrock. 1-platform (deck); 2-relieving
platform (superstructure); 3-vertical pile; 4-batter pile; 5-sheet-piling; 6, 7-anchor sys-
tems;· 8-original seafloor; 9-protected slope; 10-bedrock; 11-protection of seafloor from
erosion in front sheet-piling.
Piled Waterfront Structures 699

(i)

. ..
'• ',,
.· :: 1, ::

~ 12

{k )

6
6
5

1
12

Figure 7-1. Continued.

structure is typically founded on a pro- Some soil pressure, however, is trans-


tected sloping bank (Figs. 7- le, 7- ld, 7-lf, ferred to these structures where the sheet-
and 7-lh). In most practical cases, piled pile wall is incorporated into the pile foun-
structures with a suspended deck carries dation on the back of the structure (Fig.
little, if any, earth pressure and are de- 7-lf). This type of construction can be con-
signed to resist vertical loading and hori- sidered as a transition from the piled struc-
zontal berthing and mooring forces only. ture with suspended deck to the relieving
700 Piled Waterfront Structures

platform. This arrangement as well as those typically includes vertical, andjor batter
depicted in Figures 7-1d and 7-1h allow piles and concrete superstructure. The lat-
the width of the suspended deck construc- ter transmits all kinds of loads (e.g., dead
tion to be reduced. The latter, however, is and live loads, soil pressures, mooring
dependent on dock loading, foundation soil, forces, and others) to the pile foundation.
seismic conditions, as well as on hydraulic Structurally relieving platforms are clas-
conditions (e.g., range of tide and wave cli- sified as those with high-level deck struc-
mate). Under certain wave conditions, the ture and these with a deep-level relieving
presence of a vertical wall of any construc- structure (platform). The former are nor-
tion at the deck land side may result in mally used for overwater construction. The
increased water agitation and diminished latter can be built where a substantial tidal
stability of the underdeck slope armor stone range permits construction of the relieving
stability. The aforementioned conditions can platform at low water level, or where the
affect type, number, and size of piles to be whole structure can be constructed in dry
used; for example, just vertical piles or ver- conditions. The advantage of the deep-level
tical in combination with batter piles, driven relieving structures is that size of sheet
or drilled piles, regular precast concrete or piles can be greatly reduced; therefore,
prestressed piles, regular steel piles such as lighter sheet piling can be used. This is
H-piles or pipe piles, or large-diameter possible due to reduced lateral soil pressure
cylindrical concrete or steel piles, and so on the sheeting. The basic purpose of the
forth. platform is to distribute the vertical and
It should be noted that in general, driven horizontal loads among piles included in
piles are less expensive than drilled ones; the wall's pile foundation and to reduce the
the latter is usually used only in the cases earth pressure on the structure by reducing
where foundation conditions make pile- the soil pressure on the sheet piling. The
driving impractical. Where the depth of latter is basically attributed to the shield
pile-driving is limited by the close location effect provided by the superstructure. Fur-
of the bedrock stratum, then piles can be thermore, piles included in the relieving
socketed into the bedrock. Alternatively, the platform walls can reduce lateral soil pres-
number of piles, both bearing and in ten- sure acting on the sheeting. Details on this
sion, can be increased, or the structure can phenomenon are provided in Chapter 4 and
be anchored by conventional means. The illustrated in Figure 4-21.
relevant structural schemes are illustrated Results of model tests reported by Streck
in Figures 7-1i-7-1k. (1950) and discussed in Grundban-
Tidal range and wave climate are very Taschenbuch (1955), and by Tschebotarioff
important factors in dock design because of (1962) indicated that when the ratio dja =
their effects on the underdeck slope stabil- 0.5, where d is the pile diameter and a is
ity and potential erosion around piles. Also, the center-to-center pile spacing, the
the above factors may have a pronounced screening effect can be almost complete so
effect on the value and direction of the that the entire outside pressure could be
mooring forces applied to the structure. taken up by this row of piles. In such cases,
Relieving platforms are usually used as a granular soil confined between the row of
an alternative to the anchored bulkheads, piles and the sheeting would tend to act as
or where sheet-pile bulkheads cannot be if it were inside a silo.
constructed economically. Relieving plat- In some cases, the fill beneath the reliev-
forms are typically comprised of the follow- ing platform is finished to a slope to further
ing basic structural components: waterside reduce soil pressure. Much of the credit for
located sheet-piling, pile foundation that the early development of this type of struc-
Piled Waterfront Structures 701

ture goes to Danish engineers and contrac- nents. These structures are used more
tors who used them often, beginning in the often.
early 1920's. Since then, this type of wharf It should be noted that in countries where
construction has been used extensively cheap labor is readily available, it is more
worldwide. The advantage of this type of economica! to use cast-in-place concrete to
wall is that it can be constructed on weak construct the deck.
soils and carry very heavy loads resulting Advantages of prefabrication are as
from cargo handling and hauling equip- follows:
ment or from superimposed loadings such
as iron ore, packaged steel products, and • Reduction of construction time over water
others. • Usually good quality of deck components
due to efficient quality control at the cast-
As a final note, relieving platforms are
ing yard
also used where piles must be protected • Minimizing the costly formwork and cast-
from impact by large ice floes and from in-place concrete
erosion by moving sediments. • Less dependency on weather conditions
• Structure assembled from standard ele-
ments that are cast in many times reusable
customized formwork
7.2.2 Prefabrication • Deck units typically precast in a controlled
environment at yards with established
quality control procedures
In modern marine engineering practice,
prefabrication is commonplace worldwide.
The disadvantages of prefabrication are
It is commonly used for the construction of
listed below:
most of port related marine structures. Ex-
amples of structures built from precast
• Requirements of suitable heavy lift equip-
components are found elsewhere in this text. ment
Prefabrication is used as an effective mea- • Requirements of a good pile-driving guid-
sure for reducing both the cost and time of ing system to achieve the required align-
port construction. The choice to use precast ment
components in deck structure is based • Small tolerances required for fabrication
mainly on economica} considerations. His- and installation of the precast compo-
torically, piles and sheet piles were the first nents.
precast elements used in marine construc-
tion. Subsequently, with increased avail-
ability of heavy lift equipment and a means 7.3 OPEN PILE STRUCTURES
of transportation of heavy prefabricated WITH SUSPENDED DECKS
blocks, use of prefabricated components for
dock construction is commonplace in mod- As pointed out earlier, this type of structure
ern marine construction practices. can be designed as offshore piers or
In many practica! cases, structures are marginal wharves adjacent to natural banks
assembled from precast components and the or shores, or can be constructed on a man-
cast-in-situ concrete is used only to join made slope. Structurally, they vary widely
these components together. The deck struc- and are distinguished from each other basi-
tures of composite construction also are cally by type and materials used to con-
commonplace. These structures are typi- struct the pile and deck structure and by
cally comprised of cast-in-place pile capping the arrangement of the land side of the
beams and the prefabricated deck compo- structure (e.g., presence of the sheet-piling,
702 Piled Waterfront Structures

conventional retaining wall, or just a Despite the heavy weight of concrete piles
riprap-type transition from platform to at this wharf, their cost was less than the
land). supply price of equivalent steel pipe piles,
Depending on the site's geological and even though larger equipment was required
hydraulic conditions, the size of vessels to to handle, pitch, and drive the heavier con-
be handled at the dock, the type of crete piles. A secondary benefit of concrete
cargo handling and hauling equipment, the piles is found in the reduction of mainte-
availability of some inexpensive, locally nance cost.
available construction materials, as well as The piles for this wharf were manufac-
serviceability requirements (e.g., is the tured with a high level of prestress (8.3
structure designed for short-term or long- MPa) so that a zero-tension condition is
term operation, and other conditions), the maintained when moments are introduced
structure can be constructed from compo- by lateral loads from mooring force, ship
nents made from wood, steel, conventional, impact, wave forces, or earthquakes. ·Fur-
or prestressed concrete. Often structures are thermore, this design feature was beneficia}
built from components made from different for pile durability in the highly corrosive
materials, for example, timber sheet piles, marine environment. Because high stresses
concrete, steel or both vertical andjor bat- are known to develop in long concrete piles
ter piles, and concrete deck structure; other during their driving through soft soils, there
combinations are also used. Some typical were concems that these stresses can cause
pile cracking beyond acceptable values.
examples are depicted in Figures 7-2
Therefore, a dynamic stress testing pro-
through 7-34.
gram was implemented to monitor the
stresses in the pile shaft during driving;
measured dynamic tensile stresses were
7.3.1 Open Piled Offshore Piers suitably below (""' 7 MPa) the design pre-
stress level.
The lateral stability of the loading pier
Examples of these structures are depicted
and the access trestie was ensured by the
in Figures 7-2 through 7-8. The recent installation of batter piles. The deck struc-
example of this type of construction was ture of both pier and the access trestie was
completed in 1983 and is illustrated in Fig- constructed from precast prestressed con-
ure 7-2. This deep water offshore pier was crete structural units; box girders and dou-
constructed at Roberts Bank Delta in ble tee beams were used to span between
British Columbia, Canada (Anonymous, the pile bents. Pile capping beams were cast
1986). It was designed to accommodate very in place and constructed in two stages: Dur-
large bulk carriers. To provide the required ing the first stage, the actual seat for pre-
depth of water and to avoid the extensive fabricated deck components was cast and
dredging, this offshore pier was constructed cured; during the second stage, the concrete
approximately 0.5 km from the shoreline. was poured to fill the space between the
The wharf included the offshore pier (30.5 installed precast beams. This helped elimi-
m high, 335.5 m long, and 35.2 m wide) and nate costly forms for the cast-in-place deck
the approach trestie. Both structures are structure, thereby minimizing time and cost
comprised of the deck structure assembled of over-water construction.
from precast prestressed components in- The box girders used in the access trestie
stalled on high-strength precast prestressed are supported on neoprene bearing pads
concrete cylindrical piles 915 mm in diame- designed to resist the beam's dead load.
ter, each weighing 41 tonnes. This arrangement allowed rotation at the
Piled Waterfront Structures 703

(a)

18.25Cfc CraneRails (b)

Box Glrder

Neop.-
Bearing

(c) (d)

o
,....
N
,....

o
(O

Preeast Panel
Post TensQ!ing
Anchorage

Figure 7-2. Deep water dry bulk loading tenninal at Roberts Bank Delta, British Columbia,
Canada: (a) typical cross section through pier; (b) approach trestie; typical cross section through
pile capping beam; (c) typical detail of fender system; (d) typical cross section through crane
beam.

beam ends until the secondary stage of the upper part of the pile cap (Fig. 7-2b). This
pile cap had been poured. The load transfer provided an efficient connection while using
from the deck structure to piles is accom- the precast girder as a side form for the
plished through the reinforcement extend- second stage of the pile capping beam. The
ing from the first-stage pile capping into the deck components were designed for use with
704 Piled Waterfront Structures

an asphalt cover only, thus eliminating the Bank Delta terminal structure described
need for a costly concrete overlay. The pier above, was completed in 1982 at Hadera,
deck structure is a combination of cast-in- Israel (Figs. 7-4 through 7-6) (Yaron et al.,
place crane girders, pile capping beams, and 1982). This offshore coal unloading facility
precast double tee beams; the latter were included a 300 X 24-m pier structure, 1700-
designed as simply supported units while m-long and 12-m-wide access trestie, and a
under dead load only (construction phase), system of mooring and breasting dolphins
and to perform as a continuous beam for installed on both sides of the pier. An addi-
service loading. This has included loads tional mooring dolphin was installed off-
generated by the 25-tonne-capacity mobile shore, 50 m from the pier head. The pier
cranes and surcharge load. At their ends, structure was placed in a water depth that
the double tees had a diaphragm to transfer varied from 20 to 24 m. It was designed to
shear force to the pile capping beams. accommodate 140,000 DWT bulk carriers
The complete deck structure is ade- with a provision to serve 170,000 DWT ves-
quately stiff to effectively redistribute sels in the future.
between adjoining pile bents the greatest The pier consists of a structural system
design lateral load; that is, ship berthirig comprised of a precast, prestressed concrete
impact. Precast concrete panels with incor- deck structure placed on large-diameter
porated posttensioning ducts were installed steel piles. A typical cross section through
at the pier berthing face to accommodate the pier is depicted in Figure 7-4. Bearing
large buckling-type rubber fenders (Fig. in mind the open-sea conditions, the pier
7-2c). The precast fender panels were lifted deck was elevated to 14.0 m above the mean
into place and posttensioned onto the un- sea level. The pier deck was basically corn-
derside of the front crane track beam. The posed from the precast prestressed un-
joint between the panel and beam was sub- loader tract girders and precast prestressed
sequently grouted. Posttensioning of the deck tee units placed on precast pile cap-
fender panels was eccentric so that under ping beams, monolithically joined with
maximum design load a zero-tension condi- large-diameter steel pipe piles through
tion is maintained across the joint; thus, cast-in-situ joints. The deck components
the resulting shear force is transferred by were monolithically connected with each
friction as well as through shear keys. other, thus forming a continuous space
It should be pointed out that adequate frame structure. The piles used for pier con-
strength is required for the pile capping struction were open-ended steel pipe piles
beam as well as that reliable connections 1525 mm in diameter. The pile bents were
between the piles and the deck structure 20 m center to center. Piles were driven
are crucial for reliable performance of the through a layer of fine sand approximately
pier. For better connection with deck struc- 4-5m thick that overlays a relatively soft
ture, the concrete piles are provided with clay approximately 4-6 m thick, overlaying
an adequate number of embedded dowels, a uniform sand with layers of calcareous
and in the case of steel piles, the dowels are sandstone. The design capacity of the pile
welded to the pile head. For batter transfer (ultimate), 1800 tonnes, was expected tobe
of the bending moments, as well as to pre- achieved at about 30-35 mm penetration
vent the dowels from potential exposure to into the foundation soil. However, a pile
effects of the marine environments, the pile test carried out at the site has indicated
head must be adequately embedded into the that at 31 m penetration depth, the pile's
capping structure. An example is illustrated capacity is equal to approximately 1100
in Figure 7-3. tonnes. It was also found that the soil plug
An offshore pier, similar to the Robert did not form inside the pile. Furthermore, it
Piled Waterfront Structures 705

Figure 7-3. Cast-in-place beam: 1- steel pipe pile; 2-dowel; 3-capping beam reinforcing
steel; 4- form.

was found that the desired pile bearing ca- this pier was provided by batter piles, as
pacity could be achieved only if the piles are depicted in Figures 7-4 and 7-5.
were driven to the depth equal to approxi- The pier structure was designed to resist
mately 70 m. This was basically attributed all relevant dead and live loads, environ-
to the fact that the pile has cut through the mental, seismic, and construction related
sand and calcareous layers with insufficient loads, and load combinations. The design
resistance. Further investigation has indi- was based on the Israeli National Building
cated that the design bearing capacity at a Code, as well as on the relevant interna-
penetration depth of approximately 35 m tional standards.
can be achieved through installation inside The pier was protected from ship impact
the pile of the cast-in-place concrete plug, by fiexible dolphins made from steel pipe
approximately 12 m long or a special inter- piles, 2600 and 2800 mm in diameter with
na! steel annulus ring welded to the pile 15 wall thicknesses ranging from 34 to 70 mm.
m from the pile open end. The latter solu- The dolphins were placed on both sides of
tion was finally adopted for pier construc- the pier and were provided with cylindrical
tion; it was far less expensive than driving rubber fenders, 2.6 m outside diameter, 1.5
piles to the approximately 70 m penetration m inside diameter, and 4 m long. Dolphins
otherwise needed to develop the desired located on opposite sides of the pier have
bearing capacity. been connected with each other by steel
Transverse and longitudinal stability of pipe struts. Thus, the combined capacity of
706 Piled Waterfront Structures

Figure 7-4. Hadera, Israel-Offshore coal unloading terminal. Typical cross section through
pier: !-vertical steel pipe piles 1525 mm in diameter; 2-battered steel pipe piles 1525 mm in
diameter; 3-flexible dolphin (steel pipe piles 2600 and 2800 mm in diameter); 4, 5-precast
components of deck structure; 6-precast reinforced concrete pile capping beam; 7-mooring
accessories; 8-fender system; 9-catwalk; !O-design ship super bulk carrier 170,000 DWT.
(Courtesy ofYaron, Shimoni and Partners Consulting Engineers, Ltd., Tel Aviv, Israel)

120 IJNT 60 IJNT

GENERAL LAYOUT OF PIER LEGEND'


MD. - tviOCI'lN3 [JQPHIN
B O. - ER:ASTii'G OOLPHIN
WCK RELEASE HOOKS
QRH. •
B. - BOLLARD
ALL I»-ENSS()olS ARE METRIC.

ELEVATION

Figure 7-5. Hadera, Israel-Offshore coal unloading terminal. Pier general layout and
elevation. (Courtesy of Yaron, Shimoni and Partners Consulting Engineers, Ltd., Tel Aviv,
Israel.)
Piled Waterfront Structures 707

Figure 7-6. Hadera, Israel-Offshore coal unloading terminal. Access trestie during construc-
tion. (Courtesy ofYaron, Shimoni and Partners Consulting Engineers, Ltd., Tel Aviv, Israel.)

both dolphins was adequate to absorb picted in Figure 7-7. Here the pier's lateral
berthing energy of the 170,000 DWT ship and longitudinal stability is provided by
approaching the berth at a maximum dock- conventional 350-mm square prestressed
ing speed 20 cmjs. batter piles, whereas the vertical loads are
The 12-m-wide access trestie was con- transferred to the prestressed concrete
structed of approximately 200-m-long sec- piles with enlargements. The deck struc-
tions. It was designed to accommodate a ture in this pier is of cast-in-place type
4.5-m-wide roadway and the two 1500-mm construction.
conveyor belts. The deck structure of the Practical examples of pile arrangements
access trestie was of composite construction used for construction of the open-type off-
that included precast concrete slabs on steel shore piers are illustrated in Figure 7- 8. In
plate girders, supported on pile bents placed case (a), the pier deck is founded on just
at every 40 m. The typical pile bent has two vertical steel pipe piles; whereas in cases
large-diameter steel batter piles providing
(b) and (c), the pile foundation consists of
the trestie transverse stability; the longitu-
vertical and batter prestressed concrete
dinal stability was provided by the struc-
[case (b)] and steel [case (c)] piles. As it is
tural system that included four batter piles
located in the middle of each typical section seen from these and the earlier discussed
200 m long. Some details of the access tres- examples, as well as from many other ex-
tie are seen in Figure 7- 6. amples of the offshore pier constructions,
Where required, the capacity of both the piling and deck systems of these struc-
bearing and tension piles included in any tures, as well as methods of protection from
pier structure can be enhanced by providing ship impact, vary from project to project
the piles with enlargements of miscella- and, to a great extent, are site-specific. This
neous shapes and sizes. An example of an is best illustrated by the recent construc-
offshore pier founded on such piles is de- tion of two piers built for the U.S. Navy.
708 Piled Waterfront Structures

(a)
(b)

Figure 7-7. Offshore pier: 1-prestressed concrete piles 350 X 350 mm with blisters;
2-concrete pier to support fender system; 3-battered (3 : 1) concrete piles 350 X 350
mm; 4-cast-in-place concrete deck; 5-wooden fender pile; 6-timber fender system;
7-pier outline. [From Gorunov (1984).)

One of these piers was built off Treasure end. Beneath the silt are soils with rela-
Island, located approximately 9.3 km east tively high strength and low compressibil-
of the Golden Gate Bridge on the west coast ity. It was concluded that under existing
of the United States, the other was con- site conditions, a pile system comprised of
structed at the new Strategic Homeport prestressed 670-mm octagonal vertical and
Naval Station on NewYork's Staten Island batter piles is the most functional, con-
(Lo and Beckwith, 1987; Buslov et al. 1988; structible, and cost-effective. To ease han-
Post, 1989). The former pier, 284 m long dling and driving of raked piles, the batter
and 37 m wide, was constructed in 11.6 m used was 1 : 6. Although requiring more
of water. The critica! deck loads of this piles, this system is more effective than
structure were calculated based on the use that where large-diameter prestressed con-
of the 82-tonne mobile crane. Other loads crete cylindrical piles would be used.
included ship impact, mooring forces, and The local contractor found that the fabri-
severe earthquake-induced loads. Founda- cation cost of large cylindrical piles would
tion soils are represented by loose to have been about twice as much as for 610-
medium dense sand underlain by dark gray mm octagonal piles. Besides, this contractor
claylike silt. The lower material reaches a also concluded that the large and heavy
depth of about 33 m near the midpoint of cylindrical piles would have created han-
the pier, tapering to 17 m at the end of the dling problems and construction delays and
structure, and to about 24 m at its land would have required the use of special han-
Piled Waterfront Structures 709

(a) (b )

(c )

Figure 7-8. Typical pile foundation for offshore piers: (a) Japan- Construction of offshore pier
comprised of large-diam.e ter vertical steel pipe piles; (b) offshore pier for container terminal in
China; pile foundation comprised of concrete vertical and barrier piles; (c) construction of
offshore pier in Japan; pile foundation comprised of vertical and batter steel pipe piles.

dling and driving equipment. These prob- ter depth to serve the battleship Iowa and
lems and cost considerations were taken its accompanying ships. The subsoil condi-
into account for selecting the pier piling tions at pier site are as follows: close to
system. shore, there was compact fill underlined by
In the second case, the 430-m-long, 27.4- glacial outwash sitting on weathered rock;
m-wide pier was constructed in 13.7 m wa- beginning at the middle third of the pier,
710 Piled Waterfront Structures

piles were driven through organic clay into 1989). However, in this case, the concrete
weathered rock. To resist horizontal and pier deck was founded on steel pipe piles
vertical design loads, the original pier de- 1067 mm in diameter. Again, no batter piles
sign called for an array of vertical and bat- were used.
ter piles. The piling system included 610-
mm square prestressed concrete piles and
915-mm-diameter cylindrical prestressed
piles with the batter piles installed parallel 7.3.2 Piling
and transverse to the pier.
The contractor, who had access to a 310- It should be pointed out that the choice
tonne-capacity fioating crane, proposed and between steel and concrete piles for marine
eventually constructed the new piling sys- application is always a controversial issue
tem that included just vertical prestressed and, as it is seen from the preceeding exam-
concrete cylindrical piles 1372 mm in diam- ples, the solution to the problem is usually
eter. Some details of this structure are il- site-specific.
lustrated in Figures 7-32 and 7-47. Concrete piles, and particularly pre-
The new pile system offered adequate stressed concrete piles, are usually consid-
resistance to the design loads without bat- ered as being relatively maintenance-free in
ter piles. The pier was designed as a frame severe marine environment. However, con-
with its columns performing within an elas- crete piles, and particularly large-sized-pipe
tic foundation. The spring constants re- piles, are heavy to handle. Furthermore,
quired for the frame analysis were deduced long concrete piles can develop high tension
from pile tests carried out at the site. From stresses if driven into soft soil where ten-
these tests it was concluded that shorter sion waves bounce back up. It has also found
piles at the pier's land portion were stiff, that large-diameter cylindrical piles in-
fiexing essentially as a short rigid column; stalled by a vibrator can be exposed to the
however, long piles driven to about 30-35 considerable interna! hydrodynamic stress
m below the dredge line went through coun- (up to 2-3 MPa) developing within the pile
terfiexure to develop complete fixity at their during the driving process. If the pile is not
embedded part. The piles in the pier middle properly reinforced, the latter phenomena
section behaved somewhat between the two may ·result in severe pile cracking (Savinov
extremes. and Lavrinovich, 1969; Buslov, 1974). Natu-
Due to the fact that the bending strength rally, if concrete cracks, the surface will
ofthe proposed 1372-mm-diameter piles was spall after freezejthaw cycles, oras a result
four times that of the originally designed of the reinforcement corrosion. Prestressing
915-mm-diameter piles and eight times that helps greatly to overcome the pile-cracking
of the 610-mm square piles, the number of problem. Long concrete piles are usually
required piles was cut almost in half, also precast in sections which could be conve-
eliminating the need for batter piles. U se of niently spliced during the installation pro-
just yertical piles simplified the pile cap cess. Splicing enables use of the shorter
geometry, opening the way for extensive and, thereby, lighter pile sections and re-
use of precast deck components. In total, duces the danger of cracking due to exces-
the new structure was more cost-efficient sive handling. The most typical splices are
than originally proposed and was effectively welded and bolted; mechanical splices of
executed. miscellaneous designs are also used.
Similar experience has been gained on Sometimes the use of a steel pile in a
the construction of another pier built for saltwater environment is rejected .because
the U.S. Navy in New York harbor (Green, of concerns about corrosion. Rusting and
Piled Waterfront Structures 711

corrosion of steel in salt seawater is in- life. In some cases, the sacrificial material
evitable and can be a serious problem un- can be added to the dock components in a
less continuously protected by cathodic pro- zone of maximum corrosion only. In such
tection, coatings of miscellaneous nature, or zones, special protection can also be pro-
a combination of both. Cathodic protection vided locally (e.g., concrete jacketing, plas-
along or in combination with a protective tic facing, etc.). For more details on corro-
coating can be effective where the steel pile sion protection of steel components of the
is in the soil or in the submerged zone. It is dock in severe marine environment, the
only partly effective in the tidal zone and is reader is referred to Tsinker (1995).
ineffective in the splash or atmospheric Despite the need of protection from cor-
zones. In these zones, the use of concrete rosion, in some form, the advantage of steel
jackets around piles or the use of a coating piles and, particularly, large-diameter pipe
of coal-tar epoxy are the most frequently piles is that they provide the contractor
used methods for pile protection. In recent with flexibility in choosing the driving
years, polyethylene coats developed by sev- equipment, allowing relatively easy pile in-
eral Japanese steel makers has been intro- stallations, both vertical and batter, and
duced and successfully used as protection of allowing for the relatively simple field ad-
steel piles from corrosion (Tominaga and justments to pile length. Furthermore, steel
Heidengren, 1988). is less susceptible to damage in storage and
The economy of use of a certain type of during shipping and installation. In some
steel protection from corrosion should be cases the aforementioned advantages of the
compared with an allowance for corrosion large-diameter steel pipe piles, as compared
added to the thickness of dock steel compo- with alternatives such as concrete cylindri-
nents, such as piles and braces, as well as cal piles, makes the use of steel piles more
deck components, used with no protection cost-effective, faster, and safer in construc-
at all. In the absence of specific data on tion. Thus, to summarize, selection of the
rates of corrosion in particular areas, the most attractive structural solution, that is
corrosion rate on the order of 0.075-0.1 mm always site-specific, must be made on the
per year of service is often considered. hasis of alternative designs. More informa-
Hence, a steel thickness increase by about 5 tion on pile structures and pile design is
mm for 50 years of service is commonly provided in Section 7.5.
considered. Another approach used is to de-
sign the steel components of the pier based
on the mild steel properties but, adopting
the high yield steel in practice. This method, 7.3.3 suspended Deck
however, proved to be too uncertain and too structures for Marginal
inefficient to cope with potential problems. Wharves
Sometimes some increase in stress value
due to corrosion has been accepted, pro- Suspended deck construction for marginal
vided that the highest stresses were not in wharves have been used worldwide. They
the zone of maximum corrosion. are typically constructed .in harbors pro-
Initial pile protection by coal-tar epoxy, tected from heavy waves and where there is
or by other methods, may well protect the no substantial ice movements. In some in-
structure for some time (typically for 5-10 stances, where ice movements can damage
and more years). Therefore, it may be possi- piles, the front row of piles is designed to be
ble to combine the initial steel protection strong enough to absorb impact from drift-
from corrosion with a further addition of ing ice features such as ice floes, ridges, and
sacrificial steel for the rest ofthe structure's so forth. For details, consult Tsinker (1995).
712 Piled Waterfront Structures

Essentially, suspended deck structures are Where the open piled structure is built
indispensable where poor soil conditions or exposed to waves and current locations,
substantial depth of water prevent con- the underdeck slope would require heavy
struction of structures such as sheet-pile armor protection from errosion. This usu-
bulkheads, gravity-type quay walls, and ally adds considerable cost to dock construc-
others. tion. Despite this, open pile structures can
These structures are built to cover the stiH be cost-effective at exposed locations
width of a sloping bank which is usually and can be more economica! than other
protected from scour effects and wave ac- structures suitable for construction on soils
tion. Typically, the structure itself carries that permit pile-driving (e.g., sheet-pile
little, if any, lateral earth pressure load and bulkheads, steel sheet-pile cells, and
is designed to resist verticalloading in com- others).
bination with berthing or mooring forces
only. In some cases, it supports sheeting at
the back, which retains fill. Sheet-piling or
the other types of constructions on the land 7.3.4 Basic Design Principles
side of the platform (e.g., conventional re-
taining wall), are normally used to reduce Depending on the intended use of the dock
the width of the deck structure. As pointed structures (e.g., for long-term or short-term
out earlier, depending on foundation soil service, type of cargo andjor cargo han-
and loading conditions, the deck can be dling and hauling equipment, soil condition,
founded on just vertical piles or a combina- availability of certain structural materials
tion of vertical and batter piles. or prefabricated components, and site hy-
At locations with substantial seasonal draulic conditions), the structure can be
water fiuctuations, or with high tides, the constructed from components roade from
pile's effective freestanding length can be timber, steel, concrete, or combinations of
reduced by means of bracing. Substantial all of these materials. Severa! characteristic
water fiuctuations at the dock site can ag- examples are illustrated in Figures 7-9
grevate the uriderdeck slope stability prob- through 7-15.
lems. It may also create problems with dock The structure that is depicted in Figure
fendering; use of a special fender system for 7-9a is the simplest form of the open-type
protection of the dock and, in particular, its construction roade from wood. It was de-
piled foundation from ship impact is nor- signed to serve small ships, mostly river
mally required in large tidal range areas. In barges up to 600 DWT. To resist ship im-
some instances, the latter can increase the pact and mooring forces, it is braced at
cost of construction considerably. three levels. The structure's deck is de-
Also, where the tidal range is substantial signed to support loads from mobile cranes
and the bottom of the water edge (fascia) up to a 5-tonne lifting capacity. Clusters of
components such as craneway beam or con- piles are driven in front of the structure to
tinuous fender panel are submerged deep protect it from impact from ships and ice
under the water level, air, pressure en- fioes. These piles are integrated into the
trapped under the deck at the high tide can structure and also serve as a mooring de-
produce uplift load on the deck. Alterna- vice. This type of construction could be ef-
tively, it may lead to pavement erosion just ficient for construction at remote sites on
behind the land end of the deck (Leitass, inland waterways where wood is available
1979). This can be prevented by the instal- locally at low cost and a local labor force is
lation of vent pipes, either at the fascia readily available. In the other example de-
member or in the deck structure. picted in Figure 7-9b, the structure was
Piled Waterfront Structures 713

(a)

(b)

Figure 7-9. Braced marginal wharf: (a) constructed in Russia


from wooden components; (b) founded on screw piles 127 mm in
diameter (screw 1220 mm in diameter), Sumatra, Indonesia.

constructed on weak soils using steel com- The structure's longitudinal and trans-
ponents. In this case, the steel frame made verse stability is ensured by steel braces,
up from steel pipes 127 mm in diameter some of which were installed underwater;
and braced by steel pipes 44.5 mm in diam- the middle level of horizontal braces was
eter is founded on screw piles 127 mm in installed just above the minimum water
diameter with screws 1220 mm in diameter. level. The deck was built from timber planks
For protection from corrosion, the piles were capable of handling light mobile cranes and
wound around by steel wires and then cov- trucks.
ered by a layer of cement mortar 4-5 cm Screw piles derive their support from the
thick. large area of the screw and are usually
714 Piled Waterfront Structures

economica! where poor foundation soil is with the addition of special reinforcement;
present. These piles have been successfuliy this created a continuous moment-resisting
used in the past. However, at present, due frame. The ductility of this structure was
to considerable improvements made in achieved through the ability of ali struc-
equipment for pile-driving andjor installa- tural components of the frame to deform
tion of boring piles, in most practica! cases elasticaliy; that is, through the frame's abil-
it is more economica! to use conventional ity to translate a certain distance under the
but longer piles, instead of shorter screw lateralload that mobilizes elastic resistance
piles. while retaining structural integrity. This
In modem marine engineering practice, was achieved by the addition of ductile rein-
conventional and large-diameter cylindrical forcing steel to the frame component and by
concrete piles, and steel piles of miscella- the use of heavy, closely spaced stirrups
neous cross sections, are normaliy used for andj or spiral reinforcement in prestressed
construction of open pile structures. De- concrete piles, and adding confinement steel
pending on load conditions, the pile system in the concrete deck in areas of high bend-
may include vertical piles only, or be con- ing moment.
structed with vertical and batter piles. In Several studies (Sheppard, 1983; Park
seismic-active zones, the ali-vertical pile, a and Falconer, 1983; and others) have
ductile moment-resisting space frame sys- pointed out the importance of correct detail-
tem with no batter piling, was recently ing of prestressed pile reinforcement, par-
most often used. Structures of this type ticularly the provision of heavy closely
have been constructed in Oakland and Los
spaced spiral reinforcement in regions near
Angeles, California and for the Squamish
the pile-deck (pile cap) connection and in
Terminals, British Columbia, Canada,
the soil where the pile achieves fixity. Natu-
where a concrete deck was cast in place
raliy, longer piles are more flexible than
on vertical prestressed octagonal piles
shorter piles of identica! cross section.
(Shrivastava and Hunt, 1989; Birdy et al,
Therefore, under certain design conditions
1989; Cooper, 1991), in Boston Harbor,
where the deck was placed on steel pipe by isolating the structure from the shore by
piles 610 mm in diameter (Ericson et al., adding the transitional structure (Fig.
1993), and at many other locations. An ex- 7 -lOc), the designer is able to achieve an
ample of the ductile space frame, construc- optimum balance between longer, and
tion with deck structure composed from therefore more flexible, piles offshore and
prefabricated prestressed concrete compo- the shorter, stiffer piles near the land. This
nents is shown in Figure 7-lOa. Here the may help to avoid overstressing the shorter
structure width, as dictated by the rela- piles under lateral loads. The land-based
tively narrow crane gauge, was reduced by abutment for the transitional structure may
means of blockwork retaining wali placed be constructed in a variety of ways. Depend-
on the land side of the deck. An ali-vertical ing on site geology and hydraulic condi-
pile system was designed to resist lateral tions, it may be designed as a gravity-type
loads on the wharf, including ship berthing wali, piled abutment, or freestanding or an-
forces, mooring forces, forces from wind on chored sheet-pile wall. The transition span
cranes and trains, as weli as seismic forces. should bear on the wharf structure on a
Prefabricated pile caps provided support for low-friction bearing which serves to isolate
the concrete deck assembled from the pre- the more flexible piled structure from the
cast concrete slabs. All prefabricated com- stiffer land-based abutment structure. Also,
ponents were joined by welding together as pointed out by Weidler et al. (1987) and
dowels projected from these components Dailey et al. (1987) under a certain con-
Piled Waterfront Structures 715

(a)

(b)

Figure 7-10. Open piled marginal wharf. Typical cross sections. (a)
The deck structure comprised from precast concrete component is
placed·on prestressed concrete piles, surmounted by precast concrete
caps; (b) the deck is cast in situ on prestressed concrete piles; (c)
structure with transitional deck section at land side. 1-prestressed
concrete piles 400 X 400 mm; 2-precast pile cap; 3-concrete deck
comprised of prefabricated elements; 4-rebars and dowels extended
from deck components and pile; 5-concrete filter; 9-backfill;
10-octagonal prestressed piles 460 mm in diameter;
11-460 X 460-mm prestressed piles; 12-quarry muck; 13-rock
armor; 14-cast-in-place concrete deck; 15-ballast material;
16- 230-mm-thick paving; 17-blockwork wall; 18-crane track;
19- land-based abutment; 20-steel pipe piles; 21-prefabricated
transition concrete deck.
716 Piled Waterfront Structures

(c)

Figure 7-10. Continued.

struction conditioning the pile fatigue must 9.0


be taken into consideration.
For safe portal crane operation during
heavy dynamic loads imposed on the dock
structure and during a seismic event or
heavy ship impact, both the waterside and
land-side crane niils must be provided with
a reliable support. For example, in the cases
illustrated in Figures 7-lOa and 7-lOb, this
has been achieved by the installation of two
closely spaced piles andjor by installation
of batter piles. Other altematives (e.g., 91600 l=9mm

large-diameter piles, pile clusters, etc.) are


also used to support crane rails. Essen-
tially, ali piles in the ductile frame should
be stronger than if they had to support
gravity loads only. The underdeck riprap
prism helps to anchor the structural frame
effectively. Naturally, it must be seismically
stable. Also, rocks included into the prism Figure 7-11. Marginal wharf of open-type construc-
must be heavy enough to resist effects of tion with retaining wall comprised of prefabricated
waves and natural andjor ship-propeller- counterfort sections constructed in Japan.
induced currents.
The ductile space frame with only two of the deck was shortened by the installa-
vertical piles per bent (Fig. 7-11) was built tion of the prefabricated concrete counter-
in Japan (Fucute et al. 1990). In this case, fort retaining wall on the land side of the
the cast-in-place concrete deck has been structure; the space between both struc-
constructed on a pile system consisting of tures was covered by prefabricated concrete
steel piles 600 mm in diameter. The width slabs that could slide at the platform end
Piled Waterfront Structures 717

when it translates the effects of lateral ture and the underdeck slope) are almost
loads. identica!.
As previously stated, the wall's trans- The deck is typically constructed in a
verse displacement can be limited by either form of a rather thin (0.45-0.5 m) cast-in-
the increased number of vertical piles or by situ flat slab overlaid by the granular bal-
the introduction of batter piles. The latter last approximately 1.0 m thick. A granular
solution is illustrated in Figure 7-10b. Be- ballast is normally used to distribute the
fore the 1980s, this type of construction was concentrated loads on larger areas of a deck.
widespread along the west coast of the It is also used for damping the dynamic
United States. The structure depicted in loads generated during the operation
Figure 7-10b was constructed in the early of some cargo handling and hauling
1980s in Port-of-Los Angeles for the con- machinery.
tainer wharf. The cast-in-situ flat concrete The practice, however, has indicated that
deck of this structure was supported on ballast placed on a concrete deck often does
457 -mm square and octagonal concrete piles not have any practica! merits. Quite con-
with lateral forces resisted by a system of trary, it increases the dead load on piles,
batter piles. Similar structures were con- traps moisture, and prevents visual inspec-
structed at a great many ports in N orth tion ofthe deck upper surface. Hence, where
America and elsewhere in the world. Recent not justified by practica! requirements, the
observations of a number of dock structures use of the ballast should be avoided. In
some instances, a ballast of sufficient depth
in Oakland, California damaged by earth-
could be useful to accommodate services
quakes has indicated that structures with
piping and electrica! conduits placed above
batter piles included in pile arrangement
the structural deck. Also, in some cases, a
were more susceptible to damages by sud-
ballast could be effective in insulating the
den earth movements than those founded
deck structure from direct exposure of the
on vertical piles only.
sun and for maintaining its temperature
In the former structures, the batter and
level close to the water temperature.
some vertical piles and their connections to A very important component of the
the deck have been severely damaged dur- marginal open pile structure is its under-
ing earthquake shocks (Cooper, 1991). Pile deck slope. In the aforementioned case
damage basically resulted from the low duc- histories, the underdeck slope was approxi-
tility of the relatively stiff batter piles. The mately 2.5 : 1 seismically stable. It was pro-
latter resulted in higher dynamic loadings tected with a riprap having the exposed to
during earthquakes than might have oc- water slope 1.5 : 1.0 (horizontal to vertical).
curred in the total vertical pile system. In one specific case, piles were installed
During retrofitting of the damaged by a standard hammer. The pile installa-
wharves, ali the batter piles were removed tion was assisted by a water jet. These piles
from the structures and replaced with verti- have been installed after the slope had been
cal piles, essentially a ductile moment- dredged and prior to placement of the rock
resisting system. It should be pointed out prism. Subsequent installation of the rock
that in the wake of the experience gained prism displaced the pile heads laterally by
during several recent high-intensity earth- about 180-380 mm with about 40 mm of
quakes that hit California, several struc- this movement occurring after the fiii mate-
tures in the are a were constructed on verti- rial was placed behind the newly con-
cal piles only (Wittkop, 1986). With the structed wharf. The cause of prism move-
exception of the piling system, however, ali ment was not apparent. Tentatively, it was
other dock components (e.g., the deck struc- assumed to be the result of a prism sliding
718 Piled Waterfront Structures

due to insuf:ficient support at the toe, incor-


rect sequence of riprap placement, settle-
ment, or combination of all the above. In
another case, the prism was placed :first and
then piles were driven through the riprap.
This has resulted in substantial pile dis-
placements. As pointed out by Wittkop
(1986), pile-driving through riprap resulted
in settlement of the rock:fill during the pile
installation process.

7.3.5 suspended Deck


Structures Founded
on Large-Diameter
Cylindrical Piles
The chief advantage of this type of con-
struction is the simpli:fication of the piling
system due to a substantial reduction in the
number of piles. This piling system allows
for the relatively easy use of prefabricated
components, and, in particular, precast con-
crete capping beams, for deck construction.
Four characteristic examples of these
structures are illustrated in Figures 7-12
and 7-13. In all four examples, the concrete
deck, assembled from precast concrete com-
ponents, is founded on prestressed, vertical,
large-diameter, concrete cylindrical piles. In
each representative case, the number of
piles per bent, the pile diameter, as well as
the space between adjacent bents were se-
lected on the hasis of alternative designs
related to the particular site conditions. In
the case illustrated in Figure 7-12a, the Figure 7-12. Open-type structure with deck assem-
wharfs deck is founded on concrete piles 1.0 bled from prefabricated components founded on large-
m in diameter. In this particular case, the diameter concrete tubular piles: (a) typical structure
width of the deck is reduced by the use of for construction on competent foundation soils (water
precast sections of a counterfort retaining depth from 10 to 15 m); (b) Port-of-Kaliningrad, Rus-
sia; Wharf constructed on dense sand overlaid by weak
wall installed at the land side of the deck. soils. 1-prestressed concrete tubular piles 1.0 m in
The wall sections, each 5 m long, were ăiameter; 2-precast prestressed deck components;
placed behind the deck on previously in- 3-precast pile cap; 4-concrete paving; 5-pre- ·
stalled rock:fill slope protection. The piles' stressed concrete tubular piles 1.2 m in diameter; 6-
strength and their penetration into medium portal crane track; 7-train track; 8-slope protection
(riprap); 9-granular material as substitution for local
dense sand were sufficient to support verti- weak soil; 10-sand; 11-precast prestressed pile cap-
cal loads produced by portal cranes with a ping beam; 12-prefabricated concrete retaining wall;
lifting capacity of 30 tonnes, and by two 13-slit with pit layers; 14-sand clay; 15-loam.
Piled Waterfront Structures 719

(a)

·. 1:· :. : .... : : .. :· . ·.

. . . """'=.L............ n:.. H}~r~\/•······-·····.·.·


(b)

Figure 7-13. Open-type structure with deck assembled from prefabricated


components founded on large-diameter cylindrical concrete piles: (a) typical
structure for construction on competent foundation soil for a water depth of
10-15 m; (b) wide pier at Port-of-Novorosiysk, Russia. 1-prestressed
concrete tubular piles 1.6 m in diameter; 2-prefabricated prestressed
T-shaped concrete beam; 3-cast-in-aitu prestressed continuous concrete
beam; 4-concrete retaining wall assembled from precast sections; 5-cast-
in-place concrete cope; 6-sand; 7-riprap; 8-granular filter; 9-portal
crane track; 10-train track; 11-precast prestressed components of deck
structure; 12-rock amor; 13-precast prestressed concrete pile capping
beam; 14-concrete cover plate; 15-concrete plug; 16-concrete socket;
17 -clay; 18-marl; 19-dense clay with inclusion of gravei and small
cobbles; 20-concrete topping.
720 Piled Waterfront Structures

trains sitting on the deck. Piles were also as the use of a limited number of a stand-
designed to resist lateral forces imposed on ard precast components.
the dock by ship impact and by mooring Alternatively, the deck structure can be
loads. constructed as a combination of precast pile
The deck structure was comprised of pre- capping beams with a precast :flat one-way
cast prestressed pile capping beams and slab as illustrated in Figure 7-13b. In this
precast T -shaped girders used to span the case the prestressed concrete piles 1.6 m in
space between pile bents. The weight of diameter were fabricated by the centrifugal
prefabricated blocks varied between 50 and method in 8 m long pieces then welded
100 tonnes. The advantage of this deck sys- together to form piles 24 m long. Piles were
tem is that the webs ofT-girders were used driven by a powerful vibrator to bedrock
as supports for crane and railway tracks. A and socketed into it. Extremely accurate
further advantage was that the :flat bottom pile alignment in ali directions was achieved
of the deck did not trap moisture and, through the use of a special :floating tem-
therefore, was less susceptible to long-term plate that allowed for pile installation in
deterioration. two adjacent bents while remaining at-
Another advantage is that this deck sys- tached to the previously installed piles. The
tem is accessible for inspection and mainte- pile tips were protected by steel shoes. The
nance. A similar structure has been con- space between bents was 12 m and the
structed for a general cargo berth in length of the typical wharf section was equal
Visakhapatnam Port in India (Ansari and to 36 m. After the erection of the pile cap-
Kumar, 1986). There the wharfwas built in ping beams were completed, and they were
water 16.5 m deep. The wharfs deck of a secured in place by reinforced concrete
composite construction was founded on con- plugs, the one-way prestressed concrete
crete cylindrical piles 1.22 m in diameter, slabs, 0.7 m thick, were installed to form
four piles per bent, at 6 m center to center. the deck structure. To allow for lateral
The precast capping beams were placed on movements due to temperature effects, the
piles, thus forming stiff bents at 7.4 m cen- 12-m-long deck sections, comprised of typi-
ter to center; the secondary T-beams and cal one-way slabs were placed between two
special facial beams were installed on cap- adjacent wharf sections.
ping beams and used as the formwork for In the case depicted in Figure 7-13a, the
the cast-in-place deck slab. In this case, structure was designed to support a very
however, the T-beams were reversed to pro- heavy crane load as well as two railway
vide a smooth working surface for placing tra.cks. Longitudinal prestressed concrete
the concrete overtop. girders were designed as cast-in-place con-
The structure illustrated in Figure 7-12b tinuous beams with the standard span 16 m
was constructed on relatively weak founda- long and the end span at both sides of the
tion soils that overlay a deep stratum of beam equal to 8 m; precast prestressed con-
dense sand. Four tubular piles per bent crete beams installed between the longitu-
were required there to resist design vertical dinal girders were designed to support two
and lateralloads. Precast concrete pile caps railway tracks and the surcharge load 40
2 X 2 m were installed on each pile head kN jm2 • A precast prestressed L-shaped face
after which the 5.25 X 5.25 X 0.6-m precast beam was designed to accommodate the
prestressed concrete slabs were installed wharfs services, as well as to provide sup-
and joined together and with piles via a port for dock fender system. The deck is
reinforced concrete plug. The chief advan- overlain by a layer of cast-in-situ concrete
tage of this system is its simplicity as well 10 cm thick.
Piled Waterfront Structures 721

The advantage of this type of construc- comprised of tubular steel piles surmounted
tion is its ability to accommodate very heavy by a concrete deck constructed from prefab-
cargo handling equipment, such as portal ricated components. The design lateralloads
cranes, bulk loaders, unloaders, and so are resisted by piles and an anchor slab,
forth, and large surcharge loads. Its disad- placed on the rear beam of the deck, and by
vantage is the use of the T -beams that are the so-called friction slab. According to
not very suitable for inspection and mainte- Eriksson and Klinbenberg this system is
nance. This can lead to corrosion of rein- often used in N orwegian designs.
forcement and, thus, deterioration of struc- AII deck components on this N orwegian
ture. The latter can be mitigated by the use site, such as the pile capping beams, front
of although less economica! however, more and rear beams, deck slabs, and the compo-
practica! from maintenance view point the nents included in the anchor system, were
flat concrete slabs. The advantage of deck prefabricated in a yard and transported on
structures illustrated in Figures 7-12 and a barge to the construction site. After their
7-13b is that they are almost totally com- erection, the deck components were joined
prised from prefabricated components; pre- together by cast-in-situ concrete. Prefabri-
fabrication of these structures reaches cation helped to reduce the construction
70%-80%. time of the wharf considerably.
Two unconventional structures have been The structure's anchor system that in-
recently built in Scandinavia (Eriksson and cludes the prefabricated anchor and friction
Klinbenberg, 1986). The structure depicted slabs is capable to resist lateralloads acting
in Figure 7-14 was built near Stavanger, toward the land (e.g., ship impact), as well
on the southwestern coast of Norway. It is as toward the water side (e.g., mooring

r+A
t 2000 J• 5300 11700 3500

r
3500 ~
1
1

Pcecast transverse beams

Section A-A

Figure 7-14. Wharf constructed from precast concrete components at


Stavanger, Norway. [From Eriksson and Klingenberg (1986).]
722 Piled Waterfront Structures

force). In the latter case, a reliable articu- ated with the underwater pile-driving pro-
lated connection between the anchor and cedure-the need for the diver assistance
friction slabs, and between the anchor slab for installation of prefabricated pillars and
and the deck structure is required. Further- holding them in place before concrete grout
more, the friction slab must be placed on a can gain the adequate strength to resist the
dense foundation base with low consolida- construction loads.
tion (settlement) rate. Additional interesting case histories are
The second structure (Figure 7-15) was found in recent publications by PIANC and
constructed at Lulea, Sweden for an iron-ore proceedings of specialty conferences such as
exporting wharf. In this case, the prefabri- ASCE PORTS, Ports and Harbors under
cated concrete components also have been Arctic Conditions (POAC) and others.
used extensively. The deck structure was
supported on precast tubular pillars (three
per bent), each installed on a group of four
concrete piles. Mter installation on piles, 7.3.6 Protection from Ship
the expanded footing of the pillar was filled Impact
with concrete. The structure lateral stabil-
ity was provided by two batter piles per Open pile structures are vulnerable to ship
bent. The land-based footing was also sup- impact and, therefore, must be designed in
ported on a concrete pile that was an inte- a way to keep the front row of piles out of
gral part of the bent. The deck structure the reach of the ship's bow. These struc-
consisted of prefabricated transverse and tures are usually protected by utilizing mis-
longitudinal beams and slabs overlaid by a cellaneous fender systems which are also
cast-in-situ concrete slab. The front row of used to dissipate the energy of a berthing
the pillars had a conical cap which enables ship and to reduce the ship's impact load.
them to reduce ice loads that act on the Besides the size of the vessel and the
structure. vessel's berthing maneuver, the fendering
The advantage of structure constructed of open pile structures with suspended decks
at Lulea is its ability to carry heavy deck is influenced to a great extent by the tidal
loads and to resist substantial horizontal range that usually is a major factor in fender
loads, such as ice, mooring loads, and ship system design.
impact. Its obvious disadvantage is the Typically, at sites with no substantial
complicated construction procedure associ- variation in water level, the conventional

Figure 7-15. Wharf for export of iron ore, constructed at Lulea,


Sweden. [From Eriksson and Klingenberg (1986).]
Piled Waterfront Structures 723

fender systems such as commercially avail- type of fender pile is not quite new; they
able rubber fender units of miscellaneous have been used in the late fifties at com-
designs are used to absorb berthing energy mercial wharves in Singapore and Kuwait.
and to protect both the ship and the struc- With the cost of creosoted wood piles in-
ture from damage. These fenders are nor- creasing, as well as concerns over health
mally attached to the deck face beam above aspects of the creosote becoming apparent,
the water level. Where the tidal range is the prestressed concrete fender piles are
substantial fendering of the dock may be being given renewed interest in the United
required at more than one level within the States. One of the first applications of pre-
tidal range. stressed piles for dock fendering in the
In older structures, large tidal variations United States occurred in 1985, when these
are usually accommodated by the fender piles were installed at a tug berth at a
systems comprised of creosoted wooden piles naval station in Norfolk, Virginia.
supported by rubber units at the deck level These piles were designed for two condi-
(Fig. 7-16). Where wood was scarce or use tions of berthing impact, namely frequent
of creosoted wooden piles raised environ- loading and extreme (one-time) loading. It
mental concerns, steel or prestressed con- was thought that under relatively frequent
crete piles were used. The pile fender sys- loadings, cracks would appear in the fender
tem, however, proved to be vulnerable to piles prior to reaching the ultimate moment
damage inflicted on it by berthing ships, capacity; however, these cracks closed when
particularly by those with bulbous bows. the impact load was removed. Alternatively,
Furthermore, underwater inspection of an ~xtreme loading was assumed as one
these piles and repair or replacement are severe enough to spall the cover over the
also difficult and very costly. reinforcement; if this level of pile damage
In recent years, research sponsored by occurs, then the pile must be either re-
the U.S. Navy has resulted in a new class of paired or replaced by the new one. These
prestressed concrete fender piles capable piles were installed by water jetting.
of high-energy absorption for use in a In the process of operation at N orfolk
wide range of pier fendering applications some of the piles have cracked and subse-
(Zinserling et al., 1987; Fotinos, 1986). This quently closed as anticipated (Zinserling et
al., 1987). To the best of this author's
knowledge, no significant problems with
these piles have been reported to date. Rub-
bing strips or a high-density polyethylene
cover on the piles at Norfolk were used to
prevent abrasion to both the ship and con-
crete surface. As suggested by Fotinos
(1986), the behavior of prestressed fender
piles can be improved by providing saw cuts
in the concrete cover over the reinforcement
located on the compressive side of the pile.
Ţhese cuts placed at 600- 900 mm (on cen-
ters) in the area of maximum bending mo-
ment will allow the high compressive stress
to be carried by the concrete located away
from the unconfined surface. The saw cuts
Figure 7-16. Tradition fender system comprised of
should be filled with an epoxy material to
creosoted wooden piles supported by rubber units at maintain proper cover over the reinforce-
deck level. ment.
724 Piled Waterfront Structures

Again, it must be pointed out that pre-


stressed concrete piles could hardly resist ••
~ :.
o

the impact load produced by a large vessel. ! .o '


-----.:.o ..
o

These piles can normally be used to absorb


ship impact energy in the neighborhood of . ~. : :.
up to 4 tm. In modern marine engineering
practice, piled fender systems are more of-
ten replaced by the efficient rubber fenders.
Where the open piled construction with a
suspended deck is used in areas with large
tidal ranges, the rubber fenders are usually
mounted on a concrete face panel sus-
pended from the deck structure. Naturally,
the height of these panels depends on the
tidal variation. In recent projects, to mini-
mize the expensive over-water works, these
panels were prefabricated and subsequently PIPE 8JIACt:

stressed by posttensioning techniques to the


underside of the deck structure. A typical
example of such type construction is illus-
trated in Figure 7-2.
Where the tidal range is substantial, rel-
atively thin and long fender panels can be
supported by braces similar to that indi-
cated in Figure 7-17. Here the lateral shear
load at the panel's upper end is resisted by
the concrete shear key provided at the in-
terface, and at its lower end, the panel is Figure 7-17. Precast concrete fender supporting
panel posttensioned to concrete deck or pile capping
propped by a steel pipe brace. Longitudinal structure. [From Shrivastava and Youdale (1986).]
shear produced by ship rubbing along
the face of the dock during the berthing
maneuver is resisted, when required, by the
secondary bracing system installed parallel hard soils with inclusions of large boulders.
to the dock's face. This system was used for To develop the required uplift capacity, piles
fendering the breasting dolphins con- can be anchored in competent soil or bedrock
structed for the Ridley lsland Coal Termi- by one of the following techniques. One of
nal (Shrivastava and Youdale, 1986). these methods is to drill a socket into the
foundation that is of slightly smaller diame-
ter than the inside diameter of the pile.
Then the insert is installed in the socket
7.3. 7 Pile Anchoring in hole and grouted there and in the pile.
Foundation Soil and the Another method is to anchor the pile to
oeck Structure the foundation by using conventional
ground-rock anchor techniques. The combi-
Sometimes piles cannot be driven deep nation of both methods can also be used.
enough into the foundation soil to develop Where piles are required to be anchored,
an adequate uplift capacity. This may occur then pipe piles are normally used. The work
where bedrock is close to the surface or in is typically carried out in the following
Piled Waterfront Structures 725

sequence: Because of the nature of the socket, the


pile load is distributed through the sides of
1. Piles are driven through the soft overbur- the socket to a much larger rock area than
den to the competent soil-rock layer. As would be if the pile merely rested on the
the pile is advanced, the overburden ma- rock surface. Very high axial loads could
terial is cleaned out from the pile's inte- therefore be applied on these piles. In prac-
rior. In deep water conditions where a tice, this load would be controlled by pile
long freestanding pile is required andjor buckling and its compressive strengths.
where the overburden layer is too thin to
In a second method, the pile is driven
support pile, then steel templates or jack-
ets are used to assist both pile-driving
into the overburden with a preinstalled con-
and anchoring. Here the template or jack- crete plug or is completely filled with con-
ets are positioned and secured to the bot- crete. Next, a deep hole is drilled through
tom. The piles are lowered down through the pile and beyond the pile tip through the
template sleeves or jacket columns and tubing preinstalled at the pile center. This
driven to required penetration. is followed by installation of a steel tendon
2. With the pile in place, the socket is drilled and its grouting to the soil-rock interface.
and cleaned from inside the pile. The tendon is then stressed, anchored, and
3. Steel pins made of H-piles or other sec- grouted to ensure its complete protection
tions, a rebar cage, or a combination of from corrosion. The disadvantage of this
both are installed within the socket and method is that it requires extreme care in
the lower end of the pile; the system is controlling the quality of tendon installa-
grouted up, thus bonding the insert to the tion and grouting. Lack of reliable quality
walls of the socket hole and to the pile. control may result in a complete failure of
the anchor system, as has occurred during
In both methods, the grout is delivered to the construction of breasting dolphins for
the socket by utilizing either the tremie the Ro-Ro terminal in Boston Harbor
method, or by pumping concrete through a (Blowers et al., 1992).
grout pipe. The rebar cage should be ade- A typical example of a pile socket system
quately sized to provide ample room to per- with a steel insert is illustrated in Figure
mit the free flow of concrete or grout be- 7-18. As it is also seen in this figure the
tween the bars. To create more space for pile connection at the deckjpile cap struc-
concrete, the vertical bars can be bundled ture is provided by the rebar cage extended
in groups of two and more. A spiral rein- from the concrete plug inside the pile. This
forcement system should also be well can also be accomplished by welding steel
spaced; if necessary, it can be roade of bars to the pile as seen in Figure 7-3.
heavier-gauge steel. The space and clear-
ings between rebars should be at least four
times the size of coarse aggregate. Alterna-
tively, the insert can be roade of a precast 7_4 RELIEVINO PLATFORMS
reinforced concrete plug which contains all
the required reinforcement. The walls of As noted earlier, these structures are very
this plug should be rough enough to ensure effective where the foundation soils are
a good bond with the grout. Mter installa- weak and the tide range is substantial. Also,
tion of the plug into the socket, the annular as previously stated, the relieving platforms
space between the plug and socket, and the are usually used as a viable alternative to
plug and pile is grouted via preinstalled sheet-pile anchored bulkheads.
grout pipes; the grouting is carried out with Practically, there is no limit on water
no interruptions. depth into which the relieving platforms
726 Piled Waterfront Structures

ther advantages of these structures is that


the wall piling is well protected from im-
pacts by ships and ice and effects of envi-
ronmental loads such as waves and cur-
rents. Because in most practica! cases of
relieving platform construction, the piles are
always surrounded by soil along their full
lengths, the buckling stability of these piles
and, hence, their structural strength is
enhanced.
Relieving platforms can be comprised of
ali basic structural materiala such as wood,
steel, and concrete, and/ or combinations of
these. Severa! typical examples of relieving
platforms constructed elsewhere in the
world are illustrated in Figures 7-19
through 7-25.
An all-wood structure built in Russia is
depicted in Figure 7-20a. This structure
was built in the early 1960s in the northern
Russian Port-of-Archangelsk, White Sea. Its
piled foundation is composed ofwooden piles
30 cm in diameter and wooden sheet piles
22 cm thick. Pile bents are spaced at 1.0-m
centers. The deck structure and superstruc-
ture are constructed in the dry at a low sea
level. The deck structure is composed of
longitudinal and transverse members that
embrace the piles; the 10-cm-thick timbers
are placed on these members to form a flat
deck. The retaining superstructure was de-
Figure 7-18. Pile socket and pile to deck or pile signed and constructed in a form of a rock-
capping beam connection. [From Shrivastave and filled gravity-type timber crib wall.
Youdale (1986).]
In a freshwater environment or in cold
regions, wood structures can be quite
can be constructed economically. For durable. In fact, if structural wood compo-
example, quay walls 30.5 and 31.0 m high nenta are properly pressure treated, the
to accommodate 400,000 DWT vessels structures can serve as long as 20-30 and
(draught 22.86 m) have been successfully more years. They can represent a competi-
built at Port-of-Rotterdam, Netherlands in tive solution in dock construction where
the 1980s (Parent, 1986). The cross section wood is a cheap and locally available
of a somewhat smaller structure is depicted material.
in Figure 7-19. Normally, to prevent wood decay, the
The relieving platform structures can wooden components should be submerged
carry very heavy live loads (e.g., high- below the mean minimum water level and
lifting-capacity portal and mobile cranes, extended above the water level by ineans of
iron ore, packaged steel products, and oth- steel or concrete componenta to allow for
ers) without unacceptable deflections. Fur- construction in the dry of a concrete plat-
Piled Waterfront Structures 727

(a) (b)

Figure 7-19. Deep water quay wall constructed at Port-of-Rotterdam: (a) Typical cross section;
(b) sheet-piling details; (c) cast-iron saddle. 1-composed-type steel sheeting; 2-prestressed
concrete bearing piles; 3-M.V. (Miiller Verfahrem) tension piles (steel grouted piles); 4-con-
crete superstructure; 5-granular drainage; 6-pavement; 7-fender system; 8-conventional
sheet piles; 9-steel pipe piles; 10-reinforcing plates; 11-roofmate; 12-cast-iron saddle;
13-dowel.

form (superstructure). The relevant exam- crete, the head of the tension piles is typi-
ple is shown in Figure 7-20b. Here, the cally made in the form of a truncated cone
piled foundation is composed of wooden with its larger base at the top. Additionally,
sheeting, wooden bearing piles, and con- a spiral reinforcement made from 10-12-
crete batter piles. The sheet piles are sup- mm-diameter bars is normally placed
ported against prefabricated concrete face around the cone. Obviously, if the structure
panels incorporated via extended steel dow- is constructed "in dry" conditions, the
els into the cast-in-situ concrete platform, wooden piles are embedded directly into the
and the wooden piles are extended up above concrete platform.
sea level by means of concrete cylinders In the case depicted in Figure 7-20b, the
joined with the platform structure via steel lateral loads are resisted by concrete batter
dowels. After installation on piles, the con- piles. The timber sheeting is designed as a
crete cylinders are :filled with concrete. Nor- conventional sheet-pile bulkhead to retain
mally, to ensure reliable wooden pile- the back:fill soil. In permanent structures,
concrete structure interaction, the bearing wooden sheeting is normally used only
piles should be embedded into the concrete where there is no marine organisms able to
to the depth equal to at least the pile head damage the wood; otherwise, steel or con-
diameter; the piles in tension should be crete sheet-piling is employed. The relevant
embedded to a minimum of0.7-0.9 m. Fur- example is illustrated in Figure 7-20c. In
thermore, for a better connection to con- this particular case, the structure is corn-
728 Piled Waterfront Structures

(a) (b)

(c)

Figure 7-20. Examples ofrelieving platforms: (a) constructed totally from


wooden components; (b) concrete relieving platform with masonry superstructure
constructed on vertical and bearing wooden piles, and batter concrete bearing and
tension piles with timber sheet piling; (c) concrete relieving platform with regular
concrete sheet-piling and all-wooden regular piles. 1-Creosoted wooden piles;
2-timber criblike superstructure filled with rocks; 3-timber relieving plat-
form; 4-timber sheet piling; 5-prefabricated concrete extension for wooden
piles; 6-precast concrete panel to support timber sheeting; 7-masonry
superstructure; 8-cast-in-situ concrete platform; 9-concrete pile; !O-concrete
sheet-piling; 11-granular ballast; 12-granular backfill; 13-original surface;
14-granular drainage.
Piled Waterfront Structures 729

prised of all-wooden piles embedded into between the pipe piles as indicated in Fig-
the concrete platform at the mean water ure 7-19b. All sheet piles were installed by
level, and concrete sheet-piling. vibration down to a layer of dense sand; to
Described in the following examples are achieve adequate bearing capacity the pipe
the other combinations of piles and sheet piles were driven further into the dense
piles: prestressed concrete piles in combina- sand by an impact hammer. For better
tion with steel sheet-piling; prestressed alignment, a steel guiding frame was used
concrete bearing piles and steel tension piles during installation of the piles. In some
in combination with steel sheet-piling; pre- instances, a water jet was used to assist in
stressed concrete bearing piles and steel the installation of the pipe piles; the water
grouted piles in combination with steel jet was not used during driving into the
sheet-piling. dense sand layer. Installation of pipe piles
As noted earlier, several exceptionally was followed by driving of the conventional
high dock walls (up to 31.0 m) have been steel sheet piles.
recently constructed at Port-of-Rotterdam. During sheet-pile-driving, several inter-
One of those structures constructed for the locks were ruptured. Damages to interlocks
European Bulk Goods Transhipment Com- could not be easily detected from vibration
pany in Maasulakte, Netherlands is de- records during pile installation; subse-
picted in Figure 7-19. This low-level reliev- quently, they have been found by divers
ing structure was designed to provide the during the wall inspection carried out after
berthing facility for ships up to 350,000 dredging the harbor basin in front of the
DWT with a design draught of about 22.0 wall was completed. The underwater repair
m. It was constructed "in dry'' on land and of damaged interlocks proved to be very
followed by soil excavationjdredging in costly.
front of the berth. As mentioned earlier, in order to reduce
The pile foundation of this structure is the bending moment in the sheeting, the
composed of a raked steel sheet-pile wall, load from the superstructure was applied to
prestressed concrete batter bearing piles, the sheet-pile wall eccentrically via special
and steel grouted tension piles. Piles were cast-iron saddles. Furthermore, the effects
surmounted by a cast-in-situ concrete su- of soil arching between the pipe piles were
perstructure. The steel sheet-piling of com- taken into account in the design of the con-
posite construction was designed to carry ventional sheet piles installed between pipe
vertical loads produced by heavy bulk han- piles. In further developments at Port-of-
dling equipment and by the relevant weight Rotterdam, other combinations of the con-
of the superstructure (backfill soil included) ventional steel sheet piles with fabricated
and to resist lateralloads produced by hori- steel sheet piles were used (De Gijt et al.,
zontal soil pressure and by mooring forces. 1993).
To reduce bending moments in the sheet- Prestressed concrete piles, 450 and 500
pile wall, the load from the superstructure mm in diameter, used for constructing the
was transferred eccentrically to the sheet- quay wall depicted in Figure 7-19 varied in
ing via cast-iron saddles installed on the length from 22 to 30 m. These piles were
sheet-pile heads (Fig. 7-19c). driven by diesel hammer through miscella-
Depending on the wall's height at a par- neous soils into the stratum of dense sand.
ticular location, the sheet-pile wall was Pile installation through soils that overlaid
composed of open-end steel pipe piles 904 the lower dense bearing sand was assisted
mm-1420 mm in diameter, 30-35 m long, by a water jet; the water jet was not used
and a set of three steel sheet piles, Larssen while piles were driven into the dense sand
Ils or Ills, approximately 24 m long driven to the desired penetration of 4-7 m.
730 Piled Waterfront Structures

Use of the water jet helped reduce pile


skin friction so that pile-driving into the
dense sand proceeded without damaging the
pile heads by heavy blows. Furthermore,
water jetting loosened soil around the pile,
which resulted in a reduction of negative
skin friction with a subsequent increase in
pile bearing capacity.
Special features of the wall in question is
the tension element called the M.V. pile
(M.V. stands for Miiller Verfahrem) in-
stalled at the top of sheet piling and embed-
ded into the concrete base of the super-
structure. The length of these piles varied
from 30 to 36 m. The piles were designed to
carry tensile loads equal to 3000 kN, which
was transferred to the concrete superstruc-
ture by means of 80 dowels, each approxi-
mately 16 mm in diameter. The M. V. pile is Figure 7-21. Eemhaven Quay, Rotterdam, Nether-
essentially a steel grouted pile, or a grouted lands: 1-steel H-piles; 2-steel sheet piles; 3-pre-
steel anchor. It is designed to transfer the stressed concrete piles; 4-precast concrete beam;
horizontal load that acts on a structure to 5-cast-in-situ concrete platform; 6-granular ballast;
7-pavement; 8-riprap placed on natural slope;
the competent layer of subsoil. The anchor 9-granular fill.
has an enlarged tip and is equipped with
two grout pipes. The steel pile's cross sec-
tion depends on its required drivability as for container ships. The foundation soils at
well as on the design tensile load and the the construction site consist of very w.eak
load due to the soil that may hang on the clay and peat layers, underlain by firm,
pile. During pile-driving, a space is formed dense sand. The top of the wall was ele-
behind the enlarged tip that is immediately vated 3.35 m above mean sea level and the
filled with pumped-in grout. It is essential underside of the relieving platform was
that pile-driving and the grouting process located approximately 65 cm above the
continue without interruption and that the minimum low sea level.
grout pressure is kept as necessary to en- The wall's foundation was comprised of
sure that grout level always stays at the steel sheet piles and prestressed concrete
ground surface level. Similar to conven- piles, 390 X 457 mm, driven into the firm
tional ground (rock) anchors, M.V. piles used sand stratum. The pile-driving rig was
for wall construction were load tested. The mounted on a special platform. The water-
drainage system, as indicated in Figure side of this platform along with a pile-driver
7-19, was provided to equalize the hydro- track sat on a floating structure, and the
static pressure around the wall. land-based track was placed on sleepers
The high-level relieving platform con- embedded in the natural slope. The sheet-
structed in the mid-1960s at Port-of- ing was comprised of conventional steel H-
Rotterdam is shown in Figure 7-21. piles and conventional steel sheet piles
It was constructed for expansion of the driven in sets of two between H-piles; to
Eemhaven basin, which was required to accommodate the conventional sheet piles,
cope with the increased demand for driving of the H-piles had tobe carried out
berthage for general cargo freighters and with utmost accuracy. After pile installa-
Piled Waterfront Structures 731

tion, the sheet-pile wall was cut accurately structures constructed later at Port-of-
to design elevation. Following installation Rotterdam (Parent, 1986; De Gijt et al.,
of the prestressed concrete piles, the riprap 1993). One of the most interesting develop-
was placed on the face of the natural slope ments at Port-of-Rotterdam is construction
behind the sheeting, and the concrete level- of Swarttouw Quay (Figure 7-22). This
ing pad was cast in place at the top of the structure, which carries heavy portal cranes
riprap to provide the temporary support for and superimposed loadings, was con-
the precast concrete beams to be installed structed on poor foundation soils. Here, the
on this pad and the sheet-pile wall. These soils consisted of layers of soft clay and fine
beams were installed on top of each H-pile sand, underlain by firm clay. Similar to the
and subsequently connected to these piles previously discussed structure constructed
by means of special cast-iron saddles. Every at Maasvbakte (Fig. 7-19), this structure
fourth precast beam was extended beyond was also constructed on land and under dry
the face of sheet-pile wall to provide sup- conditions. The pile foundation of this
port for vertical hardwood timber fenders. structure was composed of raked sheet-
As depicted in Figure 7-21, the rest of piling, prestressed concrete bearing piles,
superstructure was cast-in-situ. It should and steel tension H-piles. The composite
be pointed out that originally this structure sheet-piling included fabricated steel
and its fender system were designed to ac- sections made up from two H-piles and con-
commodate fairly large ships approaching ventional steel sheet piles driven between
the dock with the assistance of tugs. In H-piles in sets ofthree. One basic feature of
practice, however, some ships approached sheet-piling design was the provision of ad-
the dock without tug assistance and some- ditional fiange plates in zones of high bend-
times with a speed that exceeded the design ing moments.
approach velocity. Furthermore, the dock Support for the waterside crane track
was used by barges that loaded the fender was provided entirely by steel sheet-piling
system in an unfavorable way. This re- which was raked toward the land so that its
sulted in substantial damage to the prefab- top lies under the crane raii. The predomi-
ricated beams used for installation of the nantly cast-in-situ concrete superstructure
fenders and to the longitudinal face beam. had a soil-retaining wall erected from pre-
The damage to wharf components was in a cast concrete panels. The panels are sup-
form of deep, long cracks spreading almost ported at counterfort concrete walls that
in all directions. One of the reasons for were designed to provide stiffness to the
these cracks was an inadequate amount of superstructure as well as to support moor-
reinforcing steel. ing bollards and the fender system.
In 1988 this structure underwent exten- The latest development at Port-of-
sive rehabilitation work during which the Rotterdam was constructed of a deep
cracks were injected with grout, and the water quay wall (Delta Container Terminal
beams were strengthened by use of postten- at Europahaven) with no tension piles at
sioning techniques (Parent, 1986). Finally, ali; the lateralload acting on this structure
it should be pointed out that the underdeck was resisted by the anchor system com-
(underwater) fill was completed using riprap prised of steel sheet-pile walls and pre-
placed on the natural slope. This resulted in stressed high-tensile-strength steel anchor
a significant reduction in lateral pressure cables (Fig. 7-23).
against the sheeting. The pile foundation of this structure con-
The experience gained in rehabilitating sists of fabricated steel box piles driven at 2
the dock shown in Figure 7-21 helped m cjc (center to center) and conventional
to refine designs of other waterfront steel sheet piles driven between the box
732 Piled W aterfront Structures

top elements of the superstructure were cast


in 13.5-m pours.
The anchor sheet-pile wall was also used
as a foundation for the land-based crane
rail. The anchor wall and the Delta super-
structure were connected by high-tensile-
strength posttensioned steel cables.
Experience gained at Port-of-Rotterdam
and elsewhere in the world indicate that
composite sheet-pile walls, comprised of
conventional high-tensile-strength steel
sheet piles alternating with large-diameter
high-strength pipe piles or made up sec-
tions, can be a very economica! solution for
construction of relieving platforms. In the
above combinations, steel sheet piles are
normally used in groups of shorter than
ring pile lengths. The advantage of these
walls is that they can withstand a very
heavy vertical load in combination with
lateral loads from the backfill. The attrac-
tiveness of this system increases with wall
Figure 7-22. Swarttouw Quay, Rotterdam, Nether- inclination.
lands. Typical cross section: 1-double steel H-piles; 2 Sometimes quay walls need to be con-
-steel sheet piles; 3-prestressed concrete piles; structed in populated areas where local con-
4-steel H-piles; 5-concrete superstructure; 6-pre-
cast concrete panels; 7-granular backfill; 8-pave-
ditions such as neighboring buildings or the
ment. presence of sensitive soil vibration equip-
ment may necessitate strict avoidance of
noise, or shock and vibration of the ground
which is normally associated with pile-
piles. As in the earlier discussed examples,
driving.
the sheet-piling is raked toward the land
Under the aforementioned site condi-
and is loaded eccentrically to reduce the
tions, the "quiet" slurry wall techniques can
design bending moment in the sheeting. be successfully utilized for wall construc-
Penetration of raked box piles, and also the tion. The example presented in Figure 7-24
raked prestressed concrete bearing piles, illustrates this type of construction. This
varied, depending on local soil conditions. deep water structure was built for
The bearing piles were driven to a mini- the Burchardkai Container Terminal,
mum 4.0 m penetration into the dense sand Germany, located on the bank of the north-
layer (the required cone resistance was ern Elbe (Hofmann, 1989). There, to pre-
specified to be not less than 30 Njmm 2 ). vent potential settlement of the closely
The Delta-shaped concrete superstructure located Elbtunnel, shocks and ground
was designed to support the waterside track vibration had to be avoided. Also, because
of the heavy portal cranes and to accommo- of the close proximity of residential housing,
date one railway track. The expansionjoints the substantial noise associated with
along the berth were at 40.5 m spacing and construction activities was unacceptable.
the concrete was placed by concrete pumps; Furthermore, the suspected (and confirmed
the inclined walls, fender (face) wall, and later during construction) presence of large
Piled Waterfront Structures 733

2000 j_
r-------~~------·

Figure 7-23. Delta Container Terminal at Europahaven, Rotterdam, Netherlands: 1-fabri-


cated steel box pile; 2-conventional steel sheet pile; 3-prestressed concrete piles; 4-hollow
concrete superstructure; 5-granular drainage; 6-steel anchor sheet-pile wall; 7-concrete
wall; 8-concrete beam to support crane track; 9-prestressed high-tensile-strength anchor
cables; 10-pavement.

boulders and other obstacles in the subsoil from the face of a new wall and space
also roade sheet pile and conventional pile- between the natural slope, and the design
driving impractical. grade was hydraulically filled with granu-
The bidding competition resulted in many lar material.
• Next, a very short steel sheet-pile guide
interesting proposals. However, the pro-
wall was installed and the reinforced con-
posal that included construction of a slurry crete slurry wall was completed. This wall
wall with a relieving platform resting on was extended down to marl bedrock. Then
both cast-in-situ piles and the slurry wall conventional bored concrete piles, both
was selected for construction. It is worth bearing and tensile types, were con-
mentioning that the generally economica! structed. Finally, the structure was com-
solution such as a slurry wall anchored by pleted with a cast-in-situ concrete reliev-
raked posttensioned ground anchors at this ing platform, after which the previously
particular location happened to be very ex- installed semivertical wall was dismantled
and soil in front of the wall was dredged to
pensive and was ruled out.
the design elevation.
The wall in question was constructed on
built-up land in the following sequence: As mentioned earlier, each project is site-
specific and therefore depends on many con-
• First, the semivertical cantilever sheet-pile straints that affect the wall design in one
wall was constructed offshore about 12 m way or another.
734 Piled W aterfront Structures

1r-- 1&00-11
r-21,00~

Figure 7-25. A deep water quay wall constructed at


Figure 7-24. Port-of-Hamburg, Germany. Burchard- Port-of-Hamburg; typical cross section: 1-sheet-pile
kai Container Terminal; Typical cross section. wall comprised of steel H-piles (king piles) and conven-
1-Guide wall constructed from steel sheet piles; tional steel sheet piles driven between king piles;
2-natural bank ofthe Northern Elbe River; 3-built- 2-drilled concrete piles; 3-grouted steel tension pile;
up grade (removed after the wall is built); 4-slurry 4-concrete platform; 5-ballast; 6-pavement.
diaphragm wall; 5-concrete drilled piles; 6-concrete
relieving platform; 7 -granular ballast; 8-pavement;
9-fendering; 10-granular hydraulically placed back-
fill. .
strate the variety of pile arrangements and
the potentials of the relieving structures for
use in port development.
The example presented in Figure 7-25
illustrates the variety of piles which, de-
pending on site conditions, can be used eco-
nomically for waterfront construction. This
7.5 STRUCTURAL ELEMENTS
deep water wall was completed in 1980 in
Port-of-Hamburg, Germany (Hofinann, Piled waterfront structures (e.g., open piled
1989). Its pile foundation included sheet- construction with suspended deck, or reliev-
piling comprised of bearing (king) piles ing platforms) are typically comprised of
(heavy H-piles) and filler elements that in- the following basic structural elements: pile
clude conventional steel sheet piles driven foundation, superstructure, and protected
between the king piles. The bearing load from erosion underdeck slope. In some in-
from heavy cargo handling equipment and stances, as discussed in Section 7.1 an an-
the surcharge load was transferred to chor system can be added to the structure
high-capacity cast-in-situ (drilled) bearing to support all or some lateral forces that are
piles, and the lateral load was resisted by acting on structure.
grouted steel piles similar to those depicted The reliable performance of all structural
in Figure 7-19. elements is paramount to an effective dock
The examples discussed in this section operation and its longevity. Damage of one
certainly do not represent all types of re- or several of these elements or premature
lieving platforms available for construction deterioration may substantially hamper
of waterfront walls; they illustrate some. dock operation, resulting in limited permis-
characteristic case histories and demon- sible loads allowed to act on the structure
Piled Waterfront Structures 735

and possibly a drastic reduction in the current, earthquake) and dock operations
structure's usefullife. (e.g., ship impact and mooring loadings).
In the following sections, details on the As discussed in Chapter 3, these loads
structural elements that are used in con- are not necessarily cumulative. For exam-
struction of piled marine structures are ple, wind, wave, and current forces can oc-
discussed. cur simultaneously and in the same direc-
tion, whereas the forces due to ship impact
and mooring forces can act in opposite di-
7.5.1 Pile Foundation rections. Furthermore, wave loads or ice
impact cannot coincide with ship impact,
7.5.1.1 Pile Arrangements and so on.
Piles of any type of construction and ma-
As discussed in previous sections, a piled
terial used in the harsh marine environ-
foundation can be comprised of vertical piles
ment are vulnerable to deterioration. As
or a combination of both vertical and batter
stated earlier, the premature deterioration
piles. Additionally, as in the case of a reliev-
of piles may drastically reduce the useful
ing platform or sometimes in the case of an
life of the structure or, at least, may result
open pile construction with a short deck
in a limiting of the permissible loads that
structure, sheet-piling is used as a part of
may act on the structure. Deterioration of
the pile foundation. Basic structural ar-
piles in a marine environment may occur as
rangements of pile foundations and some
a result of the presence of aggressive sub-
specific arrangements are discussed in Sec-
stances in the foundation soil (rock), sea
tions 7.2-7.4. Additionally, details on sheet
andjor river water, the presence of marine
piles (design, structure, material, and in-
borers, the exposure to aggressive atmo-
stallation methods) are discussed in Chap-
spheric conditions, fouling, abrasion in-
ter 6. Some information on piles and sheet
flicted by sediments movement, ice, and
piles is also given in Chapter 2.
other factors. For details on the aggressive
In this section, only basic information on
nature of the marine environment, as ap-
conventional piles and conventional meth-
plied to marine structures, and necessary
ods of pile installation are presented. For a
precautions against the effects of these ag-
detailed discussion of this subject matter,
gressive substances on marine structures,
the reader is referred to standard texts on
the reader is referred to Tsinker (1995).
Foundation Engineering, standard hand-
To prevent pile deterioration, a thorough
books on the same subject (e.g., Foundation
investigation is usually carried out to detect
Engineering Handbook, 2nd edition, edited
the presence of harmful substances in the
by H-Y. Fang (1991), or special publications
foundation soils and in the sea (river)
such as Chellis (1961), Prakash and Sharma
water. The obtained information should be
(1990), or a more recent text by Tomlinson
carefully evaluated to determine the degree
(1994).
of potential aggressive action against the
structure and its components (e.g., as piles).
7.5.1.2 Piles Foundation piles can be vertical or bat-
Piles used for construction of marine ter. They can also be prefabricated (precast)
structures are normally designed to carry or cast-in-situ (bored or drilled). Cross-
loads from the dead weight of the structure, section-wise piles can be square, rectangu-
cargo handling and hauling equipment, lar, circular, octagonal, hexagonal, and even
miscellaneous surcharge loads, as well as triangular; they can also be H-shaped, solid
all kinds of lateralloads attributed to envi- or hollow. Pile shafts can be prismatic,
ronmental phenomena (e.g., wave, wind, cylindrical, or conical; the latter can be
736 Piled Waterfront Structures

smooth tapered, or step tapered. The pile from used railway rails, or other materials
leading end (toe, tip, or point) can be blunt, are used.
open, or pointed; where hard driving is ex- Steel piles have the advantage of being
pected, the pile toe can be furnished with a easy to handle on the construction site. They
special shoe usually made of a hardened do not have to be an exact predetermined
steel. With respect to their ability to carry length before driving, as an excess length
loads, piles can be defined as friction, end can be quickly cut off by a welder's torch or
bearing, or of composite action. The former an extra length added by welding or by
developed its capacity through the friction means of special splices. Furthermore, the
that is developing between the pile surface previously cut pile length may be used for
and the soil through which it is driven. End extension of the other pile. The obvious dis-
bearing piles are those resting on a firm advantage of steel piles is their susceptibil-
bedrock stratum or any other competent ity to corrosion. The basic methods for pro-
soil. These piles derive their load-carrying tection of steel elements from corrosion used
capacity through both friction and the toe in the marine environment are the use of
resistance. protective coatings that range from simple
Materialwise, piles may be made from coal tar to a very durable polyethylene film,
wood, concrete, or steel. Combinations of installation of protective concrete jackets in
the three materials are also employed. tidal and splash zones, installation of sacri-
ficial anodes in submerged zone, and pro-
7.5.1.1.1 Wooden pi/es viding some additional steel as corrosion
allowance. The phenomena of corrosion of
These piles are discussed in detail in
steel in marine environment and its protec-
Chapter 2. Piles made from wood represent
tion are discussed in detail in Tsinker
the oldest type of pile in existence. When
(1995).
fully submerged, they will not decay and
are likely to last for centuries. Wooden piles,
however, would deteriorate rapidly if sub- 7.5.1.1.3 H-Piles
jected to alternate drying and wetting. Im-
Due to their small soil displacement,
pregnation of wooden piles by creosote
these piles are suitable where piles must
andjor other chemicals prolongs the useful
penetrate hard material (e.g., dense sand,
life of these piles but cannot entirely protect
gravei, or even sloped riprap) to the design
them from marine borer attack, especially
elevation below the dredge line. Penetration
in warm waters where borers are particu-
of 3-4 m in a weathered rock or up to 5-6
larly active.
m in cemented sand and gravei are quite
Wooden piles are usually not permitted
common. Because of their small displace-
to carry loads in excess of 250-300 kN per
ment of foundation soil, H-piles are very
pile, which prevents their use for construc-
useful for driving where close spacing is
tion of deep water, heavily loaded modern
required. They are also indispensable for
waterfront structures.
enhancement of existing structures, where
they can be driven through small spaces in
7.5.1. 1.2 Steel piles
an existing structure, or where piles are
Steel piles have been in regular use since required close to an existing structure.
the 1890s (Schwartz, 1958). They are fabri- Where large lateral or tensile loads are
cated in a variety of cross sections. Those exerted or where substantial seismic loads
used more often are H-piles and pipe piles. may occur, the large tensile and bending
Sometimes box piles fabricated from con- strength capacity of H-piles is also of great
ventional sheet-pile sections, piles made value.
Piled Waterfront Structures 737

Furthermore, to reduce the weight of H- transfer. H-piles can also be projected suf-
piles, they can be easily spliced on-site so as ficiently far into the superstructure to
to have heavier sections where required by transfer the load and moment to the pile.
the load distribution or where more al- The embedment of the pile has the further
lowance for corrosion is needed (e.g., in the advantage of enabling it to take uplift loads
splash or tidal zone). and ensuring the pile's fixity within the
H-piles do not require special care in concrete superstructure. The latter is par-
handling unless they are very long. In this ticularly important where long unsupported
case, they are supported with the web ori- piles act as columns in deep water. The
ented vertically andjor with attachments bond between concrete and pile should be
made at several points. Handling holes may kept as low as practical. Normally, the bond
be furnished at the pile's centerline, if re- stress should not exceed 0.02 f~. Where pile
quired. H-piles can be extended easily and embedment into the concrete is limited, then
quickly to any length to meet the required a better fixity may be provided with short
driving resistance. They can be spliced by reinforcing bars installed in holes burned in
bolting, welding, or commercially available the pile fl.anges andjor web.
splicers. It is essential that the splice be
compatible with the strength of the pile. 7.5.1.1.4 Steel pipe pi/es
Typical pile splicers and splicing procedures
are depicted in Figure 7-26. To avoid pile The advantage of these piles is that for
damage when it is driven through riprap, any given weight of steel, they represent
very dense gravei, or soils in which occa- the stiffest shape possible. Essentially, the
sional boulders are encountered, it must be pipe pile's strength and stiffness are equal
protected by using a commercially available · in all directions. A further advantage of
cast steel shoe of miscellaneous design or by pipe piles is that in a marine environment,
using reinforcing steel plates welded to the its circular shape minimizes drag forces
pile toe. from waves and current. Additionally, they
Caution, however, must be exercised can be readily inspected for damage andjor
when H-piles are driven into hard bedrock. deviation from plumb from inside.
Excavations have occasionally revealed that Pipe piles can be driven either as closed
H-piles can be overdriven and split when ended or open ended. Closed-ended piles
the pile hits a sloped bedrock or the edge of can be furnished with either a standard
a boulder (Tschebotarioff, 1973). The bear- conical point or with a plate welded to the
ing capacity of H-piles that do not reach a pile tip (Fig. 7-27a-7-27c). Closed-ended
hard bearing stratum can be increased by piles can be driven through rock riprap
welding steel plates on both sides of the pile when furnished with conical points. The
web and flanges. conical 60° point is welded to the tip of the
The connection between the concrete su- pile and provides dependable protection for
perstructure and the bearing H-piles is of the pipe pile driven into difficult foundation
considerable importance. It must be such to soils. These soils or bedrock are never
ensure that the pile bearing against the smooth or encounter the pile at right an-
concrete does not exceed 0.25{~, where f~ = gles. The conical point centers the reaction
28 day concrete compressive strength. This force when the hard stratum is reached or
is also true for other pile types. when natural or man-made obstructions
Steel H-piles may be cut off 100 mm within foundation soil are encountered.
(min.) above the underside of the concrete When pipe piles are driven into soft,
and furnished with a suffi.cient number of obstacle-free, materials, the end closure can
dowels to ensure an adequate axial load be fabricated with a simple flat plate welded
738 Piled Waterfront Structures

(b )

Figure 7-26. Splicing of steel H-piles: (a) details of assembly: step-by-step


procedure; (b) splicer HP 30 000. (From Associated Pile and Fitting Corporation,
Bulletin GP 8512.)

to the pile's end. The latter, however, should pulls the pile down. Details on pile installa-
not be used where the pile is expected tobe tion methods are discussed later in this
driven through diffi.cult foundation soils; chapter.
here, the plate closure may concentrate the The required wall thickness of the pipe
impact forces onto a small arc of the pipe pile is dependent on the pile's diameter, the
(Fig. 7- 27b). This may lead to complete pile driving energy delivered to the pile's head,
failure. difficulties encountered in driving, the pile's
Closed-ended pipe piles displace an length, the imposed design loads, the ex-
amount of soil equal to their embedded vol- pected effects of the ambient environment,
ume; therefore, they should not be used and protection from corrosion. Sometimes,
under conditions when excessive displace- for better buckling stability and strength,
ment of the soil may result in damaging the steel pipe piles are filled with concrete.
loads on closely located structures (e.g., ex- It should be pointed out that in marine
isting sheet piling, soil-retaining walls, and applications, large externa} hydrostatic
similar structures). In these cases, the prob- pressures can be exerted on empty closed-
lem can be solved by either driving low- ended pipe piles. These forces must be taken
displacement H-piles or using open-ended into account. In most practica} cases, how-
pipe piles and periodically cleaning out the ever, this pressure does not present a major
soil from within the pile. problem for a steel pipe pile.
Closed-ended piles can be head or toe In soft soils with no obstruction to pile
driven. In the former case, the pile is forced penetration, open-ended pipe piles are com-
into the ground by an impact force applied monly used. A soil plug that is developing
to the pile's head, and in the latter case, the inside the pile during penetration effec-
pile is driven by means of a hammer im- tively turns the pile into an end-bearing
pacting the pile toe via a mandrel from pile, similar to closed-ended piles. The ef-
inside the pile, creating a tensile force that fective plug, however, will not develop
Piled Waterfront Structures 739

(a) (b ) (C)

(d )
{e)

Figure 7-27. Steel pipe pile: (a) standard conical point; (b) end fiat plate: when contact
sloping hard stratum an unsymmetrical stress distribution will most lik.ely start crippling the
pile; (c) symmetrical pattern of stress distribution by conical point; (d) damage to open-ended
steel pipe pile after encounter with large boulder; (e) damage to open-ended steel pipe pile after
hitting the sloping bedrock.
740 Piled Waterfront Structures

within a pipe pile if the internal friction from steel plates rolled into cans (cylinders)
between the plug and the pile interior is not whose seams are closed by longitudinal
adequate to balance the full end-bearing welds. The cans are joined at their ends by
pressure. This may occur in the case where girth butt welds. Seams of two adjacent
steel pipe piles are of a very large diameter. cans are usually rotated 90° apart.
In this case, the pile, as described by Yaron In the process of rolling the individual
et al. (1982), will cut through soils such as pile cans, variation may occur both in the
sand and calcareous layers "like a cookie diameter and roundness, particularly near
cutter through the batter." To prevent this the longitudinal seam weld. At the point
phenomenon from occurring, the pile's in- where the plate edges meet in the rolling
terna! diameter should be reduced by process, it is difficult to deliver the full
installing horizontal places, as discussed in curvature to the can; therefore, the plate
Section 7.1. near the edges may remain somewhat flat.
In some practica! cases, to facilitate pile- This may result in a bending moment dis-
driving and for deeper penetration, the soil continuity with a localized stress concentra-
plug inside the pipe pile was removed by tion. Furthermore, when the cans are joined
drilling it out or by using the airlift method. together, the localized distortions in can
To restore the pile's end-bearing capacity, diameter and roundness will produce a mis-
the removed soil plug should be replaced match at the girth butt weld.
by a concrete plug. The plug is installed
Hence, great care must be exercised dur-
underwater by the tremie method or by
ing pile fabrication. Failure to meet the
pumping in concrete. Essentially, this plug
very stringent tolerances for pile fabrication
must have a sufficient length to develop full
may result in structural failure.
bonding capacity inside the pile. It must be
Commercially fabricated pipe piles are
remembered, however, that in large-
available in lengths up to 21 m (Stelco) and
diameter piles (1 m or more), cement-rich
are supplied with beveled ends to facilitate
concrete or grout plugs can develop exces-
splicing, if required. In marine applications,
sive heat. This may cause concrete cracking
as the plug cools. To reduce the heat of the prebeveled sections are normally joined
hydration, a blast furnace slag~cement is together by full penetration welds with
normally specified. Furthermore, inclusion backup plates that are usually built into
of synthetic fibers in a concrete mix will the stabbing guides. However, where soil
restrain the development of the intrinsic has to be cleaned up from the pile's interior,
cracks and inhibit crack propagation. interna! backup plates cannot be employed,
Pipe piles are commercially available in and the stabbing guides must be external to
a variety of lengths, diameters, and wall the pipe pile.
thicknesses. For example, a Canadian Piles can also be spliced together by spe-
steel-producing company (Stelco) offers pipe cial splicers made in a form of a sleeve. The
piles in a range from 219 to 1524 mm in sleeve is slipped onto the head of the previ-
diameter with wall thicknesses that vary ously driven pile length and welded to it by
from 4.8 to 28.6 mm. These piles are avail- a circumferential weld of adequate size. The
able in 241-, 310-, and 345-MPa minimum next length is set into the splicer and accu-
yield strengths. In offshore engineering, rately aligned with the previously driven
large-diameter steel piles (cylindrical piles) length. Finally, the weld is made on top for
ranging from 1 to 2 m (and up to 4 m) in the splicer, and driving continues as soon
diameter have been commonly used. In con- as the weld cools. For long piles, with sev-
ventional offshore construction practice, erallengths added, great care must be exer-
large-diameter steel piles are fabricated cised to ensure accurate axial alignment of
Piled W aterfront Structures 741

each length, so that the pile will be as piles in lieu of lifting lugs. Holes are burned
straight as possible. undersized and then reamed. U se of burned
The primary type of fabrication which holes alone, without reaming, may result in
occurs on most steel-pile-driving jobs is the dangerous stress concentrations during pile
welding of pile splices. These splices must handling and driving.
be thoroughly checked for alignment of the Where the pile is expected to encounter
pile and compliance to specifications. The hard layers, and therefore excessively hard
engineer should ascertain whether the driving is expected, for example when driv-
welder performing the job is properly certi- ing through limestone or similar, the pipe
fied to carry out the specified welding. piles should be furnished with an open-
In most jurisdictions, properly qualified ended cutting shoe. For small to moderate-
welders would carry certification cards. The diameter piles, cast steel shoes are readily
welding procedure is of paramount impor- available. The cutting shoe protects the pipe
tance; hence, materials must be carefully pile and makes it practica! to use a
selected with regard to the pile steel quali- thinner-walled pipe for driving through a
ties and the temperature at which driving hard media. In large-diameter piles, the re-
is performed. This is especially critical when inforced shoe should be at least one diame-
pile-driving is carried out at low tempera- ter in length and have a wall thickness 1.5
tures. In this case, low-hydrogen electrodes times the minimum thickness of pile section
are normally used. in that pile (American Petroleum Institute,
In some instances the head of a thin- 1989). Experience in driving large-diameter
walled piled section on which driving is pipe piles through weak limestone, contain-
carried out may be deformed during driv- ing embedded ballast cobbles, indicates that
ing. This pile must be reheated before splic- such a shoe should be two diameters in
ing is performed. Allowance for pile reheat- length to prevent pile buckling (Gerwick,
ing is usually considered. Conversely, 1986). Where most of the pile's capacity is
heavy-walled piles normally do not need to developed through skin friction, the thick-
be reheated. Essentially, the wall fitting ened shoe section is usually made flush with
driving heads would minimize head the exterior surface of the pile, so as not to
damage. reduce the shaft resistance. However, where
Large-diameter pipe piles are normally the pile is intended to be driven into the
handled with the help of lifting lugs welded hard stratum and socketed there, the pro-
transversely to the pile section. The pile tective shoe is normally the same interna!
lifting points are usually designed for the diameter as that of the pile. This approach
initial pickup load and for the load that is does not provide obstructions for drilling
developed when the pile is rotated to align- through the pile where the pile's interior is
ment with its final axis. In most practica} required to be cleaned of soil or where
cases, the lifting lug is designed to support socketed into the bedrock.
the entire hanging weight. In all design The presence of a driving shoe affects the
cases, allowance for the impact that varies conditions of soil plug formation thin an
from 50% to 100% is usually considered. open-ended pipe pile which, in turn, affects
Essentially, the pile toughness must be pile-driving. For example, Akagi et al. (1985)
adequate to maintain intact its shape dur- found that piles with both external
ing handling and driving procedures. Lift- and interna! thickened sections driven in
ing lugs, if required, are removed by medium-dense to dense submerged sand re-
flame-cutting 5 mm from the pile's surface. quired fewer blows to obtain the required
Sometimes it is followed by grinding the cut penetration relative to piles with no shoe at
smooth. Holes are usually used in smaller ali. The latter investigators also found that
742 Piled Waterfront Structures

the soil plug is developing in aU types of 7.5.1.1.5 tnsert steel piles


experimental piles. However, the longest
When proven very difficult or impossible
plug was developed in a pile having exter-
to drive pipe piles to the required penetra-
nally fiushed shoe. For more details on the
tion, then in order to reach the required tip
effects of pile shoe geometry and its effect
elevation, an insert pile can be used
on plug formation and pile drivability, the
(Gerwick, 1986). In this case, first the soil
interested reader is referred to Raines et al. plug is removed from the primary pile, then
(1992). Where the pipe pile is expected to a smaller-diameter pipe pile, or a pile of
encounter an obstacle within the founda- other suitable cross section is placed inside
tion soil or hard bedrock, the absence of tip primary pile and driven ahead. The insert
reinforcement can result in severe damage pile, which is free from skin friction over
to the pile's end. Typical examples are de- the length of the primary pile, can be driven
picted in Figures 7-27d and 7-27e. to a substantial additional penetration. The
The D jt ratio is critica! for local pile connection between both primary and in-
buckling and it must be limited so as to sert pile is achieved by filling the annulus
preclude pile overstress. Here, D is the pile between both piles with a good quality
externa! diameter and t is the wall thick- grout. Naturally, the smaller section will
ness. Where moderate driving resistance is produce a reduced moment and skin friction
anticipated, the pile is usually designed as capacity. Grouting of the insert pile, along
a steel cylindrical member and checked for with its deeper penetration, helps to in-
buckling stability due to the combined ac- crease the overall pile skin friction capacity.
tion of both axial compression and bending. A grout pipe with exit nipple(s) may be
By a rule of thumb, the latter is not critica! installed inside the insert pile.
when D jt is less than or equal to 60. When
D jt is greater than 60, the piles should be 7.5.1.1.6 soi/ plug
checked for buckling stability by using an
The behavior of an open-ended pile is
in-depth analysis, such as one recom-
infiuenced greatly by the soil plug that is
mended in API RP 2A (American Petroleum developing inside the pile during driving.
Institute, 1989).
Usually, when a pile is driven into the soil,
The performance of axially loaded pipe a soil column is created inside the pile. As
piles is affected by many parameters [e.g., penetration continues, this inner soil col-
soil parameters, pile characteristics (com- umn may develop sufficient frictional re-
pressibility, displacement ratio, diameter, sistance to prevent further soil intrusion,
length), installation method, loading, and causing the pile to become plugged with
others]. Kraft (1991), on the hasis of a de- soil. This prevents entry of additional soil
tailed review of available literature, offers a during penetration. When plugged, the
comprehensive discussion of these parame- open-ended pile penetrates in a mode char-
ters and their influence on the performance acteristic of a closed-end pile. In conven-
of axially loaded pipe piles in sand. tional design practice, the soil plugging
In the case of open pile construction, pile mechanism requires that a balance between
buckling stability can be enhanced by filling the interna! skin friction (due to soil
it with concrete, if required. Concrete-filled column-pile interaction) and the pile re-
piles are also used to resist lateral load sistance over the full pile cross section is
such as moderate ice impact. achieved.
More recent information on versatility Paikowsky (1990) suggests that, in sand,
and effectiveness of the large-diameter steel the plugging mechanism is based on soil
pipe pile is given in Gerwick et al. (1995). arching phenomenon. According to
Piled Waterfront Structures 743

Paikowsky, the arches are built-up and de- investigators conclude that in the open-
stroyed during pile penetration; this contin- ended pipe piles driven into a layer of dense
ues to the point where stable arches are sand plug slippage may occur during driv-
developed. ing at approximately equal rates that are
Brucy et al. (1991) studied plugging and up to pjd = 21 for all pile toe geometrics
driving resistance in plane-toe piles in satu- (e.g., a pile with no shoe or with the toe
rated sand as a function of driving energy. section thickened externally or internally.
They found that longer plugs were present Here, p is the pile penetration and d is the
when the energy of the hammer was the pile diameter.
greatest, hammer mass was lowest, and On the other hand, Yamagata et al.
where single-acting hammers were used. (1985), on the basis of driving tests on
They established that the incremental fiU- open-ended piles without shoes in mixed
ing ratio varies as a function of the soil profiles of clay and medium-dense sand,
density and compressibility. The latter au- conclude that plugs were not formed at all
thors also point out that the modes of pile during driving and that the only significant
penetration, and therefore these modes of resistance to penetration was in toe bear-
failure, are completely different under static ing. Here, the interlayered clay may have
and dynamic loads. In an open-ended pipe served to "lubricate" the interior pile sur-
pile loaded to failure, either the full end face during driving.
bearing at the pile tip is mobilized or the It follows from these and other reported
soil plug slips. The weakest component in observations that the soil plug interacts
this scheme is determination of the mecha- with the pile in a complex manner. This
nism of the soil plug-pile interaction. Con- may be further complicated where an inter-
ventional design practice, as stipulated by na! water jet is used to reduce the friction
American Petroleum Institute API 2A between the plug and the pile. The problem
(1989), recommends that the interna! plug is complicated by the fact that if a pile does
resistance be estimated by using the same not plug or plugs only partly during driv-
unit shear components used for estimating ing, it may stiU plug solidly and behave as a
the external skin friction. closed-ended pile during static loading
Randolph (1985) has shown that in the (Heerema and DeLong, 1980). These obser-
case of dense sand, a soil plug can have vations also suggest that the conventional
much greater capacity than predicted by one-dimensional wave equation analysis,
the conventional approach. This has been which is routinely used to model pile-
confirmed analytically and experimentally driving and to assess pile capacity, be used
by Murff and Raines (1990). The latter au- with caution. For more discussions on this
thors suggest that under drained conditions subject, the reader is referred to Raines et
in dense sand, the plug length equal to a al. (1992) and Randolph et al. (1991).
few pile diameters will be effectively locked
up, generating pressures sufficient to crush 7.5.1.1.7 Box pi/es
the sand grains.
Raines et al. (1992) conducted driving These piles are typically fabricated from
tests on 102-mm-outside-diameter steel pipe the conventional steel sheet piles or from
piles driven by the impact method into a sheet piles with the addition of steel plates,
saturated sand column. The column was as depicted in Figure 7-23. The advantage
confined in a pressure chamber to simulate of this type of pile is that they can be
a near-isotropic effective stress state in a incorporated in a row of conventional steel
dense layer of sand 30-38 m below the sheet-piling to carry heavy axialloads, such
surface of a submerged soil deposit. These as those produced by portal cranes operat-
744 Piled Waterfront Structures

ing along the edge of a wharf. Steel box length and are used for a wide range of
piles resemble steel pipe piles in that they loads. Precast piles are widely used for con-
can be driven either open ended or provided struction of wharves, piers, and other port
with shoes; also, they can be filled with related structures where piles are required
concrete or left hollow. Typical examples of to be extended above the soil level in the
box piles fabricated from conventional sheet form of structural columns, or perform as a
piles are illustrated in Figure 7-28. member in tension.
The structural design of precast concrete
7.5.1.1.8 Precast concrete piles piles is usually governed by the stresses
caused by pile lifting and handling, as well
Concrete piles used in waterfront con- as by stresses caused by pile-driving. In
struction are basically of two types: precast pile-driving, the stress wave after each flow
and cast in place. The former is usually travels from the pile head to the toe and
used for overwater construction and the lat- back, causing tensile stresses in the pile.
ter is generally employed for construction The greatest stresses, however, to which
on land where precast piles are unaccept- regular precast pile is usually subjected are
able, either due to geotechnical or environ- due to its handling prior to driving.
mental conditions. Experience has demonstrated that unless
Precast concrete piles have been used stress in the reinforcing steel is kept suffi-
extensively worldwide since the 1930s. Con- ciently low, fine cracks may develop in the
ventional precast piles are constructed surface of precast pile during handling.
either square with constant cross section, These cracks will generally close and be
tapered, or octagonal, with or without a almost invisible when the pile reaches an
circular cored hole to save material and upright position before driving, but they
reduce weight. Solid concrete circular piles will usually become visibly enlarged during
of constant cross section, cast horizontally pile-driving as a result of the impact stress
in round metal forms, have been used also. set up by the hammer. In salt water, these
Depending on foundation soil conditions and cracks may lead to corrosion of the rein-
method of driving, the tip of these piles may forcement. Hence, piles must be designed
or may not be protected by a metal shoe. and handled in such a manner that crack-
Typically, conventional precast concrete ing will not occur. Typically, precast piles
piles with non prestressed reinforcement are handled with two or more pickup points.
are economically fabricated up to 30 m in In figuring the weight of the pile for calcu-

(b) (c)

Figure 7-28. Typical steel box piles: (a) Larsen box pile; (b) Frodingham double-box pile; (c)
Frodingham plated box pile.
Piled Waterfront Structures 745

lating the bending moment due to its 3. Mechanicallocking


handling, a minimum of 25% of the weight 4. Connector ring
is normally added for impact. 5. Wedge
Because precast piles are usually cast at
6. Sleeve
specialized yards, they are subject to close
control and inspection during their fabrica- 7. Dowel
tion. Typically, these piles can carry loads 8. Posttensioned
up to 50-80 tonnes and can take fairly
heavy punishment during driving. These authors present 20 structural
A disadvantage of precast piles is that sketches illustrating the above type of pile
casting and storage require appreciable splices; the interested reader is referred to
space at both the casting yard and the job their work. Each splice used for a particular
site. At a job site lacking sufficient storage job must be job-specific. Therefore, before
space, the piles are usually stored on dedi- the job is started, the owner, the pile manu-
cated barges or pontoons. Another disad- facturer, the driving contractor, and the en-
vantage is that the required length of the gineer must thoroughly evaluate the pro-
precast pile has to be accurately deter- posed method of pile splicing, associated
mined in advance of installation, which is costs, and the long-term effects on pile per-
not always possible. Adjustments in the formance. These authorities should also
precast pile length after it has been driven agree on the proposed method of pile head
are expensive and time-consuming. If the and tip protection, the driving equipment to
pile is too long, then the excess length must be used, and the step-by-step pile installa-
be knocked off, and if it is too short, then its tion technique. Particular caution must be
length must be increased or additional piles exercised when the structure is constructed
must be added to the wharf system. In- in a moderate to severe seismic environ-
creasing the pile length is a complicated as ment. In this case, embedded steel plates
well as time-consuming effort. clamped together with tensioned high-
The above shortcomings can be effec- strength bolts offer a reliable solution, pro-
tively overcome by the use of piles made up vided that reduction in bolt tension and,
from the relatively short precast pieces therefore, reduction in friction force be-
spliced in the process of pile-driving. Effec- tween embedded plates are prevented. A
tive pile splicing can reduce or eliminate useful discussion on connections in precast
many problems associated with the instal- concrete structures operating in seismic
lation of long precast piles. Among these zones is given in Stanton (1987).
are the elimination of potential waste due Precast piles with prestressed reinforce-
to an inability to forecast the pile length ment are among those most coinmonly used
accurately, and the reduction in stresses in marine application. These piles, ranging
associated with pile handling and trans- in size from 300 to 900 mm square (or
portation. U se of proper splicing method diameter in the case of octagonal cross sec-
eliminates the need to predetermine exact tion) are usually cast in one length. They
pile lengths and allows extension of the pile have been used extensively as bearing piles,
when necessary. piles in tension, and moment-resisting piles.
Bruce and Hebert (197 4) categorize pile Larger piles [generally those above 500 mm
splices as follows: square (diameter)] are usually hollow-cored.
Because of their lower dead weight and
higher resistance to bending moment and
1. Wel.ded tensile load, the prestressed piles can be
2. Bolted made longer than is practica! with a regu-
746 Piled Waterfront Structures

Iar precast pile with nonprestressed rein-


forcement. Prestressing is usually designed
to give a minimum effective prestress of at
least 5.2 MPa in order to permit handling
and driving without damage. For very long
vertical piles, piles in tension, or moment-
resisting piles, higher prestress values (e.g.,
6.9-8.3 MPa) are usually required (Gerwick,
1971, 1974).
Better quality prestressed piles (spun
piles) are roade in a centrifuge. Spun piles
are roade in individual steel molds designed
to resist the pretensioning force during
Figure 7-29. Enlarged ends of concrete piles. [From
casting and curing of the pile. The manufac- Plantema and Nolet (1957).]
turing procedure consists ofplacing the ten-
dons and reinforcing cage in the molds,
stressing the tendons, and placing the mold lower level, resulting in an enhanced bear-
on revolving wheels that turn the mold as ing capacity. Here, the point bearing pile
the concrete is placed. The centrifugal force acts as a column, transmitting the load
compacts the concrete and forces excess down to a level at which it can be safely
water from the plastic concrete. The pile is carried. Piles of this type were extensively
used in Russia and Europe (Tsinker, 1972,
then cured and stripped from the mold.
1988; Gorunov, 1984).
Prestressed piles are more durable than
regular precast reinforced piles because the
concrete is under continuous compression.
7.5.1.1.9 cvlinder concrete pi/es
This prevents concrete spalling during driv- Concrete cylinder piles have been used in
ing. Compression also keeps the hairline Europe and in the United States since the
cracks closed. This prevents or at least lim- 1950s. Although these piles have a very
its contact between the salty seawater and good record, their utilization has been lim-
reinforcing steel. The concrete cover over ited basically by heavy self-weight and
longitudinal prestressed reinforcing steel is bending moment capacity. However, the
typically 75 rom thick. continuing improvements in concrete tech-
Basic requirements for materials used for nology as well as the application of a new
pile fabrication, as well as pile manufactur- generation of pile-driving equipment and
ing and installation are given in revised enhanced methods for pile installation roade
ACI 543 "Recommendations for Design, large-diameter cylindrical piles a viable and
Manufacture and Installation of Concrete economica} solution for marine applications.
Piles", by American Concrete Institute In the past 20-25 years, concrete cylinder
(1973), in Gerwick (1971), Tomlinson (1983, piles ranging from 600 to 2000 mm in diam-
1994), in American Concrete Institute eter (and more) have been used extensively
(1993) and many other relevant articles. for construction of marine terminals and
Sometimes prestressed piles are used other marine related projects such as off-
with enlarged points (Fig. 7-29) or enlarge- shore platforms and bridge piers. Piles of
ments (sometimes called "blisters" or this type have been used to carry design
"widenings") roade on a lower part of a loads of up to 550-600 tonnes and more.
pile's shaft. This usually involves the as- Generally, cylindrical concrete piles can
sumption that the enlargement will be be divided into two classifications: postten-
driven into a competent layer of soil at a sioned and prestressed.
Piled Waterfront Structures 747

d.A--1
Posttensioned piles are multielement piles (a) 'A'
comprised of precast cylinders joined together
using posttensioning techniques. Individual
cylinders are typically manufactured in sec-

~A~
tions approximately 5-8 m long with wall
thicknesses ranging from 100 to 250 mm.
Holes are formed longitudinally through the
walls at the time the sections are cast. After
the precast sections have cured, they are 1 7

/~ ~
aligned and the posttensioning tendons are
threaded through the holes in the walls,
IQ
stressed, and grouted in place. In this man- 'A' 'a'
ner, piles up to 45-50 m long can be made.
The disadvantage of this method of pile fabri-
cation is that because the piles are manufac- (c)
tured in certain predetermined lengths, it is (b)
difficult and costly to change the lengths on
site if required.
Prestressed cylinder piles are usually assem-
bled from sections connected together at the
yard or during the process of pile installation.
Typically, sections are made 5-8 m long with
an embedded interlocking system. Pile sec-
tions 600-2000 mm in diameter are fabri-
cated in a manner similar to that used for
previously discussed prestressed spun piles.
Piles of larger diameters (sometimes referred
to as caissons) have been fabricated in a spe-
cial vibrating steel mold. Large-diameter ~€_ F-F
cylinder piles have been manufactured also
with nonprestressed reinforcement. A typical Figure 7-30. Cylinder concrete pile. (a) Typical ele-
cylinder pile and its details are illustrated in vation, cross section, and details: 1-pile; 2-spiral
reinforcement; 3-cutting edge; 4-steel joint system;
Figure 7-30. Some useful data on prestressed
5-longitudinal regular or prestressed reinforcement;
concrete piles and cylinder piles of 915-1370 6-high-tensile-strength belt. (b, c) Bolted and welded
mm in diameter are given in Libby (1984) connections: 1-pile; 2-spiral reinforcement; 3-stif-
American Conerche Institute (1993) and fener; 4-circulai steel bearing plate; 5-steel transi-
American Petroleum Institute (1989). tion ring; 6-longitudinal regular or prestressed rebar;
7-high-strength bolt; 8-continuous fillet weld;
9-steel transition from concrete to steel; 10-steel
In the United States, posttensioned cutting edge.
cylinder piles have been manufactured by
Raymond International, Inc. They were as-
sembled from lightly reinforced concrete alignment. A plastic joint compound having
units 915 mm and 1370 mm in diameter a high strength when set was first applied
and 100 and 120 mm in wall thickness. to seal the joints between sections. Next,
Longitudinal holes for later prestressing by steel tendons were posttensioned and
steel tendons were left open in the walls by grouted to form a single unit. The excess
rubber cylinders on 22-mm-diameter rods. length of this pile can be cut off at any
Once cured, the short cylinders were assem- elevation. Under favorable soil conditions,
bled end to end horizontal to the desired Raymond posttensioned cylindrical piles
pile length with tendon holes in complete have been used for design loads exceeding
748 Piled Waterfront Structures

1.8 MN and in lengths up to 63.5 m. They In the case of large-diameter piles (2 m


were installed using heavy hammers. and larger), pile-driving can be assisted by
Cylindrical concrete piles are installed by water jets applied both externally and in-
a wide variety of techniques, including driv- ternally. If the combined capacity of the
ing, jetting, vibration, or a combination of available water jets is inadequate, then the
some of these. They may also be set in soil plug that is forming inside the pile
predrilled holes and then seated to the de- should be removed. This can be accom-
signed penetration by driving. plished by drilling it out or by using airlift
Typically, however, the relatively thin- techniques. Normally, plug removal may be
walled large-diameter, concrete cylindrical required when piles are installed into hard
piles, particularly those with diameters clay, sandy clay, or similar soils.
greater than 1000 mm, are not suitable for Typically, cylindrical piles composed of
hard driving by the impact hammer. In this prefabricated sections of limited lengths are
case, pile-driving may be aided by water driven in the following sequence (Fig.
jetting, in which case jet pipes could be 7-32a).
installed either from inside or outside the
pile. Water jetting is usually used in sand 1. The initial pile length consisting of sev-
or sandy type of soils; it is generally ineffi- eral prefabricated sections is driven to
cient for large-diameter piles driven in clay. limited penetration into the foundation
In the latter case, pile-driving can be aided soil. Then the next section of pile is in-
by excavating soil from inside the pile. stalled on a previously driven length
Multiple outside jets are usually in- and both are reliably joined together
(Phase 1).
stalled around the pile as indicated in Fig-
ure 7-31; four or more symmetrically lo- 2. The jet pipes are installed as required
cated jet pipes are usually employed for and the extended pile is driven further by
the hammer or vibrator with the aid of
installation of vertical piles, and at least
water jets. At this stage, the pile is driven
three pipes placed as indicated in Figure to the depth that is equal to the length of
7-31b are used for installation of batter the added section. This operation contin-
piles. The distance between adjacent jet ues until the pile is driven to the desired
pipes installed around the pile peripheral is penetration (Phases 2 through 5).
usually 1.0 mor less. Normally each pipe is 3. Finally, the hammer (or vibrator) and the
designed for discharge of 50-100 m 3 of jet pipes are removed from the installed
water per hour at 0.5-1.0 MPa; lower water pile and the pile-driving system is moved
pressures are used for loosening of fine to to a new location and pile-driving equip-
medium sands, and higher pressures are ment is reinstalled on the next pile.
used for aiding pile-driving in dense sand.
The jets must be run continuously during A practica! example of pile installation is
driving in order to prevent plugging. illustrated in Figures 7-32b-7-32e. In this
In order to prevent soil liquefaction in- particular case, precast prestressed cylin-
side a pipe pile, the jet pipes are usually drical 915-mm-diameter piles, 67 m long,
placed 0.5-1.5 m above the pile cutting edge. have been used for the construction of the
lf the pile is driven to a depth in excess of U.S. Navy pier in the New York harbor
20-25 m, then the effectiveness of the (Stapleton site). Figure 7-32b illustrates the
water jet can be enhanced by adding com- process of pile handling by crane for instal-
pressed air at a rate of no less than 2-3 lation at the design location; the process of
m 3 jmin. Air pipes are usually attached to pile jetting is illustrated in Figure 7-32c;
the jet pipes, at approximately 1 m above Figure 7-32d illustrates the process of re-
the water discharge nozels. moving a soil plug from inside the pile; the
Piled Waterfront Structures 749

final penetration of the pile is obtained by An investigation of the pile cracking re-
impact hammer, as shown in Figure 7-32e. vealed that it was caused by substantial
For complete details on pier construction, hydrodynamic pressure developed insi de the
the reader is referred to papers by Buslov et pile as a result of pile vibration (Savinov
al. (1988) and Damaschi and Bonasia (1989). and Lavrinovich, 1969; Buslov, 1974). This
For more details on pile jetting techniques, resulted in hoop stresses up to 3 MPa high.
the reader is referred to Chapter 6. The mechanism by which the high hydro-
In the past 25-30 years in Russia, con- dynamic pressures were created was ex-
crete cylinder piles made both from conven- plained as follows: when a pile stops at the
tional reinforced concrete and prestressed bottom of its vibrating cycle, the column of
concrete have been successfully used for water moving up inside the pile produces
construction of seaports. Most often, two an interna! pressure within the pile. It was
types of a cylinder pile were used: a pre- concluded from the investigation that low-
stressed pile with outer diameter 1600 mm frequency vibrators are preferable for pile
and wall thickness 120-150 mm, and a pile installation. Furthermore, for better pile
with regular reinforcement and outer diam- protection during installation by vibrators,
eter of 1200 mm (wall thickness 100-120 the use of air-powered shock absorbers of
mm). Piles of both types were assembled to various designs was recommended. Aerat-
required lengths from precast sections 4 ing the water column inside the pile can
and 8 m long. These piles were installed by reduce the density of water considerably
powerful vibrators. The subsequent under- and thus provide good damping. Also,
water inspection revealed long longitudinal special shock absorbers placed inside
cracks 1-50 mm wide, mainly in the middle the pile can effectively reduce the afore-
part of the pile. mentioned hydrodynamic forces (Savinov
and Lavrinovich, 1969).
When hollow cylindrical piles are used
(a) (b) for the construction of marine structures in
cold-temperature regions, the formation of

m
~
ice inside these piles may result in the high
i
A
~-6--
S .
hoop stresses with subsequent pile crack-
ing. This has been observed on severa! ma-
2
A-A
rine projects constructed in northern re-
gions of Russia (Lavrinovich et al., 1967).
To prevent ice formation in the pile exposed
to low temperature, the pile interior may be
filled with nonshrink concrete. Alterna-
tively, to reduce the excessive hoop stresses,
closely spaced spiral reinforcement may be
considered.
B-B
Normally, driving stresses are extremely
high. To mitigate the pile-driving process
and to prevent formation of cracks associ-
ated with hard driving, piles should be
Figure 7-31. Placement of jet pipes to aid driving properly reinforced and the hammer or vi-
large-diameter cylindrical concrete piles: (a) vertical
pile; (b) batter pile. 1-Steel cutting edge; 2-pipe
brator must be in strict alignment with the
pile; 3-steel jet pipe keeper; 4-rubber hose; 5-jet pile. Furthermore, adequate cushioning of
pipe; 6-steel strip to guide jet pipe. the pile head must be used.
750 Piled Waterfront Structures

(a)

1 2 3 4 5

·.: :::.: .=. ·: ,: ·:::: .:·~· .: ;'


... : ·:.. ·:·.' : ~ .

(c)

(b )

Figure 7-32. Sequence ofinstallation oflarge-diameter concrete cylindrical piles. (a) Standard
procedure; (b- e) practica! example. (b) Preassembled pile is lifted by crane; (c) pile is installed by
means of a water jet; (d) removal of soil plug from pile interior; (e) final set by impact hammer.
1- Crane; 2-prefabricated cylinder section to be installed on the top of a previously driven
pile length; 3- cylinder pile assembled from prefabricated sections; 4- templet; 5- jet pipe;
6- hammer or vibrator; 7-complete pile; 8- soil plug ejecting system. (Photographs courtesy of
Dr. V. Buslov, at the time a principal designer to Mueser Rutledge Consulting Engineers, who
together with Frederic R. Harris, both of New York City, designed the pier; Contractor: Yonkers
Contracting, N.Y.jWeeks, Stevedoring, N.J.)
Piled Waterfront Structures 751

(d ) (e)

Figure 7-32. Continued.

Where piles are installed in deep water, the mudline; therefore, adequate temporary
sufficient pile support during construction support must be provided before the pile
should be provided to prevent pile vibration head is embedded in a superstructure. Sim-
and buckling during driving. Raked piles ilarly, when piles are tobe driven in rivers
could be overstressed due to the dead weight with substantial flow velocity or in harbors
of unsupported overhanging length above with high tidal currents or waves, they may
752 Piled Waterfront Structures

require a temporary support to prevent


severe vibration before a superstructure is
in place.

7.5.1.1.10 Belled pi/es


Concrete bells may be constructed at the -···---:==.c...• :-.c.o.:-.-1--Hr------

tip of cylinder piles, both steel and concrete,


to give increased bearing and uplift capac-
ity through direct bearing on the founda-
tion soil (Fig. 7-33). Drilling of the bell is
carried out through the pile by underream-
ing with an expander tool. In a typical in-
stallation, the pile is driven down to seal it'
into the bearing stratum. A hole is then
drilled and belled, a reinforcing cage or the
reinforcing inserts (e.g., steel tubular or
structural member) are installed and the
hole is filled with concrete. Where piles are
designed to resist uplift load, posttension-
ing tendons may also be used for connecting
the bell to the pile shaft or capping super-
structure.
In reinforcing cages, the reinforcing bars
are usually bundled by a spiral reinforce-
ment; naturally, space between the rein-
forcing bars must be such as to permit the
free flow of concrete into the beii. The bell
body beyond the cage is usually left unrein-
forced because the hoop steel cannot be
placed there. The annulus between the re-
inforcing cage or other inserts and the pile
interior must be 6-10 times the maximum
size of coarse aggregate so as to permit flow Figure 7-33. Underreamed beii footing: 1-large-
around the inserts. diameter pipe (steel or concrete) pile; 2-reinforcing
In soft soils, reverse circulation of a cage; 3-underwater concrete; 4-loose sand, soft clay,
drilling mud (usually a bentonite slurry) is or the like; 5-hard clay, gravei, dense sand, or the
employed to keep the empty drilled space like.
from collapsing. Reverse circulation must
have sufficient velocity to remove the cut- tremie pipe is usually about 10 times the
tings. The drilling mud should (preferably) diameter of the maximum size of the coarse
be converted to a calcium base so as to aggregate used. In the pumping method,
avoid coagulation on contact with the the fine rounded aggregates, with an ade-
cement. quate sand content (at least 50%), are used
Placement of concrete can be carried out in concrete mix to ensure satisfactory place-
either by the tremie method or by pumping. ment of concrete. Because the flow rate of
Both methods have been used in the past concrete is normally controlled by friction
and prove to be reliable. The diameter of in the pipe, a field test is usually run to
Piled Waterfront Structures 753

determine the actual friction that is devel- in 1839 and were first used for the founda-
oped while pumping the actual mix. This tion of a lighthouse in the Thames estuary
helps to size the delivery pipe properly. To (Mitchell, 1848). There, nine wrought iron
prevent the formation of a vacuum in the piles, 125 mm in diameter and 7.95 m long
dawn-leading pipe, a relief valve is usually with 1.22-m-diameter cast-iron screws, were
installed at the top. screwed 6. 7 m into a sandbank and an iron
Placement of large amounts of confined frame was erected on them.
concrete can generate a great deal of heat In the past, timber, wrought iron, and
due to hydration. Resulting expansion and cast iron have all been used for the pile
subsequent cooling may result in severe shaft used with cast-iron screws. In recent
cracking. Hence, the cement selected for the years however, only steel and concrete have
concrete mix should be a low-heat-type ce- been used for screw piles. There are numer-
ment; for example, ASTM Type N or Type ous examples in which these piles have been
II with Pozzolanic replacement of up to 50% successfully implemented. Screw piles have
of the cement. Alternatively, a low-heat- the advantage that the large-diameter heli-
generating coarse ground slag-portland ce- cal blade gives a high bearing capacity or
ment (70: 30) mix may be used. The mix high uplift resistance to the pile because of
should be as cool as practical at the time of the large bearing area of the blade. This
placement. The aggregate should be pre- makes the best use of a low-bearing-
cooled with a cold water spray, orice could capacity soil.
be added to the mixing water to facilitate Furthermore, the shaft of a screw pile is
cooling. Additionally, liquid nitrogen can be usually of a robust construction, which en-
injected into the mix. As stated earlier, the ables it to resist substantial lateral forces,
addition of synthetic fibers to the concrete as in piers and dock structures. For exam-
mix can prevent formation of the intrinsic ple, the piles included in the structure de-
cracks. picted in Figure 7-34 have a shaft diameter
For heat reduction, Gerwick (1986) rec- equal to 915 mm.
ommends a mix which incorporates sand or Screw piles can be used with either closed
small aggregates (e.g., pea gravei 8-10 mm or open ends. The latter type is used in
in diameter). This type of cement mortar, difficult soil conditions when water jetting
according to Gerwick, has relatively low is needed to loosen the soil below the blade.
heat properties and good tensile strength To assist pile sinking, the soil plug can be
and yet is stiU able to flow readily. For more removed from inside the hollow cylinder
details on underwater placement of con- and from below its base. Typically, the screw
crete, the reader is referred to Chapter 2. has a complete turn, with a pitch large
The effects of the bell geometry on belled enough to pass the stones expected to be
pile in uplift performance are discussed by encountered. Screw piles normally have
Dickin and Leung (1992). been installed employing a capstan head
fitted to the pile head (upper end). Torque is
7.5.1.1.11 Screw pi/es and screw applied to the capstan via a steel cable to
cvlinders winch, or by electric screwing capstans op-
erating within a specially designed head
Steel or concrete screw piles and cylin- frame.
ders are a form of pile in which the shaft at It is very difficult to install screw piles
its lower end is fitted with a large-diameter through stiff clays or dense sand and gravei,
horizontal helical blade. Examples are even when aided by a water jet. In fact, the
shown in Figures 7-9b and 7-34. These water jet, although helping facilitate pile
piles were invented by Alexander Mitchell sinking through the hard clay, may cause
754 Piled Waterfront Structures

t - - - - - - - - - 2 . 2 _ 3 _ . 3_ _ _ _ _ _ _-----.;

Figure 7-34. Pier in Scotland founded on screw cylinders: 1-915-mm-


diameter cylinder; 2-2440-mm-diameter screws; 3-crane track support-
ing beams; 4-road supporting beams; 5-transverse beams; 6-deck and
railway track supporting beams; 7-406-mm-diameter wooden fender pile;
8-2155-mm-diameter screws; 9-natural seafloor level; 10-dredged level.
[From Morgan (1944).]

excessive loosening of the clay and there- where no suitable bearing stratum for regu-
fore a substantial reduction of its bearing/ lar piles exists within a reasonable depth.
uplift properties. To some extent, this also Essentially, screw piles are much shorter
applies to dense sands and gravels. Screw than regular piles because they can achieve
piles function most efficiently in weak soils a desired carrying capacity either in com-
(e.g., soft clay or loose sand) where they are pression or uplift at a much shorter pene-
easy to install. tration. On the other hand, some additional
In general, screw piles are best suited to settlement of these piles may be expected
conditions where they can be sunk through due to the disturbance of a soil media caused
soft materials to a very limited penetration by the screwing action.
into firm stratum (e.g., stiff clay or dense A further advantage of the screw piles is
sand). They penetrate soft soils without that they can be safely installed without
great difficulties and will push aside boul- adversely affecting adjacent structures, and
ders that are not too large. Screw piles and the process of pile installation is practically
cylinders are most useful in conditions noiseless.
Piled Waterfront Structures 755

A detailed discussion on screw (helical) mined. Furthermore, installation of drilled


piles is beyond scope of this text. For more piles in water-bearing soils normally re-
information on the subject, the interested quires the use of casings.
reader is referred to Chaly and Hanna In N orth America, drilled pile construc-
(1992, 1994) and Chaly et al. (1991). The tion should conform to the requirements
information on a so-called "spin-fin" pile, specified in the latest publication of ACI
which is essentially a screw pile with a 336.1-79 (revised 1988) published by the
large pitch, is given in Pierce and American Concrete Institute (1988). Uplift
Martinson (1992) and Gilman and capacity of small-diameter drilled piles is
Nottingham (1991). discussed by Lutenegger and Miller (1994).
For comprehensive information on this
7.5.1.1.12 cast-in-ptace piles type of pile, the reader is referred to Chellis
(1961) and Tomlinson (1980, 1994). Infor-
These piles are occasionally employed in
mation can also be found in almost any
marine applications. Examples were dis-
standard text on foundation engineering.
cussed earlier and are illustrated in Figures
7-24 and 7-25. There are a variety ofthese
piles. They can be formed and installed 7.5.1.1.13 Plastic pi/ing
practically in ali ground conditions. This is, perhaps, the latest innovation in
In marine engineering applications, these marine engineering. Some "plastic" piles are
piles have been used as drilled cast-in-place in fact piles of composite steel-plastic con-
piles. The size ofthese piles can vary widely struction; they are fabricated from steel
from small (300 mm in diameter) to very pipes encased in recycled plastic (Heinz,
large (up to 1500 mm in diameter and more). 1993).
Casing or drilling mud is normally used to At present, these piles are fabricated in
keep a drilled hole in water-bearing soils sections about 6 m long with a 280-330-mm
open. After installation of the reinforcement outside diameter. Both ends of the 150-
cage, the concrete is pumped in and is placed mm-diameter steel insert pipe are female
by the tremie technique. threaded, which enables the joining of pile
Drilled piles can also have an enlarged sections with a threaded steel coupler. The
bottom (beii) for an additional bearing area. adjacent sections must be tightly joined; the
The beii is created by special auguring tools. joint must be tight enough to ensure a solid
The advantages of drilled piles are as contact across the interface of the plastic at
follows: each joint. A gap between plastic surfaces,
particularly at the outer edge, will allow
1. They do not require soil displacement and, the joint to fiex or tip. Insignificant move-
therefore, do not produce any heave.
ment at the joint could result, causing the
2. The noise and soil vibration are minimal. pile to run to the side during driving, and
3. They allow inspection of the bearing thus go out of alignment.
stratum. A driving plate and protective steel shoe
4. They penetrate buried debris and large are installed at both ends of the pile simply
boulders. by screwing the attached nipples into the
5. They can be constructed in any practical pile steel core.
size. The driving plate is required to prevent
separation of the plastic cover from the steel
The disadvantage of this pile is that skin pipe core. Without the driving plate, under
friction may not be fully developed and the the impact force of the hammer, the pipe
bearing capacity cannot be readily deter- can be driven into the wood cushion block;
756 Piled Waterfront Structures

therefore, substantial loads can be trans- that for cylindrical piles with the same vol-
ferred to the plastic encasement. This can ume and the same mean radius. Zilberberg
produce a significant shear stress at the and Sherstnev investigated performance of
interface between steel and plastic parts, a tapered pile with rectangular cross sec-
resulting in a bond break between the pile tion. These piles were driven in sandy soils.
and the plastic. Axial capacity of these piles was 200-300%
Another type of plastic pile that is used higher than that of cylindrical piles with
in marine application is discussed by Taylor the same volume and mean radius. In the
(1995). These are fabricated in a single piece majority of the aforementioned field tests
with no joints to make up or seals to install. the additional side resistance continued to
These piles are available in lengths as re- develop with further axial deformation, in
quired up to 31 m. They are fabricated from contrast to conventional cylindrical piles
100% recycled plastic and are structurally where side resistance generally reaches a
reinforced with fiberglass bars. limiting value. It should be noted that most
At present, plastic piles up to 27 m long of these tests have not been carried out up
penetrating soil up to 7.6 m have been used to pile failure.To-date two basic types of a
as substitutes for creosoted timber fender tapered pile have been used for construc-
piles. They have been installed at six U.S. tion of miscellaneous geotechnical projects.
ports and, according to Heinz (1993) and These are the step taper piles (a propri-
Taylor (1995), are functioning well. etary system of Raymond International
The advantages of plastic piles is ·that Builders, Inc., Houston, TX) and piles in
they are environmentally friendly, are im- which the top cross section is greater than
pervious to marine borers, need low mainte- the bottom one. In both types an increase in
nance, and are made from recycled plastic. the pile side resistance is expected when
Also, these piles are highly ductile, possess there is some slip of the pile relative to the
a high energy-absorption capacity and can ground.
therefore be successfully utilized for con- The typical step taper pile is composed
struction of fender systems. · Plastic piles from approximately 7.9 m long cylindrical
also can be effectively used as bearing piles. steel jackets of different diameters that are
connected to each other to form a long
7.5.1.1.14 Tapered pi/es casing that steps downward in diameter
from top to bottom. At each pile location a
Under certain site conditions, e.g., when shaft is drilled part way to the design depth.
the structure such as relieving platform is Then the casing is placed in this shaft and
constructed in dry and excavated later, sim- driven with a mandrel to its full depth. The
ilar to that illustrated in Figures 7-24 and casing is then filled with concrete. The sec-
7-86a, use of a tapered pile can have poten- ond type of pile is usually fabricated from
tial cost advantage. Earlier investigations regular or prestressed concrete. Note, that
carried out by Norlund (1963) reveal that the long wooden piles are naturally tapered
having a taper on the pile wall can substan- piles. As can be expected the axial capacity
tially increase the compressive axial load of the compressed tapered pile depends on
capacity of the pile. Most recently Rybnikov taper angle, ground strength and ground
(1990) and Zilberberg and Sherstnev (1990) modulus of deformation. It is obvious that
reported results of a field test conducted in the uplift capacity of the tapered pile should
the former Soviet Union. Rybnikov has re- be substantially less than its compressive
ported on the behavior of a bored cast-in- capacity.
place tapered cylindrical pile. These piles Despite their advantages the tapered
had bearing capacity 20-30% higher than piles in most practica! cases are not consid-
Piled Waterfront Structures 757

ered as a potential design option for con- Normally, the choice of pile type is based
struction of marine structures. The most on the following considerations.
apparent reason for this is a lack of aware-
ness of a tapered pile performance charac- (a) Characteristics of the ground through
teristics under effects of axi.al and particu- which the piles are to be driven and on
larly lateral loads. Also, to-date there is a which they are to be founded.
lack of a well developed analytical proce- (b) Pile materials, which are typically wood,
dure for predicting the pile response to ax- steel, concrete, or a combination of these.
ial and lateralloads. The simplified theoret- (c) Destructive infl.uences, which include ef-
ical model for evaluation of a tapered pile fects of marine borers, wave action, salt
spray, presence of strong current, chemi-
performance is developed by Kodikara and
cal attack, corrosion, scour around piles,
Moore (1993). and others.
(d) Design life that can be considered as tem-
7.5.1.3 Selecting the Pile Type porary or permanent.
(e) Investment justification.
When selecting a suitable type of pile it
should be remembered that no single type Selection of the most economica! and tech-
can be best suited to meet all of the great nically feasible pile can be very difficult in
variety of conditions likely to be encoun- many cases. Hence, the designer must care-
tered in the construction of marine struc- fully investigate advantages and disadvan-
tures. tages of each altemative type of pile to
The pile selection is usually affected by avoid the decision that can compromise his
site geological and environmental condi- or her design. The following are brief char-
tions, the type of superstructure, and the acteristics of piles of different types that
value and character of the design live loads. are employed in marine applications.
In addition, as has been shown earlier in
this chapter, the economics of pile selection Timber Piles-Suitable for moderate loads,
depend greatly on the piling contractor's temporary wharves, or where wood is locally
ability to perform a certain type of pile available at competitive prices. They are un-
work. suitable for use in deep water structures and
On a large project, more than one type of for heavy loads. Timber piles are subject to
pile may meet the technical requirements of attack by marine borers and to decay due to
fluctuating water levels. When driven too
a given job. The availability of a certain heavily, they are also subject to brooming or
material at a given time and place and the unseen splitting.
cost of the material may prove to be the Concrete Piles-Suitable for all practica!
decisive factors. Cost of labor, which varies ranges of loading and readily adaptable to
from country to country, and availability of various sizes and shapes. They can be fabri-
pile handling and driving equipment also cated from regular reinforced concrete or from
can be important factors in the selection of prestressed (posttensioned) concrete. If prop-
a pile type. In countries experiencing dif- erly designed and installed, these piles are
ficulties with producing or importing cer- practically maintenance-free in corrosive
tain materials (e.g., steel pipe piles, steel marine environments.
Disadvantages of these piles are as follows:
sheet piles, and other metal products suit- heavy weight which requires heavy handling
able for marine application), the engineers and driving equipment; unseen cracks which
are compelled to use less economica! but can develop during handling and driving; de-
locally available materials, or even to switch lay between casting and installation; large
to a nonpiled type of structure. area needed for storage. In addition, some
758 Piled Waterfront Structures

types of concrete piles, both regular and pre- Hammer ram point
stressed concrete, cannot be readily cut down Hammer base
or extended. Jp.z:iZ?Z~J_- Cap block shield
Steel Piles-Suitable for ali practica! ~ ~W4~~~= Cap block or hammer cushioo
~ Drivehead
ranges of loading. They can be readily ad-
Pile cushion
justed to the required penetration length (cut
down or extended by welding of additional
lengths). They can be hard driven without
being damaged. Some types of steel pipes have
small ground displacement and, therefore, can
be safely installed close to existing structures
(e.g., retaining walls). A disadvantage of steel
piling is that in marine environments it re-
quires elaborate protection from corrosion,
such as paint coating, use of cathodic protec-
tion, installation of protective jackets, or a
combination of these.

7.5.1.4 Pile Jnstallation


This section provides brief background
Figure 7-35. Principal component of the pile-driving
information on piling equipment and pile system. [From ASCE Deep Foundations Committee
installation methods and requirements. (1984).]
Some relevant information is also provided
in Chapter 6. For detailed information on
pile installation, the interested reader is Normally, the pile head, particularly a con-
referred to any standard text on the subject crete one, is protected by the pile cushion
of foundation engineering. from possible damage due to hammer im-
As follows from the information provided pact. There are different systems of pile
in previous sections, there are four basic hammers that are used to drive piles. The
methods for pile installations used for the selection of pile-driving hammer is based
construction of marine structures: mainly on the type and size of the pile being
used on a job. The selection also depends on
1. Driving with impact hammer the experience of the piling contractor and
2. Installation by vibrators on the availability of the required driving
3. Forming piles in situ equipment. ·
The latest developments in pile-driving
4. Installation by water jet
equipment are the introduction of hydrauli-
cally driven hammers and sophisticated
A combination of the first two methods with
control systems. This has resulted in a more
a water jet has also been used extensively.
efficient pile installation and often in a
smaller crew needed on site.
7.5.1.4.1 Pile driving with impact
Hydraulic pile hammers like conven-
hammer
tional hammers are available in single- and
The principal arrangement of a pile- double-acting modes. The single-acting
driving system that includes the pile, ham- hammer is basically an hydraulically oper-
mer, and other components is depicted in ated drop hammer where the ram is raised
Figure 7-35. As discussed in Chapter 6, the hydraulically; when the required stroke is
hammer impact load is delivered to the pile reached, the hammer ram is released and
head via the cap block or hammer cushion. descends under the force of gravity to strike
Piled Waterfront Structures 759

the pile. The double-acting hammer uses ing time, it is usually advisable to use as
hydraulic power on the downstroke as well. heavy a hammer as possible without over-
The advantage of the hydraulic pile ham- stressing the pile.
mer is that it delivers a less punishing blow lnteresting and helpful observations on
to the pile head, which is particularly im- the effects of the impact pile-driving sys-
portant when driving in soft and variable tems on drivability and capacity of concrete
soils. This is achieved by the precise stroke piles are given in Thompson and Thompson
control that gives this hammer increased (1979). For a general discussion on pile-
efficiency of the blow. The other important driving hammers, the reader is referred to
advantage of the hydraulic hammer is that Chapter 6. Detailed information on pile-
it can operate in air and underwater. This driving equipment is found in recent pub-
makes it especially valuable for underwater lications by Gerwick (1986), Prakash
installation of piles driven within jackets. and Sharma (1990), Tomlinson (1994), and
The main disadvantage of the hydraulic others.
hammer is that it requires a separate power Noise is an obvious disadvantage of im-
source connected by hydraulic hoses which pact driving equipment, and in some situa-
can, in certain conditions, cause handling tions, noise restrictions may prohibit the
problems, particularly when driving steel use of impact pile-driving for pile installa-
sheet piles. tion. When noise is a problem, a special
Hydraulic hammers are generally avail- field procedure for reducing the noise should
able with ram weights varying from 2 to be considered. For example, most of the
over 10 tonnes, with energy outputs of 12 noise can be avoided or reduced by using a
tonne meter and greater. lnstruments added softer cushion or by installing baffie cham-
to the hydraulic hammer system measure bers around the driving rigs to intercept the
and display the amount of blow energy de- sound. This obViously makes pile-driving
livered to the pile and the displacement of more difficult and costly. It should be noted
the pile into the ground. This allows the that manufacturers of pile-driving equip-
operator to keep the stresses in the pile ment are developing less "noisy" pile-
under control. driving equipment.
In general, there is no single type of
impact hammer that is best suited for ali 7.5.1.4.2 Pile-driving rigs
classes of work; in some instances, more
than one type of hammer may be suitable The hammer is operated and guided on a
for a particular project. For example, when rig. Floating rigs are normally used for off-
driving displacement piles (e.g., precast shore operations. They typically include a
concrete into dense strata such as stiff clay, barge or special pontoon with a high driving
dense sand, or gravei), a heavy blow with a frame designed for installation of long off-
heavy ram, short stroke, and low impact shore piles. The frame has stiff leaders that
velocity has been found most satisfactory; serve the dual purpose of holding the pile in
the low impact velocity allows more energy position and maintaining the axial align-
to be transferred into the motion of the pile. ment of the pile and the hammer. The bot-
On the other hand, when driving light- tom part of the driving frame is fastened to
weight piles in soils of average consistency, a special moonbeam that allows the frame
the higher speed of double-acting hammers to move in a lateral and fore-and-aft direc-
is thought to keep the pile in motion better tions, thus permitting pile-driving over a
by reducing inertia, skin friction, and point wide range of side batter and in-and-out
resistance. batter. Examples are illustrated in Figures
Since the size of the hammer affects driv- 7-Se and 7-36b.
760 Piled Waterfront Structures

Figure 7-36. Floating pile-driving rig: (a) vertical steel pipe pile driven by a
diesel hammer; (b) batter steel pipe pile driven wi~ negative slope; (c) typical
pile-hammer system. ·

Piles can be driven also from the rig erally defined as 75% of the tipping load.
mounted on a commercial crawler or other For truck cranes, it is usually taken as
suitable mobile crane. Depending on site 85%. Detailed discussions on stability of
condition, this system can operate from crane-mounted pile drivers are given in
land, pontoon, or barge (Fig. 7- 37). For bet- Rang (1969).
ter pontoon or barge floating stability, the In cases where water jets are used to aid
crane usually operates from stern or bow. pile-driving, the jet pipes are mounted on
Obviously, both the crane used for pile the leaders next to the pile as indicated in
installation and the adversely loaded Figure 7-40.
crane- pontoon system must be sufficiently
stable during flotation and pile-driving. 7.5.1.4.3 Mandrel driving
Normally, the allowable crane load is es- technique
tablished as a fi.xed percentage of the load
that would tip the crane when working with Closed-ended cylinder piles and, particu-
a given boom length at a certain radius. In larly, thin-walled piles fabricated from reg-
North America, the allowabie load for ular reinforced or prestressed concrete are
crawler crane-mounted pile drivers is gen- usually driven via a steel mandrel (Fig.
Piled Waterfront Structures 761

( C)

(b )

Figure 7-36. Continued.

7.5.1.4.4 Floating template


7-38). The advantage ofthis method is that
heavy driving impact does not affect the Where the dock superstructure is erected
pile head. During driving, the mandrel ac- from precast concrete components, a high
tually pulls the pile into the ground. The degree of accuracy in pile insulation is re-
mandrel is extended when a new pile sec- quired. In some cases, the high degree of
tion is added to a previously installed pile accuracy is achieved by using floating tem-
length. At the final penetration, when the plates. An example is depicted in Figure
pile reaches the desired capacity, driving is 7-39. This type of template can be used for
stopped and the mandrel is withdrawn. If installation of large-diameter steel and con-
required by the design, the reinforcement crete cylinder piles. It consists of a steel
cage can be lowered inside the pile and the pontoon (5), four pile holding mechanisms
(2), and three pile guides (leads) (4), one at
shell filled with concrete. The mandrel driv-
the pontoon stern and two at the bow. The
ing technique also permits installation of
depth of guides is equal to about one-fourth
thin-walled (corrugated) steel shells; the
of the pile length. This helps to keep the
shell is filled with concrete following with-
pile firmly in the leads and to drive them
drawal of the mandrel.
accurately. The rear half of the lead (4) can
The advantage of the mandrel driving
be released and swung open, as shown in
technique is that it allows lightweight piles
Figure 7-39.
to be driven to high resistance. In certain Piles are installed in the following se-
situations where the pile head needs to be quence:
driven below the water level, a rigid steel
follower is used as an extension of the pile. 1. The template is firmly and accurately an-
762 Piled Waterfront Structures

(a) (b)

(c) (d)

Figure 7-37. Pontoon-mounted crawler crane with pile-driving leaders: (a)


plump; (b) out-batter; (c) in-batter; (d) side batter; (e) typical crane-mounted
pile-driving system: general view. 1-crane-based pile driver; 2-pontoon
(barge); 3-counterweight,

chored in place and the first seven This method of pile installation guarantees
piles-three on each side of template and the best accuracy possible. At deep water
one at the pontoon stern-are installed. sites, which are relatively :f:I.exible and
2. The holding mechanisms are released and therefore susceptible to effects of waves and
the template moves one bent ahead (e.g., currents, piles must be temporarily braced
from position "a" to position "b") and is before the superstructure is constructed.
hooked at four previously installed side
piles, as depicted in Figure 7-39. 7.5.1.4.5 Driving accessories
3. Three new piles-one at the pontoon stern
and two at the template sides-are in- As shown in Figure 7-35, the pile-
stalled; the whole operation is then re- driving system includes pile-driving acces-
peated after the template is moved one sories such as a pile cushion, a drive head,
bent ahead. and a hammer cushion capblock shield.
Piled Waterfront Structures 763

Figure 7-37. Continued.

The hammer cushion is 200-250 mm periphery in order to provide full contact


thick and is designed to protect the ham- with the pile. For example, a drive head
mer and the pile by reducing peak forces used for driving H-piles must have a mat-
while effectively transmitting the hammer ing H-pile shape which will fit snugly on
energy to the pile. The hammer cushion is pile top.
usually made from hardwood or laminated A pile cushion is 150 mm thick and is
timber materials; mechanical cushions of placed between the top of precast concrete
miscellaneous designs also have been used. piles and the drive head. It is usually made
A drive head is a steel cap installed on from layers of hardwood and softwood
the top of the pile that distributes the ham- boards or plywood. The purpose of the pile
mer blows uniformly to the pile. It must be cushion is to protect the pile top from dam-
fit snugly to the pile head around its entire age during driving and to assist in effec-
764 Piled Waterfront Structures

canceled out and movement occurs verti-


cally. The vibrator is usually suspended by
a crane and the vibration is transmitted
directly to the pile by clamping the unit to
3 the pile top. Pile vibration breaks down
(a) skin friction and tends to displace soil
particles away from the pile.
In the past 25-30 years, vibrators have
(b) been used often for pile installation. The
method is best suited for installing low-
displacement piles, such as open-ended
pipe piles and H-piles, into granular soils.
It is particularly advantageous in saturated
granular soils. However, it also has been
used efficiently for driving piles and sheet
piles in soft clays. The advantages of driv-
ing piles by vibrators include faster driving,
less noise, and less damage to the pile due
to lower impact driving energy. For exam-
ple, during construction of the bridge over
the Yangtze River at Hank.ow, China,
Russian-made heavy vibrators of 4.5-11.25
tonnes with 400-1000-rpm frequencies have
been used successfully for sinking 1.55-m-
diameter cylindrical concrete piles through
sand to the bedrock. These piles were sunk
with the aid of a water jet to a depth of
approximately 25 m; it took about 4 hours
to install one pile.
Figure 7-38. Cylinder pile-driving with mandrel: (a)
The disadvantages of pile installation by
driving through soft strata; (b) pile installed in bearing vibrators are the potential effects on adja-
stratum. 1-cylinder pile; 2-joint between pile sec- cent structures, which may include disrup-
tions; 3-pile-driving system; 4-mandrel; 5-bearing tion of the operation of sensitive equipment
stratum; 6-soft stratum. or machinery andjor consolidation of gran-
ular soils under existing foundations and
tively transmitting the hammer energy to behind retaining structures. The latter may
the pile and distributing the load unifonnly result in excessive settlement, structural
over the pile head. damages, and liquefaction of water-bearing
fine granular soils. Most of these problems
can be avoided by locating the new piles
7.5.1.4.6 Pile installation by
away from existing structures or selecting
vibrators
piles and installation techniques which
In this method, piles penetrate the avoid vibration.
ground through rapid vertical vibrations of Many different types of vibrator are
small amplitude, which are created by the presently available for pile-driving. The se-
rotations of eccentric weights mounted on lection of the vibratory pile driver depends
the vibrator. The rotation of the weights is on geological conditions and experience of
synchronized so that lateral vibrations are both the engineer and the piling contractor.
Piled Waterfront Structures 765

12,0 12,0

,;:t,·l

·=:lt
1

3 b 1 .
·-·-"':·.!
~
Figure 7-39. Floating templet for pile-driving: 1-cylinder pile; 2-pile holdingjguid-
ing mechanism; 3-pile; 4-guide; 5-pontoon.

A comprehensive list of impact and vibra- Driving of concrete piles through sand
tory pile drivers and their basic parameters and gravei deposits is often associated with
are given in Prakash and Sharma (1990). heavy resistance that sometimes results in
Despite their advantages for pile-driving damage to the piles. As an alternative to
in favorable soil conditions, the use of vi- pile-driving, water jetting may be adopted.
brators can lead to reduced pile capacity. For successful pile jetting, a sufficient sup-
Barkan (1957, 1962) demonstrated that the ply of water at an adequate pressure is
vertical vibration of a pile can markedly required. Jet pipes are usually placed sym-
reduce the pile skin friction (shearing re- metrically about a vertically driven pile (Fig.
sistance between the pile shaft and the sur- 7-40) and in nonsymmetrical fashion when
rounding soil). Rodger and Littlejohn (1980), used for installing batter piles (Fig. 7-31).
Bernard (1968), and other workers investi- Symmetrical arrangement of jet pipes for
gated many parameters that influence the driving vertical piles is required to keep the
vibratory pile-driving and their impact on pile on a vertical course. Hence, a minimum
pile-soil interaction. These included the vi- of two pipes is usually used.
brator power, peak acceleration, frequency, Jet pipes can be free hanging at the pile
pile cross section (contact perimeter), size of sides (Fig. 7 -40a) or guided by special
soil grain, angle of internal friction, mois- guides that are installed on the pipe shaft
ture content, and others. The most recent (Fig. 7-40b). In some cases, the jet pipe has
discussion on behavior of vibro-driven piles been installed centrally within the pile shaft
in sand was offered by Vipulanandan et al. (Figs. 7-40c and 7-41).
(1990), and soil-pile interaction in vibro- It is not advisable, however, to set the jet
piling is discussed by Billet and Sieffert pipe outlet in the center of the pile tip, as it
(1989). may become blocked by soil in that position;
if the jet pipe is set centrally in the pile
7.5.1.4.7 Pile jetting shaft, then the outlets should be projected
A detailed discussion on pile jetting is sideways, as indicated in Figure 7-41. In
given in Chapter 6. Several specific details ali cases where a water jet is used to aid
related to use of water jet for pile installa- pile installation, the piles should be driven
tion are discussed here. the last 1-1.5 m to their final set by an
Pile jetting can be used as a sole method impact hammer.
of pile installation or in combination with As pointed out in Chapter 6, the capacity
other methods, such as the impact hammer of driven piles could be much greater than
or vibrator. those installed by jetting with return flow
766 Piled Waterfront Structures

(a) (b) (c) (Tsinker, 1972, 1977, 1988; McClelland,


1974).
Pile jetting may not work in the presence
of large boulders and it may also be inef-
fective in thick soft clay stratum because
clay can blanket the return water so that it
cannot escape from the jets. Mouchel and
Partners (1957) have developed a system
intended for use where a clay stratum can
blanket an underlaying sand layer in which
jetting would be advantageous otherwise
(Fig. 7-41). While driving through the clay
stratum, the ends of the channels installed
at the pile sides are closed by steel scrolling
plates. When the jet pipes reach the level of
the sand stratum, a slotted steel wedge is
carried down by a 32-mm-diameter steel
Figure 7-40. Pilejetting. Conventional arrangement bolt and forces the scaling plates open, thus
of jet pipes: 1-pile-driving system; 2-fiexible high-
pressure hose; 3-jet pipe; 4-pile; 5-nozzle; 6-8-jet
forming an escape channel for the return
pipe guiding accessories. water.

~ 11 M.S. plate welded


lo Y." dia. bar. '""dia. M.S. bars
welded lo channels

s"x 2Y:,"" to·22 lb.


f !4" dia. M.S. bolt

~" M.S. cover plate

r---- f6 11 ---1
li;;;;::;;:::=;:;;;:i
1
ta"

i dia. bers
'/. 11
welded to
channel web
l .."J"~.~..-~-,.~·; .:;._;bed~-..~-.~- -..~k. ,
TYriCAL CAOSS SECUOH OF rtiL

Figure 7-41. Jetting arrangement with path for return water. [From Mouchel and
Partners (1957).]
Piled Waterfront Structures 767

7.5.1.4.8 Ground heaving The technical requirements section out-


lines requirements of materials to be used
When large displacement piles are driven
for pile fabrication, pile handling and stor-
in groups into dense cohesive soils they may
ing conditions, pile-driving criteria, instal-
cause ground heaving. This phenomenon
lation tolerances, and other relevant techni-
may cause previously driven piles to lift off
cal requirements. The materials section
their bearing stratum, thus eliminating or
governs size and quality of materials and
reducing their point resistance; to restore
procedures for checking the quality of mate-
it, lifted piles must be redriven at consider-
rials supplied. The fabrication section out-
able expense. To avoid ground heaving, it is
lines requirements applicable to welding of
advisable to switch to smaller displacement
pile splices in steel piles, placement of rein-
piles, such as open-ended piles or H-piles.
forcing steel and concrete in precast piles,
To avoid this from happening in hard clays,
and other relevant requirements aimed at
the soil can be preaugered prior to pile
obtaining good quality piles. This section
installation. Selecting a wider pile spacing
also should include the quality control re-
or using the proper pile-driving sequence
quirements.
(e.g., working outward from the center of a
The piles handling and storage section
pile group) can also effectively limit ground
stipulates conditions which will prevent
heaving.
damage to piles after they arrive on site
and prior to their installation. For precast
7.5.1.5 Installation Specification concrete piles and for long prestressed piles,
Good pile installation practice begins specifically, the location ofthe pickup points
with a good specification which explains and must be specified.
controls ali aspects of pile installation. The Pile~driving criteria is the most impor-
specification represents the written transla- tant aspect of the piling specification. It
tion of the designer's intention that the stipulates pile penetration resistance which
piling contractor and the construction must be achieved using the energy of the
inspection should follow. A well-written specified hammer. Alternatively, when piles
specification can help the inspector carry are installed by vibrator or water jet, only
out his responsibility if it is clearly defined the required penetration, which includes the
in the specification; a carefully executed de- pile final setting achieved by use of an im-
sign can be destroyed by a poorly written or pact hammer, is stipulated.
ambiguous specification. Pile installation tolerances stipulate al-
A typical pile installation specification lowed pile deviation from the design posi-
usually incorporates the following items: tion, plumpness or batter, cutoff elevation,
and amount of pile curvature. Piles exceed-
(a) Preamble ing the curvature tolerance should be re-
jected upon arrival on site.
(b) Technical requirements
The pile locations and their compliance
(c) Recording and payments with project requirements must be checked
and confirmed by the surveyor and inspec-
The preamble usually includes general tion staff in time to allow any necessary
project information such as a brief descrip- changes to be made without causing job
tion of a project, and it outlines the scope of delays and certainly before commencing
work by the contractor and by others. It construction of the superstructure. The
also defines the extent of the contractor's specification must also include the condi-
responsibility regarding the supply of labor, tions, number, and method of pile load
equipment, and materials. te sting.
768 Piled Waterfront Structures

7.5.1.6 Pile Load Test detail the arrangement and procedure of


the pile test.
The pile load test is the surest way to In a standard pile test method, the pile is
determine the ultimate design loads which loaded in eight equal increments up to a
can prove that the pile, when installed ac- maximum load, usually twice as much as
cording to the specification, will not fail due the design allowable load. Each load level is
to excessive settlement when loaded. It is maintained until the rate of settlement has
usually carried out to confirm results of decreased to 0.25 mmjh, but not longer
theoretical pile analysis, to evaluate pro- than 2 h. The finalload equals 200% of the
posed techniques of pile installation, and design load and is maintained for a dura-
to check the competence of the piling con- tion of 24 h. After the required holding
tractor. Results of pile test are used to re- time the load is removed in decrements of
fine and finalize the design. 25%' with 1 h between decrements. After
Sometimes, pile load tests are conducted the l~ad has been applied and removed, the
on existing structures to verify their real pile is reloaded to the test load in incre~
capacity and to investigate their ability to ments of 50% of the design load, allowing
carry an increased live load. 20 min between load increments. Finally,
The decision on the number and location the load is increased in increments of 10%
of piles to be tested is usually based on the of design load until the failure load is
variability ofthe site soil conditions; at least achieved, allowing 20 min between load in-
one test should be carried out for each vari- crements. This method of pile test is gener-
ety of soil condition. Piles are usually tested ally used for site investigation prior to in-
by statically applied loads, or dynamically. stallation of contract piles. Results of the
These two methods are discussed in the pile test are used for writing specifications.
following sections. This method is time-consuming, requiring
from 30 to 70 h (and more) to complete.
7.5.1.6.1 static axiat pile test A faster, and less expensive, test is the
so-called "Quick Test" stipulated by ASTM
Besides standard pile testing procedures, 1143 as an optional method. In this load
as outlined in the ASTM Standards, there test, the maximum load is normally kept on
are additional pile testing methods that the pile not longer than any other preceding
have been reported in literature. Examples load. The Quick Test is carried out in the
are found in Butler and Hoy (1977), following steps.
Fellenius (1975, 1980), Mohan et al. (1967),
New York State DOT (1974), Swedish Pile 1. The pile is loaded in 20 increments to
Commission (1970), Weele (1957), Witaker 300% of the design load (i.e., each incre-
(1957, 1963), and Witaker and Cooke (1961). ment is equal to 15% of the design load).
Chellis (1961) lists 17 methods employed Each load is maintained for a period of 5
by various individuals, organizations, and min, with readings taken every 2.5 min.
building codes for the determination of uiti- 2. The load is added in increments until
continuous jacking is reqliired to main-
roate or working loads of piles under com-
tain the test load, or the maximum test
pressive loads. load has been reached.
In N orth America, pile te sting is usually 3. The pile is unloaded in about five steps
performed in compliance with the latest is- with a duration of no greater than about
sue of ASTM Standards D-1143, D-3689, 1 min.
and D-3966 for static testing of a single pile
in axial compression, axial uplift, and lat- U sually the Quick Test is completed within
eral loading, respectively. These standards 3-6 h. Fellenius (1991) suggested that from
Piled W aterfront Structures 769

a technical, practica!, and economica! point secondary reaction beam as is shown in


of view this method is superior to the Figure 7-42b.
standard pile testing procedure. Backup load measurements are also
Another method recommended by ASTM made by a manometer or gauge calibrated
D1143 as option is the 'Constant Rate Pene- together with the jack. These measure-
tration Test.' This method is also recom- ments, however, sometimes can produce
mended by the Swedish Pile Commission substantially overestimated result
and the New York State Department of (Fellenius, 1984). The reason for the errors
Transportation. During this test, the pile are that the laboratory-calibrated jacking
head is force to settle at a rate of 1.25 system when exposed to field conditions
mmjmin to a total penetration of 50-75 (e.g., eccentric loading, bending moments,
mm. The force required to achieve the pene- or temperature variations) reacts differ-
tration rate is recorded. The main advan- ently, mainly because of effects of ram fric-
tage of this method is its speed and econ- tion occurring along the sides of the ram
omy. It may be of particular value for fric- (primarily at the location of the seal). The
tion piles, both in compression and uplift, friction may result in errors as large as
but may not be practica! for end-bearing 30-40% according to Fellenius. This is why
piles because of the high force requirements many engineers usuaiiy prefer to use load
cells for measurements of the test load and
to cause penetration through the hardbear-
to calibrate the jack-load cell system as a
ing stratum. For a detailed discussion on
unit. A complete discussion on instruments
Constant Rate Penetration Test, the reader
that are used for pile load testing is given
is referred to Feiienius (1980).
in Dunnicliff (1988).
It should be pointed out that there are no
Davisson (1970, 1989) suggests that most
conceptual differences in methodology be-
of the adverse effects associated with the
tween the axial compression and puii-out
jack system can be completely eliminated
pile tests. The only difference is that in the by use of special spherical bearings placed
latter case the test load is applied in the between the ram and the load cell (Fig.
upward direction on the test pile. Details on 7-43). Naturally, the pile loading system
the puii-out test are provided in ASTM must be designed for the anticipated maxi-
D3689. mum load; a conservative design of a reac-
tion beam is usually recommended.
Load App/ication Arrangement and As an additional note, the hydraulic jack
tnstruments should have sufficient travel to provide for
anticipated pile settlements, deflection of
A typical setup for pile testing consists of the reaction beam, and elongation of con-
a minimum of two anchor piles located on necting anchoring details.
either side of a test pile, a reaction beam Sometimes, the actual field conditions
attached to the top of each anchor pile, or dictate the use of modified pile jacking sys-
equivalent arrangement, hydraulic jack tems; for example, weight platform, loading
placed as required for pile compression or frame, or the reaction beam of a more com-
puii-out test, and instruments for measur- plicated design than illustrated in Figure
ing loads and pile movements (Figs. 7-42 7.42. Three relevant examples are illus-
and 7-43). In the case of an axial compres- trated in Figure 7-44.
sion test, loads applied by an hydraulic jack The arrangemenf illustrated in Figure
are measured either by a calibrated load 7-44a was used for a pile load test during
cell placed between the jack and the reac- construction of a shiplift facility at Seward,
tion beam, or between the jack and the Alaska. In this particular case, because of
770 Piled Waterfront Structures

"' .. (a)

" F"

1 1
r6 r-5
t--4
J
~ i
~t
~

1--2 1-
=
*=i ,,
==j~

j
i
'
l
1
(b)

~7
<;;

s--=: ?
4--1

1 1

J t-6

1 1
~ -
2-~
~
-1
r--
i
1
1
i
1

Figure 7-42. Typical setup for pile testing: (a) application of axial
compression load; (b) application of puii-out load. 1-Test pile; 2-anchor
pile; 3-reaction beam; 4-hydraulic jack; 5-load cell; 6-anchor strip
(bar); 7-secondary reaction beam.

the specifics of the existing pile foundation out on a pile included in the seven-pile
and the relatively shallow pile penetration, mooring dolphin. For this test a steel cap-
seven anchor piles were employed to test ping plate was cut at its central part, thus
the 406 mm-diameter open-ended concrete separating the centrally located, 762-mm-
filled steel pipe pile. For this test a double diameter vertical steel pipe pile from the
reaction beam system with a secondary dolphin's pile system. A special reaction
cross-beam was employed. The maximum frame was prefabricated and welded to the
test load reached approximately 340 tonnes. dolphin's capping plate. In this test the pile
The purpose of the pile load test illus- was loaded by four hydraulic jacks to a load
trated in Figure 7-44b was to obtain infor- up to approximately 700 tonnes.
mation on the real capacity of open-ended In the example depicted in Figure 7-44c,
steel piles included in existing mooring and the pile load test was performed 2000 m
breasting dolphins and in a coal-loader sup- offshore on a 1524-mm-diameter steel pipe
porting platform that are part of the coal- pile included in a mooring dolphin at
loading terminal constructed at Seward, Hadera coal unloading terminal, Israel. In
Alaska. The pile load test has been carried this test the pile load from several hy-
Piled Waterfront Structures 771

nated system. The plotted data are used to


estimate the failure load so that an allow-
able pile capacity can be estimated. By def-
inition, the failure load applied on the pile
is reached when rapid movement occurs un-
der sustained or slightly increased load.
However, this definition may be inadequate
because pile plunging may require large
Hydraulic pressure
movements that may exceed the acceptable
range of the pile-soil system.
Dial stern In general, the definition of pile failure is
reaction very much judgmental and depends on an
plate
individual's personal experience. Many en-
gineers define the failure load at the point
Wire
of intersection of the initial tangent to the
load-movement curve and the tangent to, or
the extension of, the final portion of the
curve. Other failure definitions consider ar-
Figure 7-43. Typical instrumentation for pile test- bitrary settlement limits; for example,
ing. [From Davisson (1970).] sometîmes the pile is considered to have
failed when its head has moved 10% of the
pile end diameter, or the gross settlement
draulic jacks was applied against a weight of 40 mm and net settlement of 20 mm
platform loaded with six rectangular con- occurs twice when reloading to the design
crete blocks. The pile failed at a test load load.
equal to 1100 tonnes. In theory, in order to be independent of
A typical setup of pile test measuring individual interpretation, the definition of
instruments is depicted in Figure 7-43. As failure load should be based on some math-
mentioned earlier, the pile load is measured ematical rule resulting in a repeatable
by either a load cell, a manometer, or both. value. The various interpretation methods
Movements of the pile butt, both axial are presented in detail and discussed by
and lateral, are typically measured by a Prakash and Sharma (1990). Among these
dial gauge system, the wire, mirror, and
are methods proposed by Davisson (1972),
scale arrangement, and the conventional
Chin (1970, 1971), De Beer and Walleys
surveyor's level system.
(1972), Hansen (1963), Mazurkiewicz
Furthermore, a telltale andj or an elec-
tric strain gauge measuring system are in- (1972), Fuller and Hoy (1970), Buttler and
Hoy (1977), and Van der Veen (1953).
stalled for incremental strainjstress meas-
urements along the pile length. For further Fellenius (1980) carried out failure load
details on pile movement measuring sys- interpretation on load-movement curves ob-
tems, the reader is referred to Dunnicliff tained from a static load test on a 40-m
(1988). long, 300-mm-diameter pile in clay and silt
(Fig. 7 -45). All of the above interpretation
methods have been considered. The results
Jnterpretation of Results of Pile obtained by Fellenius varied widely from a
Axial Load Test conservative 240 tons [Davisson's (1972)
The results from a pile load test are usu- method] to 330 tons [Chin's (1970, 1971)
ally plotted in a load-pile movement coordi- method]. The latter is invariably higher
772 Piled Waterfront Structures

(n )

(b)

Figure 7-44. Pile test installation. Practica! examples: (a) double


reaction beam system for pile test used at shiplift facility, Seward,
Alaska; (b) space reaction frame used for pile test at coal-loading
terminal, Seward , Alaska; (c) weight platform used for pile test at
Hadera, Israel, offshore coal unloading terminal. (Courtesy of Yaron,
Shimoni and Partners, Consulting Engineers Ltd., Tel Aviv, Israel.)
Piled Waterfront Structures 773

(c)

Figure 7-44. Continued.

than the actual test failure load. The other piles ranged from 12 to 32. Load-movement
seven values were grouped around 270 tons curves for ali test piles indicated that the
indicating that in the case in question the failure load was achieved. AII of the previ-
Fuller and Hoy, DeBeer, and Buttler and ously mentioned nine failure load interpre-
Hoy methods appeared to yield reasonable tation methods were used. The latter inves-
estimates of the failure load. tigators reached the following conclusiOJJ.S.
Joshi and Sharma (1987) carried out fail-
ure load interpretations on five different 1. For concrete bored piles and pile with
load-movement curves obtained using the expanded bases (e.g., belled end), the
standard pile load test method. In these Fuller and Hoy (1970) method provides a
tests, the length-to-diameter ratio of test reasonable estimate of the failure load.
774 Piled Waterfront Structures

site geological condition and pile loading,

300
-:::=··
]l *~·~
-+- 290 90 run1
the factor of safety can be reduced to 1.5-2.0.

1
J -t-~M -HOYCZ 7.5.1.6.2 Lateral pile Joad test
-+- 260 DE BEER.
1 ...,._ lSO BUTLER-HOY
1 -+- 240 DAVISSON This test is usually carried out to confirm
1
1 the design parameters used in piled struc-
1
1 ture design where the individual piles are

2
1 loaded axially and by the lateralloads, and
the whole structure performs as a ground
fixed frame. This test is particularly impor-

7
tant when the piles are installed in clay. As
pointed out by Sastry and Meyerhof (1987),
1
1
the pile installation method can have a dra-
1
1
matic effect on the behavior of rigid piles in
1
0 ~o-L~,~.oo~-.2.koo-L~3~.oo~~4.koo-L~s.oo
clay subjected to moments and horizontal
-l l-o.2S loads. Although the net lateral soil pressure
Movement, inches at pile failure along the pile shaft is unaf-
fected by the method of pile installation, the
Figure 7-45. Load-movement diagram from a quick pile displacement does. For example, the
maintained-load static loading test with measurement
of pile compression and toe movement. [From Fellenius
displacement of bored piles are, in general,
(1980).] 1.5-3 times those of driven piles.

2. For concrete expanded-base compacted rest procedure


(Franki) piles, the Davisson, Buttler and In the standard pile load test procedure,
Hoy, and Fuller and Hoy methods provide
the total test load is applied in 10 steps to
reasonable estimates for failure load.
200% of the design load (e.g., 25%, 50%,
3. For driven steel H-piles, Hansen's 90%
criterion and Fuller and Hoy's method 75%, 100%, 125%, 150%, 170%, 180%,
predicted a failure load similar to the 190%, and 200%). The 25% and 50% of the
actual failure load. design load increments are applied for 10
min each, and the 75% load increment is
Detailed discussion on all above nine meth- maintained for 15 min. AII other loads are
ods is found in Prakash and Sharma (1990). maintained for 20 min each. The 200% load
is maintained for 1 h, after which the pile is
Factor of safetv unloaded in steps of 50% of the design load
The working load on a pile is defined as (e.g., 150%, 100%, 50%, and 0%), maintain-
the ratio of pile ultimate capacity to the ing each decrement for 10 min.
appropriate factor of safety. Generally, the. Each test is project-site-specific. On some
safety factor for a single pile, 2.5 is consid- projects, the load testing procedure may in-
ered as appropriate. However, where there clude cyclic, surge, reverse, or other loading
is a sufficient number of pile loading tests, to suit the specific project requirements.
or where a large body of load experience is These and other loading procedures are de-
available, the safety factor of 2.0 is nor- tailed in ASTM D3966-81 (American Soci-
mally considered. ety for Testing and Materials, 1989).
In cases where the soil characteristics
are uncertain or large impact or vibratory
Jnstruments
loads are expected, the safety factor of up to The test load is normally generated by a
3.0 and more can be considered. standard hydraulic jack. As to the axial pile
For temporary structures, depending on load test, the load cell is used for measure-
Piled Waterfront Structures 775

ments of the applied load. The jack is in- It must be pointed out that pile lateral
stalled horizontally between the test pile stability may be greatly influenced by sta-
and a suitable reaction system. The latter bility of the underdeck slope into which the
should be a relatively stiff system (e.g., pile is embedded, and vice versa. Therefore,
heavy gravity block, vertical batter pile sys- the effects of lateral loads imposed by piles
tem, and others), which will not yield ap- against the sloped bank or underdeck riprap
preciably under the test load. Davisson material should be considered in the design
(1970) recommends use ofspherical bearing of a piled marine structure.
in lateral load tests because it inherently One of the main purposes of the lateral
involves pile rotation. pile load test is to obtain a real value of soil
Instruments for measuring pile lateral reaction per pile unit length for subsequent
movements, associated bending moments use in the design of the structure. The de-
and the bending moment-induced stresses sign could be carried out either manually or
are commonly the same as those used for by using available computer programs
pile axial tests (e.g., reference-beam- [e.g., PILEDG by Geosoft (1984), LPILE
mounted dial gauges, wire-mirror and scale by Ensolft (1989), or others] which are
system, and others. Inclinometers installed typically based on a conventional beam on
in the pile's interior are often used to deter- elastic foundation theory.
mine pile deflection below the ground sur-
face. An example is provided in Hunter et 7.5.1.6.3 ovnamic pile Joad test
al. (1992).
In the case of solid concrete test piles, a Where the piled structure is exposed to
steel tube or a duct could be installed inside dynamic loading, the pile response is of a
or outside the pile to accommodate an incli- paramount importance. Dynamic loads may
nometer to measure the pile's lateral shift have significant effects on soil-pile bound-
along it embedded part. In this case, stiff- ary conditions, thus affecting the pile's dy-
ness of the test pile must .be compatible namic response. This basically includes the
with that of the production pile used in the response to vertical vibration, horizontal
real structure. Other means for the investi- displacements, and, in some cases, torsional
gation of pile deflection are also available vibration. Normally, the dynamic pile test
(Tsinker, 1972). Because of substantial dis- is performed to study all of the above
placements of test pile, the dial gauges effects.
should have at least 75-100 mm travel. It should be pointed out that at the pre-
Rotation of the pile head can be meas- sent time in North America, there is no
ured by surveying the position of the pile standard procedure for testing piles by dy-
before testing and during each phase of namic methods. Where required, it is done
load application; the difference between two according to the customized procedure
readings is a measure of the change in the adopted by the individual investigator.
pile head's slope. Several full-scale dynamic pile load tests
have been reported to date (Prakash and
Jnterpretation of test data Sharma, 1969; Gle, 1981; Gle and Woods,
1984; Prakash et al., 1985; Blaney et al.,
The generally accepted criterion for esti- 1987). In each of these tests, piles were
mating the ultimate (failure) lateralload on tested by different methods.
a test. pile is to consider the point of inter-
section of tangents on the load-movement Equipment and Jnstruments
curve. When this load is effectively reached,
the test pile usually fails to rebound after A steady-state sinusoidal force for the
removal of the load. pile test can be generated by conventional
776 Piled Waterfront Structures

mechanical oscillators applying a variable As suggested by Gle and Woods (1984),


alternating force in both the vertical or hor- the dynamic response of the soil-pile-mass
izontal planes. The magnitude of this force system can be represented correctly by
is controlled by adjusting the phase angle equations of motion written for each of the
between the unbalanced masses in the os- translation and rotation degrees of freedom
cillator. The speed of the oscillator is con- and solved simultaneously for the dynamic
trolled by a variable-speed electric motor response. For details, the reader is referred
connected to the oscillator through a flexi- to work by these 1nvestigators.
ble shaft. The motor can usually generate a
variable force within a frequency range 7.5.1.6.4 Pile integritv test
of approximately 5-55 Hz.
This test is performed on piles to confirm
In experiments by Gle and Woods (1984) their integrity after installation. Low-strain
the pile was excited with a Lazon oscillator integrity testing procedure is usually used
with ·added mass, created by steel plates on ali types of piles, but in particular on
tightened together by threading steal rods, concrete piles installed by driving methods,
which added vertical load to the pile, thus or on bored piles. In this test, a compressive
reducing its natural frequency within the impact wave is generated by impacting the
range of operation of the oscillator. Pile dis- pile head, and the resulting acceleration
placements were measure<;l with two veloc- and velocity of the impact waves are stud-
ity transducers mounted on each side of the ied. It is performed using a high-sensitivity
added mass. This provided an independent accelerometer placed on the pile head, an
check on the displacement amplitude. The amplifier-receiver, and a portable computer.
output signals from these transducers were In testing the pile, a slight blow is delivered
recorded on a dual-channel, strip-chart to the pile head by means of a hand-held
recorder. hammer. This initiates a small strain wave
that travels down the pile at the speed of
Test Procedure and lnterpretation sound. When the strain wave reaches a
crack or void in the pile, a reflection in the
As pointed out by Gle and Woods (1984), form of a tensile wave is sent back up to the
for the determination of the lateral dynamic pile head. This is picked up by the signal
response of the mas caused by soil-pile in- amplifier and sent to a computer for pro-
teraction, the mass added to the oscillator cessing.
should be located as close to the ground
surface as possible, but not in contact with
the soil. 7.5.2 superstructure
Essentially, as the distance between the
ground surface and the added mass is in- Structurally, the superstructure performs
creased, the dynamic response can be mod- as the top horizontal element of an essen-
eled as a structural response of a mass on a tially rigid frame. This implies that the
cantilever. However, this results in less in- superstructure must have a fixed-end
formation being obtained for the soil-pile connection with the piles to provide for good
interaction. With this in mind, usually five interaction and load distribution.
to seven tests are conducted on the same In most practica} cases, the superstruc-
pile. From the measured free-vibration re- ture is designed for continuity and is con-
sponse, damping and the damped natural structed either as cast-in-place structure,
frequency of the soil-pile-mass system are assembled from precase concrete compo-
determined. nents, or as a combination of both. The
Piled Waterfront Structures 777

former is usually used for the construction with attached formwork, are installed on
of relieving platforms and in regions of high the pile capping beams and filled with
seismic activity; the latter is generally used concrete.
for the construction of deck of the open pile Essentially, each truss that is included
structures. in the reinforcing cage is designed for two
purposes: In the initial construction stage,
7.5.2.1 Deck Structure in the it is designed to carry loads produced by the
Open Pile construction weight of fresh concrete, formwork, con-
struction equipment, and other relevant
Deck structures are designed to maintain loads; in the final stage, the truss is de-
the minimum clearance between the under- signed as a part of the reinforced concrete
side of the deck and the expected maximum deck structure to carry all designed dead
(extreme) wave crest height, or maximum and live loads. The advantage of this method
tide level during the design life of the wharf. is the reduction of time needed for over-
Failure to provide a "safe" clearance may water construction.
result in the application of uplift loads to Alternatively, the superstructure can be
the deck structure and, more importantly, designed as a two-way slab incorporating
in damage to the upper part of the under- the concrete pile caps, edge beams, curbs,
deck slope protection. This phenomenon is crane and train track beams, and other
discussed further in Section 7.5.3. structural components. The concrete pile
The simplest, although not necessarily, capping beams must be designed to accom-
the most economica! scheme of deck con- modate pile-driving tolerances, yet assure
struction is a cast-in-place fl.at slab of uni- proper structural behavior by providing the
form thickness placed on piles. This type of appropriate fixed-end connected for both
construction minimizes and simplifies the tension and bearing piles.
required formwork. Furthermore, this type Sometimes, when piles deviate slightly
of construction can accommodate the most from specified installation tolerance they
adverse combination of misplaced pile. are forcibly brought back into alignment as
In a cast-in-place deck structure, the con- illustrated in Figure 7-46. This action,
ventional formwork can be replaced by however, can result in excessive bending
shallow precast planks that are eventually stresses in these piles. The effect of these
incorporated into the deck structures as stresses on the overall pile performance
load-carrying members. These planks typi- must be carefully evaluated. In some cases,
cally span longitudinally between trans- performance of such overstressed piles can
verse pile capping beams and are usually be enhanced by filling them with concrete.
designed to carry the weight of fresh con- In general, however, when precast concrete
crete and miscellaneous construction loads. pile capping beams are used, it is a good
The advantage of this type of construction practice to make them large enough so as to
is that it significantly enhances the quality enable them to accommodate the most ad-
of the deck underside that is exposed to the verse deviation of piles from their design
most severe effects of the marine environ- locations.
ment. It also, in general, accelerates the The selection of the structural type of the
construction process. capping beam is site-specific; as well as
Occasionally the cast-in-place deck struc- project-specific. It usually depends on the
tures are constructed by using trusslike re- number of piles in the pile bend and on the
inforcement blocks prefabricated from availability of a fl.oating or mobile heavy lift
structural steel components. These reinforc- equipment. For example, in the case of the
ing cages, consisting of space truss sections construction of the Navy Pier, discussed by
778 Piled Waterfront Structures

(a)
the precast portion is followed by cast in
place of a central part of the capping beam.
Shear keys have been provided in both pre-
cast sections to ensure the good connection
with the cast-in-place portion. Reinforcing
cages installed into the two middle piles
were extended above the beam's underside,
which ensured a reliable joint between these
piles and the cast-in-place portion of the
capping beam (Fig. 7-47d). After the rein-
forced steel was installed, the concrete was
poured and cured. This was follows by in-
stallation ofthe precast concrete planks and
(b ) the principal deck reinforcement (Fig.
7-47e), after which the deck concrete was
installed and cured. Finally, the pier was
completed by installations of precast com-
ponent of the utility trenches (Fig. 7-470.
This case history serves as a good exam-
ple of a well-balanced use of both prefrabri-
cated components and cast-in-situ concrete
for composite deck construction.
Normally, the pile capping beams are
connected to each other by longitudinal
beams which serve as the edge beams. These
beams sometimes provide support for the
Figure 7-46. Pile capping beams formwork: (a) Steel
piles in bents are brought in to a proper alignment for
crane track or accommodate service
installation of pile capping formwork; (b) the pile is trenches. They are also used for mounting
pulled into alignment with other piles (heavy flexural of fenders; in the case of marginal wharves,
stresses may result). the land-side longitudinal edge beam may
be used as the fill retaining bulkhead also.
Buslov et al. (1988) and Damaschi and The deck structure composed of precast
Bonasia (1989), the capping beams were of components is usually designed to perform
a composite construction: The edge portion as a solid unit. As pointed out earlier, the
of these beams on both sides of the pier, system, comprised of precast elements, usu-
which incorporated utility trenches, were of ally consists of precast components sup-
precast construction, whereas the middle ported on either cast-in-place or precast pile
part was cast in place (Fig. 7-47). The pre- caps. It is often topped with cast-in-place
cast components at the aforementioned pier concrete.
are connected to the concrete cylindrical Precast units are normally joined to-
piles via a reinforced concrete plug installed gether by splicing negative reinforcement
within the cylinder pile and are extended mechanically (Fig. 7- 470 or by field weld-
into t he opening left in the pile capping ing through a splice angle (Fig. 7-48). Then
beam (Fig. 7- 47a). the concrete is placed in between the pre-
The plug reinforcing cage is seen in Fig- cast members. This increases the effective
ure 7-47c. The process of installation of the depth of the capping beam and produces a
pile capping beam is depicted in Figure strong monolithic joint between all prefab-
7-47b. Installation and fixing in place of ricated components and piles (Fig. 7-48).
Piled Waterfront Structures 779

(c)

( b)
(d )

Figure 7-47. Construction of N avy pier deck of composite design. 1- Cylindrical concrete
piles; 2-precast pile capping beam; 3-cast-in-place portion of pile capping beam; 4-concrete
joint between pile and capping beam; 5- precast deck planks; 6-precast fender panel;
7- utility trench; 8- cast-in-situ concrete deck; 9- reinforcing cage for joint between piles and
precast capping beam; 10-reinforcing cage for joint between piles and cast-in-situ part of
capping beam; 11-mechanical joint of rebars of precast component of utility trenches. (a)
Typical cross section; (b) installation of precast pile capping beam (2) on piles (1); (c) reinforcing
cage (9) installed in both pile and precast capping beam; (d) reinforcing cages (10) installed into
piles and extended above pile heads to provide solid connection between piles and cast-in-situ
part of the capping beam (3); (e) precast deck planks (5) installed on completed capping beams
and reinforcement (12) of cast-in-situ concrete deck is installed; (0 precast components of utility
trenches (7) are installed and dowels extended from these componenta are joined with each other
by couplers (11). (Photographs are courtesy of Dr. V. Buslov, at the time a principal designer to
Mueser Rutledge Consulting Engineers which together with Frederic R. Harris, both of New
York City, designed the pier. Contractor Yonkers Contracting, N .Y./Weeks Stevedoring, N.J.)

Birdy et al. (1989) suggest that in cases The development and use of large mate-
where a cast-in-place deck slab supported rial handling equipment in port operation,
on prestressed concrete piles is used, con- such as fork lifts and log loaders in capaci-
ventional dowel connections at the pile top ties of up to 40-50 tonnes and mobile
can be replaced with flared strands without truck-mounted cranes in capacities of 100
compromising the pile- deck fixity charac- tonnes or more with outrigger loading of up
teristics. to 150 tonnes per outrigger, dictate that
780 Piled Waterfront Structures

(e)

TYPICAL WHARF ANO


ACCESS PIER SECTION

(t)

PAECAST PILE CN>


DETAIL 1

SI'I.ICE CONNECTION
DETAIL 2

Figure 7-48. Joint between precast deck units and


pile capping beam. [From Sundberg and Stubbs,
(1992).)

Figure 7-47. Continued.


Hence, it can transmit vertical shear; how-
deck structures must be strong enough to ever, it cannot transmit transverse mo-
support heavy concentrated loads. ments across the joint. Essentially, at pile
Precast elements used for deck construc- bents, rigidity of pile caps prevents any
tion are usually designed in the form of separation of the panels at the shear key.
one-way fl.at or hunched concrete panels, Typically, precast panels are made fully
T-beams or double-tee beams, two-way fl.at continuous over the pile caps by postten-
slabs, or others. In most cases, the deck sioning or, as pointed out earlier, by joining
components are prestressed, and, in some the negative reinforcement projected from
instances, they are posttensioned after in- adjacent panels. This reinforcement, if re-
stallation. quired, can be prestressed by placing the
Precast deck components are usually hydraulic jacks between panels; when the
provided with longitudinal joints that are required tension in the reinforcing bars is
typically made in the form of large grouted achieved, the wedges between panels are
shear keys (Fig. 7- 71). The key helps to installed, jacks are removed, and concrete
eliminate differential movements of adja- is placed as discussed earlier.
cent panels by creating a sort of a large The results of research carried out at
structural diaphragm. However, there is no Lehigh University (Roesli, 1955; Walther,
reinforcing steel across the shear key. 1956) indicate that the presence of longitu-
Piled Waterfront Structures 781

dinal joints does not impair the deck's over- ter. This dictates that the deck structure
all torsional stiffness, which is quite effec- should be designed in a way to prevent the
tive in distributing the concentrated .loads formation of cracks of unacceptable width.
transversely, and the lateral load distribu- Sufficient cover over reinforcing steel, as
tion is very similar to that in a monolithic discussed in Chapter 2, must always be
slab. It must be pointed out, however, that provided. This is especially important for
where the concrete topping over precast structures designed for long-term service.
panels is nominal, say 50 mm or less, the The quality of concrete used in deck con-
heavy point load placed directly over the struction is also of paramount importance.
shear key can produce some stress concen- Concrete mix, placement, densification, and
tration in the slab located under this load. curing must be carefully addressed in the
Typically, precast deck components are concrete specification. Congested reinforce-
designed to work in composite action with a ment should be avoided where possible, and
poured-in-place top slab, thus permitting where chloride concentration can be re-
the attainment of continuity in both longi- garded as severe, then the use of coated
tudinal and transverse directions.
reinforcement, concrete surface coatings, or
Sometimes, two-way precast concrete
the use of membranes, use of waterproofing
slabs are used for deck construction. The
admixtures, polymer in prejunction, and
relevant example is depicted in Figure
cathodic protection should be considered.
7 -12b. This type of construction has the
In good design, all structural components
advantage of being flexible and simple. It
should be readily accessible for inspection
eliminates the need for pile capping beams.
and maintenance. From an operational
The capping beams are replaced by individ-
ual pile caps that, in final analysis, allows viewpoint, steel chafe plates, or rubbing
for greater tolerances in the driving of piles. strips, should always be provided where
This is an important consideration when mooring lines pass over any part of a struc-
piles are driven without special templates. ture andj or structural elements are sub-
In such cases, pile misalignment is not un- jected to relative movements. These compo-
common, especially when driving heavy long nents should also be designed with extra
piles. material to allow for wear.
Naturally, the use of bigger precast ele- Last but not least, in the overall design
ments is better, because it helps reduce the concept, details like positive drainage to get
number of precast deck components, as well the stormwater water off the deck quickly
as the number of joints between these com- through gravity flow, proper design ofjoints
ponents. between adjacent deck sections, and others
The pier deck is normally designed simi- must be given proper attention.
larly to a bridge deck structure, and in
accordance with the standard bridge speci- 7.5.2.2 Relieving Platform
fications (e.g., AASHTO, CSA Standard S6,
British Standard Institute (1979) and oth- In the past, most of the relieving plat-
ers), with some additional requirements as forms were constructed in the form of a
applied to offshore structures and stipu- cast-in-place structure of miscellaneous de-
lated in ACI 318-89/ACI 318R-89, ACI sign. Sometimes, however, they are con-
357R-93, API RP 2A, and other relevant structed as a combination of cast-in-place
standards. techniques with the inclusion of precast
As stated earlier, the major concern for concrete components.
marine structures is their vulnerability to Typically, the approach to relieving plat-
corrosion due to their exposure to salt wa- form design is site-specific and is greatly
782 Piled Waterfront Structures

influenced by the assumed sequence of the tween piles and the sheet piles included in
construction procedure. the pile foundation. In the past, very stiff
In the case of a "dredged" type of con- relieving platforms, with little reinforce-
struction, the backfill is placed behind the ment, were used. These structures proved
sheeting and the relieving platform is con- to be easy to build and durable. They also
structed thereafter. Naturally, in the case require little maintenance. These struc-
of "backfilled" construction, the sheet-piling tures, however, were very heavy and, there-
must be properly supported to sustain the fore, produced heavy loads on piles. In mod-
lateral pressure from the backfill soil (Fig. ern marine engineering practice, "flexible"
7-86). Alternatively, in a "backfilled" type heavily reinforced structures are used more
of construction, the relieving platform can often.
be constructed prior to backfilling. This pro-
vides an effective anchorage for the sheet-
piling. In this case, the backfill material is
installed later through special openings left 7 .5.3 Underdeck Slope
in the platform.
Subsequently, to provide for the expected The underdeck slope is a very important
soil lateral pressure relieving effect these component of the open pile wharf system. If
openings are either filled with concrete or not property designed, it can be easily dam-
covered by precast slabs. The obvious disad- aged by the effects of natural or vessel-
vantage of this type of construction is induced waves andjor currents. Slope dam-
needed for the use of expensive formwork age typically consists of displaced armor
installed over water. For a detailed discus- stones, especially at the upper part of the
sion on both "dredge" and "backfill" type of slope. The latter usually results in a deep
construction, the reader is referred to Chap- hold in the pavement just behind the wharf
ter 6. It is also briefly discussed further in structure and in pavement settlement in
this chapter (Section 7.8.2.2) and illustrated adjacent wharf areas. Additionally, the
in Figure 7-86. washed-away material settles in front of
Where sheet-piling is stiff enough and is the wharf, thus reducing the underkeel
driven deep enough to sustain the full lat- clearance. Sometimes, this can severely
eral soil pressure without significant dis- hamper dock operation. Three characteris-
placement andjor overstress, then the re- tic case histories of slope damage are dis-
lieving platform can be cast-in-situ on the cussed by Leitass (1979). In the above cases,
backfill material; therefore, the need for damages observed at several wharves lo-
costly formwork is avoided. cated in breakwater-protected harbors con-
It must be pointed out that to reduce sist basically of displacements of stones
total soil pressure on a structure, some- from the upper part ofthe slope, and forma-
times the volume of the backfill material is tion of 0.5-1.0-m-diameter, 1-m-deep holes
reduced, as illustrated in Figure 7-21. This just behind the wharves.
technique is usually used in structures that At some locations, the pavement settled
include robust sheet-piling able to sustain up to 0.2 m. These damages were attributed
ship impact without significant local dam- to the combined action of waves and tides,
ages that may lead to catastrophic failure of with subsequent settlement of the rubble
the whole structure. The advantage of this mound.
structure is that the reduced quantity of a Shin and Anastasiou (1992) point out that
good quality backfill material is required. the basic reason for the kind of damages
Essentially, the stiffness of the relieving listed above can be attributed to the effects
platform affects the load distribution be- of air entrainment, wave characteristics,
Piled Waterfront Structures 783

and the amount of clearance between the the backfill sand. Mter all the piles were
wave crest and the underside of the deck. driven behind dike 1, dike 2 was placed
between pile bents spaced at 6.1 m and
7.5.3.1 5/ope Design backfilled hydraulically. The same proce-
dure was follows for dikes 3, 4, and 5.
It is usually governed by two factors, Testing of soil samples taken from the
namely requirements for global stability and hydraulic fill has indicated that fines in the
the protection from erosion by waves and fill was more than twice as much as in the
currents of miscellaneous nature. As pointed samples taken from the original source of
out earlier, the steeper and, therefore, backfill sand. Presumably it was difficult to
shorter slope results in a shorter, hence borrow from only the sand stratum, thus
more economica!, deck structure. On the unsuitable material was mixed in and de-
other hand, a steeper slope is less stable posited with the fill. A value of <P = 30° was
and more vulnerable to waves and current used for a review of slope stability, and the
effects. Normally, a steep slope needs tobe most unfavorable slip circle produced a fac-
protected by heavier armor stones, relative tor of safety of only F = 0.87 when a sur-
to a fiatter one. Also, a steep slope is less charge load was considered and F = 1.02
stable than the fiat one. without any extra loading. It should be
It follows from the above comments on pointed out that due to the high silt content
slope geometry that geotechnical as well as in the fill, the actual value of the soil angle
hydraulic conditions are the principal gov- of internal friction, </>, and, therefore, of F
erning factors in slope design. should be even smaller. Apparently, a com-
An interesting case history of underdeck plete sliding failure was prevented by the
slope design is discussed by King and dowel action of the piles.
Gavaris (1961). In this case, a new wharf Lateral displacement of the wharf was
915 m long comprised of 12.9-m-wide con- attributed to one-dimensional orientation of
crete platform · supported on 46 cm square most of the batter piles (Fig. 7-49b). This
prestressed concrete piles, with a transition was confirmed by measurements of width
deck supported on a platform and on a changes in the nine expansion joints which
land-based foundation, with a 1 : 1.25 (V: H) progressively opened up at one end of the
underdeck slope was displaced laterally and wharf and closed toward the other end,
longitudinally by approximately 5 cm well thereby refiecting the cumulative effect of
before construction was completed. The sea- the longitudinal displacement component.
ward movement has reached appro:ximately To ensure the wharfs stability, rock dikes
8.1 cm when the fiii was placed behind rock 4 and 5 were removed, the apparently too
dike 5 (Fig. 7-49a) and the concrete of the steep slope beyond dike 3 was fiattened,
lower part of most of the pile caps and of and the wharf deck was extended further
the edge beams had been poured; at this toward the land, as indicated in Figure
stage, the signs of structure overstress were 7-49c.
observed, the construction was stopped, and A similar case history is described by
the investigation has begun. The project Tawil and Tanner (1979). In this case, the
called for removal of the poor soil that over- failure of the underdeck slope occurred due
laid a good sand deposit and replacement of to excess hydrostatic pressure caused by a
poor soil with a good sand dredged out from large (8.5 m) tidal fiuctuation which im-
a borrow area in the vicinity and hydrauli- posed a severe rapid drawdown effect in the
cally placed as required. harbor sediments. This hydrostatic loading,
Rock dike 1 was placed on exposed sand coupled with a low soil shearing resistance,
stratum in front of the structure to contain led to the redesign of the open pile marginal
784 Piled W aterfront Structures

,. H.WL.
yHI.11/'IIJ.Lif
(11.8m)

(a)

·9f.lm

tttltl!!t!!ltttt Direction of lateral earth pnssures


(b)

,. H.W.L

y•l>.11fM.LLW
f•IUm/

(c)

Figure 7-49. Effects of geotechnical conditions on slope stabil-


ity: (a) a too steep underdeck slope caused global instability of
the structure; (b) original pile arrangement that permitted
longitudinal movements; (c) modification of original design. [From
King and Gavaris (1961).]

wharf at the Rodney Container Terminal, proper evaluation of ali factors that affect
St. John Harbor, New Brunswick. At this underdeck slope stability. In this sense, the
wharf, the underdeck slope was flattened, slope stabilizing effect produced by the pile
which created an intervening body of water dowel action should be considered within
behind the wharf which was spanned by reason; this can result in more economical,
severa! access bridges. yet reliable, wharf design. It should be
Both of the slope failure examples given pointed out, however, that the pile dowel
above illustrate the importance of the action, in general, should not be considered
Piled Waterfront Structures 785

when piles are installed in soft and sensi- dent soil displacement, in the form of verti-
tive clays, because relatively large move- cal ground heave or lateral displacement,
ments are required to mobilize the dowel may occur during installation of large dis-
action in these soils (Tschebotarioff, 1978). placement piles or closely spaced piles.
D'Appolonia and Lamb (1971) observed the
7.5.3.2 Slope lnstability caused vertical ground heave phenomenon at a dis-
by Pile-Driving tance of approximately 30 m from the loca-
tion of pile-driving.
Unstable slope conditions (natural or Sowers and Sowers (1967), D'Appolonia
mad-made) can develop during construction (1971), and Broms and Stille (1976) report
as piles are driven. Pile-driving contractors examples in which pile-driving displaced the
can describe many examples of outward backfill soil laterally toward the sheet-pile
movement of piles driven in slopes. Some- bulkheads, causing them to fail. The details
times, however, the possibility of the pile- are provided in Chapter 6.
driving causing slope movement is over- Low-displacement pipe piles can also
looked. This phenomenon has been dis- cause some lateral soil movement when they
cussed by many workers (D'Appolonia, 1971; get plugged. Naturally, the volume of dis-
D'Appolonia and Lambe, 1971; Crooks et placed soils can be minimized by using
al., 1980; Massarsch and Broms, 1981; Riker low-displacement pipe piles, provided that
et al., 1992; Anderson et al., 1992 and the soil plug is cleaned out from within the
others). piles. Vesic (1972) determined that the plas-
The Canadian Geotechnical Engineering tic (nonlinear) zone around driven piles can
Manual (1985) specifically emphasizes the extend 5-7.5 pile diameters for a range of
potential adverse effects of pile-driving in soil rigidity ratios E 1Tr = 200 to 500, where
slopes. Vibration of the soil caused by pile- E is Young's modulus and Tr is the shearing
driving may at least temporarily reduce the strength. In practice, for example, this
shear strength of the clay, especially if its would mean that a 500-mm-diameter pile
sensitivity is high. A reduction of up to 50% could displace soil within a range of 250
in the shear strength has been observed mm at the interface between the pile and
by Broms and Wong (1985). Broms and soil to several millimeters at a distance
Bennermark (1968) have described a land- 2.5-3.75 m from the pile.
slide in a very sensitive clay that had been However, in practice, predicting the mag-
initiated by pile-driving near the top of a nitude of soil movement is very difficult,
slope. and the usual recourse has been to avoid
Soil movement during pile-driving in the problem to the extent possible. The lat-
slopes may be caused either by soil dis- ter usually includes increasing the space
placement during pile installation, develop- between adjacent piles in both longitudinal
ment of high pore pressure in the soil re- and transverse directions. Typically, pile-
sulting from compression and shearing driving is starting from the top of the slope
strains in the soil displaced from its origi- and working down the slope so as to mini-
nal location by piles, vibration, or by any mize the effects to any installed piles.
combination of all the above. Comprehensive and shearing strains
Anderson et al. (1992) suggest that vibra- within the soil associated with pile-driving
tion associated with pile-driving is not a can lead to the buildup of pore-water pres-
primary concern. Shearing strains associ- sure. The increase in pore pressure occurs
ated with vibration attenuate so rapidly in clayey and silty soils, as well as in sands.
that their direct effect, from the standpoint In sandy soil structure, the pore pressure
of movement, is very limited. The most evi- normally dissipates rapidly and, therefore,
786 Piled Waterfront Structures

its cumulative effects on slope stability in many computer programs currently avail-
general is insignificant. However, in clayey able on the market.
soils, pore pressures do not dissipate Plots of pore pressure buildup, such as
quickly. Consequently, as the piles are those suggested by D'Appolonia (1971) or
driven into the soil, a progressive increase Massarsch and Broms (1981), can be used
in pore pressure can be expected. to obtain an initial estimate of the pore
As indicated by Anderson et al. (1992), pressure increase. In conventional methods,
the most critical cases are slopes consisting several issues must usually be dealt with
primarily of cohesive soils; however, cohe- separately or in combination. These include
sive soils with sand and silt seams and the amount of pore pressure buildup and
layers, or slopes consisting primarily of silt the zone over which it may act, the effects
materials can also be critical. of the increased pore pressure on the engi-
Pore pressure builds up in these layers neering properties of the soil, and the ef-
during pile-driving, but the presence of clay fects of the method of pile installation on
or low-permeability silt does not permit slope stability.
rapid pore pressure dissipation. Conse- The amount of slope movement is usually
quently, the soil profile will have layers and predicted by finite element methods. This
lenses with little shear strength. These lay- is, however, a very problematic undertaking
ers serve as potential failure surfaces. that involves considerable uncertainty, pri-
Quantifying the amount of pore pressure marily because of the three-dimensional
buildup and predicting the consequences of nonlinear nature of the problem, and its
pore pressure increase is very uncertain and high dependence on very accurately esti-
certainly a difficult task. In cases in which mated soil engineering properties.
the difficult soil conditions exist (e.g., sand During pile-driving, a contingency pro-
and silt seams and lenses), the observa- gram is usually developed. It includes con-
tional approach that includes a monitoring sideration of potential alteration of driving
program is usually recommended. This in- sequence or schedule until the pore pres-
cludes installation of pore pressure trans- sures are reestablished to acceptable levels.
ducers, piezometers,. slope indicators, sur-
face movements, and other devices to moni- 7.5.3.3 S/ope Protection
tor the slope performance during pile in-
stallation (Dunnicliff, 1988). The relevant Because of variability in the wave and
examples are reported by Tawil and Tanner current regimes in the vicinity of a wharf, it
(1979), Anderson et al. (1992), and others. is unlikely that any "universal" technique
As part of the observational approach, ana- can be recommended for underdeck slope
lytical methods, both conventional and the protection. Normally, each situation is site-
finite element method, are used to identify specific and, thus, has to be treated individ-
the critical point for the slope stability lev- ually.
els or pore pressure. A large variety of materials and tech-
The effects of built-up pore pressure on niques have been used worldwide for under-
slope stability are normally evaluated using deck slope protection. In most practica!
a conventional effective stress method a dis- cases, rock has been used routinely for slope
cussion of which is found in standard texts protection, especially where it could be ob-
on soil mechanics. Some of conventional tained cheaply and easily transported to
methods consider the effects of pore pres- the site. In general, however, the cost of
sure buildup generated as soil shear begins high quality carefully graded rock can be
to occur (Wright, 1991). The pore pressure very high, especially if the source of mate-
analyses are best accomplished by one of rial is distant from the construction site.
Piled Waterfront Structures 787

This has led to the construction of slope bank protection and use in wharf construc-
protection from rocks of miscellaneous sizes, tion. Concrete filled bags are environmen-
back-tipped from lorries. Although subse- tally acceptable and can be successfully em-
quent topping-up with heavier armor stones ployed in areas with low to moderate wave
may be required, this technique can be sub- activity. It must be pointed out, however,
stantially less expensive than those of more that sometimes, despite the porous fabric
sophisticated designs, such as gabions, used for bags, the bags may not allow the
concrete-filled fabric bags, concrete pre- concrete to leach out, thus, bag-to-bag adhe-
fabricated units, asphalt-based mats, and sion may be not very good. The stability of
others. concrete-filled bags used for slope protec-
Gabions, which are essentially wire tion can be enhanced by short steel rebars
cages filled with stone, have been used in driven though bags while the concrete is
areas where large rocks could not be ob- still fresh.
tained at a reasonable cost. In freshwater, The temporary structures designed to
the wire cages are usually protected from serve a relatively short period of time (e.g.,
corrosion by galvanization, those used in a wharves used for construction of hydropro-
saltwater environment are protected by jects, bridges or other marine-related pro-
coating the wire with plastic. However, the jects), sand-filled bags can be considered for
wire protection of any kind, and the wire underdeck slope protection. However, be-
itself, could be vulnerable to abrasion by cause sand-filled bags can be damaged by
movement of coarse sediment; the latter, in waves or miscellaneous floating objects, the
combination with corrosion, can lead to a low capital cost has to be weighed against
very rapid failure of the gabions. Occasion- the risk of likely high maintenance costs.
ally, gabion failures result from incorrectly Sand-filled bags can be successfully used for
packed cages in which stones move within quick, temporary repairs to damaged per-
the cage, thus causing wire abrasion from manent structures.
within. The gabions installed on an under- Sometimes the sandbags can be used in
layer made up from a coarse granular mate- combination with concrete-filled bags. Here
rial (e.g., gravei, crushed stone, or similar the dike formed by the sandbags is pro-
materials) offer significant hydraulic ad- tected against wave action by the concrete-
vantages by permitting water movements filled bags placed on the top and at the toe
in and out of the slope material without ofthe dike.
loosening fine particles from within the fill. Concrete prefabricated units are par-
In addition, the gabions are usually suffi- ticularly useful in areas where rock materi-
ciently flexible to allow them to adjust to als are scarce. There are many types of
minor settlement of the underlayer. The concrete units available for slope protection
disadvantages of slope protection with with a wade variety of uses. These units are
gabions are as follows: an underwater in- typically designed as primary armor in-
stallation requires the use of divers, it is stalled on an underlaying level of a coarse
labor-intensive, and the replacement of a granular material. These units are usually
damagedj deteriorated unit is very difficult. designed to dissipate wave energy and com-
Concrete-filled fabric bags have been bat scour due to the effects of natural
used for slope protection as substitutes for andjor ship-propeller-induced currents.
large rocks. They are readily available and Mattress comprised of prefabricated
can be used on short notice to replace armor concrete blocks. In recent years, a num-
rocks removed by waves or stabilize the ber of patented systems have been devel-
toes ofbadly eroded riprap structures. There oped and used for slope protection. Some of
is a wide variety of fabric bags designed for these systems use concrete blocks joined by
788 Piled W aterfront Structures

steel ropes or chains to form large mat- action of natural and vessel-induced waves
tresses. These mattresses are prefabricated and currents.
onshore and can be rapidly installed under- Wave-riprap interaction has been exten-
water for slope protection by conventional sively investigated in the past 30-40 years
equipment such as mobile cranes. The block and various stability formulas that relate
cabling allows the wave impact force to be the median weight (W50 ) ofthe riprap to the
distributed over a wide area of the revet- wave height and the slope, have been pro-
ment, thus allowing for use of relatively posed by many investigators. For the theo-
lightweight blocks. Furthermore, a good retical stability of riprap, the U.S. Army
block mattress design, by virtue of its Corps of Engineers Shore Protection
roughness and permeability, help to dissi- Manual (1984) recommends the following
pate wave energy in a progressive manner, formula proposed by Hudson (1959):
provided that the individual blocks are fab-
ricated from a good quality concrete and the 1 ) 1/3
Hv = ( Pr W50 KRR(SR - 1) 3 cot a (7-1)
joining steel cables are provided with suf-
ficient protection from corrosion. This type
of slope protection can be reliable and where
durable, particularly in basins sheltered H v = = design wave height
from large waves. A further advantage of W50 = median stone weight
this type of slope protection is that it can K RR = stability coefficient
accept a limited degree of settlement of the SR = specific gravity of stone (SR = Prl Pw),
underlayer. where Pw is the water mass density
The major disadvantage of this system is a = slope angle
that in cases of serious damage to the blocks Pr = stone mass density
or cables, it is difficult to repair the ele-
ments without replacement of the whole KRR values suggested for design corre-
mattress section. spond to a "no damage" condition where up
Asphaltic-based mixture mattresses to 5% of the rock units may be displaced.
have been extensively used in coasta! de- Because no tests with random waves had
fense works as well as in the protection of been conducted, the Shore Protection Man-
seafl.oors from erosion by natural and ship- ual (1984) recommends Hv = H 10 , the av-
propeller-induced currents. These mat- erage of the highest 10% of ali waves. The
tresses have a number of advantages over value of KRR for breaking waves was re-
the less resilient forms of slope protection. vised downward from 3.5 to 2.0; for non-
By virtue of their elasticity, they can with- breaking waves, KRR is equal to 4.0.
stand short- and long-duration loading and The main advantage of the Hudson for-
high-impact forces. The basic disadvantage mula is its simplicity. Although this for-
of asphaltic mixture-based mattresses is mula is lacking the strong theoretical hasis,
their low permeability and, therefore, sensi- it is supported by a large number of tests.
tivity to the uplift hydrostatic forces. Lefebvre et al. (1992) carried out the
evaluation of the field performance of a
7.5.3.4 Riprap Design number of steep ripraps used for shore pro-
tection and concluded that the Hudson for-
As stated earlier riprap is the most widely mula, in general, realistically predicted the
used solution for protecting the underdeck behavior of steel riprap that is free of fine
slope of the open piled wharf structures. It material and, therefore, could serve as a
is normally designed to resist the combined good design tool. These investigators have
Piled Waterfront Structures 789

also concluded that the presence of fine ma- For plunging waves ( gm < 2.5 to 3.0)
terial in steep-slope riprap is harmful to
stability. Generally, they found that riprap H
D_ =
__
s )o.2 f;;,o.s
6.2P0.18 ( _ d (7-3)
containing a large portion of fine material !::.Dso IN
experiences major damage even due to weak
wave action, whereas the calculations indi-
and for surging waves ( gm ;:: 3.0)
cate that the riprap should have resisted
this wave action when the gradation is ex-
pressed by Eq. (7 -1) in terms of W50 . This _H_
D_ = s )0 ·2 Vcot a gP
p-O.l 3 ( _ d (7-4)
factor, however, does not seem to be as !::.Dso IN m
detrimental for flatter slopes.
The Hudson formula has been used since where
the early fifties along with the Iribarren
formula, which reads as follows. gm = wave steepness based on mean period
NprgHJ Tm
Wso = ---------:: (7-2) gm = tan a/ {8: and Sm = 27THn/gT;,
±
!::.3 ( JL cos a sin a ) 3
P = permeability factor; P = 0.1 to 0.6 de-
where W50 , H D, a, and Pr are the same as pending on armor, filter, and core mate-
in Eq. (7-1) rials; for example, for armor and filter
material placed on impermeable slop
g = gravity acceleration material P = 0.1; for a rubble mound
N = stability coefficient (armor stones) with no filter and no core
!::. = relative density of riprap materials, P = 0.6; for more details,
JL = friction coefficient consult PIANC (1992)
N = number of waves; the maximum num-
In both Eqs. (7-1) and (7-2), coefficients ber of waves, N, which should be used
KRR and N are based on model experi- in Eqs. (7-3) and (7-4) is 7500; after
being exposed to approximately this
ments. Moreover, their values depend
number of waves the structure should
largely on the amount of damage accepted.
reach equilibrium
Despite their differences, both Eqs. (7-1)
and (7-2) have much in common. This be-
sd = damage parameter for rock slope; de-
sign values for Sd are given in Table
comes clear particularly when the influence 7-1; the "start" level of damage, sd = 2
of the slope (cot a versus J.L cos a ± sin a) to 3 is equal to the definition of "no
is compared for values of practica! signifi- damage" in the Hudson formula
cance. It should be pointed out that the D 50 = stone diameter that corresponds to W50 ;
lribarren formula has a more theoretical D 50 = (W50 /y.) 11 3 , where 'Ys =rock unit
basis than the Hudson formula, resulting in weightl
the use of the friction coefficient J.L. The
disadvantage of both formulas is that the Table 7-1. Design values of Sa for a two-
influence of the wave period is ignored. diameter·thick armor layer in eqs. (7-3) and (7-4)
Based on earlier works of various investi-
gators and on an extensive series of model Initial Intermediate
Slope Damage Damage Failure
tests that included rubble mounds with a
wide range of corejunderlayer permeabili- 1:1.5 2 3-5 8
ties and a wide range of wave conditions, 1:2 2 4-6 8
Van der Meer (1988) derived the following 1:3 2 6-9 12
formula for plunging and surging waves: Source: From PIANC (1992).
790 Piled Waterfront Structures

The transition from plunging to surging 1.00

waves can be calculated using a critica!


value of ţm as follows:
~
\
0.90

tmc = 0.80

\
(6.2P 0 ·31 Vtan a ) 1 /(P+0. 5 ) (7-5)

For cot a > 4.0, the transition from plung-


\
0.70

ing to surging does not exist, and for these

\
0.60
slope angles, only Eq. (7-3) is used. Equa-
tions (7-3) and (7-4) are more complex than 'A'o
o.so
Eqs. (7-1) and (7-2). They include the
effects of the wave period, the storm ~
\
0.40
duration, the permeability of the slope pro-

"' ------
tection structure, and a clearly defined 0.30
damage area. For more details on the V an
der Meer method, the reader is referred to 0.20 ...........
CURjCIRIA (1991) and PIANC (1992).
Taylor (1973) and Thompson and 0.10 1----
Shattler (1979) suggest that filling the voids
between large stones in the riprap with fine o.
o o.s 1.0 L5 2.0 2.S 3.0 3.5 4.0
material may be beneficia! for riprap with a PERCENTAGE OF AIR-HOLE AREA

slope of 2 : 1 of flatter. The worldwide expe-


Figure 7-50. Reduction of air pressure in relation to
rience, however, indicates that the designer percentage of air-hole area. [From Leitass (1979).]
must be aware of potential deterioration of
steep-slope riprap protection and must be waterway. These waves include the bow or
careful in terms of tolerating fine material front waue, the transuerse stern waue, and
in the riprap. the secondary waues. The wave size essen-
Also, where applicable, for example tially depends on vessel sailing speed, mode
where a continuous front wall (fender panel) of maneuvering, closeness to the wharf, and
is submerged below the water level, the a block coefficient. The latter is the ratio of
detrimental effects on riprap by fluctuating the channel or waterway cross-sectional
air pressures attributed to wave action area to the submerged cross-sectional area
should be considered. This could be sub- of the vessel(s) at midship. For details, the
stantially reduced by the installation of reader is referred to Chapter 2.
pressure-relieving holes in the deck. Re- The bow wave is propagating at an angle
sults from the laboratory tests using air from the sides of the vessel. The shape of
pressure reductions versus area of the the bow affects the magnitude and propaga-
pressure-relieving holes are shown in tion of this type of wave. In a confined
Figure 7-50. waterway, the magnitude of the bow wave
For more useful information on a subject is generally of a lesser magnitude than the
of slope protection, the reader is referred transverse stern wave. There, the bow wave
Fuehrer et al. (1981). Span et al. (1981), may coincide with the stern wave, forming
and Balanin et al. (1981). incidental peaks with amplified wave
Vessel-induced waves can also be a heights. The transverse stern wave is the
dominant force in the design of underdeck resulting wave formed as water displaced
riprap. This can occur when marine termi- by the vessel flows around the huli to the
nals are located on one or both sides of a stern. In a narrow waterway or near a bank,
relatively narrow navigation channel or the transverse stern wave moves in the
Piled Waterfront Structures 791

same direction as the vessel and is gener- Results of recent research on navigation
ally perpendicular to the bank and can re- effects by typical commercial tugboats and
semble a moving hydraulic jump. This wave barges operating on the U.S. inland water-
is related to the drawdown. ways are reported by Martin and Maynord
The transverse stern wave, sometimes (1992). The basic theory of the propeller jets
referred to as the water-level depression, is is discussed in detail in Tsinker (1995).
the drop in the water level alongside the In PIANC (1987), the height of the trans-
vessel caused as it is moving forward and verse stern wave, ht> is defined as the maxi-
the water is displaced from bow to the stern. mum water-level depression. Its maximum
Drawdown accompanies the return cur- value is related to the average drawdown,
rents. It is a function of a ship's traveling hd, as follows:
speed, size, and draft, and the waterway
geometry. Secondary waves are those corre- (7-6)
sponding to the interference peaks formed
where
by the transverse stern and other diverging
waves propagating away from the ship. c = function of the vessel speed, V.; the eccen-
tricity about waterway centerline, y; the
Similar to the regular waves, the vessel-
midship cross-sectional area; Am; water
induced waves have the most impact on the depth, h; the vessellength, L.; the waterline
upper portion of the slope, and the wave width of the channel, bw; the angle of the
magnitude, angle of propagation, and pe- slope, a; it is calculated from the following
riod affect the run up characteristics of the expression:
waves and consequently determine the type
of slope protection.
The propeller jets, also called the screw
race, can cause slope erosion problems dur-
ing the ship berthing and departing maneu-
(7-7)
vers. Typically, during these maneuvers the
ship is moving at a slow speed while at a
high rate of propeller revolution. The char- where
acteristics of the propeller jets and the flow
impingement on slope are functions of the v.
Fh = fih and (7-8)
propeller size and type of the horsepower of
the vessel.
Yt = 0.5bw - y - 0.5h cot a
Essentially, due to the propeller's clase
proximity to the ship's keel, propeller-
induced high-speed currents tend to im- Essentially, for long ships and thus large
pinge on the lower part of the underdeck values of L., Eq. (7-7) gives higher values
slope as well as the seafloor in front of the of the correction coefficient c.
wharf. For details, the reader is referred to The drawdown, hd, that accompanies the
Tsinker (1995). return currents can be determined by the
energy conservation method proposed by
7.5.3.4.1 Quantification of
Jansen and Schijt (1953). The following
navigation-induced
basic assumptions are used in this method:
forces symmetrical trapezoidal or rectangular
This has been a subject investigated in cross section of the waterway; uniform re-
the past 30-40 years by the workers in turn velocity from vessel to banks; uniform
many countries. The experience gained drawdown from vessel to bank lines; negli-
worldwide is summarized in PIANC (1987). gible friction losses; uniform or negligible
792 Piled Waterfront Structures

ambient velocity; vessel sailing on water- where


way centerline. vrs and hds return velocity and drawdown,
=
As indicated by Martin and Maynord respectively, on the side of the
(1992), in con:fined waterways having low vessel
blockage ratios, the assumption of uniform A.= area on the side of the tow for
distribution of the return velocity and the which vrs and hds are calculated
drawdown from vessel to bank is in agree-
ment with observed data. As the blockage For stability investigations, the average
ratio increases above approximately 10 the values from Eqs. (7-9)-(7-11) should be
distribution becomes increasingly nonuni- increased by approximately 20% to account
form with the highest values of return ve- for time variations above the average val-
locity and drawdown found near the vessel. ues. Equations (7-9)-(7-11) provide maxi-
From the above conditions, the continu- mum return velocity and drawdown for a
ity requires that blockage ratio N :::; 10. Above N = 10, the
return velocity and drawdown become
V.Ac = Aw(V. + Vr) (7-9)
nonuniform, and alternate techn:icques are
where being developed. According to Martin and
v. = the vessel speed Maynord (1992), most bank stability prob-
Ac = the undisturbed channel area lems resulting from return velocity and
Aw = the disturbed channel area around the mid-
drawdown would be expected in channels
section of the vessel having N < 10.
vr = the average return velocity The bow or front wave, hr, is a function
of both the average, hd, and maximum, h 1 ,
Conservation of energy requires that drawdowns and according to PIANC (1987)
can be computed as follows:
V2 (V +V ) 2
-•- + h = 8
r + (h - hd) (7-10)
2g 2g (7-12)

where
g = gravitation acceleration Secondary waves are typically related to
h = undisturbed average water depth vessel speed. The height of these waves, H,
can be computed as proposed by Verhey and
hd = average water-level drawdown
Bogaerts (1989):
The unknown Vr and hd can be determined
from Eqs. (7-9) and (7-10). A similar ap- (7-13)
proach by Bouwmeester et al. (1977) uses
conservation of momentum to determine the
average return velocity and the average where
drawdown. s = distance between the ship's side and the
For vessels off the waterway's centerline, bank
Maynord and Siemsen (1991) developed the h = depth of water
following equation to modify the return Fh = ship's Froude number
velocity and drawdown derived from Eqs. a1 =empirica! coefficient; a 1 = 0.35 for the
(7-9) and (7-10): empty conventional ship; a 1 = 0.5 for
empty conventional barge; and a 1 = 1.0 for
V __..:!!.
h
_!:!... = = 0.36 (A ) + 0.64
_c_ (7-11) loaded conventional ships
Vr hd 2A. a 3 = empirica! factor; a 3 = 4.0
Piled Waterfront Structures 793

7.5.3.4.2 scour at the toe of stope 2. Use of rock energy-dissipating armor.


protection 3. Use of scour protection blankets; for de-
Waves and natural and/ or propeller- tails, consult Tsinker (1995).
induced currents can lead to movement of
bed material in close proximity to the struc- Note: In special cases the bed material
ture. This commonly happens in the form of could be reinforced by partial or local
a local scour in front of the structure. This grouting by cementitious or bituminous
process may eventually lead to substantial materials.
bed erosion andjor accretion of sediments Recommendations for the extension of the
at the toe of the riprap resulting in both scour protection rock blanket from the toe
local damage to the slope protection system of the riprap seaward are very limited. The
and a decrease in underkeel clearance. The Shore Protection Manual (1984) states that
former may eventually result in slope fail- in situations where severe scour is ex-
ure which may ultimately lead to a signifi- pected, the width of the blanket should be
cant operational problem. Hence, this po- about 3 to 4.5 times the design wave height.
tential problem must not be overlooked in However, in a separate paragraph, it also
wharf design. Accurate prediction of bed suggests that a width equal to four to six
scour requires a careful study of the near times the rock size should be sufficient.
wharf hydrodynamics and the seafloor re- Henceforth, where docking facilities are
sponse to it. Where local experience sug- constructed at locations well protected from
gests that scour is likely to occur, or the
the action of large waves, the latter recom-
consequences of the underdeck slope failure
mendation is more appropriate for under-
could be particularly severe, then physical
deck riprap design.
andjor numerica! modeling methods should
Figure 7-51 illustrates two examples of
be used to quantizy the potential effects and
scour protection using rock. For more prac-
design the slope protection system.
tica} examples and useful information on a
Current knowledge of wave and current
part of a slope protection design, the reader
effects on slope. protection structure is dis-
cussed in PIANC (1992); simple prediction
method is given in the Shore Protection
Manual (1984). Naturally, the stability of
the riprap toe is essential because failure of
the toe will most likely lead to failure of the
entire slope protection structure. To ac-
count for the scour, the riprap should be
extended to some distance seaward from (a)
the face of the wharf, and its tow is usually
buried below the seafloor to a depth equal
to or slightly greater than the expected
depth of erosion.
The principal methods of reducing or pre-
venting the scour of bed material can be
summarized as follows:
(b)
1. The riprap slope is made more gentle.
Flattening the riprap slope reduces wave Figure 7-51. Examples of two aprons for locations
and current forces by reducing refiection with severe scour potentials: (a) berm tow apron; (b)
effects. buried tow apron. [From PIANC (1992).]
794 Piled Waterfront Structures

is referred to the Shore Protection Manual gated extensively in the last decade. The
(1984) and PIANC (1987, 1992). key element in the scour process is the
Most recently Hamil et al. (1996) re- formation of a horseshoe-type vortex around
ported results of a study on the influence of the pile. The vortex, which is formed due to
a revetment on diffusion of a propeller wash. the presence of rotation in the incoming
Obviously, during the design stage a fun- velocity profile just above the bed level, is
damental choice should be made with re- able to erode a significant amount of sedi-
spect to capital cost of slope protection. Nor- ments away from the area around the pile.
mally, higher capital investment will mean Normally, scour "equilibrium" is achieved
lower maintenance costs and vise versa. for given parameters when the volume of
Furthermore, depending on local require- material removed by the vortex turbulence
ments, either the wharf structure has to be is equal to the volume introduced into the
designed to withstand the consequences of pit through the bed and suspended loads.
the underdeck slope failure due to scour, or At this point, a constant volumetric flux is
concrete steps have to be considered to pre- achieved, and further enlargement of the
vent or reduce the bottom erosion. pit requires longer periods of turbulent
surge.
7.5.3.4.3 scour around pi/es By rule of thumb, the equilibrium scour
depth in a current lies in the range of 1-1.5
When a pile is installed in an erodible times the pile diameter. The scour results
bed, the scour will take place around it due in an increase in the bed shear stress around
to the action of current and waves of miscel- a pile exposed to the current (Hjorth, 1975).
laneous nature. This process could be of a In his work, Hjorth presented a comprehen-
great importance for the stability of marine sive review of the subject state of the art.
structures supported on piles. Modern ships In recent years, with the introduction of
are often equipped with high-power engines a large number of offshore oii exploration
and side thrusters for faster navigation and production platforms, the scour around
speed and for better maneuvering. Such piles in waves and in combined waves and
ships during a berthing or departure ma- current environments has attracted much
neuver, can cause a serious bed erosion, attention. Palmer (1969) has reported some
damage to the underdeck slope, or scour field data on the magnitude of scour around
around piles. Bergh and Cederwall (1981), obstructions exposed to oscillatory wave-
McKillen (1985) and many other workers induced fl.ows; Das (1970) presented a com-
reported numerous ship and propeller- prehensive literature review on a subject of
induced damages to the port related a local scour around vertical cylinders in an
marine structures. oscillating fl.ume; Kawata and Tsuchiya
Extensive scour around piles may reduce (1988) reported results of tests carried out
their load-carrying capacity which eventu- with real waves and also in combined waves
ally may lead to failure of the structure. and current environments; Herbich et al.
Bed erosion around piles can lower the level (1984), Eadie and Herbich (1986), Herbich
of the pile fixity within foundation soil, re- (1991) and Halfin (1990) reported results of
sulting in an increase in pile's freestanding an extensive study of scour around piles
portion. Essentially, this will result in an exposed to waves and current; the most
increase in bending movement in these piles recent studies on scour around vertical piles
which should be taken into account. This is in waves and propeller-induced current car-
a particularly important consideration in ried out by Hamill and Johuston (1993),
the case of offshore piled piers and moor- Sumer et al. (1992, 1993), and Chin et al.
ings. Scour around piles in steady strong (1996). The quantification of scour pit di-
currents andjor waves has been investi- mensions is very difficult and the best re-
Piled Waterfront Structures 795

sult could be obtained from physical model coastal are as andj or estuaries, where, due
tests or on the hasis of relevant local experi- to the history of their formation, soil condi-
ence. State-of-the-art review of ship pro- tions can be very variable within short
peller-induced scour is given in Tsinker distances.
(1995). In practice, the soil investigation pro-
gram cannot cover the whole project area
with sufficient detail because of spatial
7.6 PILE- SOIL INTERACTION variations in soil types and their character-
istics. Besides, it is usually a very difficult
Several important features of piles with task to obtain undisturbed underwater soil
emphasis on their use in marine structures samples and, as stated earlier, the soil pa-
are as follows: rameters obtained from these samples are
only valid for the location of the borehole.
(1) Typically, piles are widely separated and, Hence, the sensitivity of the selected ana-
therefore, react independently. lytical solutions and design methods to
(b) Because individual piles are exposed to variations in soil characteristics should be
high concentrated loads, significant pile carefully examined.
elastic settlement should be expected. The general attitude toward theoretical
(c) Because marine structures are normally analysis of pile foundations was exemplified
exposed to heavy horizontal loads, signif- by Terzaghi and Peck (1967) who stated
icant bending moment-induced stresses in " ... theoretical refinements in dealing with
piles at the foundation soil level and pile problems . . . are completely out of place
within the deck structure must be ex- and can be safely ignored."
pected. In the past two decades, due to the avail-
ability of the powerful desktop computers,
In the following subsections, the aforemen- attempts to develop practical theoretical
tioned factors of pile-soil interaction are methods of pile analysis based on the finite
discussed. element method and the boundary element
method have been made. Several such anal-
ysis methods have been proposed. These are
7.6.1 General evaluated in detail by Poulos (1989) in his
Rankine lecture. As he pointed out, the the-
At the present time, the design of piles and oretical methods lead to a significant in-
pile foundations are normally based on es- crease in understanding of the mechanics of
sentially empirica! methods of evaluation of pile-soil interaction and to improved appre-
pile load-carrying capacity, experience, and ciation of the factors which in:fluence pile
good judgment. Difficulties in obtaining ac- behaviors. He also pointed out that
curate input data which are needed to per- " ... theory should not be developed in isola-
form a sound theoretical analysis usually tion from controlled experiments and field
make the attempts to compute the pile observations, and should not be applied
load-carrying capacity with mathematical without due recognition of the importance
precision irrelevant. of proper characterization of the site geol-
One should realize that a calculation ogy."
method cannot be more accurate than the In general, the parameters that affect
available input data (e.g., load and soil soil-pile interaction are as follows:
characteristics). Marine piers, dolphins, and
other port related marine structures, de-
signed to accommodate large vessels are 1. Soil characteristics
usually located in relatively deep water in 2. Pile characteristics
796 Piled Waterfront Structures

3. Method of pile installation the structural engineer during his initial


4. Type of pile loading schematic design. Some of the factors af-
fecting his choice are as follows. The first is
AII of the above parameters are interrelated pile availability that may vary widely in
in one way or another; for example, soil different parts ofthe world. Next, the engi-
characteristics before and after pile instal- neer will try to minimize pile size in order
lation, or changes in soil parameters de- to hold down drag forces by waves and cur-
pending on method of pile installation. A rent, minimize weight, and reduce cost of
detailed discussion on this and the interpre- pile installation. On the other hand, in some
tation of the available data are given in cases it may be more economica! to increase
Kraft (1991). pile size in order to reduce the number of
piles to a practica! minimum. The overall
7.6.1.1 Soi/ Characteristics geometry of the structure is, therefore, a
strong controlling consideration. Essen-
Basically, it can be expected that the tially, the pile cross section must be large
ultimate shaft resistance increases with an enough to allow stresses to be held within
increase in the soil friction angle, effective allowable limits. The question of drivability
overburden stress, and effective soil-pile of the selected pile, which is a soil related
friction angle. Soil stress-strain parame- factor, is another problem to be solved by
ters, in addition to strength parameters, the engineer.
can also affect the shaft resistance and the Once pile type and size have been se-
deformation response of a pile subjected to lected, its penetration into the foundation
load. soil is determined. This is primarily a mat-
A decrease in soil compressibility can re- ter of designing the pile to develop ade-
sult in larger lateral stresses acting on the quate axial (tension and compression) and
pile shaft after installation and can affect bending capacities, given due consideration
changes in the lateral stress during loading to the pile installation procedure and its
as the principal stresses rotate when the influence on pile capacity.
soil next to the pile is sheared (Nauroy and In displacenient piles (e.g., closed-ended
LeTirant, 1983). pipe piles and precast concrete piles), the
Also important are stratigraphic changes pile size may affect the value of the ulti-
and changes in the soil properties with mate unit shaft resistance. Vijayvergiya and
depth due to increased overburden pres- Focht (1972) and Grutman (1971) suggest
sures and pile installation effects. that soil displacement during pile driving
mobilizes the lateral pressure in the sur-
7.6.1.2 Pile Characteristics rounding soil that in some instances can be
In terms of pile-soil interaction, the most compatible with passive pressure.
important pile characteristics are pile size Detailed investigation of the extent of
and configuration, compressibility, and dis- compaction of sand and the increase in rela-
placement ratio. tive density around the pile have been car-
ried out by Meyerhof (1959) and Robinsky
and Morrison (1964). The latter investiga-
7.6.1.2.1 Pile size and tors found that in the initially loose sand
configuration (relative density Dr = 17%), soil movement
Because of the high interaction between extends 3-4 pile diameters from the side of
piles and the structure of which they are an the pile and 2.5-3.5 diameters below the
integral part, the choice of pile size for pile tip. In a medium sand (Dr ""' 35%), the
marine structures is usually preempted by extent of soil movement · corresponds to
Piled Waterfront Structures 797

4.5-5.5 diameters from the side to 3.0-4.5 tively weak, a plug may not develop at all
diameters below the tip. These figures are (Yaron et al., 1982).
in broad agreement with those obtained by In the case of H-piles the actual pile
Meyerhof (1959). On the hasis of field obser- surface does not always represent the fail-
vations and model test results, Kishida ure surface and if considered as such may
(1967) suggests that the diameter of the result in a grossly overestimated pile capac-
compacted zone around a pile would be con- ity. As reported by Hegedus and Khosla
sidered equal to 7d, where d is the pile (1984) in overconsolidated clays the un-
diameter. drained shear strength approach for pre-
The pile's length (penetration) plays an dicting of ultimate pullout capacity of H-
important role in the process of pile-soil piles resulted in a good agreement with the
interaction. The unit shaft resistance typi- observed values, then the 'effective' pile
cally increases more rapidly near the sur- surface was computed with inclusion of 'soil
face than at depth. According to Hanna and plug' located between pile fl.anges. In this
Tan (1973) and Tan and Hanna (197 4) who case use of the actual pile surface would
carried out pile model tests in a loose sand, result in gross overestimation of the pile
the average unit shaft friction increased pullout resistance. Since granular (nonplas-
approximately linearly and pile penetra- tic) soils are not likely to adhere to the
tions to about 40 pile diameters and re- entire pile surface to force formation of the
mained constant for the pile penetrations 'soil plug', thus in such soils the majority of
exceeding this depth. Additionally, the gen- the pile's perimeter could serve as a failure
eral configuration of the shaft friction dia- surface.
gram was more parabolic than triangular in
shape. Other investigators (Kerisel, 1961; 7.6.1.2.2 compressibility
Vesic, 1970; Das and Seeley, 1975;
The relative soil-pile compressibility is a
Chaudhuri and Symons, 1983; and others)
function of the material and cross section of
have observed a similar trend in distribu-
the pile and the stiffness of the surrounding
tion of average unit shaft friction primarily
soil. A key parameter in the relative
in the case of driven piles. In contrast, a
soil-pile compressibility characterization is
more linear distribution of average unit
the load-displacement behavior once the
shaft friction with penetration depth can be
peak load transfer is reached.
expected in piles installed by water jet.
The conventional methods used for com-
Similar to the limiting increase in shaft puting the load-deformation response of ax-
friction, the unit toe resistance often does ially loaded piled are based on a classical
not increase linearly with depth, as implied assumption that a soil can be represented
by classical theory (Neely, 1990). Limiting by a linear elastic continnuum (Mindlin,
values of unit toe resistance have been re- 1936) or by a series of unconnected linear
ported at penetrations of 10-30 pile diame- elastic springs (Winkley, 1967). Kraft et al.
ters (Vesic, 1964; Kerisel, 1961). (1981) developed a theoretical procedure
Furthermore, the size of the pile diame- for developing the equation for load-
ter may affect the formation of a soil plug in deformation curves which takes into ac-
the open-ended pipe piles. As indicated by count the nonlinearly of the soil.
O'Neill and Raines (1991), in dense sands Armaleh and Desai (1987) developed a
the plug may be fully developed when toe procedure in which the pile-soil system is
penetration is as small as 15 pile diame- idealized by using a one-dimensional finite
ters; on the other hand, when the pile diam- element technique. This proposed procedure
eter is very large and the soils are rela- according to the latter workers can be used
798 Piled Waterfront Structures

to predict load-deformation behavior and methods such as an impact hammer or vi-


bearing capacity for axially loaded piles in brator in the final 1.5-2.0 m penetration
cohesionless soils. The proposed method ac- (Tsinker, 1972, 1977, 1988).
counts for the elastic properties of both soil Data summarized by Meyerhof (1976) on
and pile materials. piles in sand indicates that shaft resistance
is highest for driven piles, intermediate for
7.6.1.2.3 Pile displacement ratio jacked piles, and the lowest for bored piles.
Furthermore, Tan and Hanna (197 4) found
Essentially, low-displacement piles cause
that the method of pile-driving also affects
less disturbance during installation and,
residual stresses in piles.
consequently, cause smaller changes in lat-
eral stresses than full-displacement piles.
As pointed out earlier, the open-ended pipe 7.6.1.4 Pile Loading
piles usually behave as low-displacement
The character of pile loading can affect
piles before formation of a soil plug inside of
pile-soil interaction significantly. For ex-
these piles; they perform as displacement
ample, during pile loading by compressive
piles when the plug is formed. The latter
or uplift loads, the lateral stresses along the
results in subsequent changes in pile shaft
pile shaft can increase or decrease, respec-
resistance. Beringen et al. (1979) found that
tively, depending primarily on the direction
shaft resistance of a closed-ended pipe pile
of load application, as well as on the stress-
could be about 50% greater than an open-
strain and compressibility characteristics of
ended pile driven to about 6. 7 m into a
the soil.
dense, overconsolidated sand. Differences in
It should be pointed out that there is not
the shaft resistance of open- and closed-
much data available for the evaluation of
ended pipe piles have also been found by
pile shaft capacity when subjected to a com-
Vesic (1964), Nauroy and LeTirant (1983),
pressive or uplift loading. Besides, the
and many other investigators. This phe-
available data are, to some extent, conflict-
nomenon is basically attributed to the fact
ing. For example, test results reported by
that soil displacements cause substantial
Ireland (1957) on piles driven into fine sand
increases in lateral stresses and, ulti-
suggests that the average skin friction for
niately, in increased shaft resistance.
uplift loading is equal to that for compres-
sion loading; Vesic (1970) suggested that
7.6.1.3 Method of Pile the limiting skin friction along the pile shaft
lnstallation is the same in tension and compression,
This is discussed in detail in Section and it basically depends on the relative
7.5.1.4 and in Chapter 6. It should be density of the surrounding soil and mode of
pointed out that in the literature very little pile installation; O'Neill and Raines (1991)
information on the direct comparison of pile carried out large-scale-model tests on piles
behavior for different methods of installa- installed in very dense pressured sand from
tion is available. To date, published data which slightly lower values were observed
from pile model tests and from field obser- in uplift loading; from compression and
vations suggest that pile installation using pull-out tests carried out on instrumented
water jetting may have detrimental effects precast piles driven into a sand deposit,
on pile capacity in comparison with Altaee et al. (1992a, 1992b) found no dif-
driven pile (Mansur and Hunter, 1970; ference in pile shaft capacity. De Nicola and
McClelland, 1974). However, the capacity of Randolph (1993) summarized results of sev-
jetted piles can be improved dramatically, era! field tests conducted on piles installed
provided they are installed using dynamic in sands. In these tests, the ratio of tension
Piled Waterfront Structures 799

to compression pile capacity varied in the ally unsaturated) soil layers have been re-
range of 0.44-0.85. Furthermore, results of ported.
uplift tests on piles installed by different Most investigators tend to agree that, in
techniques reported by Levacher and Sief- general, tensile shaft friction is lower than
fert (1984) and McClelland (197 4) show a in compression. On the hasis of numerica!
wide variation in pile uplift capacity; in analysis, De Nicola and Randolph (1993)
these tests, the average ratio of ultimate conclude that the ratio of tensile to com-
pulling resistance of driven piles to ulti- pressive shaft capacity is a function of the
mate resistance of statically driven piles slenderness ratio Dr!d of the pile, but it is
was equal to 0.5. The former investigators largely independent of the relative stiffness
also found that the above ratio of vibro- of the pile; here Dr is the pile penetration
driven to statically driven piles is 0.67. and d is the pile diameter.
Data summarized by Sowa (1970) and De Nicola and Randolph (1993) also con-
Downs and Chienrzzi (1966) indicate con- clude that for most of the pile length inves-
siderable variation in average skin friction tigated, the shear stress mobilized in com-
found in different tests, especially in those pression is greater than that mobilized in
carried out on cast-in-situ piles. McClelland
tension and this varies with pile embed-
(1974) reported results of field tests carried
ment, Dr (Fig. 7-52).
out on four piles, each installed by different
Despite .the lack of a sound theoretical
procedures in uniform sand deposits. In this
hasis for the explanation of the discussed
set of experiments, the pull-out capacity of
phenomenon, there are plenty of good rea-
the driven piles was 9.4 times greater than
that of the pile installed by jetting with an sons for expecting the tensile shaft capacity
external return :flow. The nature of this to be lower than the compressive shaft ca-
phenomenon was explained by Tsinker
(1972, 1977, 1988). As stated by Tsinker,
the piles installed by water jet in sand can Normalised interface shear stress
regain almost full shaft capacity after expo- o 0.2 0.4 0.6 0.8 !.2 1.4

sure to dynamic affects, such as vibration, o


on driving a couple of meters by a hammer. -0.1 ~
Randolph et al. (1993) point out that the
uncertain results obtained by different in-
\\1\.
\ f\ \r-.
-0.2
vestigators, when considering both installa- -0.3
tion methods and type of pile loading, are
partly related to different methods of evalu- ?. -0.4 f\\ \
ation of the pile's ultimate capacity. For jj
:1- -0.5 \\ ~Ideal

example, some investigators do not recog-


nize the phenomenon of residual strains
... -0.6 b\ \
\ 1\
~
being locked into the pile during fabrica- Tension [\ Compression
-0.7
tion. This can lead to large changes in true t[\ \ \
'
zero reading when instrumentation is in- -0.8
stalled along the pile shaft. The latter can
-0.9
(\\
lead to uncertainties in estimating the load ''lrS ~
distribution along the shaft. By reviewing -1
the available database, Randolph et al.
Figure 7-52. Correlation between normalized shear
(1993) identify a number of cases in which stress and depth to pile length ratio at shaft failure in
anomalous results with surprisingly high tension and compression. [From De Nicola and Ran-
values of shaft friction in the upper (gener- dolph (1993).]
800 Piled Waterfront Structures

pacity for piles in free-draining soils. In the data on time-dependent strength gains in
absence of sufficient information, Poulos and sands.
Davis (1980) recommend evaluating uplift Lowery et al. (1969) tentatively suggest
shaft resistance by reducing it to two-thirds that a setup factor (i.e., the ratio of soil
of the calculated shaft resistance for down- strength a considerable time after driving
ward loading. A more conservation ap- to that immediately after driving) of 3 might
proach is recommend by Tomlinson (1983), be appropriate for soft clays, 2 for firm and
who suggests that the ultimate skin friction stiff clays, and 1 for other soils. On the
on piles subjected to uplift loads should be other hand, McClelland et al. (1969) con-
reduced by one-half to allow for di:fferent clude that for piles driven into hard clay or
mechanisms of failure where uplift loads sands, a decrease of soil strength and adhe-
are carried. Additional information on pile sion with time can occur.
uplift capacity is found in the work by In fine-grained sedimentary deposits it is
Chattopadhyay and Pise (1986), Chaudhuri not uncommon for piles to indicate initial
and Symons (1983), Hegedus and Khosla low capacity, as determined by dynamic
(1984), Hanna and Afram (1986), driving formulas, but later show excellent
Narasimha Rao and Prasad (1993), and supported values under static loads. This is
others. a result of the dynamic effects caused by
As pointed out by Chellis (1961), Kraft the hammer blow that produced the tempo-
(1991), and others, factors that affect the rary quick condition in soil close to the pile.
shaft resistance mostly during loading in- This condition may disappear rapidly, re-
clude setup time, soil properties after in- sulting in the pile gaining much more
stallation, resistance between the soil and capacity; in some instances, it takes some
pile material, and the state of stress in the time.
soil at the pile-soil interface prior to load- The capacity of such piles may not
ing and during loading. Furthermore, the always be ascertainable by redriving,
axial pile behavior is also inftuenced by the although the first few blows may show
rate of loading, previous load history, and increased resistance before the soil quick
other factors. condition effect reestablishes itself. Under
the above conditions, the static load test is
7.6.1.4.1 setup the surest way to determine the pile axial
capacity.
Depending on the rate of pore pressure Similar effects are observed on piles
change, which affects the stress-strain be- driven into soft clays and into most silts
havior with aging, piles may be subject to and silty clays. These soils remold during
setup; that is, an increase in soil strength pile-driving which results in temporary
and adhesion with time. For example, losses of a large part of their compressive
Tavenas and Audy (1972) found the com- strength and shear value. Soft clays recon-
pressive capacity of concrete piles in a solidate quite rapidly at first after the ces-
medium-dense, uniform, fine-to-medium sation of driving and continue for some days
sand to increase with elapsed time after or weeks.
installation. A maximum increase of Four case histories, in which important
50-90% above the capacity measured 72 h changes in pile bearing capacity were ob-
after installation were developed with 20 served with time, have been reported by
days of setup. Kerisel (1964) and other in- Samson and Authier (1985). In their obser-
vestigators also report an increase in pile vations, friction piles driven into deep sand
bearing capacity in sands with aging. deposits show increases in bearing capacity
Mitchel and Solymar (1984) summarized sometime after installation. The increase
Piled Waterfront Structures 801

occurs over a period longer than that needed been noted earlier, soil displacement during
for pore pressure dissipation. In one case, pile-driving mobilizes the lateral soil pres-
increases of 30% and 85% in the pile bear- sure in the pile surrounding soil, which in
ing capacity were measured 3 and 50 days, some instances could reach the magnitude
respectively, after pile-driving. In two other of a passive pressure.
cases in which closed-ended piles were Tests carried out by Robinsky and
driven to shale bedrock, the pile bearing Morrison (1964) have shown that the
capacity decreased. This was determined process of soil displacements is followed by
using dynamic monitoring during initial soil movements in areas adjacent to pile
driving and at the restrike a few days later. sides and below the tip. These movements,
The decrease ranged from 11% to 25%. particularly those in sands, tend to de-
For more information on the setup phe- crease the soil density in the immediate
nomenon, the reader is referred to Chellis vicinity of pile sides, thus reducing the ef-
(1961) Poulus and Davis (1980) and Tomlin- fect of soil densification around the pile.
son (1994). Naturally, this affects the pile-soil interac-
tion at the interface. Furthermore, pile-soil
7.6.1.4.2 Soi/ properties after pile interaction depends on the friction angle, 8,
installation between the pile and surrounding soil. Sev-
era! shear resistance studies have been per-
As discussed earlier, piles installed into
formed on the interaction between the pile
granular soils can change soil density and
and surrounding soil which suggest that for
rearrange the fabric of the soil. Further-
a smooth steel pile installed in a siliceous
more, pile-driving may result in the crush-
sand, 8/ cp ""' 0.7 could be a reasonable as-
ing of soil grains, resulting in a different
sumption, whereas 8/ cp""' 0.6 may be more
grain-size distribution, as well as in soil
appropriate for calcareous sands (Potiondy,
particles reorientation along the shaft, with
1961; Yoshimi and Kishida, 1981; Acar et
their long axis parallel to the shaft. The
al., 1982; Bozozuk et al., 1979; Datta et al.,
latter in particular may have a profound
1980).
effect of shaft resistance. The effect of grain
Yoshimi and Kishida (1981) found that
crushing is expected to be smaller for the
when the steel pile surface is very rough,
finer-sized particles and for less-angular
pile failure does not occur at the pile-soil
particles. The depth of the zone of infiuence
interface. Tsinker (1972) has observed a
along the shaft is small and is expected to
similar phenomenon during tests performed
be limited to just a few centimeters.
on large-scale-model piles roade from wood
In general, results of pile tests have indi-
and concrete. In these latter tests, piles
cated the pattern of density and fabric
were installed by means of a water jet into
changes to be very complex and nonuniform
a sand foundation. After being pulled out
(Robinski and Morrison, 1964; Tsinker,
from the soil, these piles always had a thin
1972; Davidson and Boghrat, 1983). As
film of sand that was adherent to the pile
noted earlier, cohesive soils tend to remold
shaft. This indicates that the shear failure
during pile-driving, resulting in a tempo-
occurs not at the soil-pile interface but
rary loss of the larger part of their cornpres-
rather between the film of soil adherent to
sive strength and shear value.
the pile shaft and surrounding soil. In the
above tests, the ratio 8/ cp varied from 0.7
7.6.1.4.3 Pile- soi/ interface
to about 1.0.
The condition of the soil at the pile-soil Lasebnik (1961), who carried out large-
interface plays an important role in the scale-model tests on steel sheet-pile bulk-
development of pile axial capacity. As has heads driven in sand, concludes that the
802 Piled Waterfront Structures

roughness of sheet piles has no or little for estimating these stresses that is based
effect on the magnitude of soil pressures on on a wave equation analysis. Briand and
sheeting. Again, this phenomenon is ex- Tucker (1984) have examined data for piles
plained by the fact that the shear failure at driven in sand and proposed simplified pro-
the interface between the sheeting and the cedures for estimation residual toe loads in
surrounding soil is affected not by the pile such piles. More recently, Poulos (1987)
roughness but rather by the film of soil that suggested a simple method for analyzing
is adherent to the sheet piles. the residual stresses in driven or jacking
Contrary to the aforementioned, piles and for determining the subsequent
Chaudhuri and Symons (1983) report pile axialload-deformation behavior.
scale-model test results in which the maxi- Although residual shear stress in the
mum shear stress developed on rough sur- pile-soil system does not contribute a great
face piles were approximately 4 times (for deal to the pile load capacity, knowledge of
dense sand) and approximately 3.5 times these stresses is importance for proper in-
(for medium sand) as large as those devel- terpretation of pile load test.
oped on smooth-surfaced piles. This, per-
haps can be attributed to a scale effect: In
the aforementioned tests, the embedded 7.6.1.4.5 coefficient of the lateral
depth of the model piles varied between 123 earth pressure
and 1313 mm. The horizontal effective stress on a pile
More discussions on soil parameters af- shaft prior to loading is usually related to
ter pile installation and their effect on pile the earth pressure coefficient, K. As indi-
axialload capacity are given in Olsen (1990), cated earlier, the value of K is highly de-
Briand and Audibert (1990), and Toolan et pendent on the state of stress within the
al. (1990).
soil deposit prior to pile installation, as well
as on changes in the condition of soil that
7.6.1.4.4 state of stress result from pile installation.
McClelland (197 4) reports on results of
The pile failure state can be influenced
by the magnitude of the state of stress prior pull-out tests carried out on four pipe piles,
to pile loading, which has an impact on the each installed into a uniform sand deposit
stress path during loading. The vertical ef- to the depth of 14.6 m by different methods.
fective stress next to the pile after setup The first pile was driven by a hammer, the
and prior to loading is usually assumed to second was installed by a water jet with
be related to the effective overburden stress. internal return fl.ow and final set of approx-
Kraft (1991) concludes that the rate of in- imately 4.1 m obtained by driving method,
crease in unit resistance, especially toe re- the third was jetted with external return
sistance, decreases with an increase in fl.ow and the final set of approximately 2.0
overburden stress in homgeneous sands. m obtained by driving, and the last pile was
Furthermore, in most cases, pile-driving re- installed entirely by jetting with the exter-
sults in residual stresses in the pile-soil na! return fl.ow. All piles were tested within
system. Essentially, where relevant, the 2-4 days after installation. The values of K
residual shear stress at the pile-soil inter- obtained from these tests, corresponding to
face should be considered in calculating the () = 30°, were equal to 0.4, 0.33, 0.2 and
pile load capacity. 0.05, respectively. These values clearly
The existence of residual stresses in demonstrate the effects of the pile installa-
driven piles after they have been installed tion method on the state of stress within
has been recognized for some time. the soil mass and, consequently, on pile
Holloway et al. (1978) proposed a method capacity. It should be noted, however, that
Piled Waterfront Structures 803

the aforementioned dramatic changes in the


1 1
400

values of K can be related only to the spe- 1


7
cific soil conditions which are present at the 350
e Kahl, et al. (1968) i
test site.
~- O Kerisel (1961) o 1•
o
.g "c:- t:. Muhs and Weiss 119711
"'1,--...
1 •
And again, as reported by Tsinker (1988), ~
300 ~- x Melzer (19681
q,
~

/
the capacity of jetted piles can be improved 5f
;;;
dramatically when installed to their final o.

.s"' 250
set by the dynamic method. Normally, the
degree of increase in the K value depends
:'Î
c
.,.·
200 "'
iii"'
l
on the local soil conditions as well as the ~ o J_
characteristics of the pile-driving equip-
·;;;
~
/o
"' /
ment. 8 li.'
150
il
e X
~
tl
.§'
~ 1
,/X
100
~
;3 o.

7.6.2 Piles Under Axial Static


E
o
'-' /Very loose
50
Load
o
)J~~~
7.6.2.1 Soi/ Parameters 30 35 40 45
Angle of intern al friction, +', in degrees
Soil parameters that are used for pile
design include soil strength parameters, Figure 7-53. Approximate relationship between lim-
iting static cone resistance and angle of interna} fric-
soil-pile adhesion and soil elasticity.
tion of sand (ltsf = 95.8 kN jm2 ). [From Meyerhof
(1976).]
7.6.2.1.1 strength parameters
These are angle of effective internal fric- The undrained shear strength, cu, of a
tion, 4> for cohesionless soils and the
1
, soil is usually obtained from laboratory
undrained shear strength, cu, for cohesive testing of undisturbed soil samples, or by
soils. These soil properties are obtained by field vane shear tests, where for a meas-
laboratory testing of soils for which stand- ured torque, T, the value of cu is obtained
ard methods are used. The guidelines for from
standard testing procedures are found in
miscellaneous testing manuals and other (7-14)
relevant publications [e.g., Lambe (1951),
Terzaghi and Peck (1967), NAVFAC where k is a constant, dependent on the
(Design Manuals DM 7.1 and 7.2) (1982), shape and dimensions of the vane (m 3 ).
American Society for Testing and Materi- According to ASTM D2573 American Soci-
als, (1988), Lowe and Zaccheo (1991) and ety for Testing and Materials (1989), k =
others]. 0.00000366D 3 , where D is the measured
The angle of internal friction, 4> can
1
,
diameter of the vane (cm).
also be obtained from correlations estab- For normally consolidated natural de-
lished from: the standard field penetrometer posits, cu can also be estimated as proposed
test values (e.g., N or qc, where N is the by Bjerrum and Simons (1960):
observed standard penetration test value
and q c is the cone penetration resistance Cu = u~(0.1 + 0.004Pl) (7-15)
measured in the cone penetration test). The
example of the relationship between q c and where
the soil angle of internal friction, 4> 1 , is u~ = effective vertical overburden pressure
illustrated in Figure 7-53. PI = plasticity index
804 Piled Waterfront Structures

7.6.2.1.2 Soii-PileAdhesion (caJ materials (e.g., not prestressed significantly


to pressures above the present in situ over-
In some references this is also termed
burden pressure), the CPT bearing capac-
side or skin friction. As pointed out earlier,
ity, qc, has been correlated by DeBeer
the estimation of unit soil-pile adhesion is
(1965) and Webb (1965) with modulus of
complex and is dependent on factors such
elasticity, E.:
as soil characteristics, method of pile instal-
lation, pile material, type of loading, and (7-16)
other factors. There are several theoretical
methods that are used for evaluation of The Canadian Foundation Engineering
soil-pile adhesion values. Reliable values of Manual (1985) defines E. as the "apparent
ca, however, can only be obtained by per- modulus of elasticity" and recommends
forming full-scale pile load tests in the field. the following relationship proposed by
For a preliminary design, the values of ca Schmertmann (1970):
for piles driven into clayey soils can be
obtained from Figure 7-54. (7-17)

7.6.2.1.3 Elastic soi/ parameters where C1 is equal to 1.5 for silts and sand,
2 for compact sand, 3 for dense sand, and 4
The most common elastic soil property for sand and gravei.
required for pile design is its modulus of For cohesive soils, the value of E. accord-
elasticity, E •. It can be determined from the ing to the Canadian Foundation Engineer-
cone penetrating test (CPT) which provides ing Manual (1985) can be estimated from
a convenient and rapid way of measuring the following relationship:
the soil bearing capacity and soil properties
through correlations at various depths. In (7-18)
the case of normally loaded cohesionless
where
Pc = preconsolidation pressure
1.0
c2 = 60 for firm clays, and 40 for soft clays
0.9

0.8 , . - - - Concrete and timber piles


Jamiolkowski et al. (1985) present a com-
plete review ofthe problems associated with
0.7
the obtaining of the deformation moduli
0.6 from CPT, and Meigh (1985) provided a
··.. good summary of recent work using CPT
0.4
_,.._.)· .•...•.•.••..•... tests for estimating sand deformability.

0.3 7.6.2.1.4 Calcareous sands


0.2 These are commonly found in the coastal
0.1 areas of Australia, India, Saudi Arabia, the
Mediterranean, and the Caribbean. The
oL-~--~~~--~~~~~-------
o 500 1000 1500 2000 3000 4000 construction of miscellaneous marine struc-
tures on continental shelves and in coastal
Very Soft fm1 Stiff Very stiff Hard
soft areas very often require the installation of
Figure 7-54. Variation of cafsu with cu for driven
piles in the calcareous sands. This, how-
piles [developed by Prakash and Sharma (1990) from ever, has proved to be extremely trouble-
data in Tomlinson (1963)]. some, particularly for driven piles. The
Piled Waterfront Structures 805

problems associated with pile performance For comparison, API RP 2A (American


in calcareous sands have been extensively Petroleum Institute, 1989) recommends a
discussed in the past 20-25 years. A de- limiting shaft resistance for medium-dense
tailed discussion on the state of the art is silica sand of about 80 kPa.
given in Murff (1987). Pile end-bearing resistance obtained from
Calcareous sands are soils that are basi- pile load tests show a wide range of values.
cally composed of sand-sized particles of Hagenaar and Van Seters (1985) carried
calcium carbonate. They originate from bio- out tension compression tests on the same
logica! processes such as sedimentation of piles and found the end bearing on the
skeletal debris and coral reef formation, or order of 23-36 MPa. Gilchrist (1985) car-
from chemical precipitation of particles such ried out similar tests on a plugged pile and
as oolites. More specific definitions of cal- obtained a unit end resistance of only
careous sands are found in the literature 1.3-1.8 MPa. For comparison, a limiting
[Clark and Walker, 1977; Bates and design value for a medium dense silica sand
Jackson, 1980; and others as listed by Murff recommended by API RP 2A (American
(1987)]. Petrolium Institute, 1989) is 4.8 MPa. API
The brittle , crushable nature of calcare- also suggest that great caution must be
ous sands immensely complicates site in- exercised in determining the load capacity
vestigations and laboratory testing proce- of piles driven in calcareous sands. Agarwal
dures. The physical properties of calcareous et al. (1977) recommend that sands with
sands include both the characteristics of the greater than 45% carbonates be treated as
bulk soil mass and the individual particles. calcareous and those with less than 30%
An important bulk characteristic is cemen- carbonates be treated as the "conventional
tation which affects pile installation signif- sands."
icantly. Normally, it can lead to a gap be- Other investigations are much more cir-
tween the pile and soil, thus precluding cumspect about the definition of calcareous
development of lateral stresses on a driven sands. However, there is rather widespread
pile. In fact, this is a reason for the low agreement that sands with very high car-
observed shaft resistance. Furthermore, bonate content (say 80% and over) clearly
cyclic loading can result in further signifi- act as calcareous and have the potential of
cant reductions in the pile shaft resistance. abnormally low capacities.
The assessment of cementation effects is McClelland (197 4) suggests an interim
often very difficult. Typical void ratios of limiting value of 20 kPa shaft resistance
calcareous sands is in the range of 0.8-1.4 and 5 MPa end-bearing resistance. He cau-
[Noorany (1985) and others], which is sig- tioned, however, that these values must be
nificantly higher than the values of 0.4-0.8 treated with caution. Agarwal et al. (1977)
usually found in regular sands. These recommend shaft resistance equal to 28.0
higher void ratios result in high compress- kPa for soils grading between carbonates
ibility of calcareous sands. In general, the and noncarbonates. Some more recommen-
friction angle in calcareous sand is quite dations are given in Murff (1987).
high, typically above 35° and often exceed- Unlike driven piles, the capacity of
ing 50° [Noorany (1985) and others]. grouted piles in calcareous sands in many
Murff (1987) summarizes load tests re- cases have been reported to exceed typical
sults carried out on piles driven in calcare- shaft resistance values for driven piles in
ous sands. AII tests indicated a very low silica sands (Nauroy and LeTirant, 1985).
skin friction value (13.3-17 .O kPa), result- Grouting tend to cement layers of calcare-
ing in pile capacity that typically was only ous sand which subsequently results in a
20-25% of the values used for silica sands. very high shaft resistance. Furthermore, the
806 Piled Waterfront Structures

grout which, is generally pressurized, pene- interior during driving and forms a solid
trates cracks and voids in the calcareous plug. This solid plug has a resistance to
sand. This results in increased strength of movement under static loading that is usu-
these sands and also in increased frictional ally much greater than that of the soil at
resistance. In the case of grouted piles, the the pile end. The ultimate axial pile capac-
shaft resistance is typically selected on the ity, Qu, is expressed as
order of 100 kPa (Nauroy and LeTirant,
1985). (7-19)

7.6.2.2 Piles Subjected to static where


Axial Compressive Load
QP = end-bearing capacity
The basic concept of a static load sup- Qf = frictional capacity along the pile shaft
ported by a pile customarily assumes that W = weight of the pile
this load is shared between the pile tip
(base), QP, and shaft friction, Qf, around There are many publications which pro-
the pile perimeter (Fig. 7-55a). The same vide in-depth discussions on pile design.
assumption applies to open-ended pipe pile Chellis (1961), Kezdi (1975), Meyerhof
because in most practica! cases, soil is forced (1976), Vesic (1977), Poulos and Davis
up into the pipe pile because in most practi- (1980), Tomlinson (1980, 1983, 1994), NAV-
ca! cases, soil is forced up into the pile FAC DM-7.2 (1982) Canadian Foundation
Engineering Manual (1985), and, more re-
cently, Prakash and Sharma (1990) are
some of these state-of-the-art practically
oriented publications. Various empirica! and
semiempirical methods for evaluating the
ultimate axial compression load capacity of
piles have been proposed. However, depend-
ing on the soil conditions, the ultimate pile
capacity determined by methods proposed
(b) by different authors sometimes differ sub-
stantially from each other.
Briand and Tucker (1988) reported re-
sults of field tests carried out on 98 piles
driven into different soils. In this study, the
ultimate load on the pile was defined as
that corresponding to a pile settlement
Q equal to one-tenth the pile diameter, plus
the elastic compression of the pile.
A comparison of test results with analyti-
cal pile capacity, obtained by 13 different
methods for predicting the ultimate axial
pile capacity, has indicated that no one
method can practically accurately predict
the ultimate capacity of piles driven into
Figure 7-55. Basic concept ofload support by a pile:
different soils. However, of the 13 methods
(a) loads acting on a pile; (b) simplified distribution of tests, the cone penetration method
vertical stresses adjacent to a pile in sand. (Bustamante and Gianeselli, 1983) was
Piled Waterfront Structures 807

judged to be the best one. This implies that ficult to predict those changes, the piles's
due to the extreme complexity of soil-pile ultimate bearing capacity is usually esti-
interaction, the pile capacity can only be mated based on original strength and defor-
estimated roughly. Full-scale field load tests mation characteristics of the soil. In cohe-
should, therefore, be carried out to confirm sionless soils, the effect of changed soil
the validity of results obtained by theoreti- properties is accounted for by nondimen-
cal, empirica!, or semiempirical methods. sional empirical coefficient, Nq, and mobi-
Typically, the end-bearing capacity, QP, lized skin friction, { 5 •
and the frictional capacity along the pile Because c = O for cohesionless soils and
perimeter, Q f' are expressed as follows: the term 0.5yBN1 in Eq. (7-20) is small
compared to yDfNq, then this equation can
be rewritten as follows:

where (7-22)
AP = pile point (base) area

cu= minimum undrained shear strength


where u:
is effective vertical overburden
in cohesive soil at the pile point stress at the pile base.
level The semiempirical factor Nq is a function
y = unit weight of soil of the Dr!B ratio, the angle of internal
friction cp, and the method of pile installa-
Ne, N 1 ,
tion. Theoretical values of Nq, as recom-
and Nq =nondimensional bearing capacity
mended by different investigators, are sum-
parameters that are dependent on
the angle of internal friction of soil marized by Coyle and Castello (1981) and
presented in Table 7-2. As seen from the
B = pile width or diameter
data presented in this table the value of Nq
Dr = depth of pile penetration
depends heavily on the theoretical model
used for its computation. Scattered results
(7-21) obtained by different workers (Table 7-2)
indicate that the true pile failure mecha-
where nism is not yet well understood. For design
purposes values of Nq recommended by the
p = pile parameter American Petroleum Institute API RP 2A
fs = unit skin friction over length z f (1989) (Table 7-3) are commonly used.
These terms are discussed further in this Skin Friction Capacitv, Qr
section.
The skin friction, { 8 , in Eq. (7-21) is
7.6.2.2.1 Pile capacitv in usually expressed as
Homogeneous
Cohesionless Soils fs = Ca + a~ tan o (7-23)

where
End-Bearing capacitv, Qp
ca = unit adhesion; for cohesionless soils ca = O
As pointed out earlier, the pile capacity a~ = effective horizontal stress along the pile
is influenced by many factors (e.g., soil den- shaft; af. = K a:, where a: is the effective
sity, angle of internal friction, etc.). Change vertical stress in soils surrounding the pile
of soil parameters due to pile installation and K is the coefficient of soil lateral
may result in profound changes in the pressure
soil-pile interaction. Because it is very dif- o= angle of friction between the soil and pile
808 Piled W aterfront Structures

Table7-2. Bearing capacity factors for piles in cohesionless soils

Approximate Nq Values for


Various Friction Angles, cp ', in Degrees

Theories• 25 30 35 40 45
DeBeer (1945) 59 155 380 1150 4000
Meyerhof(1953)
Driven piles 38 89 255 880 4000
Caquot-Kerisel (1956) 26 55 140 350 1050
Brinch Hansen (1961) 23 46 115 350 1650
Skempton, Yassin, and
Gibson (1963) 46 66 110 220 570
Brinch Hansen (1951) 32 54 97 190 400
Berezantsev (1961) 16 33 75 186
Vesic (1963) 15 28 58 130 315
Vesic (1972); U, = 60 b 20 27 40 59 85
1, = 200 b 29 46 72 110 165
Terzaghi (1943)
General shear 12.7 22.5 41.4 81.3 173.3
Localized shear 5.6 8.3 12.6 20.5 35.1

aVarious references are cited by Vesic (1972, 1977).


bRigidity factor.
Saurce:" From Coyle and Castello (1981).

Table 7-3. Design parameters for cohesionless siliceous soi!

Soil-Pile Limiting Unit


Friction Limiting Skin EndBearing
Soi! Angle, 5 Friction Values Values
Density Description (degrees) kipsjft 2 (kPa) Nq kipsjft 2 (MPa)

Veryloose Sand 15 1.0 (47.8) 8 40 (1.9)


Loose Sand-silt"
Medium Silt
Loose Sand 20 1.4 (67.0) 12 60 (2.9)
Medium Sand-silt"
Dense Silt
Medium Sand 25 1.7 (81.3) 20 100 (4.8)
Dense Sand-silt"
Dense Sand 30 2.0 (95.7) 40 200 (9.6)
Very dense sand-silt"
Dense Gravei 35 2.4 (114.8) 50 250 (12.0)
Very dense Sand

Note: The parameters listed are intended as guidelines only. Where detailed information such as in situ cone
tests, strength on high quality samples, model tests, or pile-driving performance is available, other values
must be justified.
• Sand-silt includes those soils with significant fractions of both sand and silt. Strength values generally
increase with increasing sand fractions and decrease with increasing silt fractions.
Source: From American Petroleum Institute (1989).
Piled Waterfront Structures 809

{a) z,/d vs 121 (b} Ks tan 0'a vs 0 ( c) Valuczs of Ks ton 00 Basad


( Orlve~n Pilczs) on MCZ)'Qrhol { 1976)
20 3·0 1·6

1 1

1
15 2·5 1·2

~/
Oriwn
.o (9_0 pilczs

.
10 2·0 $ c
0·8
c
V 2
'" ,.,-
1
.2 ,0ck<Zd
1 Pl~
V
N
5 1·5 "' / 0·4
/

V / 1,-~orod
piis
o 1 /. o
/
28 33 38 43 28 33 38 43 30 35 40
121" 0' ,
For drivan piiGs 0 - 3~ 01•10 (Fig.3.10a,Fig.3.10b)
For borad pilczs,~ o:0; -3 (Fîg.3.10a); 0,.0; (Fig.J.TOc)
WhQ:ra ~· anglcz of intczrnal friction prior to
installation of pile:

Figure 7-56. Values of zc/d and K tan <P~ versus <P for piles in sand.
[From Poulos and Davis (1980).)

For design purposes, the American Subsequently, values of Qf and QP be-


Petroleum Institute API RP 2A (1989) rec- come constant (or clase to this) beyond a
ommends values of 8 provided in Table critica! depth Zc. This has been observed by
7-3. In most design cases, a theoretical Vesic (1967) and reported later by Meyerhof
value of 8(2/3)4> can be considered. Accord- (1976). Recently, Altaee et al. (1992a, 1992b)
ing to Lehane et al. (1993), the value of 8 is reported results of tests carried out on in-
independent of the relative density of the strumented precast concreted piles driven
soil. into sand deposits and subjected to com-
For cohesionless soils, Equation (7-23) pression and puii-out tests. The observed
can be transformed into the following: load distributions appeared to suggest the
existence of the aforementioned critica!
fs = Ka~ tan 8 (7-24) depth Zc. However when the load data was
supplemented with the residual load exist-
As discussed in the previous section, exten- ing in the pile before the start of the test,
sive research (Vesic, 1967; Kerisel, 1961; the appearance of the critica! depth disap-
Hanna and Tan, 1973; Lehane et al., 1993; pears. Instead, the analysis of the tests
and others) has revealed that the unit shaft showed that the load distribution was a
and base resistances do not necessarily in- function of the effective overburden stresses
crease linearly with depth, but instead in the soil over the full pile length.
reach an almost constant value beyond a Most recently Lehane et al. (1993) di-
certain depth. In other words, it has been rectly observed a reduction in with the u;
found that the effective vertical stress, u;, increasing Dr!R ratio, where R is the pile
adjacent to the pile is not necessarily equal radius, and provided a rational explanation
to the effective overburden pressure and, in for the tendency of the average shaft shear
general, can be approximated by the dia- stress to stabilize with depth. Poulos and
gram illustrated in Figure 7 -55b. This ide- Davis (1980) evaluated Vesic's (1967) and
alized approach assumes that the effective Meyerhofs (1976) test results and pre-
vertical stress, u;, adjacent to the pile is sented values of the dimensionless ratia
equal to the overburden pressure to some Zcfd (d = pile diameter) and K 8 tan 4>~
critica! depth Zc, beyond which it remains (equivalent of K tan 8) in graphic form (Fig.
almost constant. 7 -56). As pointed out by Poulos and Davis,
810 Piled Waterfront Structures

Vesic's tests were carried out on steel tube Table 7-4. Values for K. for various pile types
in sands
piles; therefore, data depicted in Figure
7-56 may need further refinements to be Pile Type
applicable to other pile materials.
Boredpile 0.5
A critica! component in the evaluation of
unit skin friction, r.'
is the coefficient of soil
Driven H-pile
Driven displacement pile
0.5-1.0
1.0-2.0
lateral pressure, K. Severa! investigators
Source: Data from Meyerhof (1976) and NAVFAC (1982).
attempted to evaluate the coefficient of lat-
eral pressure to be used in computing of
pile shaft capacity (Ireland, 1957; Sowa, where
1970; Meyerhof, 1973; Levacher and Q( = pile skin friction capacity in the upper zone
Sieffert, 1984; and others). Recommended (depth Zc)
values of K to be used for evaluation of pile Q( = pile skin friction capacity in zone below Zc
shin friction capacity vary from K 0 to K P
and more, where K 0 is the coefficient of Theoretical values of Qr' and Q{ can be
lateral soil pressure "at rest" and K P is the obtained from the following expressions:
coefficient of soil passive pressure. The
above investigators point out that different (7-26)
techniques used for pile installation may
have a profound effect on the value of K. and
Levacher and Sieffert (1984) proposed the (7-27)
K value estimate from K = K 0 Km 0 , where
K 0 is the coefficient of lateral soil pressure
where a: = yZc, where 'Y is the unit weight
"at rest" [K 0 = (1 - sin <f>)] and Kmo is the of soil. Hence,
coefficient which accounts for the method of
pile installation. They conclude that Kmo =
2 to 3 may cover ali practica! cases of pile
installation. Subsequently, at the ground The allowable pile capacity is obtained by
level fs =O and at depth Zc and below, the applying a factor of safety which is usually
skin friction can be expressed as r. = taken as 2-3; the value of the factor of
Ktan 8a:. safety depends on the geotechnical complex-
Where detailed information, such as re- ity of the foundation, method of pile instal-
sults of pile (large model or field) test, in lation, as well as on the importance of the
situ cone tests, or pile-driving performance structure (e.g., is it permanent or is it de-
characteristics are not available, data pro- signed to serve just a short period of time?).
vided by the American Petroleum Institute The factor of safety does not apply to pile
(1989) (Table 7-3) or that recommended by weight.
NAVFAC (1982) (Table 7-4) can be used as As pointed out earlier, the capacity of the
a guideline for estimation of the pile limit- pile can be evaluated empirically by utiliz-
ing skin friction and values of K in sands. ing standard field tests based on standard
On the hasis of the preceding discussion, penetration tests and static cone penetra-
the pile frictional capacity, Q,, can be con- tion. Sometimes, a pressure-meter test is
servatively estimated from the following ex- utilized for the same purpose. Discussion on
pression. these methods are given in standard texts
on foundation engineering or guidelines
such as Canadian Foundation Engineering
(7-25) Manual (1985) and the like.
Piled Waterfront Structures 811

Dynamic Driving Resistance provide useful information to field person-


nel. Ideally, results obtained by pile-driving
Pile capacity can be estimated on the
hasis of dynamic driving resistance by us- formulas should be confirmed by pile load
ing pile-driving formulas, the wave equa- tests. Furthermore, pile-driving formulas
tion approach, or use of a pile-driving ana- can be calibrated on the hasis of pile load
lyzer (PDA). tests to suit the particular construction
Pile-Driving Formulas. In North site. NAVFAC (1982) recommends the pile-
America, the Engineering N ews Record driving formulas presented in Table 7-5.
pile-driving formula and its modified ver- Wave Equation Analysis. Smith (1962)
sions for estimation of pile capacity are pop- proposed an improved approach to dynamic
ular and often used. However, studies by evaluation of pile capacity which is based
Frageszy et al. (1985, 1986) and Lawton et on a one-dimensional stress wave propaga-
al. (1986) indicate that evaluation of pile tion. This equation is used to obtain the pile
capacity on the hasis of these formulas is capacity and the driving stresses and to
not always reliable and in general must be determine equipment compatibility. Accord-
used with caution. The advantages and ing to this method, the hummer (ram), the
shortcomings of dynamic pile-driving for- capblock, the pile cap, and the pile are rep-
mulas are discussed by Goble and Rausche resented as a series of weights and springs.
(1980). It is assumed that when a driving load is
The aforementioned studies reveal that applied to a pile, the soil compresses elasti-
most dynamic formulas have been devel- cally a certain distance, then fails plasti-
oped for specific pile types, driving equip- cally with a constant ultimate resistance.
ment, soil types, and driving conditions and On removal of the load, an elastic rebound
therefore cannot account for the variety of occurs.
pile-driving conditions that may exist at the Smith developed a mathematical formu-
specific project site. Nevertheless, despite lation suitable for numerical computation of
shortcomings of dynamic pile-driving for- pile capacity and driving stresses. It is found
mulas, they are widely used because they in standard texts or handbooks on founda-

Table 7-5. Basic pile-driving formulas

For Single-Acting For Double-Acting


For Drop Hammer Hammer Differential Hammer

a b 2WH b 2E
(Qu)all = S + 0 _1 (Qu)all = S + 0.1
2WH 2E
(Qu)!l{ = S + O.l(Wn/W) (Q)~n = S + 0.1(Wn/W)

Note: Ratio of driven weights to striking weights should not exceed 3.


(Q u>au = allowable pile load in pounds
W = weight of striking parts of hammer in pounds
H = effective height of fall in feet
E = actual energy delivered by hammer per blow in foot-pounds
S = average net penetration in inches per blow for the last 6 in. of driving set
Wn = driven weights including pile
• This is based on the most commonly used pile-driving formula, known as the Engineering
New Formula N.
b Use when driven weights are smaller than striking weights.
c Use when driven weights are larger than striking weights.

Source: From NAVFAC (1982).


812 Piled Waterfront Structures

tion engineering. The major advantage of quake, and damping factors, as well as the
wave equation analysis is its increased ac- driving stresses along the pile shaft. A
curacy over dynamic formulas. The analysis major drawback of both the PDA and
can be used to select the best pile- CAPWAP approaches is their relative com-
cushion-hammer combination and to maxi- plexity that necessitates use of specially
mize driving efficiency, and it can also be trained personnel to operate the equipment
used for prediction of driving stresses. Nat- and run the computer program.
urally, the accuracy of the analysis depends To conclude the discussion on the PDA
on the input of realistic soil model parame- approach, it should be noted that, most re-
ters of in situ soils. cently, Liang and Sheng (1993) introduced
Since the pioneering work of Smith, a an alternative method to determine the
multitude of computer programs have been Smith's wave equation parameters, includ-
developed. More notable programs in N orth ing the static soil resistance to pile-driving
America include the Texas Transportation and its distribution, the soil quake, and the
Institute (TTI) program (Hirsh et al., 1976) Smith damping factor. This method uses a
and the wave equation analysis program driven-rod for measurement in the high-
(WEAP) (Goble and Rausche, 1980). A strain test (HST). Further discussion on this
:finite-element-type program was also pro- new approach is beyond the scope of this
posed by Lee et al. (1988), Chow et al. (1988), text.
and, more recently, Liang and Husein
(1993), who present a modi:fied dynamic 7.6.2.2.2 Pile Capacity in Cohesive
method that is based on the concept of en- Soils
ergy balance (based on Smith's model). This
Typically, pile bearing capacity in cohe-
approach estimate shaft resistance of a
sive soils is critica! in the short-term be-
driven pile using the pile-driving record
cause strength of cohesive soils (e.g., clay)
(and blow-count pro:file and the transferred
normally increases in the process of consoli-
hammer energy). A detailed explanation of
dation. Therefore, for piles installed in co-
pile-driving analysis by the wave equation
hesive soils, consideration of undrained con-
and computer program to carry out these
dition is critica!. This implies the following
analyses is given in Samson et al. (1991).
Pile-Driving Analyzer (PDA). The soils parameters are applicable: cfJ = O, su =
PDA is an extension of the wave equation cu = c, N1 = O, and Nq = 1, where cfJ is the
analysis technique of piles. It is based on angle of interna! friction, su is the
data obtained from measurements of ham- undrained shear strength, cu is the cohe-
sion parameter under undrained condition
mer andjor pile performance during pile-
driving or during restrike. To obtain the (c/J = 0), and N 1 and Nq are the nondimen-
sional bearing capacity parameters. Hence,
required measurements reusable transduc-
Eq. (7-20) becomes
ers, along with a suitable data acquisition
system, have been developed to enable the
computation of pile-driving stresses, pile in- (7-29)
tegrity, hammer performance, and pile
capacity (Rausche et al., 1985). The Based on the cfJ =O, assumption, the shaft
wave equation based computer program unit skin friction, { 8 , as expressed by Eq.
(CAPWAP) developed by Rausche et al. (7-23) under the condition 8 = (2j3)4J = O,
(1972) is a further numerica! analysis will be equal to ca; therefore, the pile fric-
method for con:firming the pile capacity cal- tion capacity can be expressed as follows.
culated with data from the PDA.
The results obtained from CAPWAP in-
(7-30)
clude the static soil resistance distribution,
Piled Waterfront Structures 813

The ultimate bearing capacity, Qu, for a ate vicinity of the driven pile is displaced
pile subjected to compressive loads in cohe- and usually get remolded a certain distance
sive soil can then be obtained from from the pile side. In this zone, the pore
pressure caused by pile-driving normally
dissipates quickly and the soil may regain
much of its original shear strength after
consolidation. This, however, may not occur
where in very sensitive or stiff overconsolidated
AP = pile end (tip) area clays. In these soils, because of loss of soil
cu = minimum undrained shear strength of clay structure, the resulting shear strength may
at pile point cu = su = qu/2, where qu is be less than that in the undisturbed state.
the unconfined compressive strength; for This normally is not applicable to bored
reference, consult Table 7-6 piles. As reported by Meyerhof (1976), in
Ne = bearing capacity factory; for most practica! the case of bored piles there is no signifi-
cases, Ne = 9 is used; in the case of drilled cant changes in soil shear strength with
piles Ne = 6 is usually considered time.
Dr = effective pile length In general, for most practical purposes,
ca = soil-pile adhesion (Fig. 7-54) shear strength of a bearing stratum consist-
p = pile perimeter ing of low to medium sensitive homoge-
y = soil unit weight neous clay can be considered unchanged by
Nq = bearing capacity parameter pile-driving (Prakash and Sharma, 1990).
W = weight of pile Hence, shear strength (c) values used for
the estimation of pile bearing capacity could
In Eq. (7-31), some parameters need be obtained directly from laboratory tests
further clarification. carried out on undisturbed clay samples. In
Undrained Shear Strength of Bear- highly plastic soft clays, the undrained
ing Stratum (c =cu)· Soil in the immedi- shear strength should be obtained from the

Table 7-6. Correlation between SPT penetration N values and estimated range of soil shear strength, S

Estimated Range of su =cu


SPT Penetration N Values Estimated Consistency kPa kipsjft 2

<2 Very soft (extruded < 12 < 0.25


between fingers when squeezed
2-4 Soft (molded by light 12-25 0.25-0.50
finger pressure)
4-8 Firm or medium (molded by 25-50 0.50-1.00
strong finger pressure)
8-15 Stiff (readily indented by 50-100 1.00-2.00
thumb but penetrated only
with great effort)
15-30 Very stiff (readily 100-200 2.00-4.00
indented by thumbnail)
> 30 Hard (indented with > 200 > 4.00
difficulty by thumbnail)

Source: From NAVFAC (1982).


SPT is standard definition for "soi! penetration test."
814 Piled Waterfront Structures

field vane tests (Bengtsson and Sallfors, (a) (b)


1983).
Soil-Pile Adhesion (c). As suggested
by Meyerhof (1976), the average value of ca
for· a homogeneous saturated clay is a func-
tion ofthe average value ofundrained shear
strength, cu, within the pile embedment
length, n,, and the ratio cafcu depends on
clay strength, pile size, installation method,
and rheological (time-dependent) properties
ofthe clay.
-
Where the pile penetrates severa! dif-
r--
ferent layers, the soil-pile adhesion, ca, can
c/ay sand
be approximated by the weighted average
·values related to the individuallayers. Q
EffectivePile Length (D,). This is de-
fined as the length that contributes to the
frictional capacity of the pile. It may be
smaller than that indicated in the design
drawings because, in some cases, the top
layer of soil may not be in close contact with
'-- ____ .....__

the pile due to such factors as disturbance Figure 7-57. Pile in layered soil system: (a) pile
caused by pile installation, erosion by natu- bearing on sand; (b) pile bearing on clay.
ral and ship-propeller-induced currents,
wave action, and other reasons. case, the pile ultimate capacity will be esti-
mated from
7.6.2.2.3 Pile capacity in a Layered
Soi/ svstem Qu = pDr< 1>Ktan ?>a-; +Apa-;Nq
This situation in simplified form is illus- = a-;(pDf(l>K tan ?> + ApNq) (7-33)
trated in Figure 7-57. It exhibits two char-
acteristic cases: In case (a), the pile shaft is In a situation where the pile shaft is located
located mainly in clay and its lower part mainly in a layer of sand and is resting on a
(tip) is resting on a sand layer: in the case layer of clay (Fig. 7-57b), the ultimate pile
(b), the situation is reversed. Consequently, capacity is expressed as follows:
the ultimate pile capacity in case (a) can be Qu = pK tan ?>a-;(Df(2 ) - 0.5Zc) + ApcuNc
estimated from
(7-34)
(7-32)
The definition of the various terms in Eqs.
Relationship (7-32) assumes that the rela- (7 -32)-(7 -34) were given earlier in this
tive stiffness and strength of both clay and section.
sand layers are compatible. Conversely, for When a multilayer soil is encountered in
example, if the pile penetrates through a practice, then the multilayer soil system
layer of soft clay into a deep deposit of a can be transformed into a two-layer system
dense sand, the bearing capacity of this pile so as to enable the use of Eqs. (7 -32) and
would be derived mostly, if not entirely, (7-34). For this, ali soil parameters in-
from the frictional resistance and the end- cluded in the above equations should be a
bearing capacities of the dense sand. In this weighted average of the actual soil proper-
Piled Waterfront Structures 815

ties based on thicknesses of respective soil Ds = socketed length of pile in the sound rock
layers. cbs = allowable rock-socket friction; for prelimi-
nary design, values given in Table 7-8 can
7.6.2.2.4 Piles Resting on Rock be used

Where piles are instaiied on intact, good The aiiowable rock-socket friction, cbs> de-
quality rock, the pile bearing capacity will pends on the roughness at the rock-
be governed by the rock strength. The al- concrete interface, which, in turn, depends
lowable load on the pile is usuaiiy deter- on the strength and fracture frequency of
mined by the aiiowable working stress of the rock and the occurrence of clayey
the material in the pile shaft or, in the case weathered rock which may create a coating
of driven piles, by the stresses induced dur- on the socket.
ing the driving process. Naturally, the Qbs value should not ex-
Allowable bearing pressure on unweath- ceed the value of the bond between the pile
ered rock is usuaiiy influenced by joints and and the mortar used for pile socketing.
shear zones. For a preliminary estimate of In some cases where sound hard rock is
the allowable bearing pressure on a bedrock overlaid by fragmented weathered rock, the
foundation the values given in Table 7-7 piles can be driven through the weathered
and 7-8 can be considered. For detailed rock to sound rock.
design, the aiiowable bearing pressure on Rowe and Armitage (1987) suggest that
rock is determined from tests on rock in general soft rock should be regarded as a
cores; the average unconfined compressive very strong soil. For more information on
strength of rock core is determined accord- performance of concrete drilled in sockets or
ing to the most recent relevant national alike elements (e.g., piles and piers), an
standard; in North America, it is done ac- interested reader is referred to Horvath
cording to ASTM D2938-79. This, however, (1980), Williams (1980), Williams and Pells
is not applicable to soft stratified rock, such (1981), Armitage and Rowe (1983), Glos and
as shales or limestones (Canadian Geotech- Briggs (1983) and Tomlinson (1994).
nical Engineering Manual, 1985; Ladanyi
and Roy, 1971; Rowe and Armitage, 1987). 7.6.2.3 Piles Subject to
Where piles are terminated on strong rock, Axial Puii-Out Load
the skin friction is usuaiiy either neglected
or is reduced by a factor of 2. This is mainly The ultimate puii-out capacity, Pu, of a
because when the pile hits the sound hard pile is estimated similarly to a pile in com-
rock, the pile surrounding soil is shattered pression. Naturaiiy, in the former case, the
and may behave as a loose granular pile end-bearing capacity is irrelevant and
material. the weight of a pile, W, is added to the pile
In the case where the overlaying layer of ultimate shaft friction, Qfp· Hence,
soft material does not provide for pile fixity,
the pile can be socketed into the bedrock (7-36)
foundation. In this case the aiiowable fric-
tion capacity on socket parameter, Qbs> is 7.6.2.3.1 Puii-Out Capacitv in
based on the bond along the socketed sur- Cohesion/ess Soils
face and can be expressed as foiiows:
It is customary to assume the pile uiti-
(7-35) roate shaft friction puii-out capacity in the
cohesionless soils as two-thirds of that esti-
where mated for piles subjected to axial compres-
d = socket diameter sive load. The hasis for this are discussed
816 Piled Waterfront Structures

Table 7-7. · Estimates of allowable bearing pressure

Pres umed
Allowable
Types and Bearing
Conditions Strength of Pressure kPa
Group ofRocks Rock Material (tonjft 2 ) Remarks
(a) Massive igeneous and Highto 10,000 These values are
metamorphic rocks (granite veryhigh (100) based on assumption
(granite diorite, basalt, that the foundations
gneiss) in sound condition (2) are carried down to
unweathered rock.
(b) Foliated metamorphic Medium 3,000
rocks (slate, schist) in to high (30)
sound conditions
(1) and (2)
(c) Sedimentary rocks: Medium 1,000-4,000
shale, siltstone, to high (10-40)
sandstone, limestone
without cavities,
thoroughly cemented
conglomerates, all in
sound conditions
(1) and (2)
(d) Compaction shale and Lowto 500
other argillaceous medium (5)
rocks in sound conditions
(2) and (4)
(e) Broken rocks of any kind 1,000
with moderately close (10)
spacingof
discontinuities (1 ft
or greater), except
argillaceous rocks
(shale)
Thinly bedded limestone, See note (3)
sandstone, shale
(g) Heavily shattered or See note (3)
weathered rocks

These presumed values of the allowable bearing pressure are estimates and may need alteration upward or downward.
N o addition has been made for the depth of embedment of the foundation.
Notes
(1) The foregoing values for sedimentary or foliated rocks apply where the strata or foliation are level or nearly so, and,
then only if the area has ample lateral support. Tilted strata and their relation to nearby slopes or excavations shall
be assessed by a person knowledgeable in this field of work.
(2) Sound rock conditiona allow minor cracks at spacing not less than 1 m.
(3) Tobe assessed by examination in situ, including loading tests if necessary, by a person knowledgeable in this field of
work.
(4) These rocks are apt to swell on release of stress and are apt to soften and swell appreciably on exposure to water.

Source: From Canadian Geotechnical Engineering Manual (1985).


Piled W aterfront Structures 817

Table 7-8. Allowable rock-socket friction, skin friction cbs' and end-bearing values
for Bored Piles in Rocks.

Allowable Skin Friction Allowable End-bearing


On Rock Socket Pressure

Type of Rock (kNjm 2 ) (tonsjft 2 )

Manhattan schist 1330 13.9


Black Utica shale
(Montreal) 1120 11.7 2,620 24.4
Black Billings
shale (Ottawa) 1120 11.7
Dundas shale
(Toronto) 1120 11.7 7,850 73.2
Limestone (Chicago) 1716 17.9 10,468 96
Fragmented shale 107 1.1
Widely fissured
hard sandstone 429 4.5

Source: From Tomlinson (1977).

earlier in Section 7.6.1. Hence, 7.6.3 Pile settlement


Settlement of piles in marine structures
may result in significant redistribution of
bending moments and axial and shear
All terms in the above equation are the stresses in the pile-deck system in the case
same as in Eq. (7-28). The allowable pull- of open-type structures or in the pile-
out capacity is obtained by applying a factor superstructure system in the case of a re-
of safety which usually varies from 2 to 3. lieving platform structure. In both cases,
Again, the factor of safety does not apply for the structure is usually modeled as a rigid
the pile weight. frame founded on elastic foundation.
Prediction of pile settlement is very com-
7.6.2.3.2 Puii-Out Capacity in plex because of many factors, as discussed
Cohesive Soils in Section 7.6.1, among which are soil dis-
For cohesive soils, the ultimate pull-out placement andjor disturbance, changes in
capacity of a pile is essentially equal to the soil properties and state of stresses, method
ultimate skin friction capacity, Qf, plus the of pile installation, and others. It is also
weight of a pile. Hence, complicated by the fact that soil materials
do not show a linear relation between stress
and strain, and settlement is a function of
(7-38)
the relative stress increase. Cohesive soils,
in particular, have a distinct nonlinearity.
The allowable pull-out capacity will then be As pointed out by Fellenius (1991), the
determined by applying the safety factor of larger the existing stress before an addi-
2-3 to the ultimate skin friction capacity tional stress is applied, the smaller the in-
only. duced settlement.
818 Piled Waterfront Structures

Although maximum allowable pile settle- where


ment is usually related to the allowable s. = axial deformation of a pile shaft
settlement of the individual pile, in prac-
tice, however, differential settlement is the
sp = settlement of pile point caused by load
transmitted to it
most important feature in pile foundation
design. U sually the problem is complicated
sps = settlement caused by load transmitted
along the pile shaft
by structural stiffness of both the piles and
the superstructure, load distribution among These three components are determined
piles via superstructure, construction tech- from the following formulations:
niques, and construction sequence.
In practice, three methods are employed
for estimating pile settlement: semiempiri- (7-40)
cal, empirica!, and the pile load test. Natu-
rally, the latter, although expensive and where
time-consuming, provides the most reliable
Q~ = actual point load transmitted to the pile tip
information on pile settlement.
in working stress range (force units); Q~ =
In recent years, theoretical methods, Qp/FS, where FS is the factor of safety
based on finite element method analysis,
have been used more often for computing
Q'r = actual shaft friction load transmitted by
the pile in the working stress range Q{ =
the load settlement relationship. Reference Qr/FS
to these methods and to some available Dr = effective embedded pile length
computer programs were given in the pre-
AP =pile cross-sectional area
ceding section.
EP = modulus of elasticity of pile material
a. = factor which accounts for distribution of
7.6.3.1 Pile settlement in skin friction along pile shaft; Vesic recom-
Cohesionless Soi/ mends a. = 0.5 for uniform or parabolic
skin friction distribution along pile shaft,
In cohesionless soils, the pile displace- and for triangular (zero at the ground level
ment needed to mobilize skin friction is and maximum at the pile base) skin fric-
usually small and may not exceed 5-10 tion distribution a 8 = 0.67. Sharma and
mm. On the other hand, the displacement Joshi (1988) state that estimation of the
required to mobilize pile point resistance is total settlement based on uniform or trian-
usually larger and depends on soil type as gular distribution of skin friction are, in
well as on pile type, size, and method of general, not sensitive to a 8 values.
installation. The latter may result in uiti-
roate skin friction being mobilized sooner
(7-41)
than point bearing. Furthermore, the load
sharing mechanism depends on pile length. and
AII of the above components of the load
transfer mechanism allow for an approxi- (7-42)
mate solution to the problem.
Semiempirical Approach Proposed where
by Vesic (1977). For design purposes Vesic CP = empirica! coefficient; typical values are
suggests the total pile settlement, st, be provided in Table 7-9
broken down into three components: Q~ = net point load under working conditions or
allowable load
Q{ = pile shaft load under working conditions or
(7-39)
allowable load
Piled Waterfront Structures 819

Table 7-9. Typical values of coefficient CP soil consolidation. The latter occurs basi-
Soil Type Driven Piles BoredPiles
cally due to the fact that the imposed stress
initially carried by the pore water is gradu-
Sand (dense to loose) 0.02-0.04 0.09-0.18 ally transferred to soil skeleton as pore
Clay (stiff to soft) 0.02-0.03 0.03-0.06
pressure dissipates. In some instances, soil
Silt (dense to loose) 0.03-0.05 0.09-0.12
creep (i.e., compression without an increase
Source: From Vesic (1977). in effective stress) may add to the compres-
sion of the soil skeleton. Short-term settle-
qP = ultimate point-bearing capacity qP = ment can be determined by using Eqs.
QP/AP (7-39) through (7-44). The long-term
d P = pile diameter settlement, D...St for normally consolidated
clays can be determined from the following
expression:

In the above estimate it is assumed that the


bearing stratum under the pile tip extends
to at least 10dP below the pile tip, and
the soil below is of comparable or higher where
stiffness.
Empirica! Method. Vesic (1970) pro-
ce = coefficient of consolidation
e 0 = initial void ratio of the soil
posed the following empirica! formulation
for computing settlement of the pile ex- H = depth of clay underlained by relatively
incompressible material (e.g., rock, dense
posed to a working load:
sand, gravei, or the like)
u: = present effective vertical stress at the
middle of the layer (H - %D1)
(7-44)
u:
6. = increased pressure from pile load carried
to soil at the middle of the layer
(H- îD1); the value of c.u: depends on
where assumed geometry of a skin friction dis-
st = settlement of pile head (in in.) tribution diagram along pile shaft as dis-
dP =pile diameter (in in.) cussed in Section 7.6.3.1
Q~ = applied load (in lbs) D 1 = pile length in soil
D1 = pile effective length (in in.)
AP =pile cross-sectional area (in in. 2 ) For more details on pile settlement and
EP = modulus elasticity of pile material (in numerica! examples, the reader is referred
lbjin. 2 ) to recent work by Prakash and Sharma
(1990) and Tomlinson (1994).
As stated by Fellenius (1991), in cohesive
Pile Load Test. This is discussed m
soils the elastic modulus approach for anal-
Section 7.5.1.6.
ysis of pile settlement may not be appropri-
ate. Instead, use of compression index, Cc,
7.6.3.2 Pile settlement in and the original void ratio, e 0 , to determine
Cohesive soils the strain, E, induced in a soil should be
This usually includes short-term (elastic) considered. Fellenius points out that the
settlements that occur as the load is applied cohesive soils may be consolidated to a
and long-term settlements that are due to higher stress than the actual effective
820 Piled Waterfront Structures

stress. This higher stress is called the pre- cantly. Essentially, lateralloads usually act
consolidation stress, u;.
The compression of in combination with axialloads. In practice,
such a soil is much less for stresses below the pile head is either restrained (fixed)
the preconsolidation stress; it can be calcu- within the deck or superstructure, or free to
lated using a compression index, cer• When rotate.
in overconsolidated soils with the final Laterally loaded piles can be broadly
stress known, the strain E is calculated classifi.ed as rigid, semirigid, and fl.exible
according to piles (Figs. 7-58a, 7-58b, and 7-58c),
whereas with respect to the end restraint,
(7-46)
they can be distinguished as free-headed
(unrestrained) and restrained at the bottom
level, at the head, or at both levels (Figs.
where 7-58d and 7-58e).
IIo =original effective stress Meyerhof (1979) suggested that a free-
head pile in a soil media may be considered
II{ = new effective stress
rigid for all practica! purposes if its relative
A weakness of this equation is that the stiffness Kr ~ 10- 2 ; the pile is considered
calculation requires use of three parame- as fl.exible when Kr < 0.01.
ters: Ce, Cen and e 0 • For more information
on pile settlement consult Fleming (1992). (7-47)

where
7.7 LATERALLY LOADED PILES E P = Young's modulus of pile material
IP = moment of inertia of the pile
Eh = average horizontal soil modulus of elastic-
7.7.1 General ity along the pile
n. = effective embedment of the pile
Typical piled marine structures are ex-
pected to transmit lateral loads to the In this case, the effective embedment, D., is
ground. These loads are due to lateral soil defi.ned as the pile minimum depth that is
pressures, mooring forces, ship impact, required to resist the design lateral loads
wave loading, or a combination of some of below which the pile defl.ection can be ne-
these loads. In response to lateral loads, glected for practica! purpose. In the case of
piles are subjected to bending moments and fl.exible piles under eccentric or inclined
shearing forces that need to be evaluated in loads, D. can be approximated by
order to assess the safety against structural
failure of the pile, as well as to determine (7-48)
horizontal displacement of the pile-deck
(superstructure) system. where De is the depth of pile penetration
Where lateral forces are large they may into the foundation soil.
be distributed to the ground by a combina- Essentially, the effective length, D., is
tion of vertical and batter (raking) piles. dependent on pile stiffness, parameters of
This is an effective method of resisting lat- the surrounding soil, pile head condition
eral loads because a large proportion of (e.g., free or fi.xed), loading conditions (e.g.,
these loads are carried axially by both verti- static or dynamic), and whether or not the
cal and batter piles, thus lateral displace- lateral load is accompanied by a moment
ment of the structure is reduced signifi.- andjor a signifi.cant vertical load. Tsinker
Piled Waterfront Structures 821

(a) (b) (c) (d) (e)

Figure 7-58. Classi:fication of laterally loaded piles: (a) rigid; (b) seini-
rigid; (c) flexible; (d) unrestrained; (e) restrained. 1-Point of rotation
(fixity); 2-hinge; 3-fixed head.

(1972, 1977) found that the horizontal com- mating the pile capacity. However, the
ponent of the inclined load applied to an observed distribution of lateral soil pres-
anchor pile installed in sand considerably sures is nonlinear throughout the pile
increases the pile's pull-out carrying capac- depth and the pressure decreases rapidly
ity; the greater the angle of the applied with the smaller eccentricity and inclina-
tion of the load; for piles in clay, the
pulling out load, the greater the ability of
roughly rectangular pressure distribu-
the pile's shaft to carry the load. Similar
tion was found to be applicable in esti-
results have been reported by Rao and mating ultimate pile capacities.
Prasad (1993).
In the past 15 years, the behavior of piles (iii) In loose sand and soft clay and in the
subjected to eccentric and inclined loads absence of structural pile failure, the ul-
timate lateralload of flexible piles can be
have been extensively investigated by
expressed in terms of an equivalent rigid
Meyerhof and his co-workers. The tests pile by using an effective embedment
were carried out on relatively small but depth that depends mainly on the rela-
well-instrumented model piles. A fulllist of tive pile stiffness.
references on these investigations is given
in Yalcin and Meyerhof(1991). Some ofthe
conclusions derived from these model tests Although the aforementioned model tests
are as follows: support the proposed theoretical solution
for determination of the capacity of a pile
exposed to eccentric or inclined loads
(i) The ultimate bearing capacity of a rigid
pile varies with eccentricity and inclina- Meyerhof and his co-work~rs suggest that it
tion of the load; this type of pile loading should be verified by full-scale field tests.
has significant infiuence on the ultimate Structurally, a pile is a shaft (vertical or
bearing capacity of the pile. inclined) embedded into the soil below the
(ii) In piles in sand under eccentric or in- design surface. To resist lateral loads, it
clined loads, the lateral soil pressure can must be buried deep enough into the soil to
be approximated by the classical triangu- ensure reliable pile fixation. Essentially,
lar distribution for the purpose of esti- under lateralloads the horizontal defiection
822 Piled Waterfront Structures

of a vertical fixed field within a soil mass pile length. The input for the solution in-
decreases with increasing depth. cludes the boundary conditions that stipu-
The design of a pile subjected to lateral lates pile restraint at the head consistent
loads primarily includes computation of the with the supported superstructure and
required pile embedment in foundation soil, elastic properties of ali components in-
bending stress developed in the pile shaft, volved (e.g., the soil, the superstructure, and
selection of pile size, and determination of the pile itself).
pile deflection. Lateral loads on most piers The Reese-Matlock approach is general
and solid-type piled wharf structures, such and rigorous. It allows the supporting soil
as relieving platforms, are substantial; for to be treated as a nonlinear material. Its
an economica} structural design to be primary advantage is that it provides an
achieved, the complete solution is usually accurate assessment of the moment dia-
needed. For the preliminary design, at gram along the entire length of the pile,
which stage the soil geotechnical parame- thus allowing the designer to computer pile
ters are usually insufficiently investigated, stresses realistically below the mudline.
a simplifYing concept, such as the assump- Again, the accuracy of these analyses are
tion of a pile fixity within the foundation directly dependent on accuracy of the input
soil, is usually considered as adequate. This data involved.
approach is based on the earth pressure
theory. The most commonly used methods
of this kind are conventional (Tsinker, 7.7.2 conventional Design
1986), Hansen's (1961) and Broms'
(1964a, b) methods. Both last methods are
Methods
usually used for the design of short rigid
7.7.2.1 Lateral Load App/ied at a
piles; however, Broms' method is also appli-
cable for the design oflong piles. An evalua-
Distance Above Ground
tion of these methods in terms of their ap-
Levei
plicability for piled foundation design is As pointed out earlier, the pile is treated
given in Prakash and Sharma (1990). as a column embedded into the soil below
During more advanced design stages and the design surface. The simplest form of
particularly during the detailed design vertical pile loaded with horizontal load is
stages when sufficiently complete informa- depicted in Figure 7-59a. It is assumed
tion on foundation soil geotechnical param- that the pile would rebound to its original
eters is normally available, the analytical position after the load is removed. Natu-
solutions based on the modulus of the sub- rally, the pile must be driven sufficiently
grade reaction approach are often used. In deep into the soil to ensure a reliable fixa-
these methods it is assumed that soil acts tion. The minimum embedment (tmin) re-
as a series of independent linearly elastic quired for pile stability in a cohesionless
springs. The difference equations used for soil is determined from the balance of hori-
these analysis have been generalized for zontal load, P, and soil resistance. It is
computer solution by Matlock and Reese assumed that after the horizontal load, P,
(1960). The method was further extended is applied, the pile would tend to rotate
for piles installed in specific soils such as about its fixity point, O, within the soil mass.
soft clays and sand (Matlock, 1970; Reese et The resulting soil reaction diagram is de-
al., 1974). In this method, the pile is as- picted in Figure 7-59c and represents a
sumed to be divided into small segments, simplification ofwhat would be, in reality, a
each of which experiences lateral deflection nonlinear soil reaction (Fig. 7-59c). Further
and soil resistance in units of force per unit simplification of the soil pressure diagram,
Piled Waterfront Structures 823

KA = coefficient of soil active pressure, KA =


tan 2 (45° - c/J/2)
(a) ( b) K = coefficient that accounts for the increase in
soil resistance due to the so-called "wedge
effect" when the soil is displaced by the
shaft movement; in cohesionless soils, K is
usually assumed to be approximately equal
to 2d, where d = shaft diameter

The value of the required theoretical depth


of pile embedment is obtained by solving
Eq. (7-50) for t 0 •
The minimum required penetration, tmin•
( c)
is computed from the following:

(7-51)

where
(7-52)

Figure 7-59. Pile loaded by horizontal force; conven- where Pto is the intensity of soil resulting
tional design method. [From Tsinker (1986).]
pressure at depth t 0 ; Pto = -y(KP - KA)to;
as shown in Figure 7-59c, is achieved by
adding approximately equal trapezoids E~ = o.5ytscKP - KA)K- p (7-53)
mMNn and m'MNn' to both sides of the
resulting soil pressure diagram (Fig. 7-59c). In practice, tmin "" (1.2 to 1.25)t0 is usually
Thus, the left-hand side of the resulting considered.
diagram will have a triangular shape AnN Bending moments in the pile shaft at any
that is compatible with the ultimate soil depth x below grade is expressed by
resistance diagram; on the right-hand side,
trapezoidal diagram MFGm' is replaced by
the fictionalload E~, applied to the embed-
ded part of the pile shaft at the depth t 0 The location of the maximum bending
(Figure 7-59d). Hence equilibrium requires moment, Mmax• is found by solving
that
P(H + t0) - hCKP - KA)Kt8 = O (7-49) dMx
-- = P- 0.5y(KP - KA)Kx 02 = O ( 7-55 )
dx
from which
from which
P PH
t 3
0 -
6
y(Kp - Ka)K
t0 - 6
y(Kp - KA)K
=o
(7-56)
(7-50)

where and, therefore,


y = unit weight of soil
K = coefficient of soil passive pressure K P = 05
2( 2P )
tan 2 (45° + c/J/2), where cp is the soil angle +3
· ]
P Mmax = p [H y(KP - KA)K (7-57)
of internal friction
824 Piled Waterfront Structures

In the case of a vertical pile with a free value of (KP - KA)K in Eqs. (7-50) and
head acting as a cantilever beam from the (7-57) by the corresponding factor of safety.
fixity point, O, the pile lateral displacement, Pile length (H + x 0 ) can also be obtained
y 0 , at point C (Fig. 7-59a) is calculated empirically. One of the empirical methods
from that proved to be reliable for this calcula-
tion is recommended in the Russian
P(H + x0 )3 _ p 0 x6 ( H + x0 ) Standard SNIP II-17-80 (1980). There the
Yo =
3EI 30EI x0 pile length L = (H + x 0 ) from the bottom
of the superstructure to the fixity point, O
H + Xo (P(H + x )2- Pox8) (7-58) (Fig. 7-59) is obtained from the following
3EI 0 10 empirical formulation:
L=H+2jad (7-59)
where
p 0 = intensity of soil lateral pressure at point O where
(Fig. 7-59d); Po = yx 0 (KP - KA)K
(7-60)
E = Young's modulus of the pile material
I = moment of inertia of the pile cross section where
k = proportionality factor in kNjm 4 ; k depends
In the case of a single pile, a safety factor of on the type and condition of the soil; its
2-3 is usually considered. This means that recommended values are given in Table
the design values of t 0 and Mmax are tobe 7-10
obtained either by a corresponding increase bP = effective width of pile in meters; bP =
of value in horizontal load P by the as- 1.5d + 0.5, where d = pile diameter in me-
sumed factor of safety, or by reducing the ters

Table 7-10. Proportionality factor

Proportionality
Soi! Factor k (kNjm 4 )
Plastic to liquid clay and loam
(0.75 <IL :$ 1) 650-2,500
Plastic clay and loam
(O <IL :$ 0.75); plastic sandy loam
(O :$ IL :$ 1); silt (0.6 :$ e :$ 0.8) 650-2,500
Semisolid to solid clay and loam
(O :$IL :$ 0.5); hard sandy loam
(IL <O); fine sand (0.6 :$ e :$ 0.75);
medium sand (0.55 :$ e :$ 0.7) 5,000-8,000
Solid clay and loam;
coarse-grained sand (0.55 :$ e :$ 0.7) 8,000-13,000

Notes:
1. IL = liquidity index; e = void ratio.
2. Smaller values of proportionality factor k correspond to higher
values of IL and e. Similarly, higher values of k correspond to
smaller values of IL and e.
3. For dense gravei, dense graveljsand mixture, and very dense sand
(e > 0.55), values of k that are 30% higher than indicated for
coarse-grained sand (0.55 ;s; e ;s; 0.7) should be used.

Source: From Tsinker (1986).


Piled Waterfront Structures 825

From a pile :fixity point of view the required those depicted in Figure 7-65). Basic design
penetration depth of a conventional steel or parameters as depicted in Figure 7-60 are
concrete pile should not be less than 5.0 m; obtained from the following formulations:
for timber piles this depth can be reduced to
3.0m.
It should be added that in most practica! (7-61)
situations the design of piles to resist lat-
eral loads is based on lateral displacement (7-62)
criteria rather than the ultimate lateral
capacity. where
EI = pile stiffness
7.7.2.2 Lateral Load App/ied
at the Ground Level kh stiffness of the spring; for details, consult
=
the following sections
The empirica! method is usually used in d = pile diameter or side
the case where piles are employed to sup-
port a gravity-type structure constructed on
soft foundation materials. When the pile Maximum values of pile deflection, y 0 , soil
foundation consisting of identica! piles is reaction, p 1 , and bending moment M 1 and
symmetrical in both directions, the horizon- M 2 as indicated in Figure 7-60 are com-
tal load per pile, P, can be determined as puted from the following expressions:
P = Qjn, where Q is the full horizontal
force acting at the gravity center of the pile (7-63)
foundation and n is the total number of
piles. (7-64)
The pile is assumed to be a beam embed-
ded into an elastic soil medium and sup-
ported on a series of infinitely closely spaced
independent and elastic springs (similar to

(a) (b) (c) (d)

'"·~---~
x, P,

v= f(x)

Figure 7-60. Pile with restrained head loaded laterally at ground level:
(a) pile defiection under lateralload; (b) soil reaction; (c) shear diagram; (d)
distribution of bending moment. 1-Pile; 2-deformed line.
826 Piled W aterfront Structures

7.7.3 Broms· Method LoadQ.

This method is based on simplified assump-


tions made on soil pressure distribution
along the pile length. The method treats
both short rigid and long fl.exible piles both

ID
with free heads and fixed heads separately. (a) (b)
In Broms' method, soil is considered either
SB Mmn
as purely cohesionless (c = O) or purely co-
hesive (cf> = 0). .
.

7.7.3.1 Free-Head Short Piles .


in Cohesionless Soils
Soil reactio~~~B ~~:n,
Short piles are defined as those for which Soil reaction
~~~~
LjT::;; 2 or LjR::;; 2 (Fig. 7-61), where L (c) (d)
is the pile embedded length,

_ (EI )
l
1 5
/
T-- (7-66)
nh

R--_ (EI) 1 4
1 (7-67) Bending
~
Soilreaction
kh moment

(e) (f)

where Figure 7-61. Short stiffpiles under lateralload: (a)


nh =constant of modulus of subgrade reaction rotation of free-head pile; (b) translation of fixed-head
pile; (c) soil reaction and bending moment distribution
kh = nhxL for linearly increasing soil modulus
of free-head pile in cohesive soils; (d) soil reaction and
kh with depth XL
bending moment distribution of fixed-head pile in co-
hesive soils; (e) soil reaction and bending moment
For short piles, the possible failure mode distribution of free-head pile in cohensionless soils; (0
and distribution of ultimate soil resistance soil reaction and bending moment distribution of
and bending moments are shown in Figures fixed-head piles in cohensionless soils. [From Broms
(1964a, 1964b).]
7-61a and 7-61e, respectively.
The assumed soil pressure distribution is
depicted in Figure 7-61e. Note that to
where
account for the "wedge effect," Broms
suggests a passive soil pressure on the KP = (1 + sin c/J)/(1 - sin cp)
pile shaft to be increased by three pile B = d = pile diameter or width
diameters.
The triangular soil pressure diagram can
Relationship (7 -68) is plotted using nondi-
be replaced with a concentrated force. Then,
mensional terms LjB versus Qu!KPB 3y
taking moments about the pile toe, the fol-
in Figure 7-63a. From this figure, Q u can
lowing relationship for ultimate lateralload,
be determined, provided the values of L, e,
Qu, is obtained: B, K P, and y are known. The maximum
bending moment in pile shaft, Mmax> occurs
(7-68) at depth x 0 below ground level. At this
point, the shear force equals zero, which
Piled Waterfront Structures 827

gives

(7-69)

From Eq. (7-69), x 0 is found tobe equal to

Qu )0.5
xo = 0.82 ( yBKP (7-70)

Consequently, the maximum bending mo- Oeflection Soil


reaction
Bending
moment
Oeflection Soi!
reaction
Bending
moment
ment, Mmax• is computed as (a) (b)

(7-71)

7.7.3.2 Fixed-Head Short Piles


in Cohesionless Soils
The possible mode of failure for these
piles is assumed by Broms as a simple lat-
eral translation (Fig. 7-61b); the corre-
sponding soil reaction and bending moment
Oeflection Soil Bending Oellection Soil Bending
are depicted in Figure 7-61f. Due to the reaction moment reaction moment

assumed mode of failure, Qu and Mmax for (c) (d)

horizontal equilibrium conditions are deter- Figure 7-62. Long fiexible piles under lateralload:
mined from the following: (a) defiection, soil reaction, and bending moment dis-
tribution of free-head pile in cohesive soils; (b) defiec-
Qu = 1.5yL2 BKP (7-72) tion, soil reaction, and bending moment distribution of
fixed-head pile in cohesive soils; (c) defiection, soil
Mmax = yLsBKP (7-73) reaction, and bending moment distribution of free-head
pile in cohensionless soils; (d) defiection, soil reaction,
and bending moment distribution offixed-head pile in
Figure 7-63a provides a graphical solution cohensionless soil. [From Broms (i964a, 1964b).]
for :fixed-head short piles in cohesionless
soils.

7.7.3.3 Free-Head Long Piles computed from the following equations:


in Cohesionless Soils
Long piles in cohesionless soils are those
for which LjT > 4.0. The assumed failure
mode (de:fiection) for long piles and the dis-
tribution of ultimate soil pressure and
bending moment along the pile shaft are
shown in Figure 7 -62c. Because the maxi-
mum bending moment exists at the point where Mu is the ultimate moment capacity
where the shear force is zero, the value of of the pile shaft; Qu can be determined
x 0 can be obtained from Eq. (7-70), and the from Figure 7-63b by using expression
corresponding maximum bending moment, Qu!KPB 3y as a function of expression
Mmax• and ultimate shear force, Qu, can be Mu/B4yKP.
828 Piled Waterfront Structures

(a) Figure 7-63b provides graphical solution


for fixed-head long piles in cohesionless
soils.
,;..
~ 7.7.3.5 Free-Head Short Pile in
~1~~~~-++4~~~~
Cohesive Soils lL / T ~ 21
1
j ao~---<>-----tl--~'-.4~~ The failure mode and distribution of the
J ultimate soil reaction and bending moment
for these piles is illustrated in Figures
7-61a and 7-61c. According to Broms
o
........ L/B
(1964a), the maximum bending moment is
developed at the point located a distance
(b) (1.5B + x 0 ) from the surface of the ground.
1000 Consequently,
..... + 1.5B + 0.5x 0 )
... Mmax = Q,..(e (7-79)
d•oo
The length of the pile (L - x 0 ) is assumed
ie to resist a maximum bending moment which
1
.11 10
is given by
1 (7-80)
1
o 1.0
where
x0 = Q,../9c,..B (7-81)
Figure 7-63. Graphical solution for ultimate lateral
load capacity of short and long pile in cohensionless
soils: (a) short piles; (b) long piles. [From Broms
where cu is the undrained shear strength of
(1964b). cohesive soil (4J = 0). Equations (7-80) and
(7-81) can be solved for Qu. The graphical
solution for Qu is provided in Figure 7-64a.
7.7.3.4 Fixed-Head Long Piles From this figure, the ratio QufcuB 2 is ob-
in Cohesion/ess SOils tained for a known L jB and e jB from
The assumed failure mode and distribu- which the Q u value is calculated.
tion of ultimate soil pressures and bending
moment along pile shaft are illustrated in 7.7.3.6 Fixed-Head Short Piles
Figure 7-62d. Similar to that previously in Cohesive Soils
discussed, the ultimate shear force Q u and Failure mode, distribution ofultimate soil
the maximum bending moment Mmax are reaction, and bending moment are shown in
determined from the following relation- Figures 7-62b and 7 -62d. The following
ships: relationships are applicable for these piles
2Mu according to Broms (1964a):
Q =---- (7-76)
u e + 0.67x 0
Q,.. = 9c,..B(L - 1.5B) (7-82)

(7-77)
Mmax = 4.5c,..B(L2 - 2.25B 2 ) (7-83)

Qu and Mmax can also be obtained from


(7-78) Figure 7-64b.
Piled Waterfront Structures 829

(a) 7.7.3.8 Fixed-Head Long Piles


in Cohesive Soils
The failure mode and the distribution of
ultimate soil reaction and bending moment
for these piles are shown in Figure 7-62b.
Value of Qu is obtained from

(7-84)

where x 0 is determined from Eq. (7-81).


4 8 12 16 20
Again, the graphical solution is plotted in
Embedment length LI B Figure 7-64b.
(b)

7.7.4 subgrade Reaction


Approach
This approach treats a laterally loaded pile
as a beam on elastic foundations with a
suitably defined boundary condition. In this
method, the structural model of a real pile
is substituted by a short length of a dummy
pile loaded by forces and moment from the
overhead portion of the pile. It is also as-
Ultimate lateral resistance, M.. tc.. B3 sumed that the part of the pile embedded
into the elastic soil medium is supported on
Figure 7-64. Graphical solution for ultimate lateral a series of infinitely closely spaced inde-
load capacity of piles in cohesive soils: (a) short piles;
(b) long piles. [From Broms (1964a).
pendent and elastic springs (Fig. 7-65).
The following discussion is related to the
treatments of the problem as developed by
Reese and Matlock (1956), Matlock and
Reese (1960, 1961, 1962), Matlock (1970),
and Reese et al. (1974, 1975). This is fur-
7.7.3.7 Free-Head Long Piles ther discussed by Smith (1987), Sun (1994),
in Cohesive Soils Ooi and Dunkan (1994), Duncan et al.
(1994), McVay et al. (1995), PIANC (1984),
These piles in cohesive soils conform to Young (1991), and Prakash and Sharma
LjR ~ 3.5. Assume failure mode and the (1990).
distribution of the ultimate soil reaction and More recently Prakash and Kumar (1996)
bending moment for these piles are indi- proposed a method for predicting the load-
cated in Figure 7-61b. Broms (1964a) sug- displacement relationship for single pile
gests that Eqs. (7-79) and (7-81) are also embedded in sand and subjected to lateral
applicable to long piles. The solution in load. This method is based on using the
graphical form is presented in Figure 7-64b load-defl.ection curves that are developed on
from which for a known M ulcuB 3 , the value a premise of the soil modulus degradation
of Qu!cuB 2 and finally Qu can be obtained. with strain.
830 Piled Waterfront Structures

(a) (b) (a) (b) (c) (d) (e)


p

L V =El !!!z
clxl
p=El d'y
dz4

p Figure 7-66. Free-head pile subjected to lateralload


and bending moment: (a) lateral defiection; (b) slope;
(c) moment; (d) shear; (e) soil reaction. [From Reese
and Matlock (1956).]
1

!
X X
~ Reese, 1961, 1962):
Figure 7-65. Laterally loaded pile. Subgrade reac-
tion approach: (a) soil reaction; (b) elastic springs. (7-87)

(7-88)
7.7.4.1 Laterally Loaded
Free-Head Pile in
Cohesionless Soils (7-89)

The behavior of this pile (Fig. 7-66) is


expressed by Vx =AuQg + Bu( ~g) (7-90)

(7-85) Px = AP( ~) + BP( ;: ) (7-91)

where where
E = pile material modulus of elasticity Qg = lateralload applied to pile at ground level
I = moment of inertia of the pile section Mg =moment applied to pile at ground level
p = soil reaction per unit length of pile; p = k h y, EI = pile stiffness
where y is the pile lateral deflection and k h T = relative stiffness factor which is given as
is the stiffuess of the spring, also called the T = (Eljnh) 1 15 , where nx = khjx
modulus of horizontal subgrade reaction
Values of AY, As, Am, Au, AP, BY, B 8 , Bm,
Equation (7 -85) can be modified as follows: Vu and BP for various Z = xjT have been
computed by Matlock and Reese (1961,
1962) by using the finite-di:fference method;
(7-86)
these are given in Tables 7-11 and 7-12.
The above investigators found that the
By considering soil elastic behavior that as- pile with Zmax = 2 behaves as a rigid pile,
sumes a small defl.ection relative to the pile and the pile length beyond Zmax = 5 practi-
dimensions, the above equation can be cally does not affect the defl.ection coeffi-
solved for Yx, Mx, Sx, Vx, and Px as follows cients given in Tables 7-11 and 7-12. Val-
(for definitions, see Fig. 7-66) (Matlock and ues of AY, Am, BY and Bm for different
Piled Waterfront Structures 831

Table7-11. Coefficient A for long piles (Zmax ;;:: 5): free head

z Ay A, Am Av AP
0.0 2.435 -1.623 0.000 1.000 0.000
0.1 2.273 -1.618 0.100 0.989 -0.227
0.2 2.112 -1.603 0.198 0.956 -0.422
0.3 1.952 -1.578 0.291 0.906 -0.586
0.4 1.796 -1.545 0.379 0.840 -0.718
0.5 l.Ei44 -1.503 0.459 0.764 -0.822
0.6 1.496 -1.454 0.532 0.677 -0.897
0.7 1.353 -1.397 0.595 0.585 -0.947
0.8 1.216 -1.335 0.649 0.489 -0.973
0.9 1.086 -1.268 0.693 0.392 -0.977
1.0 0.962 -1.197 0.727 0.295 -0.962
1.2 0.738 -1.047 0.767 0.109 -0.885
1.4 0.544 -0.893 0.772 -0.056 -0.761
1.6 0.381 -0.741 0.746 -0.193 -0.609
1.8 0.247 -0.596 0.696 -0.298 -0.445
2.0 0.142 -0.464 0.628 -0.371 -0.283
3.0 -0.075 -0.040 0.225 -0.349 0.226
4.0 -0.050 0.052 0.000 -0.106 0.201
5.0 -0.009 0.025 -0.033 0.013 0.046

Source: From Matlock and Reese (1961, 1962).

Table 7-12. Coefficient B for long piles (Zmax ;;:: 5): free head

z By B, Bm Bv BP
0.0 1.623 -1.750 1.000 0.000 0.000
0.1 1.453 -1.650 1.000 -0.007 -0.145
0.2 1.293 -1.550 0.999 -0.028 -0.259
0.3 1.143 -1.450 0.994 -0.058 -0.343
0.4 1.003 -1.351 0.987 -0.095 -0.401
0.5 0.873 -1.253 0.976 -0.137 -0.436
0.6 0.752 -1.156 0.960 -0.181 -0.451
0.7 0.642 -1.061 0.939 -0.226 -0.449
0.8 0.540 -0.968 0.914 -0.270 -0.432
0.9 0.448 -0.878 0.885 -0.312 -0.403
1.0 0.364 -0.792 0.852 -0.350 -0.364
1.2 0.223 -0.629 0.775 -0.414 -0.268
1.4 0.112 -0.482 0.688 -0.456 -0.157
1.6 0.029 -0.354 0.594 -0.477 -0.047
1.8 -0.030 -0.245 0.498 -0.476 0.054
2.0 -0.070 -0.155 0.404 -0.456 0.140
3.0 -0.089 0.057 0.059 -0.213 0.268
4.0 -0.028 0.049 -0.042 0.017 0.112
5.0 0.000 0.011 -0.026 0.029 -0.002

Source: From Matlock and Reese (1961, 1962).


832 Piled Waterfront Structures

zmax = LjT are also provided in Figure at X= o


7-67.

(7-93)
7.7.4.2 Laterally Loaded
Fixed-Head Pile in
Cohesion/ess Soils
From Tables 7-11 and 7-12 for Z = xjT =
Essentially, for these piles the slope at O, -(A 8 jB 8 ) = -(1.623/1.75) = -0.93.
the ground level is zero. Hence, from Eq. Thus,
(7-89),

(QgTz)
S =A8 - - + B 8 - -
(MgT) = O (7-92)
X El El Therefore, for a fixed-head pile, Eqs. ( )
7-87

Deflection coefficient, A 1 Moment coefficient. A.


·1 o o +1 o +2 o +3 o +4 o +5 o o +02 +04 +06 +0.8

..d1 ~.:--
..........
--· ,. .........
1-

1.0
~
........... - ..........
Z.,..=2-
......... '"---
-- V
--- 3- tylf1/ "'

~r
~/

4.0
4- r-.
1- r- 5&10
5.0
Coefficients for deflection Coeff~eients for bending moment
(a)

Deflection COeffiCienl. 8 1 Moment coefficient. B.,


02.0 -1.0 o +1.0 +2.0 +3.0 +4.0 +02 +04 +06 +08 +10
..........

1.0
Z,.,..=2- ~.:: .-- . -··
~
1.0
Zrnox=2-
r---.. .........
.J
-,·::::w
-·· t' .. ...... -·- ~-~ ........ ::;;.
.... . IN
, ..., .
..- 31:> -:::. ;.... "..
...; ~2.0

t
3- 107
~ ~ r-4

Mr
~Ys
§ ~ ~-

~r
3.0
~ 4, 5 & 10 f/?
4.0 4.0
·f..- f-4
10
t---Ii v 5 'X

5.0 5.0
Coefficients for deflection Coefficients for bending moment
(b)

Figure 7-67. Coefficients Ay, Am, B 7 , and Bm for piles in cohesionless soils; (a) free-head
piles; (b) fixed-head piles. [From Reese and Matlock (1956).]
Piled Waterfront StniCtures 833

Deflection coefficient, c,
o +0.2 +0.4 +0.6 +0.8 + 1.0 +1.1

.....-:::
'/_... /
''
5&10
b:: ~· ;_-s:; ··\ 1-2
1.0
Zmax=2- 1----- ~~
~· -·· '-- 1-3
........... l~
~- ['--.. t-4
3 ~
~
/
1--'
,/ y

~r
4 ;;1

4.0
v- 10

s-.;
5.0
(a)

Moment coefficient, Cm
-0.8 -0.6 -0.4 -0.2 o
o ·1.0
.......... ... :::::-
+0.2 +0.4

1.0
::=: ~ ...
-
...... _ ... :::::::- ,..._
::-
.........
::-- ........
_.:,., ' K 4,5&10
"l
~ 2.0
Zmax=2- ~
~
3
'-../ i}
8 Q, M / 1

~r
~ 3.0
o 4- :-v V
4.0
lj
1
10 5
5.0
(b)

Soil resistance coefficient, Cp


-o 4 -o 2 o +0 2 +0 4 +0 6 . +0 8

r--r-~--- . [,r4 3
~~~k-2
1.0 1--+--+-t---+---t-1--+--+-1-+--+-f>iv~I<T--~ -v
Zmax =2 -r---._ 5&10 ~ L~ l1

.... \.----~V

rp
"l
<'.,- 2.0 l--t--+-t---+--+-l--t--t'=-1-,_.,-;:4...-.h"3:~1-+----l
~ 3 1"'
f-... - - · . , 1/,"/
~... F""

!~ 3.0 1-"-"f---+~---·-+_-+
__-+-.,...9-~-+·~_·+--+---l ~"·
4 ~z_·1/
··" _..,,
4.0 1--t--+"-t-t--:tt--+-1-t-+--+--l
y-f--10

5.0 5.- '.\


L-..L.-L--'-.L.::......J.J.lL__I._.J...._...L..__L__JL__....!..__ __J

(c)

Figure 7-68. Deflection, moment, and soil reaction coef-


:licients for :lixed-head piles subjected to lateralload: (a)
deflections; (b) bending moments; (c) soi! reaction. [From
Reese and matlock (1956).]
834 Piled Waterfront Structures

and (7-88) can be rewritten under Zmax = LjR ~ 2, the pile behaves as
a rigid shaft, and for Zmax = L / R ;;::: 4, the
(7-94) pile behaves as if it was infinitely long.
In conclusion, it should be pointed out
(7-95) that the lateral capacity of piles calculated
on the hasis of a subgrade reaction can be
where extended beyond the elastic range where a
soil yields plastically. This can be done by
employing so-called p-y curves. The non-
linear p-y in soil spring idealization in a
Values of CY and Cm can be obtained from finite difference scheme (Reese, 1977) or in
Figure 7-68. a one-dimensional finite element method
scheme (Sogge, 1983) are presently very
Note: In cases where the pile head is ex- popular modeling techniques. In this
pected to undergo some rotation at the point method, the pile is divided in elements and
of fixity within the superstructure (partly each element is characterized by a load-
fixed-head pile), the coefficients cy and cm deflection p-y curve. The shape of these
are modified as follows: curves is approximated by a hyperbola with
parameters that are determined by the
CY = Ay - 0.93..\BY (7-96) pile-soil system interaction. In the develop-
ment of the p-y curves, the effects of dy-
Cm =Am - 0.93..\B;n (7-97)
namic loading, · reloading after extreme
loading, scour, as well as effects of other
where A is a factor that accounts for degree piles have to be considered. For a detailed
of pile fixity; for example, A = 1 for 100% discussion, the reader is referred to Matlock
fixity (fully restrained head), A = O for a (1970), Reese et al. (1974), Reese and Welch
fully free head. Accordingly, A= 0.5 and (1975), Bhushan et al. (1979), and Reese
A = 0.25 for 50% and 25% fixity, respec-
(1984). Reese (1977) provides a computer
tively. program documentation COM624 that
!30lves for deflection and bending moment
7.7.4.3 Laterally Loaded Pile for a pile under lateralloading. More recent
in Cohesive Soi/
development in the use of the p-y curves
Similar to the preceding discussion on technique is given in Smith (1987), Young
piles in a cohesionless soil, the lateral de- (1991), Ooi and Duncan (1994), Duncan et
flection of a pile in a cohesive soil, y, and al. (1994), Sun (1994), and Poepsel and
the bending moment at any depth, x, can Dowd (1995). Prakash and Sharma (1990)
be obtained from provide a detailed guideline in the form of a
step-by-step procedure and numerica! ex-
amples which can help the designer to s·olve
(7-98) the problem by developing and using p-y
curves.
(7-99) Finally, it should be noted that proce-
dures for developing p-y curves are differ-
in which case R = (Eljkh)l/ 4 • ent for various soil types and it is helpful to
Values of coefficients A and B can be consult the original works on the various
obtained from curves plotted in Figures criteria before applying the method to the
7-69a and 7-69b. As is seen in this figure, particular design problem.
Piled Waterfront Structures 835

{a)
Deflection coeffiCient A.JC and moment coefficient Amc
-1.0 o +1.0 +2.0

Free head
Q8 =i, M6 =0

(b)
Oeflection coefficient B>< and moment coefficient B=
-2.0 -1.0 o +1.0 +2.0

..
...············

Free head
Q6 =0, M6 =i

4r---------+-------~Hr---------r--------~

5L---------~-----------W~--------~---------J

Figure 7-69. Coefficients of moments and deflections for


free-head pile in soils with constant soil modulus: (a)
coefficients of deflections and bending moment for pile
carrying horizontalload at head and zero moment; (b)
coefficients of deflections and bending moment for piles
carrying moment at head and zero lateralload. [From
Davisson and Gill (1963).]
836 Piled W aterfront Structures

7.7.5 Laterally Loaded


socketed Piles
Soil
A pile socketed in to rock is usually de-
signed to transmit a lateral load and bend-
ing moment to the bedrock, as well as to the
overburden material. In general, an accu-
0<2flczction Soil Rczsistanccz
rate solution to the problem is very diffi.cult
to achieve. A state-of-the-art review and the (a) ShOrt Pilcz

approximate theoretical approach for esti-


mating the ultimate lateral capacity of a
socket is given in Carter and Kulhawy
(1992). These authors assume that the em-
bedded shaft section has sufficient moment
and the shear capacity to resist the design
load and that the ultimate failure of the
shaft occurs when the surrounding rock
mass is not able to sustain any further
loading. These assumptions are critical and
must always be checked. Bcznding
There is very little published information (b) Long Piicz
Momcznt

on the performance of pile sockets in a rock.


One documented case of a field test on con- Figure 7-70. Free-head socketed piles. [From Poulos
and Davis (1980).]
crete piers embedded into rock was re-
ported by Tucker and Askari (1986).
Due to lack of sufficient knowledge on along a pile shaft can be estimated from the
socket performance, as well as difficulties theories described in Chapters 4 and 6. In
with underwater quality control design of these proposed methods, similar to that dis-
socketed piles, the design should be treated cussed earlier, the high pressure near the
with extreme caution. Socket design should pile tip is replaced by a single force, F.
be based on the application of a simple Poulos and Davis (1980) discuss two practi-
model for description of socket-rock inter- ca! situations related to "short" and "long"
action, and a sufficient factor of safety piles. According to these workers in the
should be considered. case of "short" piles, by taking moments
The following procedure for determina- about pile tip the following information is
tion of the ultimate value of lateral load, obtained.
Qu, and maximum bending moment, Mmax>
for piles that are socketed into rock or whose
tip is embedded in a firmer stratum has
Qu = e: L (IoL'Purxdx + ~:+L,Pusxdx)

been developed by Poulos and Davis (1980). (7-100)


A typical situation for a free-head pile is
illustrated in Figure 7-70. In the examples where
depicted in this figure, the assumed failure Pur = ultimate lateral resistance of stiffer
mode and moment distribution pattems are stratum
shown together with an arbitrary distribu- Pus = ultimate lateral resistance of soil
tion of ultimate soil resistance, Pu· In prac- x = vertical distance, measured from the tip
tice, distribution of the lateral soil reaction d = pile diameter
Piled Waterfront Structures 837

Hence, for the known value of Qu, a value precast concrete piles with regular or pre-
of Mmax can be calculated. If Mmax < Mx, stressed reinforcement, pile size (e.g., diam-
the pile will fail as a short pile; if Mmax > eter, penetration, length of unrestrained
M x, the pile will fail as a long one and M max freestanding portion), number of piles per
must be equal to Mx. The position of Mmax bend, requirements for batter piles, poten-
(i.e., the distance f below the surface) is tial for soil erosion around piles, and other
determined by the condition of zero shear factors.
force. Hence, Two basic types of piled structures,
namely open-type piled structure and re-
lieving platforms, are discussed further in
this section.
Then the maximum moment, Mmax> is ob-
tained from 7.8.1 Design Criteria
This is a basic document on which the
structural design is based. It usually in-
cludes a statement in which the purpose of
In Eqs. (7-101) and (7-102), x = vertical the project is clearly stated. This normally
distance, measured downward from the sur- includes the description and volume of cargo
face, and the bracketed form {Pur} applies to be handled, type of cargo handling and
only if f extends below the top of the stiffer hauling technology to be used, type and size
stratum. Similar analysis can be carried out of ship(s) to be served at the dock structure,
for long piles and piles with fixed (re- method of ship handling during berthing
strained) head. For more details, the reader and departing operations, means of protect-
is referred to Poulos and Davis (1980). ing the dock structure from ship impact,
type of mooring accessories to be used,
safety of operation, and others.
7.8 Piled Marine Structures The "design criteria" establishes the ba-
Design Methods sic configuration and sizes of the docking
structure(s). This includes the required
The design of a piled marine structure is length and width of the structure, the depth
normally aimed at the most economica! of water which depends on the size of the
pile-deck (superstructure) system able to design vessel and the required underkeel
effectively resist the design vertical and clearance, and the deck elevation. The lat-
horizontal loads with an acceptable degree ter depends on tide range, storm surge, crest
of both horizontal and vertical displace- elevation of the most probable highest
ments. This includes the design of a stiff waves, ice run-up, effect of air gap, and
and durable deck (superstructure) system others. The discussion on the required depth
supported on a pile foundation, able to sup- of water in front of docking structures and
port the design dead and live loads and their elevation above sea level is given in
effectively resist environmental loads (e.g., Chapter 2.
waves, current, wind, ice, and earthquake). The "design criteria" includes ali static
In the design process, the following items and dynamic loads to be considered in the
are usually considered as an alternative: dock structural design. These typically in-
type of piles [e.g., pipe piles (steel or con- clude dead and live loads, environmental
crete), closed ended or open ended], large- loads, and construction (temporary) loads.
diameter cylinder piles (steel or concrete), Details are given in Chapter 3.
838 Piled W aterfront Structures

Dead loads usually include the weights of mental results, these investigators propose
the structure, permanently installed equip- empirica! formulas for the calculation of the
ment, and miscellaneous appurtenant struc- aforementioned pressures. Shih and Anas-
tures. tasiou believe that the results of their in-
Live loads are those imposed on the struc- vestigation are applicable for practica! de-
ture during its operation. They normally in- sign purposes.
clude all kinds of surcharge loads, weights Construction loads are those resulting
and dynamic loadings produced by all kinds from fabrication, transportation, and instal-
of cargo handling and hauling equipment, and
lation of individual structural components,
forces exerted on the structure due to ship
maneuvers, and while moored at the dock.
or prefabricated blocks.
Although detail geotechnical site condi-
Environmentalloads are those imposed on tions are provided in geotechnical reports,
the structure by natural phenomena includ-
ing wind, current, wave, ice, and earth-
basic information on soil data is also in-
quakes. These loads are all discussed in
cluded in the "design criteria."
Chapters 1 and 3. "Design criteria" normally includes the
proposed method(s) of structural analysis,
load combinations, allowable level of
It should be pointed out that according to stresses in structural components, and as-
established practice, the deck of a piled pier sumed factors of safety. Last but not least,
or a marginal piled wharf is designed to the "design criteria" includes requirements
maintain a sufficiently "safe" clearance be- for construction materials, fabrication and
tween the underside of the deck and the installation tolerances, safety considera-
expected maximum design wave crest ele- tions, method of inspection and testing
vation. Sometimes, due to uncertainties in- standards, and guidelines and regulations
volved in determination of the maximum to be used in structural design.
design wave height, the deck clearance is
reduced below "safe" minimum, and in the
extreme case, it is in direct contact with the
wave crest. This may be possible in the case 7.8.2 Design Methods
of a pier constructed offshore with no pro-
tection from wave action; it is less likely to This section is dedicated to discussion on
occur in protected harbors or in inland conventional methods of structural analysis
waterways. used for design of free standing and
Direct contact between the wave crest marginal open pile structures, and piled
and the underside of a pier deck will result relieving platforms.
in an uplift load to the whole st11,1cture. The
potential of such occurrences must be ac- 7.8.2.1 Design of Freestanding
counted for in pier design. and Marginal Open
The background of this phenomenon and Pile structures
results of a laboratory study of the wave-
induced vertical loading on platform deck
are discussed by Shih and Anastasiou
7.8.2.1.1 Freestanding piers
(1992). These investigators point out that in Examples of these structures are illus-
the case of offshore structures the total up- trated in Figures 7-la, arid 7-lb, 7-2, and
lift load consists generally of three compo- 7-4 through 7-7. These structures are nor-
nents, namely rapidly varying impact pres- mally constructed in a form of elevated
sures and slowly varying positive and nega- above the sea-level deck that is supported
tive pressures. Based directly on experi- either on solely vertical piles or on a combi-
Piled Waterfront Structures 839

nation of vertical and batter piles. Depend- Normally, when the deck structure is
ing on the pier length, its deck can be con- comprised of precast components, they are
structed in one piece or in several pieces usually joined to each other by means of
separated from each other by the trans- large grouted shear keys. Typically, prefab-
verse expansionjoints. Traditionally, in long ricated panels are made fully continuous
piers, the expansion joints have been placed over the pile capping beams either by post-
no more than 180 m apart. However, in tensioning or by welding (alternatively by
more recently constructed navy piers in the using mechanical joints) negative reinforce-
New York harbor area, the expansionjoints ment. The prefabricated coinponents are
were placed at approximately 290 m apart joined to the capping beam by means of
(Green, 1989). It should be pointed out that dowels that protrude from the capping
normally the distance between expansion beam, and the cast-in-situ concrete is in-
joints is governed by the allowable stress stalled between ends of the deck panels.
level in the piles due to deck thermal ex- This results in a strong, massive composite
pansion or contraction. Essentially, the joint between the deck structure and pile
space between expansionjoints is a trade-off bents. The longitudinal joints between pre-
between the number of expansion joints in cast components are typically made using
the pier and the number and size of piles in large grout shear keys. There is no reinforc-
a pier section. ing steel across the shear key; therefore,
As pointed out earlier, the deck system is although these keys can transmit vertical
usually constructed either in a form of a
shear, they cannot then transmit trans-
solid cast-in-place concrete slab or assem-
verse moment across the joint. However,
bled from the precast components sup-
the connection between the deck panels and
ported on either cast in place or made from
the torsionally stiff pile capping beams pre-
precast concrete elements pile capping
vent any separation of panels at the shear
beams.
key. The presence of a shear key might be
Naturally, all vertical and horizontal
thought of as a significant reduction in the
loads acting on the deck structure are
transferred to the pile's foundation. A piled transfer distribution of a concentrated load
pier is essentially a three-dimensional exerted on the deck. However, the presence
frame in which all loads are distributed of the longitudinal shear key does not mate-
between all structural components accord- rially impair torsional stiffness of the slab,
ing to their relative respective stiffnesses. and the lateral load distribution is quite
similar to that in a monolithic slab (Roeski,
1955; Walther, 1956). Some additional
7.8.2.1.2 oeck Stiffness stress concentration, however, could occur
The deck stiffness (EI) depends on the when a load is placed over the shear key.
design and particularly on the material used A typical joint between precast deck com-
(normally regular prestressed concrete) for ponents is depicted in Figure 7-71. Strain-
the construction. Essentially, the deck mo- gauged tests on a dock deck comprised of
ment of inertia, J, depends on the deck precast haunched panels have been carried
geometry and is calculated taking into ac- out in Port-of-Seattle, Washington (ABAM
count every aspect of the deck details (e.g., Engineers, Inc., 1967). These tests prove
if it is designed as a solid slab of uniform the deck prefabrication does not cause any
thickness supported directly on piles, or on fundamental changes in slab behavior. In-
pile capping beams, or comprised of precast terestingly, the effect of the haunches in
units supported on pile caps or capping precast panels prove to be relatively minor;
beams, or other methods). they cause just a slight increase in negative
840 Piled Waterfront Structures

moments at the supports and a slight de- a National Building Code of Practice and
crease in positive moments at the midspan. other national standards applicable for de-
In summary, deck sti:ffness is a very impor- sign ofbridges. In North American practice,
tant structural parameter and must be the pier deck is commonly designed to con-
evaluated carefully in each specific case. form to the requirements of the latest issue
The distribution of horizontal loads be- of AASHTO specification, Canadian Stand-
tween pile bents depends directly on the ard S6 "Design of Highway Bridges," and
relative stiffness of the deck; naturally, the other relevant documents. Where the deck
relative sti:ffness of a longer deck is smaller structure is designed as a continuous slab
than a shorter deck with the same cross- supported on pile bents, elastic properties
sectional stiffness value. of pile bents (settlements) under vertical
loads are considered. For this, the bent
7.8.2.1.3 Deck design elasticity factor (spring characteristic) is
best obtained from the pile load tests. For a
In general, the deck should be placed preliminary design, relevant information
high enough above the highest expected published in standard texts on foundation
wave crest elevation. Where this is not fea- engineering can be used.
sible, wave uplift pressure should be consid- Calculations can be carried out on the
ered in the design. This pressure is charac- hasis of a five-moment equation (continuous
terized by a high localized initial peak pres- beam on elastic supports) or by utilizing a
sure of short duration followed by a slowly suitable computer program. The former is
varying uplift pressure (e.g., upward and found in basic standard texts on structural
downward) oflower magnitude but oflonger mechanics, and the latter is readily avail-
duration. Details are found in El Ghamry able on the market. In these analyses, the
(1963), Wang (1967), French (1979, and Lai deck structure is usually treated as a con-
and Lee (1989). As noted earlier, the uplift tinuous beam on elastic supports.
pressure can be significantly reduced even In general, any viable design method that
by a very small percentage of pressure re- is used for the deck structure design (analy-
lief openings (relative to deck area) pro- sis) is based on a number of interactive
vided in deck structure (Fig. 7-50). elements included in the criteria that de-
Structurally, the pier deck is usually de- fines satisfactory performance of the struc-
signed in accordance with the latest issue of ture. This includes design loads, structural
materials, and their properties, also a be-
havioral model and realistic, achievable
boundary conditions. Obviously, the nature
and accuracy of this model must take into
consideration the accuracy of available
foundation soil data. If the structure is de-
signed to be in service for a long period of
time, then the aggressive actions of chlo-
rides, sulfates, and other chemical agents
that are plentiful in the marine environ-
ments must be included in structure degra-
1 dation model(s).
Figure 7-71. Grouted key between precast deck
The structural design with ali con-
panels: 1-precast concrete panel; 2-grouted key; 3 straints and allowances included in the de-
-concrete topping. sign model is finally translated into con-
Piled Waterfront Structures 841

crete and reinforcing steel specifications. (a)

These include the grades of both concrete


and reinforcing steel, the water1cement
ratio, and all ingredients and admixtures
included into the concrete mix.

7.8.2.1.4 Pile foundation ana/ysis


Essentially, this is affected by all kinds
of vertical and horizontal design loads (b)
transmitted to piles mostly through the deck
structure and in some instances imposed
directly on the piles (e.g., environmental
loads generated by waves, current, and ice).
The most significant horizontalloads that
affect pier or marginal structure design are
impact loads from berthing ships and moor-
ing loads. The former is transmitted to the
deck structure via the fender system
mounted on a deck face (edge) beam, and
the latter acts on the pier deck via mooring
accessories such as bollards, quick-release
hooks, and other items installed on the deck.
It should be pointed out that the wind load
imposed against the ship in light condition
can sometimes generate horizontal load on
the dock structure that is more critica! than
impact loads produced by fully laden ship.
Obviously, these loads are transmitted to Figure 7-72. Distribution of horizontalload among
the deck structure via the fender system or piles through deck structure: (a) deck comprised from
mooring accessories strategically placed prefabricated panels placed on pile capping beams; (b)
conventional 45° load distribution from impact point;
along the pier. (c) beam on elastic supports approach. 1-pile;
Lateralloads are distributed among piles 2-cast-in-place or precast pile capping beam; 3-pre-
through the deck structure and, as stated cast panels placed on pile capping beams; 4-cast-in-
earlier, this load distribution is dependent place concrete deck; 5-spring representing stiffness
on deck structural design and its stiffness of pile bent; kb = stiffness coefficient.
(EI). In rare cases of a pier design when
the deck structure is comprised of precast cast-in-place concrete slab placed directly
components sitting freely on pile caps, the upon piles, or the like, the horizontal load
entire horizontal load (e.g. ship impact) is distribution among piles is usually evalu-
assumed to be resisted by a single pile bent ated by one of the following techniques.
(Fig. 7-72a). This structural arrangement The first and the simplest technique as-
can be practica! in the case of a bridge-type sumes that the horizontal force of a berthing
pier in which a single-span deck section ship hitting the pier projects outward at a
rests on pile bends that are able to resist 45° angle from the impact point and is dis-
large ho~zontal loads. On the other hand, tributed among piles within the angle (Fig.
in the case of a continuous deck structure 7-72b). In other words, piles within 45°
rigidly joined with pile capping beams, or distribution only are mobilized to resist the
842 Piled Waterfront Structures

horizontal force. Hence, the horizontal load 7.8.2.1.5 Pile foundation stiffness
on one pile, qP, is determined as qP = Qjn,
The parameters affecting pile stiffness
where n is the number of piles included in
are numerous and not easily simplified.
the area abcd (Fig. 7-72b). In reality, how-
They basically depend on the following
ever, the load distribution among piles de-
factors:
pends on the relative stiffness of both the
deck and the piles. In most practica! design
cases, the deck structure is rather stiff and • Number of piles per bent and their cross
has the ability to distribute the horizontal section(s)
load among much greater number of piles • Spacing between piles and pile bents
than those indicated in Figure 7-72b. As • Presence of batter piles and pile batter
• Total pile length and freestanding portion
pointed out by Padron (1985) and Padron
of a pile that is the length between the
and White (1983), in practice the pier can underside of the deck and fixity point in
resist 2.5 times or more of the impact load the foundation soil
than that assumed in the 45° distribution • Soil parameters, which includes its elastic
method. Typically, the cost of piles consti- properties in both vertical and horizontal
tutes well over half the total cost of the pier directions
structure capable of accommodating deep • The group effect, which depends on factors
draft large ships. Hence, in order to opti- such as pile spacing in both the longitudi-
mize the design of pile-supported pier struc- nal and transverse directions, type and
tures, it is essential to maximize the distri- strength of soils, pile installation method,
bution of the horizontal loads (e.g., ship and so forth.
impact) or mooring forces among the sup-
porting piles. Vertical piles tend to resist lateral load
The second most often used technique flexurally by behaving as cantilever
treats the deck of a pier in a horizontal columns; therefore, when the pier founda-
plane as a girder supported on a set of tion is comprised of vertical piles only, the
elastic supports (springs) with a horizontal pile lateral stiffness, kb, is determined by
concentrated load among the pile bents dis- methods described in the preceding sec-
tributed according to the relative stiffness tions. Naturally, more accurate values of kb
ofthe deck and the piles (Fig, 7-72c). Again, can be obtained from pile lateral load tests.
this system can be analyzed by using the The latter can be economically justified for
five-moments equation or by utilizing a a detailed design of a pier comprised of a
suitable computer program. large number of piles.
This type of analysis produces a rela- Many (not to say most) piled marine
tively accurate assessment of distribution of structures utilize both vertical and batter
horizontal loads between pile bents, pro- piles. In most cases these piles are part of a
vided that the deck stiffness and pile bend typical pile bent; however, the batter piles
spring characteristics are estimated with a can also be placed randomly elsewhere
sufficient accuracy. When the space be- within a structure as required by the load-
tween pile bents is sufficiently small, the ing condition. Naturally, the order of mag-
deck-pile system can be treated as a beam nitude of stiffness of batter piles is much
on an elastic foundation (Padron and greater than vertical ones and, therefore,
Elzoghby, · 1986). Again, the results ob- the latter contribute to bent stiffness much
tained from the above method depend heav- less than battered piles. Mechanism by
ily on accuracy in calculating deck stiffness which batter piles resist horizontal load is
and estimation of pile spring characteris- different than that of vertical piles. Batter
tics. piles tend to resist lateral loads axially by
Piled Waterfront Structures 843

behaving as columns. However, the axial and vertical piles are included (Fig. 7-73b).
pile deformations (settlement and pull-out However, because stiffness of batter piles is
due to compression and tensile loads) in- much greater than the vertical ones, practi-
evitably res.ult in lateral displacement. This cal infiuence of the latter on the k b value is
displacement is very similar to that which insignificant and therefore, where relevant,
occurs in sheet-pile bulkheads anchored by the effects of vertical piles can be conve-
racked anchor piles (Fig. 6-66d) and calcu- niently ignored.
lated using Eq. (6-56) in Chapter 6. Natu- In performing the frame analysis de-
rally, as discussed earlier, pile lateral dis- scribed above, it is assumed that the lower
placement results in soil reaction and, part of a real pile, below fixity point, O, is
therefore, in bending moments and shear substituted by three independent springs
forces in the piles. Consequently, the lateral that characterize the pile reaction at this
displacement of batter piles constitutes a point; one spring simulates the axial load
very complex phenomenon that includes the response of the pile-soil system, another
elastic deformations of piles itself and the reproduces the lateral resistance, and the
elasto-plastic displacements of foundation third gives the moment-rotation character-
soil. istic of the embedded portion of the pile.
Again, the most reliable values of lateral From these analyses, with a load equal to
displacements of the batter pile system are unity, the bent stiffness coefficient, kb, is
best obtained from large-scale model tests determined as kb = ljy 0, where y 0 is the
of field tests. The approximate solution for system lateral displacement under action of
the batter stiffness coefficient (which can unity load Q = 1.
also be defined as a pile bent stiffness coef- Numerica! values of the spring charac-
ficient), kb, can be obtained by modeling the teristics can be obtained from the pile model
batter pile system as illustrated in Figure indicated in Figure 7-65. This model can
7-73a. provide the required components of pile re-
Essentially, a more accurate solution for sponse needed with the boundary condi-
bent stiffness coefficient, k b, can be ob- tions for analysis of a frame formed by bat-
tained from the model in which both batter ter piles.

(a) (b)

Figure 7-73. Structural model for determination of pile bent stiffuess coefficient,
kb (note: horizontal springs conditionally are not shown): (a) batter pile system only;
(b) complete pile bent system that includes both batter and vertical piles.
844 Piled W aterfront Structures

Padron and Elzoghby (1986) developed (a)


curves for determination of k b in which the
value of kb is given as function ofthe Lt!A
ratio, where Lt is the pile equivalent length
and A is pile cross-sectional area. The solu-
tion is based on simpli:fied method devel-
oped by Ting (1985).
In the most complete method of open pile
analysis, the structure is not segregated
into two separate elements (e.g., the deck
and the pile foundation are treated as a
space frame). This procedure, however, is
very complex and time-consuming. It may
be used primarily in the final design stages
and also when accurate input data based on
real pile tests are available. The structural
analysis can be carried out via computer by
utilizing miscellaneous commercially avail-
able space frame programs.
Analysis of a marginal open-type piled
structure is carried out in a manner similar
to that used for open pile pier design. The
basic difference between a piled pier struc- Figure 7-74. Open pile marginal wharfwith grav-
ity-type retaining wall: (a) cross section; (b) plan.
ture and a marginal piled wharf structure 1-Piled structure; 2-gravity-type retaining wall; 3
is that in the former, relative stiffness of all -design slope; 4-"virtual slope" [according to
piles is approximately the same, whereas in Japanese Technical Standard for Port and Harbour
the latter, the structural elements are vari- Facilities (1980)]; 5-nth pile.
able. This is due to different heights of the
freestanding portion ofpiles (Figs. 7-74 and In this Japanese standard, the depth x 0 is
7 -85). The front piles are relatively more de:fined as the distance from the surface of a
fl.exible than those located closer to the land "virtual slope" to the pile :fixity in the soil
side of the structure. The freestanding pile (point 0). The "virtual slope" is de:fined as a
portion, Ln, which is de:fined as the pile slope that is stable while loaded by miscel-
length from the underside of the deck to laneous horizontal loads, including earth-
pile :fixity within the foundation soil, can be quake loads. The angle a, of a "virtual
determined as recommended by Japanese slope" with the horizontal, is determined
Technical Standard for Port and Harbour from
Facilities (1980) (Figure 7-74):
a=cp-e (7-104)
(7-103)
where
where cp = angle of internal friction of the main compo-
nent of slope material
x0 = 1/{3 in which {3 = (khdj4EI) 114 e = tan- 1k/,, where k/, is the horizontal seismic
kh = coefficient of horizontal subgrade reac- coefficient in water
tion (kgfjcm 3 )
d = diameter (width) of pile (cm) Where a gravity-type retaining structure is
(EI)n = flexural rigidity ofthe nth pile (kgfcm 2 ) placed behind the piled structure, the base
Piled Waterfront Structures 845

of this retaining structure should be located ered


beyond the edge of the "virtual slope"
(Fig. 7-74) Yn = Yo + axn (7-108)

The J apanese Standard given above also where a is the torsional rotation of the
recommends the following distribution of piled section
horizontalload, Q, among individual piles.

(a) The horizontal load, Q, is acting at the


structure's gravity center (GC); therefore,
torsional rotation need not be considered.
Hence, The bending moment at the nth pile head,
M n, is calculated from
Q~ (kh(n)~~ kh)Q
= (7-105) (7-110)

(b) The horizontalload, Q, is acting eccentri- The axial load on each pile, Pn, included
cally; hence, the structure torsional rota- two components, Pn(h) and Pn(u)
tion needs to be taken into account:
(7-111)

where
Pn(v) = axialload contributed by verticalloads
where Pn(h) = axialload contributed by horizontalloads

Q~ = horizontal load distribqtion to the nth


pn(h) = sn(n-1) + sn(n+1)
pile
M(n-1)n + Mn(n-1)- Mn(n+1)- M(n+1)n

l
kh(n) = horizontal stiffness on the nth pile, or (7-112)
horizontal spring characteristic of the
nth pile; kh(n) = 12Eln!L!, where Eln where sn(n-1) and sn(n+l) are respectively
is the stiffness ofthe nth pile and Ln is
shear forces at the head ofthe nth pile on a
the design freestanding part of the n th
pile
side of the (n - l)st pile and on a side of
the (n + l)st pile caused in the deck struc-
L:n k n = horizontal stiffness of pile foundation,
or the sum of horizontal stiffness of all ture by horizontalloads; M(n-l)n' Mn(n-l)'
piles included in the pile foundation Mn(n+l)' and M(n+l)n are bending moments

e = distance from gravity center (GC) of a in the deck structure at the pile head caused
pile foundation to load application point by the horizontal force respectively on a
xn = distance from GC to the nth pile side of an nth pile at the head of a (n - l)st
pile, on a side of an (n - l)st pile at the
Accordingly, horizontal displacements ofthe head of an nth pile, on a side of an (n + l)st
structure, y, and individual piles, y n, can pile at the head of an nth pile, and on a
be calculated using the following expres- side of an nth pile at the head of an (n +
sions (note that the deck structure is con- l)st pile; l = space between pile bents
sidered as very rigid):
7.8.2.1.6 Length of a deck
(a) Torsional rotation need not be considered
between expansion
joints
Yo = Q ~~ kh(n) (7-107) Deck movements due to changes in the
ambient temperatures cause piles to deflect.
(b) Torsional resistance needs to be consid- Naturally, the piles included in the edge
846 Piled Waterfront Structures

bents on both sides of the deck will defl.ect L..__ _ __


the most. Hence, assuming that the maxi-
mum permissible defl.ection of the pile is
equal to [ y 0 ], the maximum length of the
deck between expansion joints, Ld, can be
expressed as follows.

2[y 0 ]
L --- (7-113) L
d- cÂt

where
c = coefficient of thermal expansion of deck 1
material (Table 7-13) 1
 t = design temperature fluctuation (in degrees)
1
The permissible defl.ection of a pile, [y 0 ], is 1
governed by the permissible stress level, ft, \
induced in a pile due to displacement from
its original position by virtue of deck move-
ments. Figure 7-75. Width of expansion joint between adja-
Essentially, relatively fl.exible piles are cent deck sections.
more adaptable to defl.ections than stiff ones.
The allowable pile displacement can be ob- between adjacent deck sections (Fig. 7-75).
tained from l = 2[ y 0 ] + 5 (7-115)

Note: l and [y 0 ] are in mm.


L 2 (S.M.)ft
[yol = 3(EI)p (7-114)
Essentially, permissible stress level in a
pile due to deck expansionjcontraction is a
where part of a total stress value contributed by
L =pile freestanding portion; L = H + x0 other loads, e.g., verticalloads, ship impact
load, mooring forces, environmental loads,
(S.M.) = pile section modulus
etc.
(EI)P = pile stiffness
7.8.2.1.7 Pile structural design
Allowable displacement of a pile also deter-
mines the width of the expansion joint, l, In open piled piers, the piles are nor-
mally treated as part of a rigid frame. De-
pending on the pile arrangement (e.g.,
Table 7-13. Coefficientofthermal expansion whether it is comprised of vertical piles
(linear, per degree X 10- 6 )
only, ora combination of vertical and batter
Material c per°C piles), structural pile design can be con-
trolled by either the combined action of both
Structural steel 11.7
Concrete 10 axial and horizontalloads or by axialloads.
Fir (parallel to fiber) 3.8 As said earlier, the former is basically char-
Fir (perpendicular to fiber) 58 acteristic of the case where the pile founda-
tion is composed of vertical piles only, and
Note: Coefficients of thermal expansion given in this table
are average values obtained from various sourees. the latter is more applicable to pile founda-
Piled Waterfront Structures 847

tions that includes both the vertical and the where


batter piles. A = pile "effective" load-bearing area
For a long unsupported length, L, the r = (ljA) 0 ·5 = radius of gyration of pile cross
axial load-carrying capacity of the pile, Q, section.
is limited to that of a long column; thus,
Euler's formula for the critica! buckling load The expression kLjr is called the "effective
is applicable. Applying this formula within slendemess ratio," where kL is de:fined as
the elastic range, Q is computed from the the pile "effective length." Theoretical k
following expression: values for severa! idealized end conditions
are given in Figure 7-76. It should be
pointed out that in many practica! cases,
(7-116) the pile's reinforcement and its cross sec-
tion geometry are govemed not by axial
where loads (static or dynamic) but by the maxi-
E and 1 = the pile material modulus of elasticity mum bending stress induced during pile
and the pile cross-section moment of pick-up and handling operations.
inertia, respectively
k =pile end condition factor, which deter- Precast and Prestressed concrete
mines the pile's "effective" length Piles
L = length of pile between the deck un- These piles have been used extensively
derside and fixity point in foundation elsewhere in the world and are discussed
soil earlier in this chapter. A detailed discus-
sion on these piles as used in structural
The pile buckling stress, {p, is determined design is given in Gerwick (1968, 1971,
from 1974) and Libby (1984). Chellis (1961),
Q 7r2E
Quinn (1972), and Libby (1984) provided
(7-117) tables with typical details of precast con-
{p = A = (kLjr) 2
crete piles of different cross sections . and

(bl (el (dl (el (fi

) ~B?
(al
,..
~~ ~
. -<(,t'f'
... ~ ~
1 \
1 1 1 \ 1 1
1 1 1 1 1
BUCKLED SHAPE OF 1 1 1
1 1 1
COLUMN IS SHOWN l\ \ 1 1
BY DASHED LINE
\
1
1 1 1
l 1
1

.
1
1 1
"'1--
t i'"rr ~~
~~
~t
THEORETICAL K VALUE 0.5 0.7 1;0 1.0 2.0 2.0

DESIGN VALUE OF K
WHEN IDEAL CONDITIONS 0.65 0.80 12 1.0 2.1 2.0
ARE APPROXIMATED

1 ROTATION FIXED TRANSLATION FIXED

ENO CONDITION CODE V ROTATION FREE TRANSLATION FIXED

~ ROTATION FIXED TRANSLATION FREE


y ROTATION FREE TRANSLATION FREE

Figure 7-76. Theoretical value of effective length factor, K, for


axially loaded piles.
848 Piled W aterfront Structures

allowable maximum design loads for vari- and shop procedure, it can be expected that
ous sizes and lengths. circumferential welds between tubes will be
the controlling imperfection when they are
steel pi/es contained within a region of uniform mo-
ment.
The design requirements of these piles
are similar to those for concrete piles. The steel Piles Fi/led with concrete
task of the designer is to calculate the axial,
bending, and shear capacities and to exam- Concrete is often used to increase pile
ine the combined effect of axial and bending strength and buckling stability. Further-
stresses. Large-diameter thin-walled steel more, use of concrete-filled piles that, in
piles are basically designed as long-span general, will be of a smaller diameter than
girders, the strength of which is controlled those not filled with concrete require lesser
either by material yield or by local buckling driving effort for pile installation. Also, use
of the cylinder wall. The strength of axially of stronger concrete-filled piles may require
compressed cylinders can be taken as a fewer piles to support a pier deck.
starting point for the evaluation of local Concrete-filled piles perform similarly to
buckling strength of thin-walled pipe piles slender steel tubular columns filled with
in flexure. There is a considerable volume of concrete. Both have been used extensively
literature on this subject. The most recent in marine engineering practice; they were
discussion is given in Pinkney et al. (1983), used to support bridges and for construction
Stephens et al. (1982, 1983), and Kulak et of port related structure, in conventional
al. (1988). Criteria for design of thin-walled civil engineering practice, concrete-filled
fabricated large-diameter tubular members columns have been used extensively to sup-
are given in a variety of North American port floors in multistory buildings.
specifications [e.g., ASTM "Metal Contain- The load-carrying capacity of both axially
ment Shell Buckling Design Methods" that and eccentrically loaded slender steel tubu-
is part of the ASME "Boiler and Pressure lar piles and columns can be calculated us-
Vessel Code" (1980) from the American So- ing the design provisions contained in stan-
ciety of Mechanical Engineers]. dard codes of practice (e.g., ACI 318-891ACI
A comparison of codes is given in 318R-89, CAN 3-516.1 M84, BS 5400: Part
Stephens et al. (1982). It should be pointed 5, and others. These design guidelines are
out that current North American specifica- simple to use, although largely empirica!
tions do not distinguish between the local and conservative. On the other hand, some
buckling strength of cylindrical members, theoretical methods, such as one proposed
subjected to uniform axial compression, and by Neogi et al. (1969), are too complex.
the flexural strength of these members, The Intemational Committee for the De-
whereas the current European specification velopment and Study of Tubular Structures
"European Convention for Constructional (CIDECT), which includes members from
Steelwork" (1979) does. Kulak et al. (1988) Japan, Europe, the United States, and
carried out bending tests on large- Canada, recently issued Monograph No. 5,
diameter fabricated steel cylinders and con- "Design of Concrete-Filled Hollow Section
cluded that the local buckling strength of Columns: Calculation Method and Applica-
fabricated steel tubes loaded in flexure is tion Technology" in North America, repub-
affected by initial geometric imperfections lished by Stelco, Inc. (Stelco, 1981). This
and by nonhomogeneous material behavior document is based on research work on the
caused by residual stresses. Kulak et al. structural behavior of concrete-filled hollow
found that, assuming that fabrication of the structural section (HSS) columns done in
tube is done according to best workmanship several countries. It is important to note
Piled Waterfront Structures 849

that in this monograph, the proposed


method of analysis for concrete-filled steel (a)

columns is based on European design rules F, • 350 lo4PI


~~ - 20MJ>a

such as Limit State Design Method of the


ECC (European Committee for Concrete)
and the buckling curves of the ECCS
,.
~ -~ 10 "'
v ..
s
.
""-

(European Convention for Constructional / / "


Steelwork). V/ V "
The design method included in the afore- 1; ./ ./
mentioned Monograph No. 5 is used for /;' / ./ /"
analysis of either square or tubular steel ~ V / ./
"
shells filled with concrete. The strength of ;/,V V / /
/

concrete, f~, has obvious effect on pile (col- j, '/ V / "


umn) strength; high-strength concrete can
"
d 'i / V //
..
~ / / / 1// c ....-
appreciably increase the load-carrying ca- f/

pacity of a pile while keeping steel use to a ~V / / .... v / -----e--"'


f.--"

-- - -
-- --
minimum. When the pile is exposed to ec- .â~ / V ./ / V v /
centric loads or when end moments are pre- "
~ ~ ~ ~ ./ / / / v 1-- ....-
sent, these can be accounted for by using
~ ~ ~ ~ -; VV ....- ::::: ....- 1--
factored loading conditions, which are dis-
~ ~ ~ ~ :;...-
cussed further in this section. Dimension-
less load charts for both square and round """ '
sections, given in Figures 7-77 and 7-78,
are aimed at simplifying the design calcula- (b)

tions by providing a direct graphical equiv-


alent of numerica! results.
These charts are prepared for long-term "'
loads; therefore, creep of concrete has been / ' •"
taken into account. The load charts account /1/ "
for the overall buckling of the column and / /1v"
V/ /
"'
the hoop effect for short circular columns.
") [/
They also give the factored compressive /
~ '/ "
strength, Cr, or concrete-filled steel columns V
/,'/ /
subjected to axialloading. " /'l / / /"
The following parameters are required ~/ / v
for use of the load charts: :'iV V V v"
~V V ....-
"
1. Type of steel column cross section (e.g., " ~ V / V ./ 1-- 1--
~/ / V /
" e-- 1--f-'
square or round)
v ....- --
1-- r-
50

A f? / ./ / /
_,....
2. Specified minimum yield stress of steel, ~~ V / / V / v ....- 1--
"
FY ~ ~~ ~ ;:::: v f.-:::: ::::: 1--
:;..-
3. Specified compressive strength of con- ~ ~ ~ f'::::
crete,r: ~

4. Effective length of column, kL


Figure 7-77. Concrete-filled steel piles. Load charts
5. Outside diameter, D, or width, b, of the
for square cross sections. [From Stelco Inc., CIDECT
steel column Monograph No. 5, Canadian edition (1981).]
6. Thickness of column wall, t
850 Piled Waterfront Structures

(a)
(a) F, .. 3$0 MPa Fr • 350MPa
1~ • 20MPa
ic - 20MPa

c. c.

. '1-%
.!!!,.

1 '
D
210
/ '
' "
.
,
,.
/
70
/ " V /

." /
20
,,.
..
, V
V "
.
."
/ V
vV V "
. ,,. / /
.1/ vvv /
no / "
~.~,v V V v'" ......
k;::vv 1/ / . 1/ /
""' v
/

_....!-- "'
30
~vv v v" ""'
V
~/ V / /
v"' "
. 1--- ~~--- ~--~
"
~
~'/
/ V
V
""' f-- ~~
v" . ..
1--
1-
1--1- 1-"'
i-1-- !--" ..
1-- " 1- i-1-r-
'ţ/ /
V
i-1-
1-
"
~~ ~::::: ~~
li'
15 20 25 .
(b)
.!i Fr .. 3$0 MPa
re • 2SMPa

c.
c.
r-f,- gs
,.
y .. ~
'
.!!!,.
o
'
240

"
V V
>O
/ "

,,. v" 1/
1/
..
,. V V .. /
" "
o
/ /
1/
V / ,.
/
" vvv /
o
~ V
V ,. "
V ""' ///V V V
.
/
V •o
/ d
.
/

...
70
.....
.
" ·e: '/,...- "..,.f-" v-f-- v v",.
...
1-- V
..... 1- ~~ /
V f-I-- "
1- !!l!!v 1--~--'
~[:;: ~P- F-i-1-
"
.. . .
1-

"""
." .""
30

"
90 95 100

Figure 7-78. Concrete-filled steel piles. Load charts for round cross sections. [From Stelco Inc.,
CIDECT Monograph No. 5, Canadian edition (1981).]
Piled Waterfront Structures 851

ff-
c, (a) /1-~
Co
p. • 1~3 .{ji (c:ltcul• HSS)
Fr • 350 NPa
Mt p. - b 'f ~1 (squar.ttSS)
e--
e,
1f t"-..
" . - 2ftAI.
.. l• ..,l.,
b.,
1"/.
~
,, A.•1KAr.

l'f.. "V ''-, lf'


", 1• I~Ac
1\ .'\
3.0
'\
·'V.
.r
1'\
''\ i/ ['\.. _L "• . ,"""",
'\
r'\ ll !"""' ..
(b)
"'
.. i\ X \1 /
a
1/ IL' j
~ 1\ .~ '\
fi 1\ V lX ~ '\
... 1~

'
\ 1 !"-. ~ '\ Aa • H.~-

~"'
~ r--...' 1'-, ..
~
)
/}\
~ ~
\
.L
~ 1
IC•20M~
1

...
~~ [',. Î"-
1'--.
1\
'~"~ "ES.
~
(o;.• 25MPa

... r:::
~ [;::--- 1---- r---. t-
.!!!.•
o
l"
~ r;:: r-- t--
r--- r=:::.. ~V "':MPI
~
~V rc .. 35~•
... ~ ~ 1-- ..
F=:::::: ..
..• •..
Jl re •
..'
40MP•

'
.. .. ..
1~15
7.0 }'

... ... ... .Lor..!.


Figure 7-80. Concrete-filled steel piles correlation
between factors {3 and p. as a function of A 8 = f(Ac),
Figure 7-79. Concrete filled.steel piles: (a) correla- CIDECT monograph No. 5, Canadian edition (1981).
tion between axialload and end moment; (b) eccentric-
ity reduction factor, a. [From Stelco lnc., CIDECT
Monograph No. 5, Canadian edition (1981).] (Fig. 7-79a). Detailed information concern-
ing the theoretical data on which this ec-
centricity factor chait is based is given in
The appropriate chart is chosen according work by Guiaux and Janss (1970). As is
to parameters 1, 2, and 3. Parameters 4, 5, seen from Figure 7-79b, the eccentricity
and 6 are used to determine kL jD or kL jb factor, a, is a function ofthe real or equiva-
ratios, which are used to determine the fac- lent eccentricity, e, the effective length of
tored compressive strength, cr = yt 2 ' where the pile (column), kL, and the outside diam-
the factor y is obtained from the load chart eter or width of pile cross section, D
as a function of FY, {~, kLjD or kLjb, and or b; essentially, the favored compressive
Djt or bjt strength of the eccentrically loaded pile
As noted earlier, the eccentricity factor, equals era.
a, is used for calculating the factored com- Preliminary parameters for concrete-
pressive resistance of the pile when end filled piles are obtained with help of the
moments are present. This factor is ob- graph given in Figure 7-80. This graph is
tained from the chart presented in Figure based on the following assumptions:
7-79b. This chart is based on a linear inter-
action formula and leads to conservative 1. The load is purely axial; however, the
values of the factored axial capacity of a eccentricity reduction factor, a, is used
pile (column) of a given length when sub- when required to account for eccentricity.
jected to a real or equivalent eccentric load 2. The steel and concrete are limited to max-
852 Piled Waterfront Structures

imum stress levels of 0.9FY and 0.67{~ 8. From the manufacturer's catalogs, select
(limit state design). the nearest available preliminary diame-
3. The ratio between the area of steel, A., ter, D, or width, b, of the pile section
and the area of concrete, Ac, is in the that is as close as possible to values ob-
range 5% Ac :o; A 8 :o; 25% Ac. tained by expressions (7-118) or (7-119).
9. Check the data for the pile selected in
The preliminary parameters obtained for a item 8 against the assumptions made in
pile must then be checked by the use of item 2. lf they are not comparable, then
load charts and the eccentricity reduction correct the assumptions and repeat items
factor a. 3-9 again until convergence.
The following is a step-by-step approach 10. Obtain f.L from expressions (7-118) or
used to determine the pile axial load- '(7-119), and for known FY and {~, deter-
carrying capacity from a strength of mate- mine {3 from Figure 7-80.
rial viewpoint. 11. Compute the preliminary area of a steel
pile cross section, A., from
1. Specify pile cross-section geometry (e.g.,
square or round), minimum yield strength (7-120)
of steel, FY, and compressive strength of
concrete, {~.
2. Determine effective length, kL. 12. Using A., define a preliminary wall
3. Detine factored loading conditions in ac- thickness, t, from the manufacturer's
cordance with the appropriate national catalog.
design standard.
When the preliminary pile section is se-
The above parameters are then used for the lected, check it by using the load charts in
preliminary sizing of pile (column) cross the following sequence.
section. It is carried out in the following
sequence: 1. Compute the ratio DjT or bjt.
2. For known FY, {~, kLjD or kLjb, and
1. Calculate the factored load, e,. Djt or bjt, determine y from the appro-
2. Assume D or b, pile cross section. priate chart in Figure 7-77 or 7-78.
3. Calculate the real or equivalent eccen- 3. Calculate the factored compressive
tricity, e (Fig. 7-79a). strength of a pile under axial load from
4. Calculate ejD or ejb, and kLjD or er = yt2.
kLjb. 4. Finally, check for aer < e,, where e, is
5. From Figure 7-79b, determine eccentric- the factored load on the pile.
ity factor, a. In the case of aer< e,, it will usually
6. Determine a preliminary value of f.L from suffi.ce to increase the pile thickness by
Figure 7-80. A steel section with area one increment. Finally, it should be re-
A. close to 12% of the area of the con- membered that SI units are used
crete section Ac for a square cross sec- throughout the above procedure.
tion and 9% for a round cross section is
a good starting point for an economica!
design.
More recently, Rangan and Joyce (1992)
7. Compute a trial diameter, D, or width,
proposed a new simple method for calculat-
b, from the following relations: ing the strength of eccentricity loaded slen-
der steel tubular columns filled with con-
D = 1.13f.L(e,;a)0 .5 (7-118) crete. This method is based on the assump-
tion that the failure load is reached when
b = f.L(e,;a) 0'5 (7-119) the maximum moment at midheight of the
Piled Waterfront Structures 853

column is equal to the ultimate bending bility are required, a reinforcing cage can
strength of the cross section at that loca- be added to the concrete in-fiU to com-
tion. Another assumption is that the de- pensate for potential loss of steel due to
flection of the column due to creep of corrosion.
concrete (or other long-term effects) and
imperfections in the steel shell may be 7.8.2.1.8 Pile design sequence
treated as an additional eccentricity. Ac-
cording to Rangan and Joyce, the column The design process is essentially a trial-
strength calculated by their method show and-error procedure during which the most
good correlation with test results. economica! solution is typically found by
Where additional strength andjor dura- the following sequence.

Determine design conditions

Assume the structure geometry and pile arrangements

Perform underdeck slope stability analysis

Design preliminary deck structure

Design preliminary and do stability calculations for


soil retaining structure (if any)

Calculate axial loads, bending moments and shear forces in


piles. Determine horizontal displacement of the structure

Determine pile penetration and calculate pile bearing capacity

Do detailed design

7.8.2.2 Design of Piled Relieving


Platform the relieving platforms are typically com-
These structures are illustrated in Fig- prised of a pile foundation surmounted by a
ures 7 -le, and 7 -lg, and 7-19 through heavy concrete deck ora retaining-wall-type
7-25. As is seen from these illustrations, concrete superstructure. As stated earlier,
854 Piled Waterfront Structures

relieving platforms are attractive for con- mitted eventually to vertical andjor batter
struction where soft andjor weak founda- piles. The superstructure is designed to
tion soils, unsuitable for construction of a carry the weight of a back:fill placed on the
gravity-type quay wall or conventional platform, surcharge, and other relevant
sheet-pile bulkhead, are present, andjor loads. Because these loads are transmitted
where the wharf is designed to handle an directly to the piles, the back:fill behind the
extremely heavy surcharge or concentrated sheeting is "relieved" from them, and the
loads. These loads, along with other loads lateral thrust against the sheet pile wall is
such as lateral thrust of a backfill soil, substantially reduced.
mooring and ship impact forces, loads pro- The soil pressure "relieving effect" can be
duced by miscellaneous cargo handling and maximized by extending the deck structure
hauling equipment, and others, are trans- some distance beyond the last row of piles.
mitted to piles which, in turn, carry them This will also add some vertical load to the
directly to the foundation soil. In most prac- tension piles, in which case the uplift .load
tica! cases, the deck or soil-retaining super- on these piles can be reduced or even elimi-
structure are placed some distance above nated. The latter may be an important con-
the mean water level, which enables the sideration where pile penetration is limited
contractor to carry out the construction by closely located bedrock and, thus, the
work above sea level. Sometimes, however, tension piles are not able to develop the
when a structure is constructed on existing required uplift capacity.
land or temporary placed :fill, and the har-
hor basin is dredged afterward, the super- 7.8.2.2.1 superstructure
structure can be placed below the mean
water level. It is similar to that discussed in Section
As discussed earlier, in some cases the 7.8.2.1. In the case of relieving platform,
concrete superstructure is placed directly however, the task of calculating the super-
on the :fill installed behind a sheet-pile wall, structure's stiffness can sometimes by more
whereas in other cases, the superstructure difficult and time-consuming because of its
spans a sloping :fill behind the sheeting. The complex geometry.
advantage ofthe former arrangement is that
the superstructure can be cast directly on 7.8.2.2.2 Sheet pi/ing
the back:fill, in which case however the
sheet-piling may require temporary anchor- As is seen from the various examples of
age (Fig. 7-86b); the anchorage may not be piled relieving platforms described earlier
required if the sheeting is of a "dredged" in this chapter, sheet-pile bulkhead placed
type of construction. For details on the in front of the wall can be of miscellaneous
"dredged" and "back:filled" bulkheads, con- construction; for example, conventional steel
sult Chapter 6. sheet-piling, reinforced concrete sheet-
The disadvantage of the second arrange- piling of miscellaneous design, heavy build-
ment is that the superstructure needs to be up steel king piles placed apart with the
cast on formwork supported on scaffolding, conventional steel sheet piles installed be-
none ofwhich may be recovered and reused. tween them, wooden sheet-piling, and oth-
Alternatively, the superstructure can be ers. In some instances, the waterside bulk-
erected from precast components. The obvi- head can be constructed in a form of slurry
ous advantage of this structural arrange- wall. The example is provided in Chapter 6.
ment, particularly in combination with a The waterside bulkhead is an integral
sloping fiii, is a substantial reduction in the part of a wall and is designed accordingly.
lateral soil pressure exerted on the sheet- In most practica! cases, this wall is joined
pile wall; the reduced soil pressure is trans- rigidly to a concrete superstructure; how-
Piled Waterfront Structures 855

ever, sometimes the sheet-piling is hinged soil pressure against sheet pile wall is af-
at the superstructure's underside (Figs. fected substantially by the presence of ver-
7-19, 7-20a, and 7-23) or just leans against tical and batter piles located behind the
a special supporting member which is a sheeting. These piles reduce the amount of
structural part of the superstructure (Fig. soil thrust acting on the sheeting. In fact,
7-20b). Depending on the sheet-piling fixity because of a shielding effect of the vertical
mode, its design may vary substantially. and batter piles, the actual soil pressure
Normally, the sheet-piling is designed to against the sheeting included in the reliev-
have sufficient penetration in order to per- ing platform system is normally much
form according to the fixed-earth support smaller than that with no due consideration
mode of failure (for details, consult Chapter to piles effects.
6). Accordingly, when the sheeting is fixed When the soil pressure against sheet-pile
within the concrete superstructure, it is as- wall is determined, the following sheet-pile
sumed that the bending moments are trans- wall analysis is carried out in a manner
mitted through the sheeting in a way that similar to that used for the conventional
the fixing moments at the wall head and sheet-pile wall design. For a detailed dis-
the fixity point within the foundation soil cussion on sheet-pile wall analysis, consult
equal the mid-wall moment, whereby the Chapter 6. An example of a graphic analy-
capacity of the wall is utilized to its maxi- sis of a sheet-pile wall included in the
mum extent. In this scheme, however, the relieving-platform-type retaining wall is
location of the maximum bending moment presented in Figure 7-81. In this example,
will typically coincide with a tidal andjor the soil pressure diagram against the
splash zone which are most sensitive to the sheet-pile wall is depicted in Figure 7-81b.
material corrosio:n,. Therefore, ali attempts Further steps include calculating individual
must be roade to protect the wall from dete- loads and plotting the load vector diagram
rioration, specifically in a zone of maximum and string polygon, from which the sheet-
bending moment. pile penetration, D 0 , is obtained, depending
Stiff sheet-pile walls, such as those corn- on an assumed mode of connection between
posed of reinforced concrete sheet piles, the sheeting and the superstructure; namely
heavy steel piles, or the like, are usually fixed or hinged. Accordingly, line AB in
designed to carry both vertical loads from string polygon (Fig. 7-81c) corresponds to
the superstructure and bending moments full sheet-pile fixity within the concrete su-
from lateral soil thrust. Walls comprised of perstructure, and line A'B' is used when
conventional steel sheet piles and, in most the sheeting is hinged at the superstruc-
instances, walls roade from timber sheet tures underside. As discussed in Chapter 6,
piles are usually considered as too flexible the value of the bending moment in the
and, therefore, ineffective to carry the verti- graphic method ofthe sheet-pile wall analy-
calloads. In this case, the first row of verti- sis is computed from M 8 = pzmax> in which
cal bearing piles is usually installed as close values of p and Zmax are obtained from the
as practica! to the sheet-pile wall, and the vector diagram and string polygon (Figs.
latter is designed as an anchored flexible 7-81c and 7-810. As pointed out by
sheeting with no verticalload acting on it. Smirnov et al. (1979), values of M 8 and D 0
The basic concept of computing the lat- can be affected by the deflection of at least
eral soil thrust against the sheet-pile wall, the first row of piles located just behind the
which is part of a relieving platform, is sheet-pile wall; when deflection of the
discussed in detail in Chapter 4 and illus- sheeting is greater than the deflection of
trated in Figure 4-21. From the discussion piles, then some additional soil pressure
given in Chapter 4, one can find that the will be transferred to the sheet-pile wall.
856 Piled Waterfront Structures

(a) (b) (c)

~~ J1 'Y, cp, 11 1

z 1 1
3-+r--+--~~--~~-

Figure 7-81. Graphic analysis of ftexible sheet-pile walllocated on the waterside of relieving
platform: (a) design scheme; (b) soil pressure diagram on sheet-piling; (c) sheet-piling string
polygon; (d) soil pressure diagram on vertical pile (line EF); (e) :file string polygon; (f) sheet-piling
vector diagram; (g) pile vector diagram. 1-Sheet-piling with restrained head; 2-sheet-piling
hinged at superstructure.

An accurate determination of the addi- be distributed between the sheet-pile wall


tional bending moment in a sheet-pile wall and the piles, proportional to their respec-
due to pile deflection is a very difficult task. tive stiffnesses. This assumption is sup-
Hence, a simplifi.ed approach is usually used ported by measurements on a quay wall at
to account for the effect of pile deflection. In Baltimore where a proportion of the total
this approach, the fi.rst row of piles located active lateral loads, carried by the steel
behind the sheet-pile wall is assumed to be sheeting and by Monotube piles, appeared
exposed to the pressures exerted on it from to approximately correspond to their EI
the soil in the space between the sheet-pile values (Tschebotarioff and Ward, 1957).
wall and the piles (Zone A) and also the soil In practice, the calculations proceed in
in the space between the fi.rst and the sec- the following sequence. Initially, the values
ond rows of piles (Zone B) (Fig. 7-8la). of M; and D 0 are computed as discussed
Naturally, if the pressure exerted by the earlier; the bending moment in sheet-piling
soil located in Zone A is smaller than that produced by the pressure of soil included in
in Zone B, then the piles will tend to deflect Zone A only is obtained from M; = pzmax.
toward the sheeting, and vice versa. Thus, N ext, the pressure on the screen plane
the unbalanced part of the soil pressure can EF exerted by soil located in Zone B is
Piled Waterfront Structures 857

plotted (Fig. 7-81d), and, subsequently, The design bending moment for sheet-piling
M; = p 'z;"ax is calculated. Here, M; is the is determined from
bending moment in the first row of piles
when the balancing soil pressure from Zone
(7-126)
A is ignored. However, soil from Zone A will
apply pressure on row of these piles, which
in the case of conditional absence of soil in where K M is the bending moment reduc-
Zone B will result in bending moment in tion factor which accounts for fiexibility of
the piles equal to M;. sheet-pile wall. The value of K M is deter-
Because bending moments at place EF, mined as discussed in detail in Chapter 6.
attributed to soil pressures from Zones A For design penetration, the larger value
and B, have opposite signs; thus, the unbal- of D 0 or D'o is used. Note that when the
anced part of the bending moment is ex- sheet-pile wall is placed on the land side of
pressed as llMP = M; - M;. Naturally, un- the structure, the soil pressures on it is
der conditions where M; > M;, the pile will determined as discussed in Chapter 4 (Fig.
defiect toward the sheeting, thus exerting 4-32) and the sheet-pile wall is designed
an extra load on it. As noted earlier, the according to the guideline provided in
unbalanced moment, llM, is distributed be- Chapter 6.
tween the sheet-pile wall and the row of
piles according to their stiffnesses (El).
Mathematically, this assumption can be
7.8.2.2.3 Lateral soi/
expressed in the following terms:
pressure on relieving
platform used for
M; (EI). pile system analysis
(7-121)
As pointed out earlier; the advantage of
the relieving platform is that the screening
and effect produced by the wall's superstructure
(7-122) and piles help to significantly reduce the
lateral soil pressure exerted on the wall.
where The important question, of course, is how
M"s = part of !lMp transmitted to sheet pile wall much the lateral soil pressure exerted on
due to defiection of piles toward the wall the wall is reduced by the relieving plat-
M"p = part of !l M p acting on the pile form. Unfortunately, to date, not very many
a = distance between piles along the wall studies have been carried out to investigate
the interaction between the soil and the
From Eqs. (7-121) and (7-122) the fol- piled relieving platform wall.
lowing expressions for M; and M; are ob- Model tests perfoJ;IDed at the Franzius
tained: Institute, Germany (Streck, 1950;
Grundbau-Taschenbuch, Vol. 19, 1955)
M" = !lM/E/)sa (7-123) show that the size of piles, as well as the
" a(EI). + (EI)p space between them in both transverse and
M" = !lMP(EI)pa
longitudinal directions, have a profound ef-
(7-124) fect on soil pressure under the platform; the
P a(EI). + (EI)p
noncohesive soil present between piles, as
The total bending moment in a sheet-pile well as between piles and the sheet-pile
wall is obtained from wall, tend to act as if it· were inside of an
infinitely long silo, in which, because of fric-
(7-125) tion along vertical walls, a maximum lat-
858 Piled Waterfront Structures

erai pressure is reached at a certain depth line 1 is used when sheeting in hinged and
and remains constant below that depth. (For line 2 is considered when sheeting is joined
details on the silo effect, consult informa- rigidly with the wall superstructure. For
tion provided in Chapter 5). This compli- details on this graphic method of sheet-pile
cated phenomenon of soil-pile interaction analysis, consult Chapter 6. Because the
results in an uncertain soil pressure distri- sheet-pile wall is relatively flexible, the ac-
bution under the relieving platform. tual reaction force, Ra, is somewhat larger
To simplify calculations of the total soil than R~ obtained from the vector diagram
pressure on the relieving platform wall, the (Fig. 7-83e). It is usually computed as equal
effect of the piles is usually ignored. Obvi- toRa = KRR~, where, depending on sheet-
ously, this results in a conservative esti- pile wall stiffness (EI)P, K R = 1.2 to 1.4.
roate of soil pressure. The simplified version For a detailed discussion on K R, consult
of the soil pressure diagram on a sheeting Chapter 6.
(one soil system) is illustrated in Figure The design horizontalload from soil pres-
7-82. The soil pressure diagram as depicted sure, Ea, exerted on relieving super-
in this :figure is drawn using the following structure is determined as follows:
steps.
(7-127)
From the inner lower corner (D) of the
platform, a line DF' is drawn under the where E 1 is the soil active thrust on the
angle <jJ 2 , which is the angle of internal vertical projection of the wall's superstruc-
friction of a backfill soil. This determines ture.
the point F of the pressure diagram, above
which no pressure from beyond the plat- 7.8.2.2.4 Pile foundation
form is transmitted to the sheeting. Then a
line DN' is drawn under the angle 8 = This includes aU vertical and batter piles
45° + <Pd2, which defines the sliding place and sheet-piling on which the superstruc-
of the active wedge of soil through point D.
Below point N, opposite N', the lateral pres-
sure is taken equal to the value beyond the
limits of the platform. A straight-line tran-
sition is assumed between points F and N.
The active lateral soil pressure on the
sheeting is then represented by ABGNF.
Soil pressure ordinates F'F, N'N, and BG
are computed according to the rules dis-
cussed in Chapter 4. A relevant example is \

illustrated in Figure 4-20. The full horizon- \


.\
\
tal soil thrust on the platform is determined 1
as the upper reaction, R~, of a beam AB
having lower support below the dredge line. \
1
The determination of the soil pressure \
diagram and the horizontal reaction on the \
platform for a complex soil system that in- N' \
cludes several types of soils is illustrated in
Figure 7-83. This :figure is self-explanatory.
Figure 7-82. Conventional estimation of soillateral
As is seen from Figure 7-83d, the value pressures on sheeting for determination of soil reaction
of beam reaction, R~, depends on the as- on relieving platform (screening effect of platform is
sumed mode of fixity in the superstructure; considered, but that of piles is ignored).
Piled Waterfront Structures 859

Figure 7-83. Practica! example of conventional estimation of lateral soil pressure


exerted on sheeting, and graphic calculation of soil reaction on relieving platform in a
multilayered soil system: (a) design diagram; (b) soil pressures diagram on sheeting; (c)
vector diagram; (d) string polygon.

ture (relieving platform) is resting. Essen- ably larger for piles in loose sand than for
tially, the load-carrying capacity of this piles driven into dense sand). Accordingly,
group of piles depends on factors such as in cohesive soils, the skin resistance of a
pile spacing, type and strength of soils that pile group should not be taken larger than
are present, pile installation methods, and the sum of the skin resistance of individual
so forth. Normally, because of a relatively piles. Reductions may be required when the
large spacing between piles, the skin resis- pile spacing is small, or when the overlap-
tance of a pile group is taken as equal to ping zones of shear deformation in:fiuences
the sum of the skin resistance of individual the skin resistance of the individual piles
piles; for this, the minimum space between over a signi:ficant length of the embedded
piles must not be smaller than 3d center to part of the pile. The reduction factor may be
center, and with a minimum clear space expressed as a ratio of the outside perime-
between pile sides of no less than 1.0 m. ter of the pile group to the sum of the
Here d is the pile diameter. perimeters of the individual piles. Accord-
Det Norske Veritas (1977) recommends ing to this approach, no reduction is re-
that the skin resistance of a pile' group in quired if the relative spacing of the pile,
granular soils may be at least equal to the l j d (l is the pile spacing), is greater than
sum of the skin resistance of individual n°· 5 + 1 for square piles and 0.785(n°· 5 + 1)
piles times the ratio of the outer perimeter for circular piles, where n is the number of
of the group to the sum of perimeters of piles in the group.
individual piles. The possible increase of Pile installation, both method and se-
this minimum resistance depends on the quence of installation, should always be
initial density of the soil (e.g., is consider- given proper attention. For example, driv-
860 Piled W aterfront Structures

ing piles behind the sheeting will densify deck; structurally, it is treated as an elastic
the fill andjor natural soil, resulting in an slab founded on elastic supports (piles).
increased lateral pressure against the A conventional relieving platform-pile
sheeting. In some instances, such practices foundation system is normally analyzed as
may result in sheet-pile failures. Some a two-dimensional frame. These analysis are
characteristic case histories are discussed time-consuming. In order. to arrive at the
in Chapter 6. most economica! solution in the shortest
possible way, first a preliminary simple
7.8.2.2.5 Pile foundation analysis graphic analysis on different alternatives of
pile arrangement is performed.
In general, the piled relieving structure Ferguson (1992), pointed out that the ad-
is treated as a two-dimensional frame, con- vantages of graphical statistics are qualita-
sidering the pile group as a structural tive, presenting in the calculations a sense
system elasticly interacting with the of "what's going on"-"a feel,"-permitting
foundation soil in a manner similar to that the engineer "to built in the mind's eye a
indicated in Figure 7-73. For reliable per- vision of the forces in a complex structure."
formance, the piles should be embedded A coherent system of graphic statistics
deep enough into the load-bearing soil to have been developed in the 1800s by Karl
preclude permanent settlement under ap- Culman. Culman's graphic method is based
plied loads. Also, the number of piles and on the assumption of an equilibrium of
their arrangement must be such to preclude forces at every joint in a structural system;
significant horizontal displacement of the the graphical diagram of forces, called a
relieving platform; thereby the piles are force polygon, makes it possible to predict
subjected to axial loads and to bending forces in different components of a system.
stresses associated with the elastic interac- · These forces are read directly from the force
tion between soil and piles and due to pile polygon, being proportional to the length of
shaft elasticity. In the case of a short, the lines in the polygon. The method provides a
latter may result in significant bending built-in check of the analysis through the
stresses. closure of the force polygon (diagram) and
In most practica! cases, the piles are allows a clear distinction of the compressive
treated as being fixed within the load- and tensile forces acting in the system's
bearing soil and at the concrete superstruc- structural components.
ture. However, where piles and, in particu- The graphic analyses are carried out on
lar, sheet piles have hinged connections the hasis of the following assumptions: The
with the relieving superstructure, they are relievii.lg platform is assumed to be a very
treated as beams fixed within a foundation rigid structure, and piles are hinged at both
soil and having bearing at the relieving the platform and within load-bearing soil.
platform. Then, axial loads on piles are calculated
Relieving platforms are usually rela- graphically by using Culman's method in
tively thin and, thus, are designed to absorb the following steps (Fig. 7-84).
bending moments and shear forces occur-
ring from the verticalloads. Naturally, the
1. Each group of identica! piles (e.g., verti-
platform must be strong enough to carry cal, batter in compression and in tension)
the design loads in both transverse and is replaced by an imaginary pile located
longitudinal directions to the pile system. at the center of gravity of the group. In
In general, the relieving platform is de- the example depicted in Figure 7-84a, a
signed according to the same philosophy as group of vertical piles, that includes the
that applied to the design of the open pier sheeting and the vertical piles, is replaced
Piled Waterfront Structures 861

(a) Because of the very complex nature of


interaction among the relieving platform
(b)
system, foundation soil, and hackfill mate-
rial, accurate methods of statistic analysis
of such systems are not availahle presently.
Because the structure is treated as a frame
system, the important question is how to
determine the effective freestanding portion
of a pile, li (Fig. 7-85), or, in particular,
how to accurately determine the location of
d
pile fixity point within the foundation soil.
In the conventional design practice, the pile
fixity point within the foundation soil is
usually assumed to he located at some dis-
Figure 7-84. Preliminary analysis of load distribu- tance t; helow the line drawn at angle <P
tion between vertical and batter piles by Culman's from point a [i.e., the intersection of the row
method: (a) Design scheme; (b) load diagram. of sheet piles and the dredge line (Fig. 7-85,
line ah)]. Here, <P is the angle of interna!
friction of the hackfill material. Experience
hy the imaginary pile with axial reaction
indicates that, depending on the soil condi-
Ra.
tion, the pile fixity point can he assumed to
2. The line of action of the resultant load, R, he located approximately 1.5-3.0 m helow
is extended to intersect the line of Ra
line ah; larger values are relevant for the
action at point a.
weak soils and for the relatively rigid piles.
3. Point a is joined with point h (i.e., the The fixity plane can also he assumed as
point of intersection of the hatter piles' located at the point of maximum hending
line of action).
moment in the lower part of a pile (line cd,
4. Draw the load diagram as depicted in Fig. 7-85). This can he ohtained in a man-
Figure 7 -84h. This is done in the follow- ner similar to that depicted in the example
ing sequence: Draw lin.e ac parallel to and illustrated in Figure 7-81. Budin and
equal to resulting load R hy selecting an
Demina (1979) suggest that the fixity plane
appropriate scale. From point a, draw line
ah parallel to line ah shown in Figure
7-84a. Then from point c, draw line ch
parallel to Ra to intersect ah at point h.
From h, draw a line parallel to Rb and
from point a draw a line parallel to Re to
ohtain point d. Then Ra will he equal to
ch, Rb will he equal to hd, and Re will he
equal to ad.

These calculations will provide the designer


with a clear idea of potentialloads that may
act on each pile group. Suhsequently, the
designer should arrive at a preliminary so-
lution to the hest pile arrangement for the
particular site conditions. After a prelimi-
nary design scheme is finalized, more accu- Figure 7-85. Relieving platform. Design assump-
rate calculations are performed. tions on pile fixity within foundation soil.
862 Piled Waterfront Structures

can conservatively be considered at line ef produced by piles is usually taken into


(M = O) (Fig. 7-85), and the piles should be account.
analyzed for the unbalanced soil pressure However, because relatively large move-
in a manner to that as discussed earlier. ments are required to mobilize the dowel
In practica} analysis, it is usually as- action, the dowel effect is usually ignored
sumed that pile fixity point is determined when foundation soils are represented by
by the line cd; for this, the value of ti is sensitive and soft clays. The global stability
determined either empirically, or as sug- calculations can be carried out by methods
gested earlier; the best result is obtained similar to those used for anchor piles, as
from large-scale-model tests. As soon as the discussed in Chapter 6.
frame parameters and loadings are estab-
lished the structure can be analyzed as a 7.8.2.2.7 construction aspects
conventional frame on elastic supports, sim-
ilar to that discussed earlier and illustrated Normally, the pile placement specifica-
in Figure 7-73. tion refiecting practica! experience caUs for
Bending moments and shear forces in piles to be installed not further than 75-100
the sheeting, obtained from frame analysis, mm from their theoretical location and to
should be combined with those obtained be aligned not more than 15-20 mm from
from calculations of sheet-piling, as dis- the vertical or from the specified batter
angle.
cussed earlier and illustrated in Figure
Fuller (1983) suggests that the pile butt
7-81. A similar approach can be used for
(head) location tolerance can be increased
calculating bending moments and shear
up to 150 mm. This is a reasonable assump-
forces in vertical and batter piles. However,
tion for installation of timber piles because
because lateral soil pressure does not pro-
their nonuniform shape makes them diffi-
duce significant bending moments and shear cult to hold in position during driving. Be-
forces in piles, the latter pile analyses are sides, man.ufactured piles are not perfectly
usually ignored. Normally, bending mo- straight. Additionally, the allowable devia-
ments, shear forces, and axial loads result- tion from straightness is usually based on
ing from frame analyses are considered standard specifications, [e.g., American So-
suf:ficiently accurate for subsequent pile de- ciety for Testing and Materials (1982, 1987
sign. As pointed out earlier, the sheet-pile and 1988), and others]. A common specifi-
wall constructed from conventional steel or cation for both camber and sweep for H-piles
timber sheet piles is normally not designed (and their equivalent for other types of piles)
to carry verticalloads and, therefore, is not is that the pile should not deviate from a
included in the frame analysis, whereas re- straight line connecting the butt by one-
inforced concrete sheet piles or heavy duty tenth the pile length in feet (equivalent to a
steel sheet piles designed to carry vertical deviation of approximately 1.04 mmjm).
loads are normally analyzed for the com- These tolerances are considered sufficient to
bined effects of bending moment and axial preclude pile interference with each other
load. during driving. Essentially, the possibility
of pile interference increase with increasing
7.8.2.2.6 Global stability pile length and fiexibility. Hence, stiff piles
are used where long piles are required. The
This is carried out by using conventional other alternative is to increase the space
slip circle methods as discussed in Chapter between piles.
4. When slip circles cross piles, the effect of Severa! interesting case histories of pile
the additional resistance due to dowel effect interference are given in Fuller (1983) and
Piled Waterfront Structures 863

the U.S. Army Corps of Engineers (1990). the cone overlapping zone. Wolff has devel-
These examples indicate the fact that the oped the pile interference probability
contractor should always be aware of the method using a two-step procedure. First,
likelihood of pile interference and be pre- the probability of interference is deter-
pared for the necessary adjustments in pile mined for a single interior pile in a large
arrangement, such as changes in space be- group using Monte Carlo simulation; then,
tween the piles, installation of additional a probability distribution for the number of
piles, and others. During the design phase intersecting piles in a group is obtained
it is usually assumed that the clear space using an adaption of the binomial and Pois-
between the sheeting and the first row of son distributions. Wolff (1993) developed a
batter piles should not be smaller than 1.0 set of charts that may help estimate the
m (Fig. 7 -86). probability of pile interference during in-
Wolff (1990, 1993) attempted to quantify stallation as a function of pile length, spac-
the likelihood of pile interference and how ing, and diameter.
it may be affected by values of various pa- Again, the sequence of pile-driving has to
rameters (e.g., pile diameter, length, spac- be given proper attention. For example, in
ing, camber, batter, and the standard devi- the case of "backfilled" construction (Fig.
ation from the required placement and 7-86b) piles could be driven in any se-
alignment). Wolff assumed that the pile po- quence, which in most cases depends basi-
sition in the ground may deviate from its cally on the availability of pile-driving
"as designed" location in two directions and equipment. On the other hand, in the case
the possible driven locations fali in regions of "dredged" construction, as illustrated in
that take shape of truncated cones. Subse- Figure 7 -86a, regular vertical and batter
quently, pile interference can occur where piles are preferably installed prior to instal-
these cones overlap. If the cones correspond lation of sheet-pile wall. As pointed out ear-
to the specific locations and alignment tol- lier, this would preclude additional soil
erances, a pile can be driven at any place pressure exerted on the sheeting due to the
within a cone. Pile interference can only driving of regular piles.
occur if both adjacent piles pass through Naturally, the relieving platform con-

{a) {b)

Figure 7-86. Relieving platform. Construction stages: (a) 'dredged' construction; (b)
'backfilled' construction. 1-sheet piling; 2-vertical and batter piles; 3-relieving
platform; 4-temporary anchorage for sheet piling; 5-natural slope' 6-dredge line;
7-above water excavation; 8-below water excavation; 9-backfill-first stage; 10-
backfill-second (final) stage.
864 Piled Waterfront Structures

structed on a pile foundation makes all piles ing exceeds those produced by the final de-
work together as a group. In many practica! sign loads, then the designer must decide
cases, the relieving platform is designed in on whether the temporary anchor system
a form of the retaining wall that is cast in should be installed to support the sheet-pile
situ directly on granular backfill material wall during construction (Fig, 7-86b), or
placed behind sheet-piling. At this point, altematively, the strength of sheet piles is
the sheet-piling will be performing as a can- increased to resist the construction loads.
tilever flexible wall loaded with lateral soil
pressures, weight of fresh concrete, and 7.8.2.2.8 Design sequence
weight of construction equipment as dis-
The design of a piled relieving platform
cussed in Chapter 6.
is usually carried out in a sequence as is
If the effects of the aforementioned load-
presented in the following chart.

Determination of the design conditions

J
Preliminary assumptions on wall height and elevation of 1~
relieving platform above seallevel; preliminary pile 1-+--
arrangement and width of relieving platform 1~

~
Preliminary design of sheet pile wall and calculation of
reaction force from earth pressure acting upon relieving t---
platform

~
Preliminary analysis of pile foundation in order to arrive
at the most economica! pile arrangement ~
~
Review of wall global stability

j
Final analysis of pile foundation; design of sheet piles
and other piles included in pile foundation; determination
of pile penetration

Final design of relieving platform


Piled Waterfront Structures 865

REFERENCES AMERICAN SOCIETY FOR TESTING AND MATERIALS,


1980. "Boiler and Pressure Vessel Code."
ABAM ENGINEERS, INC., 1967. "Concentrated American Society of Mechanical Engineers.
Loads on Haunched Deck Panels."Unpub-
AMERICAN SOCIETY FOR TESTING AND MATERIALS,
lished re port. Tacoma, WA.
1988. Annual Book of Standards. "Soil and
ACAR, Y. B., DURGUUOGH, H. T., and TAMAY, M. Rock; Building Stones; Geotextiles," Voi.
T., 1982. "Interface Properties ofSand." ASCE 04.08.
Journal Geotechnical Engineering Division, AMERICAN SOCIETY FOR TESTING AND MATERIALS,
Voi. 108, No. 4. 1989a. ASTM D1143-81, "Standard Method of
AGARWAL, S. L., MALHOTRA, A. K., and BANERJIE, Testing Piles Under Static Axial Compressive
R., 1977. "Engineering Properties of Calcare- Load," Voi. 04.08.
ous Soils Affecting the Design of Deep Pene- AMERICAN SOCIETY FOR TESTING AND MATERIALS,
tration Piles for Offshore Structures." Pro- 1989b. ASTM D3689-83, "Testing Individual
ceedings 9th Annual Offshore Technology Piles Under Static Axial Tensile Load," Voi.
Conference. Houston, TX. 04.08.
AKAGI, T., MURA, K., KAwAKAMI, K., and JAKEI, AMERICAN SOCIETY FOR TESTING AND MATERIALS,
E., 1985. "On Embedded Length ofLong Steel 1989c. ASTM D3966-81, "Standard Method of
Piles with Large Diameter." Proceedings Testing Piles Under Lateral Loads," Voi.
Technical Conference on Penetrability and 04.08.
Drivability of Piles. Japanese Society for Soil ANDERSON, D. G., RIKER, R. E., and ERIKSON, B.
Mechanics and Foundation Engineering, P., 1992. "Pile Driving-Can it Cause Slope
Tokyo. Movement?" ASCE Proceedings Specialty
ALTAEE, A. EVGIN, E., and FELLENIUS, B. H., Conference PORTS '92, Seattle, WA.
1992a. "Axial Load Transfer for Piles in Sand. ANONYMOUS, 1986. "Westshore Terminals Deep
II Numerica! Analysis." Canadian Geotechni- Sea Wharf and Access Trestie." PCI Journal,
cal Journal, Voi. 29. November jDecember.
ALTAEE, A., FELLENIUS, B. H., and EVGIN, E., ANSARI, M.A.R. and KUMAR, M. N., 1986.
1992b. "Axial Load Transfer for Piles in Sand. "Construction of Some Port Facilities at
1. Tests on an Instrumented Precast Pile." Visakhapatnam Outer Harbour (India) Using
Canadian Geotechnical Journal, Voi. 29. Precast Elements." PIANC Bulletin No. 54.
AMERICAN CONCRETE INSTITUTE, 1973. "Recom- ARMALEH, S. and DESAI, C. S., 1987. "Load-
mendations for Design, Manufacture and Deformation Response of Axially Loaded
Installation of Concrete Piles." Journal Piles." ASCE Journal of Geotechnical Engi-
American Concrete Institute, Voi. 70, No. 8; neering, Voi. 113, No. 12.
revised Voi. 71, No. 10; reaffirmed in 1980. ARMITAGE, H. H. and ROWE, R. K., 1983. "Cur-
AMERICAN CONCRETE INSTITUTE, 1989. "Building rent Practice for the Design of Socketed Piles:
Code Requirements for Reinforced Concrete Response to a Survey." Research Report
and Commentary ACI 318-89/ACI 318R-89." GEOT-4-84, Faculty of Engineering Science.
Detroit, Michigan. The University of Wisbern, Ontario.
AMERICAN CONCRETE INSTITUTE, 1993. "Design ASCE DEEP FOUNDATION COMMITTEE, 1984.
and Construction of Fixed Offshore Concrete Practica[ Guidelines for the Selection, Design
Structures ACI 357R-93, Draft No. 1." and Installation of Piles." American Society of
Civil Engineers.
AMERICAN CONCRETE INSTITUTE, 1988. ACI 336.1-
BALANIN, V., BYKOV, L., METELYTSINA, G., and
79 (revised 1988). "Standard Specification for
NATALCHISHIN, G., 1981. "Means for Protect-
the Construction of Drilled Piers."
ing Waterway Slopes and Beds from Streams
AMERICAN PETROLEUM INSTITUTE, 1989. API RP and Waves, Including Ship Effect." PIANC
2A, "Recommended Practice for Planning De- Proceedings 25th Congress Edinburgh.
signing and Constructing Fix Offshore Plat- BARKAN, D. D., 1957. "Foundation Engineering
forms," 18th ed. and Drilling by Vibration Method." Proceed-
866 Piled Waterfront Structures

ings 4th International Conference on Soil Me- Structure" ASCE Proceedings of Specialty
chanics and Foundation Engineering, Vol. 2. Conference PORTS '92, Seattle, WA.
BARKAN, D. D., 1962. Dynamics of Basis and BoUWMEESTER, J., VAN DE KAA, E. J., NuHOFF, H.
Foundations (translated from Russian) A., and ORDEN, R. G. J., 1977. PIANC, Pro-
McGraw-Hill Book Co., New York, NY. ceedings 24th Congress, Section 1, Subject 3,
BATES, R. L. and JACKSON, J. A. (eds.), 1980. Leningrad.
Glossary of Geology. American Geological In- BozoZUK, M., KEENAN, G. H., and PHEENEY, P.
stitute, Falls Church, VA. E., 1979. "Analysis of Load Tests on Instru-
BENGTSSON, P. and SALLFORS, G., 1983. "Floating mented Steel Test Piles in Compressible Silty
Piles in Soft, Highly Plastic Clays." Canadian Soils." Behavior of Deep Foundations, ASTM
Geotechnical Journal, Vol. 2, No. 1. STP670, Philadelphia, PA.
BERGH, H, and CEDERWALK, R., 1981. "Propeller BRIAND, J.-L. and AUDIBERT, J. M. E., 1990.
Erosion in Harbors." Bulletin No. TRITA- "Some Thoughts on API RP2A for Vertically
VBI-107, Hydraulic Laboratory, Royal Insti- Loaded Piles." Proceedings of Annual Off-
tute of Technology, Stockholm. shore Technology Conference, Paper OTC
BERINGEN, F. L., WINDLE, D., and VAN 6418, Houston, TX.
HooYDONK, W. R., 1979. "Results of Loading BRIAND, J.-L. and TuCKER, L. M., 1984. "Piles in
Tests on Driven Piles in Sand." Proceedings Sand: A Method Including Residual Stresses."
Conference on Recent Developments of Design ASCE Journal Geotechnical Engineering, Vol.
and Construction of Piles. Institute of Civil 110, No. 11.
Engineers. London, U.K BRIAND, J-L. and TuCKER, L. M., 1988. "Mea-
BERNARD, R. K, 1968. "Pile-Soil Interaction sured and Predicted Axial Response of 98
During Vibra-Pile Driving." Journal of Mate- Piles." ASCE Journal of Geotechnical Engi-
rials, Vol. 3, No. 1. neering, Vol. 114, No. 9.
BHUSHAN, K, HALEY, S. C., and FONG, P. T., BRITISH STANDARD INSTITUTION, 1979. "Steel,
1979. "Lateral Load Tests on Drilled Piers in Concrete and Composite Bridges." Code of
Stiff Clays." ASCE Journal of Geotechnical Practice for the Design of Composite Bridges
Engineering Division, Vol. 105, No. CT8. BS 5400: Part 5. British Standard Institute.
BILLET, P. and SIEFFERT, J. G., 1989. "Soil-Sheet London.
Pile Interaction in Vibro-Piling." ASCE Jour- BROMS, B., 1964a. "The lateral Resistance ofPiles
nal of Geotechnical Engineering, Vol. 115, No. in Cohesive Soils." ASCE Journal of Soil Me-
8. chanics and Foundation Division, Vol. 90, No.
BIRDY, J. N., ELY, A. L. and HURTINNE, W. E., SM2.
1989. "Effects of Pile Reinforcement on Seis- BROMS, B., 1964b. "The Lateral Resistance of
mic Behavior on Concrete Wharf." ASCE Pro- Piles Cohesionless Soils." ASCE Journal of
ceedings Specialty Conference PORTS '89. Soil Mechanics and Foundation Division, Vol.
Boston, Massachusetts. 20, No. SM3.
BJERRUM, L. and SIMONS, N. E., 1960. "Compari- BROMS, B. B. and BENNERMARK, H., 1968. "Shear
son of Shear Strength Characteristics of Nor- Strength of Soft Clay." Discussion, Proceed-
mally Consolidated Clays." Proceedings of ings Geotechnical Conference, Oslo.
American Society of Engineers, Research BROMS, B. B. and STILLE, H., 1976. "Failure of
Conference on Shear Strength of Cohesive Anchored Sheetpile Walls." ASCE Journal of
Soils, Boulder, Colorado. the Geotechnical Engineering Division, Vol.
BLANEY, G. W., MUSTER, G. L., and O'NEILL, M. III, No. 3.
W., 1987. "Vertical Vibration Test of a Full- BROMS, B. B. and WONG, 1. H., 1985. "Embank-
Scale Pile Group" ASCE Proceedings Dy- ment Piles." Proceedings 3rd International
namic Response of Pile Foundations, Special Geological Seminar, Nanyang Technical Insti-
Publications, No. 11. Atlantic City. tute, Singapore.
BLOWERS, J. et al., 1992. "Ro-Ro Terminal in BRUCE, R. N. and HEBERT, D. C., 1974. "Splicing
Boston Harbor. Strengthening the Dolphin of Precast Prestressed Concrete Piles: Part 1
Piled W aterfront Structures 867

-Review and Performance of Splices." Jour- ASCE Journal of Geotechnical Engineering,


nal of the Prestressed Concrete Institute, Vol. Vol. 117, No. 5.
19, No. 5. CHATTOPADHYAY, B. C. and PISE, P. J., 1986.
BRUCY, F., MEUNIER, J., and NAUREY, J.-F., 1991. "Uplift Capacity of Piles in Sand." ASCE
"Behavior of Pile Plug in Sandy Soils During Journal ofGeotechnical Engineering, Vol. 112,
and After Driving." Proceedings Offshore No. 9.
Technology Conference. Houston, TX. CHAUDHURI, K. P. R. and SYMONS, M. V., 1983.
BuDIN, A. Y. and DEMINA, G. A., 1979. Handbook "Uplift Resistance of Model Single Piles."
"Waterfront Structures." Gosstroyisdat, ASCE Proceedings Geotechnical Practice in
Moscow (in Russian). Offshore Engineering.
Buswv, V. M., 1974. "Experimental Study of CHELLIS, R. D., 1961. Pile Foundations, 2nd ed.
Hydraulic Pressure Effects on Reinforced McGraw-Hill Book Company. New York.
Concrete Cylinders Driven by Vibrating CHIN, F. K., 1970. "Estimations of the Ultimate
Equipment in Marine Pile Foundation Engi- Load of Piles Not Carried to Failure." Pro-
neering." Proceedings 6th Harbour Congress, ceedings 2nd Southeast Asian Conference on
Antwerp. Soil Engineering, Singapore.
BusLOV, V. B. GoULD, J. P., and KocH, R. C., CHIN, F. K., 1971. Discussion, Pile Test-
1988. "The Precast Pier." Civil Engineering, Arkansas River Project. ASCE Journal Soil
December. Mechanics and Foundations Diuision, Vol. 97,
BUSTAMANTE, M. and GIANESELLI, L., 1983. "Pre- No. SM6.
vision de la capacite portan te des pieux par CHIN, C. 0., CHIEW, Y. M., LIM, S. Y., and LIM. T.
la methode penetrometrique." Compte Rendu M., 1996. "Jet Around Vertical Pile," ASCE
de Recherche F.A.E.R. 1.05.02.2. Labaratoire Journal of Waterway, Port, Coastal, and
Central des Ponts et Chanssees (in French). Ocean Engineering, Vol. 122, No. 2,
BUTLER, H. D. and HoY, H. E., 1977. User's MarchjApril.
Manual for the Texas Quick-Load Method for CHOW, Y. K., KARUNARATNE, K. Y., WONG, K. Y.,
Foundation Load Testing. Federal Highway and LEE, S. L., 1988. "Prediction of Load-
Administration, Office of Development, Carrying Capacity of Driven Piles." Canadian
Washington, DC. Geotechnical Journal, Vol. 25.
MEYERHOF, G. G. and FELLENIUS, B. H. (eds.), CIDECT, MoNOGRAPH, No. 5. (Canadian edition
1985. Canadian Foundation Engineering 1981) Part One: Design of Concrete-Filled
Manual, Canadian Geotechnical Society. Hollow Section Columns: Calculation Method
Vancouver, B.C. and Application Technology. Part Two: De-
CANADIAN STANDARD AssociATION, 1984. "Steel sign Charts, edited by CIDECT.
Structures for Buildings" (Limit State De- CLARK, A. R. and WALKER, B. F., 1977. "A Pro-
sign), CAN3-516.1 M84. posed Scheme of the Classification and
CARTER, J. P. and KuLHAWY, F. H., 1992. "Analy- Nomenclature for Use in Engineering De-
sis ofLaterally Loaded Shafts in Rock." ASCE scription of Middle Eastern Sedimentary
Journal ofGeotechnical Engineering, Vol. 118, Rocks." Geotechnique, Vol. 27, No. 1.
No. 6. COOPER, S. E., 1991. "Ductile Frames are Tough
CHALY, A. and HANNA, A., 1992. "Stress and for Earthquake." ASCE Civil Engineering,
Strain Around Helical Screw Anchors in August.
Sand." JSSMFE Soils and Foundations, Vol. COYLE, H. M. and CASTELLO, R. R., 1981. "New
32, No. 4. Design Correlations for Piles in Sand," ASCE
CHALY, A. and HANNA, A., 1994. "Model lnvesti- J ournal of Geotechnical Engineering, Vol. 107,
gation of the Performance of Single Anchors No. 7.
and Group of Anchors." Canadian Geotechni- CROOKS, J. H. A., MATYAS, E. L., and McKAY, H.
cal Journal, Vol. 31. M., 1980. "Excavation Slope Stability Related
CHALY, A., HANNA, A., and HANNA, M., 1991. to Pore Water Pressure Variation During Pil-
"Uplift Behavior of Screw Anchors in Sand." ing." Canadian Geotechnical Journal, Vol. 17.
868 Piled Waterfront Structures

CUR 1 CIRIA, 1991. Manual on the Use of Rock DAVISSON, M. T., 1972. "High Capacity Piles."
in Coasta! and Shoreline Engineering Centre Proceedings, Lecture Series Innovations in
for Civil Engineering Research and Cades. Foundation Construction. ASCE, Illinois Sec-
Gonda, London. tion. Chicago.
DAILEY, J. E. et al. 1981. "Pile Fatigue Failures. DAVISSON, M. T., 1989. "Foundations on Difficult
Part II: Mechanism Studies and Part III: Mo- Soild-State of the Practice Deep Founda-
tion in Seas." ASCE Journal of Waterway tions-Driven Piles." ASCE Seminar on
Port, Coastal, and Ocean Engineering, Vol. Foundations in Difficult Soils, New York.
113, No. 3, May. DAVISSON, M. T. and GILL, H. L., 1963.
D'APPOLONIA, D. J., 1971. "Effects of Foundation "Laterally-Loaded Piles in a Layered Soil
Construction on Nearby Structures." Proceed- System." ASCE Journal of Soil Mechanics
ings Fourth Panamerican Conferences on Soil and Foundation Division, Voi. 89, No. SM3.
Mechanics and Foundation Engineering, Vol.
DEBEER, E. E., 1965. "Bearing Capacity and
1.
Settlement of Shallow Foundations on Sand."
D'APPOLONIA, D. J., and LAMBE, T. W., 1971. Proceedings of the Symposium on Bearing
"Performance of Four Foundations on End Capacity and Settlement of Foundations,
Bearing Piles." ASCE Journal of Soil Me- Duke University.
chanics and Foundations Division, Voi. 97,
No. SM1. DEBEER, E. E. and WALLAYS, M. 1972. "Franki
Piles with Overexpanded Bases." La Tech-
DAMASCHI, G. L. and BONASIA, J. J., 1989. "U.S.
nique des Travaux, No. 333.
Navy Homeporting Project Stapleton, New
York." ASCE Proceedings Specialty Confer- DE GIJT, J. G., VAN DER HoRST, H., and VAN
ence PORTS '89, Boston, MA. ScHAIK, C. N., 1993. "Quay-Wall Design and
DAS, B. M. and SEELEY, G. R., 1975. "Uplift Construction in the Port of Rotterdam."
Capacity of Buried Model Piles in Sand." PIANC Bulletin No. 80.
ASCE Journal Geotechnical Engineering Di- DE NICOLA, A. and RANDOLPH, M. F., 1983.
uision, Vol. 101, No. 10. "Tensile and Compressive Shaft Capacity of
DAs, M. M., 1970. "A Literature Review on Bed- Piles in Sand." ASCE Journal of Geotechnical
Load Transport Due to Wave Action and Engineering, Voi. 119, No. 12.
Localized Scour in Non-cohesive Sediments." DET NORSKE VERITAS, 1977. Appendix F-
Final report HEL 21-6; A Literature Review Foundations. Novik, Norway.
on Erosion and Deposition of Sediment near DICKIN, E. A and LEUNG, C. F., 1992. "The In-
Structure in Ocean, Einstein, H. A and fiuence of Foundation Geometry on the Uplift
Wiegel, R. L. (eds.). Hydraulic Engineering Behavior of Pile with Enlarged Bases."
Laboratory, College of Engineering, Univer- Canadian Geotechnical Journal, Vol. 29.
sity of California. Berkeley.
DOWNS, D. l. and CHIENRZZI, R., 1966. "Trans-
DATTA, M., GULHATI, S. K., and RAo, G. V., 1980.
mission Tower Foundation." Journal Power
"An Appraisal of the Existing Practice of De-
Division, ASCE, Voi. 92, No. P02.
termining the Axial Load Capacity of Deep
Penetration Piles in Calcareous Sands." Pro- DUNCAN, J. M., EVANS, L. T., and OOI, S. K.,
ceedings 12th Offshore Technology Confer- 1994. "Lateral Load Analysis of Single Pile
ence, Houston, TX. and Drilled Shafts." ASCE Journal of
DA"v'lDSON, J. L. and BOGHRAT, A., 1983. "Dis- Geotechnical Engineering, Voi. 120, No. 5.
placement and Strains Around Probes in DUNNICLIFF, J., 1988. Geotechnical Instrumenta-
Sand." ASCE Proceedings Conference on tion for Monitoring Field Performance. John
Geotechnical Practice in Offshore Engineer- Wiley and Sons, New York.
ing. EADIE, N, R. V. and HERBICH, J. B., 1986. "Scour
DAVISSON, M. T., 1970. "Static Measurement of about a Single Cylindrical Pile due to Com-
Pile Behavior." Proceedings Conference on bined Random Waves and a Current." ASCE
Design and Installation of Pile Foundations Proceedings 20th Coasta! Engineering Con-
and Cellular Structures, Bethlehem, PA ference.
Piled Waterfront Structures 869

EL GHAMRY, O. 1963. "Wave Forces on a Dock." FRENCH, J. A., 1979. "Wave Uplift Pressures on
Report No. HEL-9-1, University of California, Horizontal Platforms." ASCE Proceedings
Berkeley. Specialty Conference, Civil Engineering in the
ENSOFT INC., 1989. LPILE-Computer Program Oceans IV, New York.
for Laterally Loaded Pile Analysis. Ensoft Inc. FuEHRER, M., RCiMISCH, K., and ENGELKE, G.,
ERICKSON, B. P., McCoLLOUGH, T. J., and SURKo, 1981. "Criteria for Dimensioning the Bottom
A., 1983. "Massport Modernization." Civil and Slope Protections and for Applying the
Engineering, April. New Methods of Protecting Navigation
ERIKSSON, K. and KLINBENBERG, H., 1986. Canals." PIANC Proceedings 25th Congress,
"Precast Elements in Port Structures in Edinburgh.
Scandinavia." PIANC Bulletin No. 54. FUKUTE, T., K:!YOMIYA, O. and MINAMI, K., 1990.
EUROPEAN CONVENTION FOR CONSTRUCTION "Steel Structures in Port and Harbour Facili-
STEELWORK, 1979. "Buckling of Shells." Rec- ties. Actual Conditions of Corrosion and
ommendations of Task Group T8-79-37. Pro- Countermeasures." PIANC Bulletin No. 68.
posed Rules R4.6. FULLER, F. M., 1983. Engineering of Pile Instal-
FANG, H.-Y. (ed.), 1991. Foundation Engineering lations. McGraw-Hill Book Co., New York.
Handbook, 2nd ed. Van Nostrand Reinhold, F'uLLER, F. M. and HoY, H. E., 1970. "Pile Load
New York. Test Including Quick-Load Test Method, Con-
FELLENIUS, B. H., 1975. "Test Load of Piles and ventional Methods and Interpretations." HRB
New Proof Testing Procedure." ASCE Jour- 333.
nal Geotechnical Engineering Division, Vol. GEOSOFT INc., 1984. PILEDG-Computer Pro-
101, No. GT9. gram for Laterally Loaded Pile Analysis.
FELLENIUS, B. H., 1980. "The Analysis of Results Geosoft Inc.
from Routine Pile Tests." Ground Engineer- GERWICK, B. C. Jr., 1968. "Prestressed Concrete
ing, Vol. 13, No. 6. Piles." Journal of the Prestressed Concrete
FELLENIUS, B. H., 1984. "Ignorance is Bliss." Institute, Vol. 13, No. 5.
Geotechnical News, Vol. 2, No. 4. GERWICK, B. C. Jr., 1971. Construction of
FELLENIUS, B. H., 1991. "Pile Foundations." Prestressed Concrete Structures." Wiley-
Foundation Engineering Handbook, 2nd ed., Interscience, New York.
Fang, H.-Y. (ed.) Van Nostrand Reinhold, New
GERWICK, B. C. Jr., 1974. "Marine Structures."
York.
Handbook of Concrete Engineering, Fintel, M.
FERGUSON, E. S., 1992. Engineering and the (ed.) Van Nostrand Reinhold Co., New York.
Mind's Eye. MIT Press, Cambridge, MA.
GERWICK, B. C. Jr., 1986. Construction of Off-
FLEMING, W. G. K., 1992. "A New Method for
shore Structures. John Wiley and Sons, New
Single Pile Settlement Prediction and Analy-
York.
sis." Geotechnique, Vol. 42, No. 3.
GERWICK, B. C., BACH, P. E., and FOTINOS, G.,
FoTINOS, G. C., 1986. "New Structural Develop-
1995. "In the Wet." ASCE Civil Engineering,
ments for Concrete Marine Structures." ASCE
May.
Proceedings Specialty Conference PORTS '86,
Oakland, CA GILCHRIST, J. M., 1985. "Load Tests on Tubular
FRAGASzy, R. J., ARGO, D., and HIGGINS, J. D., Piles in Coralline Strata." ASCE Journal of
1986. "Comparison of Formula Predictions Geotechnical Engineering, Vol. 111, No. 5.
with Pile Load Tests." Transportation Re- GILMAN, J. F. and NOTTINGHAM, D. 1991. "En-
search Rec. 1219. vironmental W ave Barriers in Small Craft
FRAGASzy, R. J., HIGGINS, J. D., and LAWTON, E. Harbors." ASCE Proceedings Specialty Con-
C., 1985. "Development of Guidelines for Con- ference, "World Marina."
struction Control of Pile Driving and Esti- GLE, D. R., 1981. "The Dynamics Lateral Re-
roate of Pile Capacity." Report Wa-RD 68.1. sponse of Deep Foundations." Ph.D Disserta-
Washington State Transportation Center. tion, Department of Civil Engineering, The
Olympia, WA University of Michigan, Ann Arbor.
870 Piled Waterfront Structures

GLE, D. R. and Woons, R. D., 1984. "Suggested liANNA, T. H. and TAN, R. H. S., '973. "The
Procedure for Conducting Dynamic Lateral Behavior of Long Piles Under Compressive
Load Test on Pile." Symposium on Laterally Loads in Sand." Canadian Geotechnical
Loaded Deep Foundations, Analysis and Per- Journal, Vol. 10, No. 3.
formance, ASTM STP 835, Kansas City, MO. HANSEN, B. J., 1961. "The Ultimate Resistance of
GLOs, G. H. and BRIGGS, 0. H., Jr., 1983. "Rock Rigid Piles Against Transversal Forces."
Sockets in Soft Rock." ASCE Journal of Danish Geotechnical Institute, Bulletin No.
Geotechnical Engineering, Vol. 109, No. 4. 12. Copenhagen.
GOBLE, G. G. and R.AUSCHE, F., 1980. "Wave HANSEN, B. H., 1963. Discussion. "Hyperbolic
Equation Analysis of Pile Driving-WEAP Stress-Strain Response Cohesive Soils."
Program." Submitted to U.S. Department of ASCE Journal Soil Mechanics and Founda-
Transportation by Goble and Associates, Inc. tion Engineering, Vol. 89, No. SM4.
GoRUNOV, B. F., 1984. Use of Prestressed Rein- HEEREMA, E. P. and DELONG, A., 1980. "An Ad-
forced Concrete for Construction of Hydraulic vance Wave Equation Computer Program
Structures. Stroyisdat Publisher. Moscow (in which Simulates Dynamic Pile Plugging
Russian). through a Coupled Mass-Spring System." ICE
GREEN, P., 1989. "Navy Permits Nontraditional Proceedings Conference on Numerica! Meth-
Design." Engineering News Record, February ods in Offshore Piling, London.
9. HEGEDUS, E. and KHosLA, U. K., 1984. "Pullout
Grundbau-Taschenbuch, Vol. 1, 1955. Ernst, Resistance of H-Piles." ASCE Journal of
Berlin. Geotechnical Engineering, Vol. 110, No. 9.
GRUTMAN, M. C., 1971. "Approximate Method for HEINZ, R., 1993. "Plastic Piling." Civil Engineer-
Computing the Ultimate Unit Shaft Resis- ing, April.
tance. Seminar Presentation. Budivelnik, Kiev
HERBICH, J. B., 1991. "Scour Around Pipelines,
(in Russian).
Piles and Seawalls." Handbook of Coastal and
GUIAUX, P. and.JANSS, J., 1970. 'Comportiment
Ocean Engineering, Vol. 2, Herbich, J. B. (ed.).
au flambement de colonues constituees de
Gulf Publishing Co. Houston, TX.
tubes en acier remplish de beton." Fabrimetal
Publication CRIF-MT65, Brussels. HERBICH, J. B., SCHILLER, R. E. JR., DUNLAP, w.
A., and WATAHABE, R. K., 1984. SeafloorScour.
HAGENAAR, J. and VAN SETERS, A., 1985. "Ulti-
Marcel Decker Inc., New York.
mate Axial Bearing Capacity of Piles Driven
into Coral Rock and Carbonate Soils." Pro- HIRSCH, T. J., CARR, L., and LOWRY, L. L., 1976.
ceedings of llth International Conference on "Pile Driving Analysis." Waue Equation User's
Soil Mechanics and Foundation Engineering, Manual TT1 Program 3 volumes. U.S. De-
Vol. 3. San Francisco, CA. partment of Transportation, Federal Highway
HALFIN, J. S., 1990. Effects of Wave Action on Administration Office of Research and Devel-
Offshore Platforms. (In Russian), Nedra Pub- opment, Washington, DC.
lisher, Moscow. HJORTH, P., 1975. "Studies on the Nature of
liAMILL, G. A. and JOHNSTON, H. T., 1993. "The Local Studies." Bulletin Series A, No. 46. De-
Decay of Maximum Velocity within the Initial partment of Water Resources Engineering.
Stages of a Propeller Wash," Journal of Lund Institute of Technology, University of
Hydraulic Research, Vol. 31, No 1. Lund, Sweden.
HAMIL, G. A., QURRAIN, R. M. M. and JoHNSTON, HOFMANN, K.-F., 1989. "New Methods for the
H. T., 1996. "The Influence of a Revetment on Construction of Quay Walls in the Port-of-
Diffusion of a Propellar Wash," PIANC, Bul- Hamburg," PIANC Bulletin No. 64.
letin No 91, June. HOLLOWAY, D. M., CLOUGH, G. W., and VESIC, A.
RANG, S. C., 1969. "Stability Analysis of Pile S., 1978. "The Effects of Residual Driving
Rigs." Foundation Facts, Vol. V, No. 1. Ray- Stresses on Pile Performance Under Axial
mond International Inc., New York. Loads." Proceedings Annual Offshore Tech-
HANNA, A. M. and AFRAM, A., 1986. "Pullout nology Conference, Houston, TX.
Capacity of Single Batter Piles in Sand." HORVATH, R. G., 1980. Research Project Report
Canadian Geotechnical Journal, Vol. 23. on Load Transfer System for Rock-Socketed
Piled Waterfront Structures 871

Drilled Pier Foundations. For the NRC of American Conference on Deep Foundations,
Canada, DSS file N10X5.31155-9-4420, con- Mexican Society of Soil Mechanics.
tract serial No. ISX79.00531. KEzor, A, 1975. "Pile Foundations." Foundation
HunsoN, R. Y., 1959. "Laboratory Investigation Engineering Handbook, Winterkorn, H. F. and
of Rubble-Mound Breakwaters." ASCE Jour- Fang, H.-Y. (eds.). Van Nostrand Reinhold,
nal of Waterways and Harbors Diuision, Vol. New York.
85, No. 3. KING and GAVARIS, 1961. Confidential Report,
HUNTER, M. F., YOURMAN, AM., DIAZ, G. M., and June.
HsucH-HSIN CHU, 1992. "Pile Lateral Load KISHIDA, H., 1967. "Ultimate Bearing Capacity of
Test in the Port-of-Los Angeles," ASCE Pro- Piles Driven into Loose Sand." Soils and
ceedings Specialty Conference PORTS'92, Foundations, Vol. 7, No. 3.
Seattle, WA KODIKARA, J. K and MooRE, 1. D., 1993. "Axial
IRELAND, H. 0., 1957. "Piling Tests on Piles in Response of Tapered Piles in Cohesive Fric-
sand." Proceedings 4th International Confer- tional Ground." ASCE Journal of Geotechni-
ence on Soil Mechanics and Foundation Engi- cal Engineering, Vol. 119, No. 4.
neering, Vol. 2. KRAFT, L. M. JR., 1991. "Performance of Axially
ISMAEL, N. F. and KLYM, T. W., 1979. "Uplift and Loaded Pipe Piles in Sand." ASCE Journal of
Bearing Capacity of Short Piers in Sand." Geotechnical Engineering, Vol. 117, No. 2.
ASCE Journal of Geotechnical Engineering, KRAFT, L. M. JR., RAY, R. P., and KAGAWA, T.,
Vol. 105, No. 5. 1981. "Theoretical t-z Curves." ASCE Jour-
JAMIOLKOWSKI, M., LADD, C. C., GERMAINE, J. T., nal ofGeotechnical Engineering Diuision, Vol.
and LANCELOTTA, R., 1985. "New Develop- 107, No. GT11.
ments in Field and Laboratory Testing of KULAK, G. L., STEPHENS, M. J., and BAILEY, R.
Soils." Theme Lecture No. 2. Proceedings of W., 1988. "Bending Tests of Large Diameter
the XI International Conference on Soil Me- Fabricated Steel Cylinders." Canadian Jour-
chanics and Foundation Engineering, San nal of Civil Engineering, Vol. 15.
Francisco, CA LADANYI, B. and RoY, A, 1971. "Some Aspects of
JANSEN, P. PH. and SCHIJT, J. B., 1953. Baring Capacity of Rock Mass." Proceedings
PIANC, Eighteenth International Navigation of the 7th Canadian Symposium on Rock Me-
Congress, Section 1, Communication 1, Rome. chanics, Edmonton, Alberta.
JAPANESE TECHNICAL STANDARD FOR PORT AND LAI, C. P. and LEE, J.-J., 1989. "Interaction of
liARBOUR F ACILITIES, 1980. The Overseas Finite Amplitude Waves with Platforms or
Coastal Area Development Institute of Japan, Docks." ASCE Journal of Waterway, Port,
Tokyo. Coasta[ and Ocean, Vol. 115, No. 1.
Josm, R. C. and SHARMA, H. D., 1987. "Predict- LAMBE, T. W., 1951. Soil Testing for Engineers
ion of Ultimate Pile Capacity from Load Tests John Wiley and Sons, New York.
on Bored and Belled, Expanded Base Com- LAsEBNIK, G. Y., 1961. "lnvestigation of An-
pacted and Driven Piles." Proceedings Inter- chored Sheet-Pile Bulkheads." Thesis pre-
national Symposium on Prediction and sented to the Kiev Politechnical Institute,
Performance in Geotechnical Engineering, Kiev Ukraine, in partial fulfillment of the
Calgary, Alberta. requirements for the degree of Doctor of
KAWATA, Y. and TSUCHIYA, Y., 1988. "Local Scour Philosophy (in Russian).
Around Cylindrical Piles Due to Waves and LAVRINOVICH, E. V., FILIPPOV, V. N., PYATETSKI,
Current Combined." ASCE Proceedings 21st V. M., and KILLAN, 1. K, 1967. "Formation of
Coastal Engineering Conference, Vol. 2. Cracks in Hollow Piles as a Result of Cold
KERISEL, J., 1961. "Foundations Profondes en Temperature." Transportnoye Stroitelstvo,
Milieu Sableux." Proceedings 5th Interna- No. 10 (in Russian).
tional Conference on Soil Mechanics and LAWTON, E. c., FRAGAS2Y, R. J., HrGGINS, J. D.,
Foundation Engineering, Vol. 2. KILIAN, A P., and PETERS, A. J., 1986. "Re-
KERISEL, J. 1964. "Deep Foundations, Basic view ofMethods for Estimating Pile Capacity."
Experimental Facts." Proceedings North Transportation Research Rec. 1105.
872 Piled Waterfront Structures

LEE, s. L., CHOW, Y. K., KARUNARATNE, G. P., nal Soil Mechanics and Foundation Engineer-
and WONG, K. Y., 1988. "Rational Wave Equa- ing Division, Vol. 96, No. 5.
tion Model for Pile Driving Analysis." ASCE MARTIN, S. K. and MAYNORD, S. T., 1992. "Riprap
Journal Geotechnical Engineering, Voi. 114, Design in Marina Terminals." ASCE Proceed-
No. 3. ings Specialty Conference. PORTS '92,
LEFEBVRE, G., ROHAN, K., BELFADHEL, M. B., and Seattle, WA.
DASCAL, 0., 1992. "Field Performance and
MASSARSCH, K. R. and BROMS, B. B., 1981. "Pipe
Analysis of Steep Riprap." ASCE Journal of
Driving on Clay Slopes." Proceedings 10th
Geotechnical Engineering, Vol. 118, No. 9.
International Conference on Soil Mechanics
LEHANE, B. M., JARDINE, R. J., BoND, A. J., and and Foundation Engineering, Vol. 3, Stock-
FRANK, R., 1993. "Mechanism of Shaft Fric- holm.
tion in Sand From Instrumented Pile Tests."
ASCE Journal of Geotechnical Engineering, MATLOCK, H., 1970. "Correlations for Design of
Vol. 119, No. 1. Laterally Loaded Piles in Soft Clay." Proceed-
LEITRASS, R., 1979. "Wave Damages to Stone ings 2nd Annual Offshore Technology Confer-
Slope Under Marginal Wharves." ASCE Pro- ence, Houston, TX.
ceedings Specialty Conference Coastal Struc- MATLOCK, H. and REESE, L. C., 1960. "Gener-
tures '79. alized Solutions for Laterally Loaded Piles."
LEVACHER, D. B. and SIEFFERT, J.-G., 1984. "Test ASCE Journal of Soil Mechanics and Foun-
of Model Tension Piles." ASCE Journal of . dation Division, Voi. 86, No. SM5.
Geotechnical Engineering, Vol. 110, No. 12. MA'l'LOCK, H. and REESE, L. C., 1961. "Founda-
LIANG, R. Y. and HUSSEIN, A. I., 1993. "Sim- tion Analysis of Offshore Pile Supported
plified Dynamic Method for Pile-Driving Con- Structures." Proceedings Fifth International
trol." ASCE Journal of Geotechnical Engi- Conference on Soil Mechanics and Founda-
neering, Vol. 119, No. 4. tion Engineering, Vol. 2, Paris.
LIANG, R. Y. and SHENG, Y., 1993. "Wave Equa- MATLOCK, H. and REESE, L. C., 1962. "Gener-
tion Parameters from Driving-Rod Test." alized Solutions for Laterally Loaded Piles."
ASCE Journal of Geotechnical Engineering, Transaction of ASCE, Voi. 127, Part 1.
Vol. 119, No. 6.
MAYNORD, S. T. and SIEMSEN, T. S., 1991. "Re-
LIBBY, J. R., 1984. Modern Prestressed Concrete, turn Velocities Induced by Shallow-Draft
3rd ed. Van Nostrand Reinhold, New York. Navigation." ASCE Proceedings National
Lo. G. K. and BECKWITH, R. D., 1987. "Treasure Conference on Hydraulic Engineering.
Island Berthing Pier." Concrete Interna-
MAzURKIEWICZ, B. K., 1972. "Test Loading on
tional, May.
Piles According to Polish Regulations." Royal
LOWE, J., III and ZAcCHEO, P. F., 1991. "Sub- Swedish Academy of Engineering Science
surface Explorations and Sampling." Founda-
Commission on Pile Research, Report No. 35.
tion Engineering Handbook, 2nd ed. Fang, Stockholm.
H.-Y. (ed.). Van Nostrand Reinhold, New York.
LOWERY, L. L., HIRSCH, T. J., EDWARDS, T. C., McCLELLAND, B., 1974. "Design ofDeep Penetra-
COYLE, H. M., and SAMSON, C. H., 1969. "Use tion Piles for Ocean Structures." ASCE
of the Wave Equation to Predict Soil Resis- Journal ofGeotechnical Engineering, Vol.100,
tance on a Pile During Driving." Proceedings No. 7.
7th International Conference on Soil Mechan- MCCLELLAND, B., FOCHT, J. A., and EMRICH, W.
ics and Foundation Engineering, Special J., 1969. "Problems in Design and Installation
Session No. 8, Mexico City. of Offshore Piles," ASCE Journal Soil Me-
LUTENEGGER, A. J. and MlLLER, G. A., 1994. chanics and Foundation Design, Vol. 95, No.
"Uplift Capacity of Small-Diameter Drilled SM6.
Shafts from In Situ Tests." ASCE Journal of McKILLEN, G., 1985. "A Model and Field Study
Geotechnical Engineering, Vol. 120, No. 8. of Ship Propulsion Induced Bed Movement of
MANSUR, C. I. and HUNTER, A. H., 1970. "Pile Berth." M.S. theses, Queen's University of
Tests-Arkansas River Project." ASCE Jour- Belfast, Ireland.
Piled Waterfront Structures 873

McVAY, M., CASPER, R., and SHANG, TE-l., (1995). NAUROY, J. F. and LETIRANT, P., 1983. "Model
"Lateral Response of Three-Row Groups in Tests of Piles in Calcareous Sands." ASCE
Loose to Dense Sands at 3D and 5D Pile Proceedings, Geotechnical Practice in Off-
Spacing," Vol. 121, No. 5. shore Engineering, The University of Texas
MEIGH, A. C., 1985. "In Situ Testing Using the at Austin.
Static Cone Penetration." Draft Report to NAUROY, J. E. and LETIRANT, P., 1985. "Driven
CIRIA, Appendix G. London. Piles and Drilled and Grouted Piles in Cal-
MEYERHOF, G. G., 1959. "Compaction of Sands careous Sands." Proceedings 7th Annual Off-
and Bearing Capacity of Piles." Japanese shore Technology Conference, Houston, TX.
Society for Soil Mechanics and Foundations NAVASIMHA RAo, S. and PRASAD, Y. V. S. N.,
Design, ASCE, Vol. 82, SMl. 1993. "Uplift Behavior of Pile Anchors Sub-
MEYERHOF, G. G., 1973. "Uplift Resistance of jected to Lateral Cyclic Loading." ASCE
Inclined Anchors and Piles." Proceedings of Journal ofGeotechnical Engineering, Vol. 119,
8th International Conference on Soil Mechan- No.4.
ics and Foundation Engineering, Vol. 2, NAVFAC 1982. Design Manuals DM. 7.1 and
Moscow. 7.2. Foundations and Earth Structures. De-
MEYERHOF, G. G., 1979. "Bearing Capacity and partment ofthe Navy. Alexandria, VA.
Settlement of Pile Foundations." ASME Jour- NEELY, W. J., 1990. "Bearing Capacity of
nal ofGeotechnical Engineering, Vol. 102, No. Expended-Base Piles in Sand." ASCE Jour-
GT3. nal Geotechnical Engineering Division.
MEYERHOF, G. G., 1979. "Pile Foundations with NEOGI, P. K., SEN, H. K. and CHAPMAN, J. C.,
Special Reference to Bridges." Proceedings 1969. "Concrete-Filled Tubular Steel Columns
Seminar of International Association for Under Eccentric Loading." Structural Engi-
Bridge and Structural Engineering, Vol. 1, neer, Vol. 47, No. 5.
Madra, India. NEW YoRK STATE DOT, 1974. Static Load Test
MINDLIN, R. D., 1936. "Force at a Point in the Manual. N.Y. DOT Soil Mechanics Bureau,
Interior of a Semi-Infinite Solid." Physics, Vol. Albany. Soil Control Procedure SCP 4/74.
7, No. 5. NooRANY, I., 1985. "Side Friction of Piles in
MITCHELL, A., 1848. "On Submarine Founda- Calcareous Sands." Proceedings 11th Inter-
tions; Particularly the Screw Pile and Moor- national Conference on Soil Mechanics
ings." Proceedings ICE VII, Vol. 108. and Foundation Engineering, Vol. 3, San
MITCHELL, J. K. and SOLYMAR, Z. V., 1984. Francisco, CA.
"Time-Dependent Strength Gained in Freshly NORLUND, R. L. 1963. "Bearing Capacity of Piles
Deposited or Densified Sand." ASCE Journal in Cohesion Soils." ASCE Journal of Soil
Geotechnical Engineering Division, Vol. 110, Mechanics and Foundation Engineering Divi-
No. 11. sion, Vol. 89, No. 3.
MoHAN, D., JAIN, G. S., and JAIN, M. P., 1967. "A O'NEILL, M. W. and RAINEs, R. D., 1991. "Load
New Approach to Load Tests." Geotechnique, Transfer for Pipe Piles in Highly Pressured
Vol. 17. Dense Sand." ASCE Journal of Geotechnical
MoRGAN, H. D., 1944. "Design of Wharves on Engineering, Vol. 117, No. 8.
Soft Ground." Journal ICE, Vol. XXII, No. 5. OLSON, R. E., 1990. "Axial Load Capacity of Steel
MouCHEL, L. G. and PARTNERS, 1957. "Founda- Pipe Piles in Sand." Proceedings Annual
tions by A. L. Little." Edward Arnold Ltd., Offshore Technology Conference, Paper
London. 1961. OTC6419. Houston, TX.
MURFF, J. D., 1987. "Pile Capacity in Calcareous Oor, P. S. K. and DUNCAN, J. M., 1994. "Lateral
Sands: State of the Art." ASCE Journal of Load Analysis on Groups of Piles and Drilled
Geotechnical Engineering, Vol. 113, No. 5. Shafts." ASCE Journal of Geotechnical Engi-
MURFF, J. D. and RAINEs, R. D., 1990. "Soil Plug neering, Vol. 120, No. 6.
Behavior of Piles in Sand." Proceedings Off- PADRON, D. V., 1985. "Pier Review." Civil Engi-
shore Technology Conference, Houston, TX. neering, September.
874 Piled Waterfront Structures

PADRON, D. V. and ELZOGHBY, H. M., 1986. PLANTEMA, G. and NOLET, P. 1957. "Infiuence of
"Berthing Impact Force Distribution on· Pile Pile Driving on the Sounding Resistance in a
Supported Pier Structures." ASCE Proceed- Deep Sand Layer." Proceedings 4th Interna-
ings Specialty Conference PORTS '86, tional Conference on Soil Mechanics and
Oakland, CA Foundation Engineering, Vol. IL
PADRON, D. V. and WmTE, S. M., 1983. "Optimiz- PoEPSEL, P. H. and Down, W. M., 1995. "Elasto-
ing Pier Design by Utilizing Deck Stiffness." Plastic Response of Deep Foundations and
ASCE Proceedings Specialty Conference Soil for Ship Impact Loading; Applachicola
PORTS '83, New Orleans, LA. River bridge, Florida. ASCE Proceedings
PAIKOWSKY, S. G., 1990. "The Mechanism of Pile Specialty Conference PORTS '95, Tampa, FL.
Plugging in Sand." Proceedings Offshore PosT, N., 1989. "Redesign Paves Navy Pier Cost."
Technology Conference, Houston, TX. Engineering News Record, March 10.
PALMER, H. D., 1969. "Wave-Induced Scour on POTIONDY, J. G., 1961. "Skin Friction Between
the Sea Floor." ASCE Proceedings of Various Soils and Construction Materials."
Specialty Conference Civil Engineering in the Geotechnique, Vol. 11, No. 4.
Oceans IL Pouws, H. G., 1987. "Analysis ofResidual Stress
Effects in Piles." ASCE Journal of Geotechni-
PARENT, M. G., 1986. "Application of Prefabri-
cal Engineering, Vol. 113, No. 3.
cated Elements in Quay Wall Structures in
the Port-of-Rotterdam." PIANC Bulletin No. POULOS, H. G., 19S9. "Pile Behavior-Theory
54. and Application." Geotechnique, Vol. 39, No. 3.
PoULOS, H. G. and DAVIS, E. H., 1980. "Pile
PARK, R. and FALCONER, T. J., 1983. "Ductility of
Foundation Analysis and Design." John Wiley
Prestressed Concrete Piles Subjected to Simu-
and Sons, New York.
lated Seismic Loading." PCI Journal, Vol. 28,
PRAKASH, S. and SHARMA, H. D., 1969. "Analysis
No. 5.
of Pile Foundations Against Earthquakes."
PIANC, 1984. Report of the International Com- Indian Concrete Journal, Vol. 43, No. 6. ·
mission for Improving the Design of Fender PRAKASH, S. and SHARMA, H. D., 1990.
"Pile
Systems. Permanent International Associa- Foundations in Engineering Practice." John
tion Navigation Congresses, Supplement to Wiley & Sons, New York.
Bulletin No. 45.
PRAKASH, S., RANJAN, G., and KUMAR, K., 1985.
PIANC, 1987. Guidelines for the Design and "Dynamic Soil-Pile Constraint for Turbogen-
Construction of Flexible Revetments Incorpo- erator Foundations." Madras Refineries Re-
rating Geotextiles for Inland Waterways. Per- port Issued by Geotechnical Division, Central
manent International Association of Naviga- Building Research Institute, Roorkee, India.
tion Congresses. Report of Working Group 4, PRAKAsH, S. and KUMAR, S. 1996. "Nonlinear
Supplement to Bulletin No. 57. Lateral Pile Defiection Prediction in Sands."
PIANC, 1992. Guidelines for the Design and ASCE Journal of Geotechnical Engineering,
Construction of Flexible Revetments Incorpo- Vol. 122, No. 2, February.
rating Geotextiles in Marine Environment. QmNN, A F., 1972. Design and Construction
Permanent International Association of N avi- of Port and Marine Structures, 2nd ed. Mc-
gation Congresses. Report of Working Group Graw-Hill Ryerson Limited, New York.
No. 21, Supplement to Bulletin No. 78/79. RAINES, R. D., UGAZ, O. G., and O'NEILL, M. W.,
PIERS, D. and MARTINSON, R. E., 1992. "Seismic 1992. "Driving Characteristics of Open-Toe
Rehabilitation of Seattle's Pier 69." ASCE Piles in Dense Sand." ASCE Journal of
Proceedings, Specialty Conference PORTS '92, Geotechnical Engineering, Vol. 118, No. 1.
Seattle, WA RANDOLPH, M. F., 1985. "Capacity of Piles
PINKNEY, R. B., STEPHENS, M. J. MURRAY, D. W., Driven into Dense Sand." Presentation to XI
and KuLAK, G. L., 1983. "Use of Measured ICSMFE, San Francisco, CA
Imperfections to Predict the Buckling of Axi- RANDOLPH, M. F., DOLWIN, J., and BECK, R. D.,
ally Loaded Cylindrical Shells." Canadian 1993. "Design of Driven Piles in Sand."
Journal of Civil Engineering, Vol. 10, No. 4. Geotechnique.
Piled Waterfront Structures 875

RANDOLPH, M. F., LEONG, E. C., and HOULSBY, G. ROESLI, A, 1955. "Lateral Load Distribution in
T., 1991. 'One-Dimensional Analysis of Soil Multi-Beam Bridges." Fritz Engineering Lab-
Plugs in Pipe Piles." Geotechnique, Vol. 41, oratory Report 223.10, Lehigh University,
No.4. Bethlehem, PA
RANGAN, B. V. and JoYCE, M. 1992. "Strength of RoWE, R. K. and ARMITAGE, H. H., 1987. "A
Eccentrically Loaded Slender Steel Tubular Design Method for Drilled Piers in Soft Rock."
Columns Filled with High-Strength Con- Canadian Geotechnical Journal, Vol. 24.
crete." ACI Structural Journal, November- RYBRIKOV, A M., 1990. "Experimental Investiga-
December. tion of Bearing Capacity of Bored-cast-in-
RAo, S. N. and PRASAD, Y. V. S. N., 1993. "Uplift Place Tapered Piles." Soil Mechanics and
Behavior of Pile Anchors Subjected to Lateral Foundation Engineering, Vol. 27, No. 2.
Cyclic Loading." ASCE Journal of Geotechni- SAMSON, C. H., HmscH, T. J., and LOWERY, L. L.,
cal Engineering, Vol. 119, No. 4. 1991. "Pile Driving Analysis by the Wave
RAUSCHE, F., GOBLE, G. G., and LICKINS, G. E., Equation." Handbook of Coasta[ and Ocean
1985. "Dynamic Determination of Pile Capac- Engineering, Vol. 2, Herbich, J. B., Gulf Pub-
ity," ASCE Journal of Geotechnical Engineer- lishing Company, Houston, TX.
ing Division, Vol. 111, No. 3. SAMSON, L. and AUTHIER, J., 1985. "Change in
REESE, L. C., 1977. "Laterally Loaded Piles: Pro- Pile Bearing Capacity with Time: Case
gram Documentation." ASCE Journal of Histories." Proceedings 38th Canadian
Geotechnical Engineering Division, Vol. 103, Geotechnical Conference: Theory and Practice
No. GT4. in Foundation Engineering, Edmonton, Al-
REESE, L. C., 1984. Handbook on Design of Piles berta.
and Drilled Shafts Under Lateral Loads, Re- SASTRY, V. V. R. N. and MEYERHOF, G. G., 1987.
port No. IP-84-11. Federal Highway Adminis- "Infiuence of Installation Method on the Be-
tration, Washington DC. havior of Rigid Piles in Clay Subjected to
REESE, L. C. and MATLOCK, H., 1956. "Non- Moment and Horizontal Load." Canadian
Dimensional Solutions for Laterally Loaded Geotechnical Journal, Vol. 24.
Piles with Soil Modulus Assumed Propor- SAVINOV, O. A and LAVRINOVICH, E. V., 1969.
tional to Depths." Proceedings 8th Texas Con- "U se of Hollow Piles for Hydraulic
ference on Soil Mechanics and Foundation Structures." Gidrotechnicheskoi Stroitelstro,
Engineering, Austin, TX. No. 7, Moscow (translated into English).
REESE, L. C. and WELCH, R. C., 1975. "Lateral ScHMERTMANN, J. H., 1970. "Static Cone to Corn-
Loading of Deep Foundation in Stiff Clay." pute Static Settlements Over Sand." ASCE
ASCE Journal of Geotechnical Engineering Journal Soil Mechanics and Foundation En-
Division, Vol. 101, No. GT7. gineering Division, Vol. 96, No. SM3.
REESE, L. C., Cox, W. R., and KooP, F. D., 1974. SCHWARTZ, J. W., 1958. "A Look at Old and New
"Analysis of Laterally Loaded Piles in Sand." Methods and Materials." ENR. CLXI (August
Proceedings 6th Annual Offshore Technology 21), p. 42.
Conference, Houston, TX. SHARMA, H. D. and JOSHI, R. C., 1988. "Drilled
RIKER, R. E., ANDERSON, D. G., and BACON, D. D., Pile Behavior in Granular Deposits."
1992. "Slope Displacement From Pile Canadian Geotechnical Journal, Vol. 25, No.
Driving." ASCE Proceedings Specialty Con- 2.
ference on Stability and Performance of Slopes SHEPPARD, D. A, 1983. "Seismic Design of Pre-
and Embankments-II, Berkeley, CA stressed Concrete Piling." PCI Journal, Vol.
ROBINSKY, E. l. and MORRISON, C. F., 1964. "Sand 28, No. 2.
Displacement and Compaction Around Model SHIH, R. W. K. and ANASTOSIOU, K., 1992. "A
Friction Piles." Canadian Geotechnical Jour- Laboratory Study of the Wave-Induced Verti-
nal, Vol. 1, No. 2. cal Loading on Platform Deck." Proceedings
RoDGER, A A. and LI'ITLEJOHN, 1980. "A study of Institute of Civil Engineers, Vol. 96.
vibratory driving on granular soils," Geotech- SHRIVASTAVA, S. M. and HUNT, S. G., 1989.
nique, Vol. 30, No. 30. "Seismic Design of New Wharf for Squamish
876 Piled Waterfront Structures

Terminals, British Columbia, Canada." ASCE 3rd International Colloquium on Stability of


Proceedings Specialty Conference PORTS '89, Metal Structures, Toronto, Ontario.
Boston, MA. STRECK, A, 1950. 23 Jahre Baugrundforsehung
SHRIVASTAVA, S. M. and YOUDALE, J. E., 1986. Hanuoverschen Versuchsanshal fur Grund-
"Precast Concrete Elements in Port Struc- bau und Wasserbau, Technishe Hochschule
tures for Ridley Island Terminal." PIANC, Hanover.
Bulletin No. 54. SUMER, B. M., CHRISTIANSEN, N., and FREDSOE,
SMIRNOV, G. N., GORUNOV, B. F., KURLOVICH, E. J., 1993. "Influence of Cross-Section on Wave
B., LEVACHEV, S. N., and SmoROVA, A G., Scour Around Piles." ASCE Journal of Water-
1979. Ports and Port Related Structures. way, Port Coasta[, and Ocean Engineering,
Stroyisdat, Moscow (in Russian). Vol. 119, No. 5.
SMITH, E. A, 1962. "Pile Driving Analysis by the SUMER, B. M., FREDSOE, J., and CHRISTIANSEN,
Wave Equation." ASCE transcription, Voi. N., 1992. "Scour Around Vertical Pile in
127, Part 1. Waves." ASCE Journal of Waterway, Port,
SMITH, T. D., 1987. "Pile Horizontal Soil Modu- Coasta[ and Ocean Engineering, Vol. 118, No.
lus Values." ASCE Journal of Geotechnical 1.
Engineering, Voi. 113, No. 9. SUN. K., 1994. "Laterally Loaded Piles in Elastic
Media." ASCE Journal of Geotechnical Engi-
SNIP II-17.80, 1980. Piled Foundations: Design
neering, Vol. 120, No. 8.
and Construction. Stroyisdat Publisher.
Moscow (in Russian). SWEDISH PILE COMMISSION, 1970. "Recommenda-
tions for Pile Driving Test and Routine Test
SOGGE, R., 1983. Microcomputer Analysis of Lat-
Loading of Piles." Royal Swedish Academy of
erally Loaded Piles, Laterally Loaded Deep
Engineering Science, Commission on Pile Re-
Foundations (STP 835), Langer, Mosley, and
search, Report No. 11, Stockholm.
Thompson (eds.), ASTM, Philadelphia.
TAN, R. H. S. and HANNA, T. H., 1974. "Long
SOWA, V. A, 1970. "Pulling Capacity of Concrete Piles Under Tensile Loads in Sands."
Cast in Situ Bared Piles." Canadian Geotech- Geotechnical Engineering, Vol. 5, No. 2.
nical Journal, Vol. 7.
TAVENAS, F. and ANDY, R., 1972. "Limitations of
SOWERS, G. B. and SOWERS, G. F., 1967. "Failure the Driving Formulas for Predicting the Bear-
of Bulkhead Excavation Bracing." Civil Engi- ing Capacities of Piles in Sand." Canadian
neering, Voi. 37, No. 1. Geotechnical Journal, Vol. 9, No. 1.
SPAN, G, J. TH. et al., 1981. "A Reviewed Rele- TAWIL, A H. and TANNER, R. G., 1979. "Geotech-
vant Hydraulic Phenomena and of recent nical Aspects of a Container Terminal in St.
developments in Research, Design and Con- John Harbor." Proceedings First Canadian
struction of Protection Works." PIANC Pro- Conference on Marine Geotechnical Engineer-
ceedings 25th Congress, Edinburgh. ing, Calgary, Alberta.
STANTON, J., 1987. "Connections in Precast Con- TAYLOR, K. V., 1973. "Slope Protection on Earth
crete Structures." Concrete International, and Rockfill Dams." Transaction 11th ICOLD,
November. III (Q.42, R.13), Madrid.
STELCO, 1981. Hollow Structural Sections. De- TAYLOR, R. B., 1995. "Composite Recycled Plastic
sign Manual for Concrete-Filled HSS Marine Piling and Timber: An Alternative to
Columns. Stelco Inc., Toronto, Ontario. Traditional Wood Products for Marine Use."
STEPHENS, M. J., KVLAK, G. L., and MoNT- ASCE Proceedings Specialty Conference
GOMERY, C. J., 1982. "Local Bucking of Thin- PORTS '95, Tampa, FL.
Walled Tubular Steel Members." Structural TERZAGm, K. and PECK, R. B., 1967. Soil Me-
Engineering Report No. 103. Department of chanics in Engineering Practice, 2nd ed. John
Civil Engineering University of Alberta, Ed- Wiley and Sons, New York.
monton, Alberta. THOMPSON, C. D. and THOMPSON D. E., 1979.
STEPHENS, M. J., KVLAK, G: L., and MONTO- "Effects of Piled Driving Systems on Drive-
GOMERY, C. J., 1983. "Local Buckling of Thin- ability and Capacity of Concrete Piles." ASCE
Walled Tubular Steel Members." Structural Proceedings of Convention and Exposition,
Stability Research Council. Proceedings of the Atlanta, GA
Piled Waterfront Structures 877

THOMPSON, D. M. and SHATI'LER, R. M., 1976. TSINKER, G. P., 1977. "Performance of Jetted
"Design of Riprap Slope Protection Against Anchor Piles with Widening." ASCE Journal
Wind Waves." CIRIA Report 61, Hydraulics ofthe Geotechnical Engineering Division, Vol.
Research Station, Wallingford, U.K 103, No. GT3.
TING, W. H., 1985. "A Simplified Method of Esti- TSINKER, G. P., 1986. Floating Ports. Gulf Pub-
mating Settlements of Individual Piles." lishing Company. Houstan. TX.
Ground Engineering, Voi. 18. TSINKER, G. P., 1988. "Pile Jetting." ASCE
TOMINAGA, M. and HEIDENGREN, C. R., 1988. Journal of Geotechnical Engineering, Voi. 114,
"Steel Pipe Foundations for Deep Water." No.3.
Civil Engineering, March. TSINKER, G. P., 1995. Marine Structures Engi-
TOMLINSON, M. J., 1963. Foundation Design and neering. Specialized Applications. Chapman
Construction. John Wiley and Sons, New York. & Hali, New York.
TOMLINSON, M. J., 1977. Pile Design and Con- TuCKER, K D., and AsKARI, S., 1986. Interm
struction Practice. Cement and Concrete As- report on evaluation of SCE BIPILE com-
sociation, London, U.K puter program, Southern California Edison
Company (unpublished).
TOMLINSON, M. J., 1980. Foundation Design and
Construction, 4th ed. Pitman, London. U.S . .ARMY CORPS OF ENGINEERS, 1984. Shore
Protection Manual, 4th ed. Coastal Engineer-
TOMLINSON, M. J., 1983. "Piles and Piled Foun- ing Research Center, Washington DC.
dations." Handbook of Structural Concrete,
U.S. .ARMY CORPS OF ENGINEERS, 1990 Engineer
Kong, F. K, Evans, R. H., Rohen, E., and Roll,
Manual EM 1110-2-2906. Oftice of Chief of
F. (eds.). McGraw-Hill Book Co., New York.
Engineers, Washington, DC.
TOMLINSON, M. J., 1994. Pile Design and Con- VAN DER MEER, J. W., 1988. "Rock Slopes and
struction Practice, Fourth edition, E & FN Gravei Beaches Under Wave Attack." Doc-
Spon, London, U.K toral Thesis. Delft University of Technology;
TOOLAN, F. E., LINGS, M. L., and MIRZA, U. A., also Delft Hydraulics Communications No.
1990. "An Appraisal of API RP2A Recommen- 396.
dations for Determining Skin Friction of Piles VAN DER VEEN, C., 1953. "The Bearing Capacity
in Sand." Proceedings Annual Offshore Tech- of a Pile." Proceedings 3rd International Con-
nology Conference, Paper OTC 6422, ference on Soil Mechanics and Foundation
Houston, TX. Engineering, Vol. 2, Zurich.
TscHEBOTARIOFF, G. P., 1962. "Retaining Struc- VERHEY, H. J. and BoGAERTS, M. P., 1989. "Ship
tures," Foundation Engineering, Leonards, G. Waves and the Stability of Armor Layers Pro-
A., (ed.). McGraw-Hill Book Co., New York. tecting Slopes." Publication No. 428, Delft Hy-
TSCHEBOTARIOFF, G. P., 1973. Foundations, Re- draulic Laboratory, Delft, The Netherlands.
taining and Earth Structures, 2nd ed. VESIC, A. S., 1964. "Model Testing of Deep Foun-
McGraw-Hill Book Co., New York. dations in Sand and Scaling Laws." Panel
TscHEBOTARIOFF, G. P., 1978. Personal Commu- Discussion, Session Il Proceedings North
nication. American Conference on Deep Foundations,
TSCHEBOTARIOFF, G. P. and WARD, E. R., 1957. Vol. V, Mexi.co City.
"Measurements with Wiegmann lnclinometer VESIC A. 8., 1967. "A Study of Bearing Capacity
on Five Sheet Pile Bulkheads." Proceedings of Deep Foundations." Final Report, Project
4th International Conference on Soil Mechan- B-189, School of Civil Engineering. Georgia
ics and Foundation Engineering, Vol. 2, Lon- Institute of Technology, Atlanta, GA.
don, U.K VESIC, A. S., 1970. "Testing on Instrumented
TsiNKER, G. P., 1972. "Rational Design of Anchor Piles, Ogeechee River Site." ASCE Journal
Piles with Widenings at the Ends for An- Soil Mechanics and Foundation Division, Voi.
chored Sheet Pile Bulkheads." Thesis pre- 96, No. SM2.
sented to Odessa Institute of Civil Engineers, VESIC, A. S., 1972. "Expansion of Cavities in
Odessa, Ukraine in partial fulfillment of the Infinite Soil Mass." ASCE Journal of Soil
requirements for the degree of Doctor of Phi- Mechanics and Foundation Division, Vol. 98,
losophy (in Russian). NoSM3.
878 Piled Waterfront Structures

VESIC, A. S., 1977. Design of Pile Foundations. stone and Shale." Canadian Geotechnical
Transportation Research Board National Re- Journal, Voi. 18, No. 4.
search Council, Washington, DC. WINKLER, W., 1867. Die Lehre uon der Elastizitat
VIJAYVERGIYA, V. N. and FocHT, J. A., Jr., 1972. und Festigkeit. Prague, Czechoslovakia.
"A New Way to Predict the Capacity of Pile in WITTKOP, R. C., 1986. "Prestressed Concrete
Clay." Proceedings Offshore Technology Con- Piling-Design vs As-Built." ASCE Proceed-
ference, Voi. II, Houstan, TX. ings Speciality Conference PORTS '86,
VIPULANANDAN, C., WoNG, D., and O'NEILL, M. Oakland, CA.
W., 1990. "Behavior of Vibro-Driven Piles in WoLFF, T. F., 1900. "'PILINT': A Software Pack-
Sand." ASCE Journal of Geotechnical Engi- age Designed for Pile lnterface Analysis." Re-
neering, Voi. 116, No. 8. search report, prepared for U.S. Army Corps
WALTHER, R., 1956. "Investigation of Multi-Beam of Engineers, Michigan State University, Ann
Bridges." Fritz Engineering Laboratory Re- Arbor.
port 223.14, Lehigh University, Bethlehem, WOLFF, T. F., 1993. "Probabilistic Assessment of
PA. Pile Interference." ASCE Journal of Geotech-
WANG, H., 1967. "Estimating Wave Pressure on nical Engineering, Voi. 119, No. 3.
a Horizontal Pier." T.R. No. R546, USNCEL, WRIGHT, S. G., 1991. "'UTEXAS3': A Computer
Port Hueneme, CA. Program for Slope Stability Calculations."
WEBB, D. L., 1969. "Settlement of Structures on Shinoak Software, Austin, TX.
Deep Alluvial Sandy Sediments in Durban, YALCIN, A. S. and MEYERHOF, G. G., 1991.
South Africa." Proceedings of the Conference "Bearing Capacity of Flexible Piles Under Ec-
on In-Situ Investigation in Soils and Rock, centric and Inclined Loads in Layered Soil,"
London. Canadian Geotechnical Journal, Voi. 28.
WEELE, A. F., 1957. "A Method of Separating the YAMAGATA, K., FUKUYA, T., and OMOTE, S., 1985.
Bearing Capacity of a Test Pile into Skin "Penetrability ofOpen-Ended Steel Pipe Piles
Friction and Point Resistance." Proceedings on Land." Proceedings Technical Conference
4th International Conference on Soil Mechan- on Penetrability and Drivability of Piles,
ics and Foundation Engineering, Voi. 2, Japanese Society for Soil Mechanics and
London. Foundation Engineering.
YARON, S. L., SHIMONI, J., TZACHAR, C., and
WEIDLER, B. et aL, 1987. "Pile Fatigue Failure
ZWEMMER, D., 1982. "The Hadera Offshore
Part 1: Damage Appraisal." ASCE Journal of
Coal Unloading Terminal: A New Approach to
Waterway, Port, Coasta[ and Ocean Engineer-
Unloading of Bulk in the Open Sea." Proceed-
ing, Voi. 113, No. 3, May.
ings Offshore Technology Conference, Paper
WHITAKER, T., 1957. "Experiments with Model OTC 4396, Houstan, TX.
Piles in Groups." Geotechnique, Voi. VII. No.
YOSHIMI, Y. and KlSHIDA, T., 1981. "Friction Be-
4.
tween Sand and Metal Surface." Proceedings
WHITAKER, T., 1963. "The Constant Rate ofPene- 10th International Conference on Soil Me-
tration Test for the Determination ofthe Ulti- chanics and Foundation Engineering.
mate Bearing Capacity of a Pile." Proceedings YoUNG, A. G., 1991. "Marine Foundation Study."
Institute of Civil Engineers, Voi. 26. Handbook of Coasta[ and Ocean Engineering,
WHITAKER, T. and CooKE, R. W., 1961. "A New Vol. 2, Gulf Publishing Company, Houstan,
Approach to Pile Testing." Proceedings 5th TX.
International Conference on Soil Mechanics ZIBBERBERG, S. D. and SHERSTNEV, A. D., 1990.
and Foundation Engineering, Voi. 2, Paris. "Construction of Compaction Tapered Pile
WILLIAMS, A. F., 1980. "The Design and Perfor- Foundations (Experience of the "Valdespets-
mance of Piles Socketed into Weak Rock." stroi" company)." Soil Mechanics and Foun-
Ph.D. Thesis, Monash University, Melbourne, dation Engineering, Voi. 27, No. 3.
Australia. ZINSERLINE, M. H., SCHLECHTON, J. R., and
WILLIAMS, A. F. and PELLS, P. J. N., 1981. "Side JULIAN, R. W., 1987. "Prestressed Concrete
Resistance Rock Sockets in Sandstone, Mud- Fender Piles." Concrete International, May.
8
Offshore Deep water
Terminals

8.1 INTRODUCTION bulk materials, such as coal, iron ore, and


other have been constructed at many
The continuing worldwide demand for en- locations in the world. The key factor in
ergy and raw materials creates a strong the operation of these terminals has been
need for specialized bulk material ports able utilization of very high-capacity loading/
to accommodate a large deep draft vessel. unloading technology. As discussed in
It has been found, however, that the con- Chapter 2, movement of millions of tonnes
ventional approach to creating an artificial of liquid and dry bulk materials is best
deep water harbor, involving dredging of achieved through construction of deep wa-
large quantities of seafl.oor materials and ter highly specialized terminals capable of
construction of breakwaters, in most practi- handling very large ships with sizes in the
ca! cases can be prohibitively expensive. At range of 150,000-750,000 DWT. At these
the same time, only a relatively few exist- facilities the loading and unloading of cargo
ing ports are able to accommodate vessels is carried out at very high rates (e.g., load-
with drafts exceeding 15 m, not to mention ing of up to 20,000 tonnesjh of dry bulk
super carriers of the 27-28-m draft range. and 220,000 m 3 of crude oii per day, with
Studies carried out worldwide indicate annual throughput of tens of millions of
that the economics of bulk transportation in tonnes has been achieved).
large quantities is such that the unit cost The conventional approach to creating a
per tonne of transported material is re- new artificial deep water harbor, or deepen-
duced considerably with the use of larger ing the existing harbor to serve a large deep
ships able to carry superloads to terminals draft vessel, in a great many cases appears
for further distribution. Consequently, deep to be cost-prohibitive. Sometimes, even if a
water facilities able to handle a large deep deep water area can be found or created in
draft vessel, such as supertankers for trans- existing waterways or ports, very often the
porting crude oii and liquid natural gas environmental concern, combined with a
(LNG) and superbulkers to transport dry lack of usable land and limited transporta-

879
880 Offshore Deep W ater Terminals

tion networks, prevents the creation of a These facilities have been linked to the
major new port complex. shore either by a bridgelike trestie designed
The practice of the past 30 years clearly to support pipe lines or conveyer systems
demonstrates that large quantities of bulk and to provide access to the terminal for
material can be effectively moved at low lightweight vehicular traffic, or by sub-
cost via offshore marine terminals not pro- marine pipelines (Yaron and Shimoni,
tected from the effects of environmental 1982; Soros et al., 1986; Youndale and
loads. At these facilities, the low berth occu- Shrivastava, 1986; Wright et al., 1987).
pancy due to rough sea conditiona is com- The seafloor area adjacent to the shores
pensated for by high rate of material han- is termed a continental shelf. A continental
dling. The high throughput is achieved by shelf is relatively shallow and typically ex-
the construction of facilities designed to tends very far from the shores. At shallow
handle only one specific kind of cargo (e.g., locations, the approach channel(s) can be
coal, iron ore, LNG, or crude oil) and the dredged to provide access for deep draft
use of very effi.cient high-capacity material vessels. However, where the coastal regime
handling technology. is such that a dredged channel would soon
In the case of a dry bulk loadingjunload- silt up and, therefore, require steady main-
ing facility, the technology includes utiliza- tenance dredging, the operation of the ter-
tion of fewer high-capacity conveyors, fewer minal can be cost-prohibitive. This problem
stackers, reclaimers, and shiploaders or un- was successfully overcome by construction
loaders. The conventional traveling loaders of transhipment islands, located far enough
have been replaced by slewing-bridge-type offshore and designed to provide a berthage
radialloaders that reduce substantially the for large deep draft vessels. The islands
cost of marine construction. Different types also have been designed to provide suffi.-
of machines that are used for handling of ciently protected loading or unloading facil-
both the dry and liquid bulk materials are ities for a smaller dedicated shuttle vessel(s)
discussed by Ferguson (1981) and Soros that can bring the cargo to and from the
et al. (1986). island.
Structurally, offshore terminals are not The design and construction problems
only site-specific but they are also different associated with offshore islands are beyond
due to the type of operation they are de- the scope of this work; this chapter is lim-
signed to perform. Typically, dry bulk un- ited to discussion of some specific aspects
loading facilities have been constructed in a associated with the design and construction
form of piers designed to support traveling of the offshore deep water marine facilities,
unloader(s) and conveyer systems that carry such as piled and gravity-type dolphins,
material to the land-based storage yard. On platforms, and piers. For discussion on off-
the other hand, most of the dry bulk load- shore transhipment island, the reader is
ing terminals and liquid bulk loadingjun- referred to Chesson (1980).
loading facilities have been constructed in Naturally adequate onshore-based facili-
the form of a platform designed to support ties are needed for handling the required
loadingjunloading equipment, flanked by amount of the transhipped material and a
breasting and mooring dolphins. These fa- shuttle vessel(s). Optimal access to the off-
cilities have been constructed far enough shore island is usually decided on the hasis
offshore where suffi.ciently deep water is of economic studies (e.g., all-weather access,
found and no maintenance dredging is re- such as trestles, submarine pipe lines, or in
quired. In some instances these terminals some cases submarine tunnels in combina-
were located .as far as 2000 m offshore to tion with well-balanced storage facilities
serve very large crude or dry bulk carriers. both onshore and offshore versus larger
Offshore Deep W ater Tenninals 881

storage facilities and access that could be gallery or shiploader can develop into a
interrupted by weather conditions), the cost chimney-stack e:ffect, causing a rapid travel
of the shuttle vessel(s) (both capital and of the flame. Hence, instead of complete
maintenance) and cost of required berthing enclosures, all galleries should be vented in
facilities must be considered. It should be order to localize fire. Also, flame-resistant
noted that construction of the subaquarims belting should be adopted in all enclosed
access tunnels in most practica! cases will co:riveyor systems. Naturally, the system
be less economica! than the construction of must be continually cleaned to remove the
a conventional trestie. Nevertheless, in trapped dust.
heavily populated areas, where local resi- When the project is located in a popu-
dents may regard the conventional trestie lated area, noise created by the transport-
as unacceptable "visual pollution" to the ing and handling of materials both o:ffshore
area, the submarine tunnel can be the only and onshore may be the worst environmen-
solution to the problem. It should be noted, tal pollution present. It usually requires a
that in general, construction of a nonaes- concerted e:ffort by di:fferent authorities (e.g.,
thetically designed facility can create objec- port, railway, roads, local councils, etc.) to
tions from local communities. bring this type of pollution to an acceptable
Access trestles, which are essentially level.
short- or medium-span bridge structures, For general information on deep water
are designed to support the relevant loads o:ffshore terminals, the reader is referred to
that in many cases include the weight of Chapter 2; some information on this subject
the loaded conveyors or pipelines and loads matter is also found elsewhere in this book,
produced by lightweight vehicular traffic. specifically in Chapter 7.
The span structure can be constructed, sim-
ilar to conventional bridges, in a form of
simple prefabricated concrete or steel
beams, or steel trusses. Again, as in conven- 8.2 LAYOUT
tional bridges, these spans are supported on
elastomeric bearings. The layout of an o:ffshore marine terminal
Naturally, the environmental protection is usually designed in a way to minimize
of bulk handling terminals must be thor- the ship-induced loads on its mooring sys-
oughly investigated. At the o:ffshore. end, tem. In practice this means that the termi-
the liquid bulk handling facilities are usu- nal mooring facility is aligned with the pre-
ally protected with floating curtains or vailing environmental forces generated by
booms installed around the ship during un- the wind, waves, and currents; these forces
loading and loading operations. Environ- a:ffect orientation of the terminal's mooring
mental pollution by dusty materials such as facility the most. Accordingly, the facility is
coal, iron ore, and so forth, is usually pre- usually placed in a way to maximize the
cluded by complete enclosure of the mate- berth availability during the year, and the
rial handling systems, as well as by sup- wave and wind, and, in some instances, the
pressing the dust by use of di:fferent types ice and current criteria are established to
of moisturizing systems (e.g., sprinklers and minimize the ship impact against the ter-
similar) or a combination of both. minal marine structures. In some instances,
As pointed out by Ferguson (1981), in this has been achieved by providing the
cases of coal loadingjunloading facilities, ship with a variable orientation at the dock
the complete enclosure of conveyors may as is shown in Figure 2-34 (Chapter 2).
create an extremely dangerous situation This type of o:ffshore terminal has been con-
with regard to fire. Fire in an enclosed structed at Punta Colorada, Argentina
882 Offshore Deep Water Terminals

(Soros et al., 1986). There, the ship is able support traveling bulk unloading towers
to moor at any one of six headings to dock (cranes). The impact load from a berthing
according to the wind and sea state prevail- ship is usually absorbed by breasting
ing at the moment. This resulted in a 90% strategically placed around the pier. Practi-
berth availability; in other words, the facil- ca! examples of such installation are illus-
ity could be used approximately 330 days trated in Figures 2-33, 7-4, 7-5, and 8-11
per year. (Chapters 2, 7, and 8 accordingly).
The Punta Colorada facility has been In some cases, piers have been designed
constructed offshore at the end of a 1400- to directly absorb the energy of the berthing
m-long combination of causeway and tres- ship, in which case the breasting dolphins
tie. There the 2-m-wide conveyer belt trans- are not required. Subsequentiy, the pier
ports iron ore pellets at the rate of 2000 tph structure is designed to support the loads
(tonnes per hour) to the slewing ship loader produced by the cargo handling and hauling
located on the variable orientation berth. equipment, and relevant environmental
The loader can pivot over a 173° arc and loads, and, where required, to resist loads
has a retractable boom with a belt conveyor from berthing ship and mooring loads.
that can extend out 38 m beyond the fender Severa! examples of such installations are
line so that the ship can be loaded regard- discussed in Chapter 7.
less of which direction they are headed. The mooring configuration illustrated in
In general, when planning the berth fa- Figures 8-1b and 8-2a are typical for bulk
cility at an exposed location, a statistica! loading facilities. This arrangement typi-
approach is usually taken in which the per- cally comprises of the following elements:
cent frequency of time the berth is occupied
and subject to conditions exceeding certain 1. Slewing ship loader(s) with a retractable
prescribed limits (e.g., limiting wave height boom pivot at one end and supported on
by direction can be compared with the eco- the seaward end on a curved rail behind
nomics of vessel turnaround time). the berthing line.
2. Breasting dolphins; usually two units, but
sometimes four or more, depending on
vessel size and its variances, as well as on
8.2.1 Dry Bulk Loading 1 the berth's mode of operation (e.g., Is the
vessel suspended in one position only dur-
unloading Facilities ing the loading operation or does it need
to be moved along the berth line in the
Regardless of their orientation offshore, dry process of loading operations?).
bulk loadingjunloading facilities typically 3. Mooring dolphins, usually four units,
include pier or a loadingjunloading plat- placed symmetrically to the centerline at
form flanked by breasting and mooring dol- 20-40 m behind the berthing line, de-
phins, and an access trestie that links the pending on the size of the ship. Two and
mooring facility to the shore. Sometimes the six units have also been used.
offshore mooring facility incorporates a
berth(s) for standby service tugs which can The facility depicted in Figure 8-1b is de-
be moored there when the sea condition signed to serve very large bulk carriers (up
allows for it. Typical mooring con:figurations to 300,000 DWT) and requires two ship
(layouts) of dry bulk loadingjunloading fa- loaders, four breasting, and six mooring
cilities are depicted in Figures 8-1 and 8-2. dolphins.
The layout shown in Figure 8-1a is typical Where capital investment in two loaders
for that used in dry bulk unloading termi- is not warranted because of insufficient
nals. Here the pier structure is used to cargo throughput, one loader can be used.
Offshore Deep Water Terminals 883

3
t3
--.1-

13
~

Figure 8-1. Dry bulk loadingjunloading terminal. Standard layouts: (a)


unloading terminal with traveling towers; (b) loading terminal with two
slewing shiploaders; (c) loading terminal with one slewing loader (alterna-
tively boom of the loader is moving just back and forth); in both cases the
vessel is pulling along the dock during loading operation. 1-Large bulk
carrier; 2-breasting dolphin; 3-mooring dolphin; 4-elevated pier; 5-
railway for traveling towers; 6-traveling tower; 7 -boom of slewing
loader; 8-loading platform; 9-approach trestie or conveyer gallery; 10-
catwalk; 11-curved rail supporting structure; 12-tug; 13-direction of
prevailing environmentalload, e.g., wind, wave, current.

The boom of this machine, however, may ment, the loader sits on a central platform
not cover the whole ship; therefore, for com- (dolphin) and a minimum two breasting/
plete loading, it must be moved along the mooring dolphins are usually placed on
dock. This is usually accomplished by in- either side of the central dolphin. Subse-
stallation of extra breasting dolphins, as quently, the ship can be moved along the
illustrated in Figure 8-lc. In this arrange- berthing line with the help of dedicated
884 Offshore Deep Water Terminals

/
b

Figure 8-2. Tanker Berth: standard layout. (a) Berth linked to a shore by the
approach trestle; (b) double-berth terminallinked to a shore by submarine pipeline or
tunnel. 1-Large crude oil, LNG, or other liquid petroleum product carrier; 2-
mooring dolphin; 3-breasting dolphin; 4-loadingjunloading, platform; 5-
trestle; 6-stem line; 7-bow line; 8-breast line; 9-spring line; 10-catwalk.
Note: L' = length of the largest design ship, L" = length of the smallest design
ship, a = 15° to 45°, f3 = 8 = 0° to 15°, (} = TJ "' 25°.
Offshore Deep Water Terminals 885

winches placed on the side dolphins or by


using winches on board. The mooring facil- !.
ity is linked to the shore by a trestie or by
the conveyor gallery.
Representative practica! examples of
mooring layouts depicted in Figures 8-lb
and 8-1c are illustrated in Figures 8-3
and 8-4. The mineral material loading ter-
minal facility shown in Figure 8-3a has
been built offshore of N ewfoundland at a
location unprotected from heavy waves. This
facility is comprised of five dolphins con-
structed from floated-in concrete caissons
installed on prepared stone bedding at 70 m
center to center. The central caisson, 19.5 X
24 m in cross section and 18.5 m high, was
designed to serve as a foundation for the
slewing material loader with a retractable
boom. The elevated seaward curved raii is
supported on a curved steel frame-like
structure. The loader can pivot over an 87°
arc, which enables it to cover a substantial
area of a 60,000 DWT design vessel. For
complete loading, the ship needs to be
moved along the berth line. This is done by
onboard winches.
The central dolphin of this facility is
flanked by four breastingjmooring dol-
phins, two on either side of the centerline.
These dolphins are 19.5 X 19.5 m in cross
section and 12.1 m high. For the sake of
economy, the dolphins are not linked with
each other by catwalks; therefore a service
boat is used when required to bring the
mooring lines in and to secure them at
the bollard located at the center of each
breastingjmooring dolphin. Ali dolphins
have been designed to withstand the heavy
ice and wave loads as well as impact load
from the berthing ship. The facility is linked
to the shore by a conveyor gallery.
A similar, however much smaller, facility
for loading crushed rock material, is illus- Figure 8-3. Dry bulk loading terminals: At these
trated in Figure 8- 3c. Here, a very simple facilities the ship is moving along the berth in the
retractable conveyor has been used. Thus, . process of loading operation. (a) Mineral materialload-
for complete loading, the ship needs to be ing facility, Newfoundland; (b) coal loading facility,
moved its fulllength along its hold area. To Seward, Alaska; (c) crushed rock-loading dock on Kre-
menchug waterstorage, Ukraine (artistic impression).
accommodate for changes in water level,
886 Offshore Deep Water Terminals

six breasting dolphins installed along the


berth line for complete loading. The space
between breasting units must be such to
ensure the ship safe berthingjunberthing
operations and also ensure safe and reliable
handling of a vessel during the loadingj
unloading operations. For this, the distance
between the outer corners of the breasting
dolphins should not be less than 35% of the
overall length of the largest design vessel,
Figure 8-4. Typical dry bulk loading facility, with a
slewing loader equipped with a retractable boom.
and the distance between the inner corners
(Courtesy of Hollandsche Beton Groep nv.) of the breasting faces should not exceed
50% of the overall length of the smallest
design ship .
the conveyor can be moved vertically. The . If a wide range of vessel sizes is required,
mooring facility is linlu;d with a shore by a then the mid-body length of the smallest
causeway. This terminal was constructed ship may control the space between adja-
on a very large waterstorage created by cent dolphins. An example is shown in Fig-
the hydroproject built on the Dnieper River, ure 8-2a. Furthermore, the length between
Ukraine. the outer ends of the breasting dolphins
The coal loading facility depicted in should not exceed 0.4 times the "length be-
Figure 8-3b was constructed at City-of- tween perpendiculars" for the largest ship
Seward, Alaska in the natural, well- to be handled. It must be pointed out that
protected Resurrection Bay, free from heavy the breasting structure must be able to
ocean waves and ice. It was designed to handle the design approach angles (the
serve large bulk carrier up to 136,000 DWT. angle between the ship side and fender
Because of a relatively small annual panel surface) of at least ± 10°.
throughput, only one loader has been in- Useful information on bulk commodity
stalled. This loader, however, could not terminals is given in Zonailo (1992).
cover the whole length of the largest design
vessel which, for complete loading, needs to
be moved along the dock; this is done with
the help of onboard winches. A total of five 8.2.2 LiQuid Bulk Loading 1
piled breasting dolphins have been in- Unloading Terminals
stalled in front of a piled platform that
supports the loader. The piled breasting Two standard layouts for these facilities are
dolphins (Fig. 8-7) are flanked by piled depicted in Figure 8- 2. As is seen from this
mooring dolphin (Fig. 8-8) and mooring figure, there is no conceptual differences
buoy. The · mooring facility is linked to the between the mooring layout designed to
shore by a trestie that is designed to carry handle the dry bulk carriers and those
the conveyer system and light vehicular needed to handle the liquid petroleum prod-
traffic. uct carriers. In the latter, the vessel can be
The facility shown in Figure 8-4 illus- handled on one side of the dock (Fig. 8- 2a),
trates the iron pellets loading dock where or on both (Fig. 8-2b). Again, the number of
the smaller vessels can be loaded via a breasting and mooring dolphins used for
slewing loader with retractable boom with- mooring and handling vessels varies, de-
out moving along the berth line; on the pending on the variety of the vessels and
other hand, a large ship can be moved along their sizes. It must be pointed out that
Offshore Deep Water Terminals 887

tankers never move along the dock during potential intersection between adjacent bat-
loading or unloading operations, during ter piles. In most cases, the platform is not
which time they must be properly moored designed to resist the impact by a berthing
to prevent any significant movements. Liq- ship. Therefore, relative positioning of
uid bulk handling facilities can be linked to breasting dolphins and the platform along
the shore by either sitting on trestie the berth line is very critica!, because when
pipelines or by subsequent pipeline(s). The a ship berths and rests against the breast-
relevant practica! examples are illustrated ing dolphins, it must not hit the platform,
in Figures 8-5a and 8-5b. even when dolphins deflect. At the same
In most practica! cases, the loadingjun- time, the platform must be placed clase
loading platform and the adjacent breasting enough to the fender line to allow hose
dolphins are placed in clase proximity to connections to be made or to stay within the
each other. As pointed out earlier, very of- reach of a ship's loadingjunloading equip-
ten these structures are supported on pile ment. Hence, great attention should be paid
foundations that may include batter piles. when establishing the berthing line (front
Hence, as a practica} matter , pile arrange- face of the breasting dolphins) or fender
ments ofboth the platform and the dolphins line and the setback for the loading/
must be designed in a way to prevent the unloading platform.
If more than two breasting dolphins are
used to accommodate the design vessel, then
to avoid a heavy overloading of individual
!
dolphin(s), misalignment of exposed faces of
all dolphins within the system must be min-
imized. Positioning of the mooring dolphins
is usually not that critica}: A meter or two
displacement each way is normally accept-
able. Where two or more terminals are con-
structed in a clase proximity to each other,
then sufficient space between two adjacent
berths must be provided to ensure safety of
operation. Safety requirements at berths
that handle liquid petroleum products
(e.g., crude oii, LPG, and LNG) are very
stringent.
Although there are no established stan-
dards, it is generally assumed that berths
handling petroleum products must be sepa-
rated from each other and from other in-
stallations by at least 150-300 m; the larger
space is required at LNG handling facili-
ties, and a smaller distance can be used at
crude oil or LPG handling berths. It must
be pointed out that for crude oii, LNG, and
LPG termin als, the design criteria is usu-
Figure 8-5. Typicalliquid petroleum product load- ally more stringent than for dry bulk han-
ingjunloading facilities. (a) Linked with a shore via
pipeline(s) sitting on a trestie. (Courtesy of Holland- dlin:g facilities. This applies to berthing and
sche Beton Groep nv.) (b) Offshore double-berth facility mooring procedures, the design of the fender
linked to the shore by submarine pipelines. system, and the structural design.
888 Offshore Deep Water Terminals

In conclusion of this section it must be smallest and the largest ships for which the
noted that there are no standard rules for berth is designed are met.
layout design of dry and liquid bulk han- Mooring structures also should be veri-
dling terminals; usually every project is fied for satisfactory mooring patterns for
site-specific and has its own distinct solu- the full range of vessel sizes in ballast and
tion based on the cargo and type of vessels in full displacement condition and for ships
the terminal is supposed to handle. Suc- in head-in and head-out berthing positions,
cessful planning of a port or terminal facil- unless the facility is designed specifically
ity for a specific purpose is the key to the for a single heading. Usually the mooring
most efficient project. structures are located in a way to ensure
Special expertise developed all over the that the length of mooring lines is about 30
world and reported by many experts in this m or more. This is needed to allow for
field has become an international commod- changes in tide and ballast conditions and
ity. This experience can provide a unique to provide for sufficient elasticity.
solution to a specific problem that may con- As a practica! matter, the bow and stern
front the designer. For additional informa- lines are attached to the respective ends of
tion on this subject matter, the reader the vessel and make a horizontal angle (a)
is referred to works by Quinn (1972), with the berthing line of around 15° to 45°
Thorensen (1988), Ferguson (1981), Bruun so that they provide some degree of both
(1989), Gaythwaite (1990), and to the many lateral and longitudinal restraint. Vertical
other publications referenced elsewhere in angles of these lines, fJ, should not exceed
this text. 25° to 30° maximum and will vary with the
stage of tide and vessel ballast condition.
The horizontal angle, {3, should prefer-
ably be equal to zero, but not to exceed 15°;
8_3 MOORING SYSTEM the vertical angle, fJ, of breast lines should
not exceed 25° to 30° maximum. The angle
of spring lines, 8, with the berthing line
As it follows from the preceding discussion, should be as small as practica!, preferably
the mooring system is comprised of piers, not exceeding 15°; the vertical angle of
breasting and mooring dolphins, and some- spring lines, TJ, should not exceed 25° to 30°.
times loadingjunloading platforms. The Note that all the aforementioned angles are
layout of a particular mooring system shown in Figure 8-2.
should be such to ensure safe and reliable The number and location of mooring de-
restraint of the ship during the loadingj vices on mooring structures should be veri-
unloading operation. fied to make sure that the mooring re-
All the aforementioned structures should straint requirements of all design vessels
be arranged in a way to provide the re- intended to be served at the berth are met,
quired longitudinal and lateral restraint to cognizant of line strength and the number
the ship without overstressing the mooring and location of the fairleads.
system. It must consider the full range of The overall adequacy of a ship mooring
ship sizes the berth is designed to accom- arrangement should be reviewed for safety
modate under the adverse combination of under design storm conditions, and also for
displacement, wind, wave, current, and ice limitation of ship movements at berth to
forces that are within the operating criteria. prevent possible damages to the ship,
Furthermore, the location of all mooring berthing structures, and loadingjunloading
structures should be checked to verify that equipment. Table 3-9 (Chap. 3) contains
the mooring restraint requirements of the approximate ranges of limiting motions for
Offshore Deep Water Terminals 889

the safe operation of various large ship and the repeated load should preferably not
types. For safe and efficient mooring sys- exceed 50% of breaking load. For more dis-
tems the following basic principles should cussions on allowable ship motions and
be considered (PIANC, 1995): mooring lines arrangement the reader is
referred to Chapter 2, Bruun (1989), Slinn
1. Mooring lines should be distributed sym- (1979), Thorensen (1988), LeMehaute
metrically with respect to the vessel's (1977), Brattleland (1974), and PIANC
center of gravity to obtain an equal load (1995).
distribution over the mooring lines. Essentially, the mooring system and
2. Mooring lines and fenders should ideally structures should be designed for the maxi-
have the same stiffness to reduce vessel mum forces generated by the largest ship,
displacements. weather in ballast, or in fully laden condi-
3. The transversal forces from the mooring tion under the combined action ofthe maxi-
lines on the bow and aft of the ship should mum design environmentalloads. It is pru-
remain symmetrical around the center of dent to consider the most severe design
gravity when the ship is displaced (surg- wind condition from any direction within
ing) along the berth from its equilibrium the operating criteria. The wind force should
situation. be computed with consideration given to
4. All mooring lines within a certain group gust factor and ship size. Details on this
should be identica!, i.e., manufactured subject are provided in Chapter 3. For more
from the same material, have the same details, consult Headland (1995).
diameter and approximate length; this
Mooring systems designed to handle large
will help to obtain an equal distribution
of focus over all mooring lines with the vessels (e.g., 100,000 DWT and more) should
group. incorporate quick-release mechanisms, each
hook of which should be capable of rotating
5. The number of mooring lines should be as
low as possible to prevent congested
freely in both the vertical and horizontal
mooring arrangements and to facilitate planes through the full range of mooring
mooring operations. lines angles. For the sake of safety, each
hook should be equipped with a locking
6. Mooring lines should be as horizontal as
device to prevent accidental release of the
possible and always kept taut in order to mooring line. Every component of the moor-
prevent impulse shock loads in mooring ing device should be designed with a mini-
lines and to create suf:ficient friction be- mum safety factor of 2.0 and should be
tween the ship hull and the fenders. factory tested with the maximum design
load. The mooring lines handling system(s)
It must be noted that in practice, despite all should be provided with a powerful capstan
precautions, mooring lines are subject to to draw in the heavy mooring lines that are
loading cycles with variable intensity which used to moor the large vessel. These cap-
results in fatigue phenomena; this should stans should be weatherproofed with totally
be taken into account for all types of moor- enclosed motors. Furthermore, mooring de-
ing lines. From longevity point ofview, steel vices must be located a minimum of 1.5 m
wire mooring lines should not be loaded in above the maximum expected water level
excess of 55% of the minimum breaking during the loadingjunloading operation.
load and synthetic lines should never be Recommended mooring practices and
loaded above 75% of their breaking load. specific guidelines for mooring of large ships
The mean load on synthetic lines should are given in OCIMF (1978, 1992),
preferably not exceed 25% of breaking load TERMPOL (1983) and PIANC (1995).
890 Offshore Deep Water Terminals

8.3.1 Basic Structural concepts The wharf structure(s) must be designed to


survive the extreme weather condition
The planning and design of any port facil- which may exist over a range of tide levels
ity, including offshore terminals, requires and wind and sea directions. The worst load
certain input data for the design criteria combination should be used for structural
that typically include the following informa- analysis. Sugin (1983) and Soros et al.
tion: (1986) describe a number of case histories
of offshore terminals and summarize design
1.The annual throughput tonnage and its criteria for severa! offshore dry bulk loading
projected growth pattern with respect to terminals. A number of practica! examples
time. is also found in Bruun (1989).
2. The size of the ships expected to use the The structural concept and its economics
facility, including arrival drafts, hold and depend greatly on the exposure of the facil-
hatch configuration for dry bulk carriers ity to environmental loads and foundation
and manifold layout and similar parame- condition. The exposure of marine structure
ters for liquid bulk carriers, layout of to heavy storm-induced waves has a domi-
mooring devices on board of a ship, and nant effect on selection of a structural con-
other special features of particular ships. cept for the offshore facility.
3. The mode of ship approach to the dock
(e.g., being assisted by tugs or self-
propelled) and the required mooring ar-
rangement.
8.3.1.1 Piled Structures
4. Environmental conditions under which
the vessel of a certain size would remain
As pointed out earlier, in most practica!
alongside the berth during loadingj
unloading operation. This criterion is pre-
cases offshore marine terminals are de-
sented in terms of the prescribed limiting signed in the form of an open piled struc-
conditions for berth operation under ture(s) in which a heavy concrete super-
which the vessel could safely remain structure is supported on concrete or steel
alongside the berth. piles of miscellaneous designs. A number
5. The most extreme storm condition (e.g., of relevant examples is discussed in
100-yr) with no vessel alongside the berth. Chapter 7.
6. The minimum expected life of the facility. Although storm waves occur relatively
It may be determined by the expected life rarely, they have a profound effect on the
of the industry that the marine facility is number, size, and configuration ofpiles (e.g.,
designed to serve; for example, expected vertical or batter) which must carry both
life of a mining operation. axial loads as well wave-induced bending
7. Site foundation conditions that may moments. A fewer relatively large-diameter
greatly effect the development of a struc- piles can offer the least frontal area for
tural concept and its economics. wave attack, relative to a large number of a
smaller-diameter piles. Furthermore, less
In the case of an offshore terminal opera- obstructed passage of waves through the
tion, standard practice is to have the vessel pile system will reduce the standoff forces
leave the berth in extreme weather condi- on a moored ship. A practica! size and num-
tions when wind andjor waves exceed cer- ber of piles are usually found from a com-
tain threshold values. parison of severa! structural alternatives.
Obviously ali loadingjunloading activi- Foundation conditions have a consider-
ties should ceased as soon as the vessel's able in:fluence on the best structural con-
motion reaches an unacceptable magnitude. cept required. With piled structures, the
Offshore Deep Water Terminals 891

piles can be either driven to the required thus bringing the steel below the optimum
depth, in the case of soft soil foundations, or welding temperature. Furthermore, in the
be socketed into the underlaying bedrock if case of a large-diameter steel pipe pile, ali
the depth of the soft soil is insufficient to welding is done in the shop, where condi-
provide sufficient axial capacity for the piles tions are optimum and proper procedures
in tension andjor proper fixity in soil ofthe and nondestructive testing can be carried
piles exposed to bending moments. out; here good quality pile splice can be
Offshore terminal piling consists, typi- ensured. This approach can prevent piles
cally, of large-diameter steep pipe piles (1 from and particularly those subjected to
mor more in diameter), 40-60 m in length, cyclic and dynamic loads, from failing due
having wall thicknesses of 20-40 mm. to fatigue stresses.
High-yield steel (fY = 350 MPa) is normally Pile positioning at exposed deep water
employed. Altematively, concrete cylinder locations is a very difficult operation. It is
piles, mostly with prestressed reinforcing usually carried out with the help of a tem-
steel of similar strength, are also used. Both plate that has sleeves through which both
vertical and batter piles are employed, in- vertical and batter piles may be set and
tersecting at the deck level, so as to react temporarily supported. Typically vertical
against each other under lateral loads. piles are driven first to support the tem-
As the size of ships increase and the plate at the predetermined elevation, then
exposure conditions become more severe, the batter piles are set and driven to the
the proportion of batter piles is increasing required penetration. Vertical piles should
so that they typically dominate the con-
not be spaced too far apart, otherwise it will
struction. Typical load per pile in compres-
be too difficult and expensive to provide
sion is between 300 and 600 tonnes and
forms for the concrete capping structure to
between 50% and 100% of that in tension.
support the fresh concrete when it is placed.
To minimize bending in the piles it is im-
Yet, in many cases the mooring dolphins
portant that the vertical and batter piles be
are designed with batter piles only. In this
accurately located so that the axes of a
adds one verti-
batter-vertical pile group intersect at a sin- case the contractor typically
cal pile, usually in the center ofthe dolphin,
gle point. It must be pointed out that the
pile-driving equipment should be suitable as a temporary support for the template.
for work at exposed sea conditions; some- This pile can be removed from the system
times conventional pile-driving equipment later.
at an exposed location can be too small and Altematively, the template could be held
limited. over the stern of a dedicated barge in proper
Naturally, construction of all types of plan location, before one or more vertical
structures at exposed locations is more dif- piles are driven through its sleeves to pro-
ficult than at protected harbors. This is par- vide lateral and vertical support to the tem-
ticularly true for construction of piled struc- plate. When the template is raised verti-
tures, where the use ofheavy piles 40-60 m cally to is proper elevation, and temporarily
long is commonplace. The reason why con- welded off to the vertical pile(s), the cement
tractors very often elect to handle and set a grout is injected into the space between the
long pile as a single piece is in the difficul- sleeve and the pile. This is required to pre-
ties associated with splicing many smaller vent potential vibration. Where a jack-up
units during rough sea conditions; piles may construction platform is used, it can provide
be vibrating due to waves and vortex shed- the initial support for the template. For
ding from the current, and the joint area more details on offshore pile installation,
may be constantly wet with ocean spray, the reader is referred to Gerwick (1986).
892 Offshore Deep Water Terminals

8.3.1.2 steel Jackets The insert-pin pile system is then grouted


to the bedrock and jacket legs. When grout-
As pointed out earlier, sometimes prefabri- ing piles to the jacket legs or when grouting
cated full steel jackets, similarin concept to between an insert pile and a primary pile, it
those used in the offshore oil drilling plat- is essential that the spaces be completely
forms, are employed for construction of off- filled. For a better bond the mud must be
shore terminals. Essentially, the jacket is a excluded from the annulus and the steel
steel space frame braced horizontally and surfaces should be free of mill scale, var-
vertically. After fabrication, jackets are nish, marine growth, and other contami-
loaded on barge(s) and transported to the nants.
site of installation. There they are lifted by A nonshrinking grout should be used
a heavy-lift fl.oating crane, installed on the (American Petroleum Institute, 1984) and it
seafl.oor and leveled. The jacket must be should develop at least 10 MPa within 24 h
self-supporting until the piles are installed. in order to reduce its sensitivity to struc-
It is important that the jacket be leveled
ture movements and vibration during the
and remain so within specified tolerance setting period. Speed of set and strength
until the piles are properly grouted within gain are controlled by the type of cement as
well as by type of admixtures and tempera-
the jacket's sleeves. Any attempt to level
ture of the water. The annulus width is
the jacket after installation may result in
usually limited to 100 mm (maximum) to
unacceptable stresses in the piles.
prevent adverse effects from heat of hydra-
The weight of the jacket with weight of
tion that may develop excessive tempera-
piles and miscellaneous auxiliary construc-
tures causing internal microcracking within
tion equipment supported by the jacket
grout.
must be safely supported on the seafl.oor
The uniform width of an annulus is
mud. The bearing pressure on the soil must maintained by special centralizers (spacers),
be within allowable limits under the combi- and packers of miscellaneous designs are
nation of direct load and that produced used to confine the grout to a certain part of
by waves and currents experienced during a jacket leg and prevent its escape from the
the construction phase. The American system. For details, the reader is referred to
Petroleum Institute API RP2A (1984) al- Gerwick (1986). Details on design and in-
lows a one-third increase in allowable soil- stallation of piles used for construction
bearing values during this phase if wave of a deep water terminal are found in
action is considered in the loading case. Chapter 7.
Furthermore, a safety factor of 2.0 should
be provided against a bearing failure. Scour
around and under the jacket's mudmats
must be foreseeing and prevented; if war- 8.3.1.3 oravity-Type Structures
ranted, the erosion control can be accom-
plished by the placement of graded rock Gravity-type structures, such as fl.oated-in
through a long tremie pipe. concrete caissons, large-diameter concrete
The piles are dropped through thejacket's or steel cylinders, or steel sheet-pile cells,
tubular legs and are driven to final penetra- ali filled with granular material may be
tion. In the case of a bedrock-type founda- employed should hard rock be encountered
tion, a large-diameter pin pile is drilled to a on the seabed. They also have some merits
predetermined depth into the bedrock, that in exposed choppy sea conditions where
is followed by a smaller-diameter (section) prefabricated structures can be quickly po-
pin pile drilled further into the bedrock. sitioned, installed, and filled in periods of
Offshore Deep Water Terminals 893

calm seas. The typical drawback of these 8.4.1 Breasting Dolphins


structures is the need for preparation of a
well-leveled stone bedding which, in a deep- These are primarily designed to take the
sea environment, can present considerable impact ofthe ship when docking and to hold
problems. the ship against a broadside wind. Nor-
In general, conditions that are in favor of mally, they are also designed to take loads
prefabricated gravity-type structures are as from spring lines, and in the case similar to
follows: that depicted in Figure 8'--lc, the breasting
dolphins also function as mooring dolphins.
1. Suitably located casting yard (or ship- To dissipate the impact energy of a
yard) with launching facilities and con- berthing ship, the breasting dolphins are
venient access to a deep water waterway :fitted with a fendering system. In a deep
2. When the use of a relatively small num- water application, depending on the magni-
ber of units is economica! tude of lateral forces and foundation condi-
tions, the breasting dolphins are usually
3. Firm foundation relatively close to the
designed in one of the following ways:
surface of the seafloor.

1. Groups of battered and vertical piles,


It should be pointed out that in the case of a capped with a concrete cast-in-place slab
single offshore berth, use of a limited num- (deck)
ber of prefabricated caissons, in most cases, 2. Group of just verticallarge-diameter steel
proves to be less economica! than other piles joined together with horizontal di-
types of construction. However, it also aphragms that are located above sea level.
proves to be an almost indispensive type of 3. Single, large-diameter, steel piles (i.e.,
construction where heavy ice loads are ex- 2-3 m and more), sometimes referred to
pected. A number of examples of offshore as "caissons"
marine terminals operating under heavy ice 4. Gravity-type structures (e.g., steel sheet-
conditions in which floated-in caissons have pile cells, floated-in-concrete caissons or
been employed are found in Tsinker (1995). large-diameter concrete or steel cylinders
Details on design and construction of 5. Prefabricated steel jacket pinned to the
gravity-type structures, namely quay wallş, sea floor by steel piles
are given in Chapter 5. Design specifics
of the gravity-type dolphin are discussed Severa! practica! examples of these struc-
further in this chapter. tures are illustrated in Figures 8-7 through
8-11 and in Figures 2-25 and 2-71d in
Chapter 2. Figure 8-6g illustrates a deep
water mooring dolphin that is comprised of
8.4 DOLPHINS ANO PLATFORMS a steel jacket with large-diameter steel piles
driven through jacket legs. At the top piles
As pointed out in previous sections, dol- are welded to the jacket. In this structure,
phins are discrete structures that are used the steel shims have been used to center
to absorb the impact of a berthing ship and the pile within the jacket leg. Shims of
to secure a vessel's mooring lines. The for- these kind are usually one-quarter to one-
mer are called "breasting'' Cor sometimes third segments of steel pipe of proper ra-
"berthing'') dolphins and the latter are de- dius. The welds from the pile to shims, and
:fined as "mooring'' dolphins. Sometimes, from ·shims to the jacket leg are usually
breasting dolphins are also designed to per- designed for the maximum shear load. In
form functions of a mooring dolphin. the illustrative case, additional shear plates
894 Offshore Deep W ater Terminals

5
4

plan

!.. !

3 .....

1-- -lr+-- --ft---- 12

Figure 8-6. Breasting (mooring) dolphins for deep water construction: Basic
structural types. (a) Piled structure, (b, c) flexible dolphins, (d, e) gravity·type dolphins, (f, g)
steel jacket-type construction. 1-Piles (steel or prestressed concrete); 2-cast-in-place
concrete capping; 3-fender system; 4-large-diameter steel pipe piles; 5-very large-diame-
ter steel pile ("caisson"); 6-floated-in concrete caisson; 7-prefabricated concrete block,
8-rock armor; 9-stone mattress; 10-cast-in-place concrete slab, 11-steel sheet-pile cell
or bottomless large-diameter steel or concrete cylinder; 12-pin piles; 13-steel jacket;
14-catwalk, 15-welded connection between steel jacket and piles.
Offshore Deep Water Terminals 895

high-performance rubber fender units. De-


tails on fender system structures and fender
system design are given in Chapter 2.
As noted earlier, the primary purpose of
breasting dolphin is absorption of a ship's
berthing energy. Most of this energy is ab-
sorbed by the fender system and in part by
the structure itself; naturally, more fl.exible
structures such a piled dolphins can absorb
a greater portion of a ship's berthing energy
than a gravity-type structure. In fact due to
very little displacement or deformation of a
gravity-type dolphins, their ability to ah-
sorb some energy from a berthing ship is
usually ignored in the design process.

8.4.2 Piled Breasting Dolphins


Essentially, flexible cantilever dolphins,
such as those shown in Figures 8-6b, and
Figure 8-7. Coalloading terminal, Seward, Alaska.
Piled breastingjmooring dolphin.
8-:-6c, 8-12, 2- 25, and 2-7ld (Chapter 2),
are the most efficient absorbers of ship im-
pact energy. These structures employ large-
diameter open-ended steel pipe piles driven
into the seabed to a penetration that en-
sures proper fixity within a competent soil
stratum.
It should be noted, however, that instal-
lation of very large-diameter steel piles
(caissons) is a difficult task. For installa-
tion, the heads of these piles must be fitted
in order to distribute the hammer blow over
the large-diameter cylinder and to prevent
weld cracking under impact. These piles
may also be preinstalled by use of water
jets, and then driven by a large offshore-
type impact hammer to the required
penetration.
Figure 8-8. Coalloading terminal, Seward, Alaska.
Piled mooring dolphin. A major construction problem for instal-
lation of this type of pile is the handling
and positioning of such large cylinders up
have been welded to the pile and shims to to 50 m, or more long, 2- 4 m in diameter,
improve connection between the piles and and weighing a couple of hundred tonnes.
the jacket. In areas with strong tidal currents, these
All of the above structural types of piles tend to be displaced in the direction of
breasting dolphins are normally fitted with the current and laterally due to vortex
896 Offshore Deep Water Terminals

shedding. However, once set into the soil, caliy in tension and in compression but are
the problems of current are reduced where also loaded by bending moments that may
firm soils are encountered (e.g., glacial till, exist in the system due to elastic settlement
overconsolidated silt, strata of weak rock, of the piles and their elastic elongation and
etc.), the pile-driving is assisted by excavat- shortening resulting in some lateral dis-
ing these soils from within the pile's inte- placement of the dolphin. Where bedrock or
rior. For more discussion on the subject of a firm bearing stratum is close to the seabed
pile driving, consult Chapter 7. Useful in- surface, the length of a tension pile can be
formation on the subject matter is given in reduced either by adding weight to the dol-
Gerwick (1986). phin's superstructure or by socketing these
PIANC (1984) suggests that, depending piles into the bedrock. Furthermore, if the
on soil condition and pile flexibility, the bedrock is near the seabed surface, then ali
energy absorbed by a fender system can piles can be socketed into it.
only constitute 20-50% of the total energy. The thickness of the capping slab must
In general, PIANC recommends the selec- be sufficient to provide for reliable fixity of
tion of the fender unit(s) in a way such that both compressed piles and piles in tension.
at a rated energy absorption capacity, the This usually required the slab thickness to
normal reaction force constitutes about 70% not be thinner than 1.2-1.5 m. The capping
of the maximum permissible load on a slab must be rigid enough to distribute hor-
pile(s). izontal and verticalloads between the piles.
Note that in addition to horizontalloads Furthermore, the slab must be thick enough
due to berthing impact, vertical andjor lon- to accommodate the fender unit and to
gitudinal rubbing loads due to friction be- provide reliable support for the frontal
tween the fender unit(s) and a ship's huli fender panel as is shown in Figure 7-17
must also be taken into account. Naturaliy, (Chapter 7).
loads such as the aforementioned rubbing
loads do not act through the center of grav-
ity of a dolphin and result in torsional mo- 8.4.3 Gravity-Type Dolphins
ments that must be resisted by the dolphin's
structure. Gravity-type dolphins, such as those de-
Suitable foundation soil conditions is an picted in Figures 8-6d, 8-6e, 8-9 8-10a,
essential precondition to the application of and 8-10b, offer an excellent solution when
a purely flexible dolphin. Flexible (canti- they are designed to withstand very large
lever) dolphins are sometimes driven at a forces from berthing ships and to resist
slight batter toward the direction of impact.
This results in a sort of pile prestressing by
aliowing for more deflection from the origi-
nal position and, thus, an increase in the
energy absorption potential of the pile.
Where subsurface soil conditions are not
favorable for use of flexible dolphins, dol-
phins comprised of both vertical and batter
piles can represent an economica! solution.
Where pile penetration is not limited by
bedrock or by the presence of a large boul-
der, the dolphin's superstructure can be
constructed on friction piles driven to the Figure 8-9. Port-of-Cadiz, Spain. Gravity-type
required penetration. These piles are basi- breasting dolphins.
Offshore Deep WaterTerminals 897

heavy ice forces. As pointed out elsewhere In most practica! cases floated-in-type
in this book, gravity-type structures are caissons are installed on prefabricated stone
usually constructed in the fonn of steel beddings protected from erosion by waves
sheet-pile cells, floated-in concrete caissons and currents in one way or another. In
that can be of square, rectangular, or circu- some instances the caiss;>n-type platfonns
lar cross section, and large-diameter steel are built and completely outfitted with ma-
or concrete cylinders. It should be pointed terial handling mechanism, fenders, and so
out that square and rectangular cross- forth in a protected hasis, then towed out
section caissons have been used more fre- and installed at the deployment site as a
quently than the other structural types. single complete unit. An example of this
· Gravity-type structures are usually filled
type of installation (Hay Point Tenninal in
with good quality, well-drained granular
Queensland, Australia) is described in
material and are capped with a heavy con-
Chapter 5. Here, for better contact with the
crete slab to which the mooring bollards or
stone mattress, the space under the
quick-release hook systems are anchored.
caisson's base is filled with grout containing
a thixotropic admixture. The scour protec-
tion of the caisson base is accomplished by
using articulated concrete block mats placed
on the periphery of the caissons.
Examples of concrete cylinder-type
floated-in caissons that are used as breast-
ing as well as mooring dolphins are given in
Tsinker (1995). It should be pointed out
that the round-shaped dolphins . are better
suited to resist environmental loads (e.g.,
wave, ice, and current). They are also very
convenient for warping or turning the ship
around the end of the dock, when required.
The example of the gravity-type breasting/
mooring dolphin constructed in a fonn of a
steel-pile cell filled with a granular mate-
rial and capped with a concrete slab is
shown in Figure 8-10b.
An unusual gravity-type dolphin struc-
ture of composite construction is illustrated
in Figure 8-12. This structure comprises a
steel frame-like superstiucture that is pro-
jected above seal level and attached to a
floating concrete base. This type of dolphin
requires temporary buoyancy tanks tem-
porarily added to the system to ensure sta-
bility and control during towing and instal-
Figure 8-10. Examples of gravity-type breastingj lation. Alternatively, the system seating on
mooring dolphins. (a) Floated-in concrete caisson
filled with granular material; (b) steel sheet-pile cell
prepared bedding may be assisted by a
filled with granular material and capped with concrete heavy-lift crane(s); in this case, the buoy-
slab. ancy tanks are not required.
898 Offshore Deep Water Terminals

8.4.4 steel Jacket-Type ally installed in the largest segments that


Structures can be conveniently handled. Naturally,
proper bearing plates and anchor bolts must
As noted earlier, construction of offshore be used to ensure the fender's adequate
terminals and its many independent struc- performance and safety. Details on fenders
tures can be accomplished by utilizing a and fender systems design are given in
steel jacket fixed by piles to the seafloor. Chapter 2.
This type of construction was successfully
completed at an offshore terminal in Iran
(Kharg lsland), in Japan (Kaiyo II Sea-
Berth terminal in Tokyo Bay), and many 8.4.6 Mooring Dolphins
other locations all over the world. Details of
these projects are found in Gerwick (1986). Mooring dolphins are normally located some
distance behind the berthing line and are
designed for mooring of ships. Structurally,
these dolphins are, in general, similar to
8.4.5 Fenders breasting dolphins; perhaps, the only con-
ceptual difference between both is that
Fenders are provided on breasting dolphins mooring dolphins are designed strictly to
to absorb the impact energy of the docking secure a vessel's mooring lines. As in the
ship. Regardless of the type of construction, case of breasting dolphins, the top of the
the fender units are usually large and mas- mooring dolphin superstructure is located
sive. They must be properly and accurately at least 1.5 m above the highest sea level
aligned along the berthing line within ac- under operational conditions (including
ceptable tolerances. Fender units are usu- tides, storm surges, and waves).

H ,,
,, 11
~,
i\1

un ::\\
•• 11 •\ \

u u uu ...:!!::.L
Figure 8-11. Iron ore loading facility. The loader supporting pier
structure if protected from ship impact by fiexible dolphins.
Offshore Deep W ater Terminals 899

Normally, mooring dolphins are designed tural design of a platform. It should be


for mooring loads from the ship only. How- noted that, in some instances, these envi-
ever, where needed, they are also designed ronmental forces may also control design of
to sustain heavy loads imposed on them breasting and mooring dolphins.
directly by environmental components such In designing to resist wave forces, it is
as wind, waves, and ice. vital to provide sufficient clearance between
Mooring devices such as bollards, quick- the crest of the design wave and the under-
release hooks, and others are normally in- side of the platform deck. Depending on the
stalled at the center of dolphin's superstruc- level of acceptable risk, the design wave
ture. Hence, there are no torsional loads from a 50-year to 100-year return period is
that affect the dolphin structure. An exam- usually selected. The assignment of an ac-
ple of a convenient piled mooring dolphin is ceptable risk level must reflect the eco-
shown in Figure 8-8. nomic consequences of a shutdown of the
facility due to damage. Naturally, each fa-
cility should be treated individually. For
8.4.7 Loading 1 Unloading example, a facility having more than one
berth should be treated differently from that
Platforms with only one berth. Another factor to be
considered is the availability of tugs of the
Platforms are designed primarily to support
required installed horsepower.
the cargo handling equipment. They are also
The platform should provide sufficient
used as operation centers for the facility. In
work area to support all wharf operations
some installations they are used as a pri-
other than mooring. This will include (as
mary breasting structure. Examples are
applicable to the particular project) cargo
shown in Figures 8-1, 8-2, 8-3, and 8-11.
handling equipment, operating oftice, wash-
In most cases, however, the platform is
rooms, storage space, fire protection equip-
used as a separate part of the berth facility
ment, boat landing, and so forth. If a sub-
to support cargo handling equipment and
marine pipeline(s) is used to link the wharf
associated services. In the latter case, envi-
to the onshore facilities, then it is necessary
ronmental forces produced by waves, cur-
to consider a reliable means of supporting
rent, or ice will usually control the struc-
the pipeline riser at both the seafloor and
the platform levels; it must be designed to
minimize potential stress concentrations.
The potential for scour around the base
of the riser as well as around all other parts
of the wharf should be considered. Scour is
3
particularly troublesome on sand bottoms
but can occur on any bottom type (i.e., silt,
clay, or the like).

8.4.8 Access Trestles and


bedrock catwalks
Figure 8-12. Dolphin of composite construction; in-
stallation procedure. 1-Steel superstructure;
The offshore facility is normally accessed by
2-fioating concrete base; 3-buoyancy tank to be re- a trestie. It may be constructed in a form of
moved after installation; 4-rockfill bedding. a short- or middle-span bridge that is de-
900 Offshore Deep Water Terminals

signed to carry conveyer loads, product by terminal personnel and are therefore de-
pipelines, vehicular traffic, weight of miscel- signed for a uniform load ranging from 200
laneous service lines, and a uniformly dis- to 500 kgjm 2 ; smaller values are typically
tributed load around 500 kgjm 2 • Trestie used for design of narrow catwalks (1.2 m
spans are typically placed over independent wide or less) and higher loads are used for
pile bents or gravity-type piers and are sup- design of catwalks up to 1.8 m wide. To
ported on elastomeric bearings (Husain, account for the dynamic nature of a pedes-
1983; EAU, 1990) or other suitable bearings trian load, a coefficient of 1.3 is usually
designed to accommodate the expected considered.
loads, displacements, and rotations of the Additionally, the stiffness of narrow cat-
span ends. walks must be checked in the horizontal
Short spans are usually designed in a plane. Structurally, long-span (30 m and
form of prefabricated, prestressed concrete more) catwalks are constructed in a form of
T -shaped beams (sometimes steel beams are a steel truss usually fabricated from hollow
also employed). Longer spans are typically steel structural sections (HSS), with a
designed in a form of a single-span steel walkway placed on struts that join the
truss bridge. Prefabrication of the trestie load-carrying trusses (Fig. 8-13c). Direct
connection between the HSS members re-
structural components results in a dramatic
quires special design considerations to en-
reduction in construction time over open
sure proper behavior. Two problems are es-
water. Both short- and long-span structures
pecially significant. First, where a smaller
are normally designed in accordance with
HSS section is connected to a larger one
general practice applicable to design and
(with no stiffeners or gussets added), it is
construction of conventional bridges.
important to be sure the wall thickness of
Furthermore, steel components of the
the larger section is thick enough to resist
trestie must be adequately protected from
very high local stresses that may be in-
corrosion.
duced. Second, in earthquake areas it is
In water depths that are less than about important to detail the connections so that
3-3.5 m, construction of a rockfill causeway they can carry high earthquake-induced
may be more economica! than a bridge-type stresses that can be several times greater
construction. Hence, sometimes the access than those during normal operation. In
structure can be constructed as a combina- some cases, for better performance, HSS
tion of both causeway portion and bridge structural components can be filled with
structure. Naturally, the bottom of a bridge concrete. For a discussion on this subject
span and top of a causeway should be ele- the reader is referred to Chapter 7.
vated above the highest water level, high In the case of a conventional road-
enough to prevent them from being over- bridge-type geometry of a catwalk, the de-
topped by waves or being damaged by signer must be concerned with the buckling
waterborne ice or flotsam. stability of the upper cords of both side
Catwalks are usually designed in a form trusses; normally they must be provided
of an open steel truss construction with with an upper cord restraint system. An
open grating deck or of prefabricated pre- economica! solution can be achieved by
stressed concrete beams placed on terminal placing both side trusses at an angle to the
structures (e.g., platform and breasting and horizontal and providing them with a
mooring dolphins). Details are illustrated mutual upper cord to form a space truss
in Figures 8-13a through 8-13c. Some- (Fig. 8-13a).
times, catwalks may require intermediate In locating the catwalks it is important
supports. to place them high enough out of reach by
Normally, the catwalks are used solely waves, floating ice and flotsam, and far
Offshore Deep Water Terminals 901

enough in back of the berthing line of the


dock so that they cannot be stuck and
damaged by ship berthing or departing
maneuvers.

8.5 STRUCTURAL DESIGN

8.5.1 Marine Foundation


and lts Effects on
Structural Design
8.5.1.1 Jntroduction
The purpose of this section is to provide
the reader with a concise overview of ma-
rine foundations and their effects on design
and construction of offshore marine facili-
ties. In most cases the quality and composi-
tion of the marine foundation plays a deci-
sive role in locating the offshore facility and
selecting the most appropriate type of ma-
rine structure. Typically, deep water off-
shore terminals are constructed in a rela-
tively shallow area of the ocean's physio-
graphic province called the "continental
shelf."
Emery (1969) states that the shelf break
is commonly defined as that area within
200 m of water depth; although in practice
it varies from 20 to 500 m (Young, 1991).
The width of the continental shelf varies
from O to 1500 km, and its average gradient
is approximately 0.2%. The east coast of the
United States and Canada has a very wide
continental shelf owing to accretion,
whereas on the Pacific coast of South
America, it is very narrow as a result of
subduction of oceanic plates. The complex-
ity of the shelf areas is due to its geological
history and the action of the various en-
vironmental elements.
Geological processes of the seafloor and
Figure 8-13. Catwalk systems: (a) space truss cat- shallow subsurface have been studied ex-
walk fabricated from steel tubular components; (b)
tensively in the past 50 years and are rea-
catwalk constructed from prefabricated, prestressed
concrete beams; (c) conventional trusslike catwalk fab- sonably well understood. A vast amount of
ricated from hollow structure steel (HSS) components. research and explorations have been fo-
902 Offshore Deep Water Tenninals

cused on the shelfs geologic setting, 3. Soil behavior related to the different
sediment type, and geologic processes. The levels of stress and strain expected in the
process includes water depth, geologic field
structure, stratigraphy, regional tectonics, 4. Soil stress-strain characteristics and
and sedimentation. pore pressure response related to cyclic
Emery (1969) points out that on the hasis loading
of e:xisting data, the continental shelfs
foundation materials can be distinguished Details on standardized laboratory proce-
as those formed by sedimentary strata and dures and equipment used to perform these
those formed by igneous rock and metamor- tests are given in ASTM (1987), Bishop and
phic rocks. Additionally, the topography of Henkel (1962), BSI (1975), Lambe (1951),
the continental shelf is essentially flat, with and most standard texts on the subject of
seafloor deposits ranging from rock out- soil mechanics.
crops, boulders, coarse sands and gravels, Measurements and techniques to deter-
to biogenic soils, carbonate reefs, silts, clays, mine in situ soil properties is discussed by
and permafrost in the arctic and subarctic Winterkorn and Fang (1991); standard soil
regions. The wide array of processes that exploration practice used in foundation en-
have accrued over thousands of years, cou- gineering is given in Lowe and Zaccheo
pled with the various environmental forces (1991); Chaney (1991) and Young (1991)
such as waves, currents, ice scouring or discuss methods of offshore geotechnical in-
gouging, earthquakes, and others have re- vestigation that include sampling and
sulted in many complex and varied seafloor preparation of marine sediments. The latter
conditions around the world. The state of author also provides an insight into geo-
the art on this topic is discussed by Young physical survey of the seafloor. For more
(1991) and Chaney and Demars (1991). information on a subject matter consult
Chapter 4.

8.5.1.2 Sediment Characteristics 8.5.1.5 sampling


Knowledge and proper understanding of Sediment samples are generally obtained
a seabed's sediment characteristics is from boreholes drilled from the jack-up
needed for the successful design of the pro- platforms or miscellaneous drill ships
ject's marine structures. Some offshore in- (McClelland, 1972). For the same purpose,
stallations may require information only on modern technology offers a self-contained
the upper few meters of sediments, whereas submersible unit that can be operated on
others, such as piled structures, may need the seafloor from almost any vessel type
information to a depth of 50 m and more. (Geise and Kolk, 1983). A state-of-the-art
Engineering properties of various soils review of soil sampling methods is given in
are obtained either from laboratory tests Young (1991).
conducted on samples of the seafloor It should be noted that despite all devel-
sediments or from in situ. As indicated by opments in the field of soil sampling, it is
Sullivan et al. (1980), the objectives of the still very difficult to obtain high-quality
laboratory program should accomplish the undisturbed samples of offshore soils for
following: laboratory testing. As pointed out by
Noorany and Seed (1965) and Skempton
1. Soil identification and classification and Sowa (1963), stress relief and distur-
bance associated with removing samples
2. Soil compressibility characteristics re· from a deep borehole may alter the stress-
lated to sustained load
strain-strength properties of a soil sample.
Offshore Deep Water Terminals 903

In the past 10-15 years, much greater environmentalloads on the structure as well
emphasis has been placed on determining as ship impact and mooring loads are
soil properties by in situ testing techniques. determined. Then the worst practical load
A wide variety of in situ testing devices combination is established and used for
have been developed for either downhole or structural analysis. Normally, this is a
seafloor modes of operation. These are sum- trial-and-error procedure. The most efficient
marized and discussed by Young (1991). and economic structure(s) is selected on the
In special cases, the validity of soil pa- hasis of evaluation of several alternative
rameters (e.g., seafloor soil stiffness and structural solutions.
bearing capacity) that are used in struc-
tural design can be verified by conducting 8.5.2.2 Berthing and Mooring
various in situ small or full-scale tests such Loads
as plate load tests (Andresen et al., 1979).
Pile design parameters are normally veri- These are discussed in detail in Chapter
fied by carrying out full-scale pile tests. For 3. The typical breasting dolphin is exposed
details, consult Chapter 7. to berthing and mooring loads that usually
As pointed out elsewhere in this book, do not act simultaneously. In a typical
the seafloor's geotechnical parameters dic- berthing maneuver the ship will approach
tate the type of structure that is most suit- the dock at a certain angle to the berthing
able for the particular geotechnical condi- line. Ship movement normal to the breast-
tions existing at the project site; these ing system can coincide with some move-
structures could be either those supported ment parallel to the berth. This results in a
by deep foundations that include relatively lateral rubbing force transmitted to the dol-
long driven piles or drilled and grouted phin structure through the fender system
piles, or designed in a form of miscellaneous (Figure 8-14).
gravity-type structures. Furthermore, depending on fender-
More information on the subject of system geometry, ship impact can result in
soil testing and soil properties is found in the eccentrically applied load. This load re-
Chapter 4. sults in torsional moments that must be
resisted by the dolphin structure. Note that
in addition to ship impact load and longitu-
dinal rubbing force, the vertical rubbing
8.5.2 Basic Design Procedures force, attributed to ship heaving during
berthing maneuvers, should also be
8.5.2.1 Sequence of structural
considered.
Design
The magnitude of a ship impact force,
When the project site is selected and en- developed during the berthing maneuver,
vironmental conditions, water depth, and should be properly evaluated. As discussed
seafloor conditions are evaluated, the facil- in Chapter 3, when a vessel approaches a
ity's mooring layout is determined based on marginal wharf, its bow (in some instances,
principles as discussed earlier. Essentially, the stern) will impinge on one or severa!
the layout of mooring structures depends on fender units first. The reaction of the fender
the type of vessel(s) to be accommodated, will cause the vessel to rotate, during which
number and size of mooring lines available a portion of the vessel's kinetic energy is
on a ship and on the nature of the bulk dissipated. The value of the berthing coef-
material to be handled at the terminal. ficient, commonly known as eccentricity fac-
N ext, the structural concept is developed tor, CE, which accounts for energy dissipa-
(e.g., piled or gravity-type dolphins) and the tion depends largely on the contact point
904 Offshore Deep Water Terminals

between the fender and the ship huli. In the indicated in Figure 3-14 (Chapter 3). This
case in which a ship contacts the wharfs can be accounted for by increasing the de-
fender system by the bow or stern, the sign lateralload by a factor of 1.2, as indi-
aforementioned contact point is commonly cated in Eq. (3-29) (Chap. 3). For details,
referred to as the "quarter point" and the consult Chapter 3.
value of CE is usualiy taken as equal to 0.5. Wind and current also produce lateral
However, in the case of a dolphin-system and longitudinal loads and a yaw moment
berth the point of contact between ship and on a moored vessel. These loads displace
fender unit may coincide with the vessel's and rotate the vessel relative to its position
center of gravity, in which case the vessel before the loads were applied (Fig. 8-15).
may not rotate; hence, the berthing energy Essentialiy, the applied loads are resisted
wili be equal to the fuli kinetic energy of by the mooring system roade up of breast-
the vessel (CE = 1.0). For complete details ing and mooring dolphins. The mooring
on this subject, the reader is referred to loads are transmitted to the mooring sys-
Chapter 3. tem through mooring lines. For static equi-
If a breasting dolphin is also used for the librium, the applied loads must equal the
purpose of ship mooring, then, in addition restoring loads along the longitudinal and
to the ship impact load, the mooring forces lateral axis of the vessel and, as weli, the
from the spring lines, and under certain sum of the applied and restoring yaw mo-
conditions from breasting lines, should be ments about the center of gravity of the
considered for evaluation of the dolphin's vessel must equal zero.
stability and its structural integrity. When The load distribution within the system
more than two breasting dolphins are used, is determined by considering the system's
possible dolphin misalignment along the elasticity (e.g., stiffness of the fender sys-
berthing line, Il, may effect distribution of tem, dolphin flexibility, and elasticity of
the berth lateral (Ps(p)) and longitudinal mooring lines). For these calculations, the
(Ps<h>) loads between adjacent dolphins as elasticity of ali individual components of a

P'
d

elevation

Figure 8-14. Breasting dolphin. Load summary sketch: Pd and P:f are loads transmitted to
the dolphin through individual fenders; Pd is the total impact force (Pd = Pd + P:f); Pw is
dolphin weight acting at the structures center of gravity (c.g.); R 1 and R w are respectively the
longitudinal and vertical rubbing forces. 1-Ship huli; 2-fender; 3-dolphin.
Offshore Deep Water Terminals 905

total mooring system should be preliminar- vessel due to wind, current, and other
ily evaluated. It is obvious that elasticity of relevant environmental forces as per
the mooring lines and fenders is consider- Figure 8-15.
ably greater than that of regular piled or 2. Determine the breastingjmooring dol-
gravity-type dolphins. In practica! design, phins layout, mooring lines geometry, and
these structures are usually treated as rigid. properties of each of the mooring lines.
On the other hand, the flexible dolphins Calculate a load-deflection curve for each
of the mooring lines.
can substantially deflect under horizontal
Note: Information on mooring line prop-
loading. This is usually taken into consider- erties provided in Chapter 2 can be used
ation in the dolphinjfender-system design. for preliminary analysis.
The analysis of the system is typically
3. Select fender system to be used and con-
carried out in the following steps (Naval sider a load-deflection curve for each
Facilities Engineering Command, 1986): breasting dolphin.
4. Assume an initial displacement and rota-
1. Determine the total lateral and longitudi- tion of the vessel (new orientation) under
nalloads, and yaw moment on the moored the applied load.

)..,,
3

5 FxT: TOTAL FORCE ALONG X-AXIS


FyT: TOTAL FORCE ALONG Y-AXIS
MxyT: TOTAL YAW MOMENT ABOUT CEN-
TER OF GRAVITY (C.G.)
âx : SURGE OISPLACEMENT
ây : SWAY OISPLACEMENT
8 e : YAW ROTATION

Figure 8-15. Vessel response to static wind and current loading; design sketch.
[After Naval Facility Engineering Command, (1986).]1-Vessel position before
loading; 2-vessel equilibrium position after loading; 3, 4-center of gravity before
and after loading, respectively; 5-mooring dolphins; 6-breasting line; 7-fender;
8-breasting line; 9-spring line.
906 Offshore Deep Water Terminals

5. Determine the elastic elongation in each When using this procedure, good judgement
of the mooring lines and defiection of a must be exercised. The first step in this
dolphinjfender system corresponding to simplified procedure is to determine the
the vessel's orientation.
maximum total applied loads, both lateral
6. Determine the forces in each of the moor- and longitudinal, and maximum yaw mo-
ing lines and dolphinjfender system from
ment; these are further resolved into lateral
the above mooring line elongations.
loads at the bow and stern of the vessel,
7. Balance the acting and resisting forces and longitudinalloads. Lateral and longitu-
and moments.
dinalloads are analyzed individually.
Three load cases are typically visualized
If the above forces and moments do not in the analysis of lateral mooring loads as
balance, then the vessel is not in its equilib- are shown in Figure 8-16. In Case 1, the
rium position under the applied load. Hence,
lateralloads at the bow and stern act in the
a new vessel orientation must be considered
same direction and puii the vessel away
and ali previous steps are repeated until
from the mooring. Consequently, lateral
the new equilibrium position is found. Es-
loads are resisted entirely by the mooring
sentially, the above procedure can be easily
programmed and solved using a computer. lines in tension. Next (Case 2), lateralloads
Mooring line analysis can be performed at the bow and stern act in the same dire<;-
using the computer program FIXMOOR de- tion and push the vessel against the breast-
veloped by the Naval Civil Engineering ing dolphins. Here, the lateral loads are
Laboratory, Port Hueneme, California. taken entirely by breasting dolphins. Fi-
Marlowe (1995) discusses a program devel- nally (Case 3), lateralloads at the bow and
oped by Exxon and its application to gener- stern act in opposite directions. In this case,
ation mooring guidelines for ship /terminal the lateral load at one end of the vessel is
mooring situation. restrained by mooring lines, whereas lat-
The Navy Facilities Engineering Com- eral loads at the opposite end of the vessel
mand DM26.4 (1986) allow for the following are restrained by a breasting structure. In
simplified procedure as illustrated in Fig- ali three cases the longitudinal loads are
ure 8-16. This approach can be used for the assumed to be resisted by spring and bow
preliminary evaluation of mooring loads (or stern) lines in tension.
acting on the wharfs mooring components. Limitations of this simplified method are

----FyB - - - . .... FyB

2
2

----Fys

Figure 8-16. Analysis of mooring loads; design sketch: (a) Case 1; (b) Case 2; (c) Case 3.
1-Breast line in tension; 2-spring line in tension; 3-breast dolphin in compression. [After
Naval Facility Engineering Command (1986).]
Offshore Deep Water Terminals 907

as follows: the load on each individual dolphin is ana-


lyzed assuming that the vessel is simply
1. The mooring layout must be reasonably supported by dolphins. If the vessel is sup-
symmetrical. ported on two dolphins only, the system is
2. Vertical and horizontal mooring line
statically determined and is modeled as a
angles are not to exceed values indicated beam (vessel) supported on two supports
in Figure 8-4a. (dolphins); the multidolphin system if stati-
cally undetermined and is modeled as a
3. Mooring lines are of identical material,
construction, and diameter and are effec- continuous beam support on a multisupport
tively grouped at the bow and stern. system. In the case of piled dolphins that
include batter piles, or in the case of
4. Some mooring lines are assumed to be
gravity~type dolphins, the supports are
zero when there is no applied load, de-
spite the fact that lines are always loaded treated as rigid; the fl.exible dolphins can
due to self-weight andjor pretension, be modeled as elastic supports. In both
which is used to take the slack out of the cases, potential dolphin misalignment
line. should be accounted for as discussed ear-
5. Synthetic and natural ropes are ignored lier. When lateralloads act in opposite di-
in the analysis if the vessel is moored rections (Case 3), the system can be ana-
with both synthetic or natural rope and lyzed as previously discussed (Case 2), with
wire rope. the addition of a yaw moment. For a certain
6. The mooring line tension is linearly pro- combination of lateral loads and yaw mo-
portional to line extension. ment, a portion of the applied lateral load
will be restrained by a breasting dolphin in
compression while the remaining portion is
The tension developed in any one mooring
being restrained by the mooring line in ten-
line is assumed to be proportional to the
sion. Analyses of load distribution within
quantity(cos a cos fJ)jl, where the angles a
and fJ are shown in Figure 8-2a, and l is the system are carried out by using proce-
the length of mooring line. The term dures discussed earlier.
(cos a cos fJ)jl is thought of as a stiffness Maximum loads obtained from all three
term; the shorter line, the higher the ten- design scenarios are used for structural de-
sion in the line. This is consistent with the sign of berthing and mooring dolphins, as
fact that shorter lines are less elastic than included in the mooring system.
longer ones. The component of tension in As noted earlier mooring loads are trans-
the lateral direction is proportional to mitted to either breasting dolphins or purely
(cos 2 a cos 2 ())jl. The totalload in a moor- to the mooring dolphins via the vessel's
ing line is the sum of the distributed ap- mooring lines, which are secured at the
plied load, including the line pretension load bollards or quick-release hooks installed on
if applicable. dolphins; the former are normally used for
The procedure in question assumes that mooring of smaller vessels and the latter
the lateral loads are restrained primarily are used to handle the large vessels.
by breast lines and, to a lesser degree, by Petroleum products and LNG handling
bow and stern lines that are grouped near terminals used by large tankers and super-
the bow and stern of the vessel. This allows tankers are typically complete with quick-
one to analyze mooring lines at the bow and release system exclusively. An example of
stern separately. lf the applied loads push mooring accessories used in offshore instal-
the vessel toward the breasting system and lations are illustrated in Figures 8-6g, 8-7
no yaw moment is applied (Case 2), then through 8-11, 8-13, and 8-17.
908 Offshore Deep Water Tenninals

tially, the pile tensile capacity excludes the


end-bearing component. The pile's weight
and soil plug weight, where applicable, are
included in computing both tensile and
compressive capacity of the pile(s).
AII types of offshore structures supported
by pile foundations are analyzed to confirm
that the piles can resist substantial bending
moments, which comes from the direct ex-
posure of piles to environmental forces such
as waves and ice, impact by flotsam, but
Figure 8-17. Typical three-hooks quick release sys- first and foremost due to elastic displace-
tem installed on a mooring dolphin. ments caused by ship impact. Therefore,
reasonably accurate estimates of pile lat-
eral deflection and bending moments are
8.5.2.3 Analysis of Piled necessary, particularly in the high-stress
Dolphins
region at the point offixity near the seafloor.
As discussed earlier, a piled dolphin con- As pointed out in Chapter 7 the location
sists of a group of long, usually open-ended, of the fixity point within the soil stratum
driven piles, or drilled and grouted piles. depends on pile stiffness, value of displace-
These piles are designed to carry, without ment, and a soil horizontal subgrade reac-
,excessive deflection and vibration, the static tion. The embedded portion of the pile is
and dynamic loads associated with facility treated as a beam on an elastic or elasto-
operation and environmental conditions. plastic foundation. Additionally, as pointed
The analysis of piled dolphins must confirm out in Chapter 7, the current procedure to
that the selected arrangement of the pile determine the fixity point consists of itera-
group can provide sufficient axial and lat- tive elastic pile solutions with the secant
eral capacity during extreme loading condi- modulus of soil reaction, E 8 , being deter-
tions. The end result of this analysis is the mined from precalculated nonlinear p-y
sizing of piles for diameter, penetration, and curves varying with depth in any conve-
structural strength. nient manner (Matlock and Reese, 1960;
In some cases where soil-pile interaction Reese and Sullivan, 1977; Bogard and
is uncertain due to insufficient information Matlock, 1977). lnput parameters to this
on soil parameters or other factors, the pile method include the pile stiffness, EI, and
capacity is determined from direct field the soil response, which is a function of the
static tests or from a large-scale-model test. deflection of the pile.
As pointed out in Chapter 7, if the load There are several computer programs
application or structural stiffness are non- available for analysis of a laterally loaded
symmetric, the piled structures should be pile using a nonlinear soil reaction ap-
treated as a three-dimensional integrated proach (Young, 1991). Obviously, the best
pile- superstructure system. Consideration result for developing p - y curves is ob-
must be given to the stiffness of the super- tained by performing field test s.
structure, as well as axial, lateral, and tor- If piles in a pile group are closely spaced,
sional response of the pile group. then the response of an individual pile
The pile's axial capacity is usually deter- within this group to lateral loading should
mined by classical soil mechanics as a sum be evaluated not only on the hasis of the
of pile end bearing capacity and shaft re- lateral load deformation characteristics but
sistance (or skin friction resistance); essen- also as to consider the effects from other
Offshore Deep Water Terminals 909

piles in the group. In this case, p-y curves In a practica! design, a limit is usually
for the single pile should be modified (Focht placed on .!1 1 due to the selected type of
and Koch, 1973). It should be noted that in fender system, and the value of .!1 2 is basi-
most practica! cases of dolphin design, piles cally limited by the strength of a pile(s).
are spaced far enough apart to be treated as In some instances, however, .!1 2 can also
individual piles rather than as a pile group. be limited by some specific operating condi-
Further details and a complete list of tions. Hence, by selecting values of .!1 1 (or
related references on this subject matter .!1 2 ) the allowable value of .!1 2 (or .!1 1 ) can be
are provided in Chapter 7. computed from Eq. (8-1). In practice, the
design process is one of trial and adjust-
8.5.2.4 Analysis of Flexible ment to determine the most efficient and
Dolphins economica! combination of fender system
and number, material, and size of piles re-
Flexible dolphins are designed as later-
quired to absorb the energy from a berthing
ally loaded piles or columns as either a
ship, or to resist mooring loads.
single pile, cluster of piles, or group of verti-
For some additional information on
cal piles joined at one or several levels of
geotechnical aspects of fl.exible dolphin de-
stiff horizontal diaphragm members, as il-
sign, the reader is referred to PIANC (1984).
lustrated in Figures 8-6 and 8-11.
Useful information is given in Vrijling and
Independent dolphins are usually al-
Oosting (1985).
lowed to defl.ect over a considerable dis-
tance (large defl.ections are the most effi-
cient means of absorbing ship berthing 8.5.2.5 Analysis of oravity-Type
kinetic energy). The associated vertical pile Dolphins
defl.ections in either a free-head form or As discussed in Section 8.4, gravity-type
head fixed within a rigid concrete deck, and caissons can be deployed in a form of a
the resulting bending moments are dis- fl.oated-in concrete boxlike or cylindrical
cussed in Chapter 7. The bending moment caisson, bottomless large-diameter cylinder,
that can be applied to a pile is limited by
allowable stresses. In the working stress
design method, these stresses can be in-
creased by 33% for the extreme load condi-
r ....n H
tion, such as impact by a ship that ap-
proaches the dolphin structure with an ab-
1 4
1 L p
=It 2
normally high velocity, or mooring load due 11
11
to an extreme wind condition. 11
h
In the case of an independent dolphin, 11
the kinetic energy of a berthing ship (Ek) ,_ 1

that is absorbed by the dolphin structure •VW'~

can be calculated from the following


(Fig. 8-18). 1
zl
1
~ tix,itY
Polnt

where
H = horizontalload Figure 8-18. Reaction of flexible dolphin: 1-pile; 2
-fender system. H is the ship berthing impact load,
11 1 = defiection of a fender L\. 1 is the deflection offender system, and L\. 2 is the pile
11 2 = distance moved by the pile or pile group deflection.
910 Offshore Deep Water Terminals

or steel sheet-pile cell, ali filled with granu- dolphins are exposed to large horizontal
lar material. loadings that create inclined and eccentric
In most practica! cases, these structures load conditions affecting the concentration
have been installed on a prepared stone of local bearing pressures on the mattress
bedding. Where required, the stone bedding and foundation materials. Load eccentricity
or mattresses have been protected against decreases the ultimate load that the foun-
scour in one way or another. Concrete mat- dation can withstand. Moment equili-
tresses of miscellaneous construction, heavy brium can be reinstated by applying the
concrete blocks, or layers of stone blanket vertical load to the centroid of the base
placed around the base have been used for on an "effective foundation area" (American
protection against scour. For the same pur- Petroleum Institute, 1984; Hansen, 1970).
pose, caisson-type structures may have a The foundation's ultimate bearing strength
skirt that penetrates the weaker near- analysis are carried out as discussed in de-
surface soils. tail in Chapter 5. In these analyses, shape
Gravity-type dolphins require stability factor components are determined as fol-
analyses that are very similar to that used lows:
for analysis of a gravity-type quay wall. For a rectangular base,
These include a stability analysis to confirm
that both foundation soil and bedding mate-
rial can provide sufficient bearing resist-
ance against static and dynamic loads im-
posed on them, and an analysis to confirm
that the structure itself possesses the slid- (8-2)
ing and overturning resistance against the
combined environmental and operational
loads. In addition, settlement of the struc-
ture should be analyzed and the effects of
cyclic loading on the foundation soils must For a circular base (centric load only),
be predicted. Where applicable, additional
analysis is required to check that the skirt
penetration resistance, developed by the
soil, can be overcome by the loads imposed
by the structure in the process of installa- (8-3)
sq = 1 + tan cp
tion. The effect·of cyclic loads primarily as-
sociated with the wave force component of s-r = 0.6
storm loadings must not be overlooked. As
discussed in Chapter 4, this type of loading
may lead to liquefaction of the foundation The definition of Nq, Ne, and cp are given in
soil due to the complete loss of shear detail in Chapter 5. Components B' and L'
strength. This may result in a sudden represent the effective bearing area A'.
downward movement of the structure The load eccentricity, e, is computed from
andjor in structural damage. e = MjQ as indicated in Figure 8-19. The
As pointed out in Chapter 5, usually sev- effective dimensions B' and L' for a rectan-
era! interactions of these analysis are re- gular foundation are computed as follows:
quired to achieve a balanced design with
respect to economy and safety of the struc- B' = B- 2e 1
ture construction, installation, and opera- (8-4)
tion. Normally, both breasting and mooring L' = L- 2e 2
Offshore Deep WaterTerminals 911

Again, when undrained conditions exist for


cohesive soils, the foundation's ultimate ca-
pacity is computed for <P = O, and for fuliy
drained conditions where an ali-sand/
gravei profile exists, the undrained shear
strength ofthe foundation soil, su, is treated
as equal to zero. It should be pointed out
that, where applicable, soil properties used
in ali analyses should be based on values
that consider effects of the cyclic loading
upon foundation soil (Fogg et al., 1978).
Figure 8-19. Eccentrically loaded gravity-type cais-
Values of ultimate bearing capacity of
sons: effective foundation area. (a) Equivalent loading; foundation soil determined by the above
(b) rectangular footing; (c) circular footing. [After method are normaliy reduced by a factor of
American Petroleum Institute (1984).] safety to ensure an adequate margin of
safety against bearing failure. A factor of
safety equal to 2.0 is typicaliy considered
where e 1 = MtfQ, e 2 = MdQ, and B and
for a normal load combination and 1.5 for
L are the actual width and length, respec-
the extreme loading case. These values
tively, ofthe base. To prevent the structure's
should be used after cyclic loading effects
base from lifting off the bedding, the eccen-
have been taken into account. If geotechni-
tricity should not exceed B j6.
For a circular base, the effective bearing cal data are sparse or site conditions are
area according to the American Petroleum particularly uncertain, increased values of
Institute (1984) is determined by consider- the factor of safety may be warranted.
ing it to be twice the area of the segment As pointed out earlier, both breasting and
ADC (Fig. 8-19). In calculations, this area mooring dolphins are subjected to combined
is assumed as quasi-rectangular with a vertical and horizontalloads resulting from
length-to-width ratio equal to line lengths ali kinds of environmental forces, ship im-
AC to BD; thus, the effective dimensions for pact, mooring forces, as weli as weight of
a circular base with radius R can be com- the cargo handling and transporting equip-
puted from ment. The horizontalloads cause a shearing
stress on the stone bedding and foundation
soil, while the vertical load and moment
A'= 28 =B'L' (8-5)
cause a nonuniform bearing pressure at the
interface between the structure's base and
where stone bedding andjor foundation soil.
The maximum value of bearing pressure
exerted on the stone mattress and the foun-
8 = 0.51TR 2 - [ e(R 2 - e 2 ) 0 "5 + R 2 sin- 1 ( ~)] dation soil beneath the mattress may be
(8-6) calculated assuming the pressure to vary
linearly over the structure's base, similar to

'-[ (R+e)0.5] 0·5


that used in the design of a gravity-type
L- 28 - - (8-7) quay wall (Chap. 5):
R-e

R-e )o.5 V M"


+- +-
My
(8-9)
B'=L' ( - - (8-8) amaxjmin = -
A- W"- WY
R +e
912 Offshore Deep Water Terminals

where as follows:
V effective vertical load acting on
=
Hu = suA'p (undrained conditions)
dolphin base
(8-11)
A= base area
M:x: and MY = overturning moments correspond- Hu = c' A~+ Q tan cf>' (drained conditions)
ing to base axes x and y about
(8-12)
the geometrica! center of the base
Wx and WY = base sectional modulus corre- where
sponding to axes x and y
su = foundation soil undrained shear strength
In the case of a centrally loaded circular c' = drained cohesion
base, cf>' = effective angle ofinternal friction
A~ = projected effective area, A'P >A'
V 4Hh (8-13)
arnaxjmin = 7TR2 ± 7TR3
(8-10)

where B~ = B' + 2h 1 and EP = L' + 2h 1 ,


where h 1 is the average depth of the rock
where
mattress. If the structure is placed directly
R = base radius
on the seafl.oor, A'P = A' is used in Eqs.
H = horizontal force (8-12) and (8-13).
h = horizontal force arm In many practica! designs the friction
coefficient concept has been used for analy-
Because the footing is incapable of exerting ses of a structure's stability against sliding
a tensile force on the soil, if the value of failure. In these analyses for concrete struc-
amin is negative, the solution obtained from tures placed on a rockfill mattress, the co-
Eqs. (8-9) and (8-10) need to be corrected efficient of friction is usually taken to be
in a manner similar to that recommended 0.5-0.65; the friction coefficient between a
in Chapter 5. More accurate solutions can rubble mattress and foundation soil com-
be obtained by analyzing the dolphin as a prised of granular material (drained condi-
rigid structure sitting onan elastic founda- tion) is equal to tan(Î<f>'), where <P' is the
tion. The alternative procedures for calcu- effective angle of internal friction of the
lating the bearing stresses exerted by foot- material used for the wall bedding or foun-
ings of various shapes on fomidation soils dation soil, whichever produces a smaller
are given in Dawson (1983) and the value. For more details on this subject, see
American Petroleum Institute (1984). Chapter 5.
To assure lateral stability, analysis is If a fully embedded skirt is part of a
carried out to confirm that the horizontal structure base, the sliding resistance devel-
component of the angled load is sufficiently ops along the full soil-structure interface.
smaller than the sliding resistance devel- In the case of a deep penetrating skirt, a
oped between the structure's base and the portion ofthe passive soil resistance against
stone bedding, and between the latter and the skirt may be included in the sliding
the foundation soil. The ultimate sliding stability analysis. For more information on
resistance, Hu, between the stone bedding this subject, the reader is referred to Young
and the natural seabed material is consid- (1991) and Young et al. (1975).
ered to be the adhesion in clay soils and a A discussion on the prediction of forces
function of the effective angle of internal necessary to allow the skirt to penetrate
friction between the soil and stone bedding into the seafloor soils is found in Chapter 5.
Offshore Deep Water Terminals 913

The overturning stability ofthe gravity-type (7- 8v)(1 + v)H


os= (8-15)
dolphin analyzed similarly to that as dis- 16(1- v)ER
cussed in Chapter 5.
To provide an adequate margin of safety 3M(1- v 2 )
(Jm = (8-16)
against failure under various loading condi- 4ER 3
tions, the ultimate bearing capacity and re-
sistance to sliding and overturning should
where
be multiplied by a safety factor. For normal
and extreme load conditions, the safety fac- ou and oh = relatively vertical and horizontal
displacements, respectively
tors are respectively considered as follows:
Bm = overturning rotation
the ultimate bearing capacity of foundation
soil and overturning stability 2.0 and 1.5, Q and H = vertical and horizontal loads, re-
spectively
sliding stability 1.5 and 1.25.
If a gravity-type dolphin structure is E = modulus elasticity of foundation
soil
placed on soft foundation materials, then in
addition to the stability analysis, a settle- v = Poisson's ratio of the soil
ment prediction may be required. The ana- R = radius of the foundation base
lytical procedures used for a settlement pre-
diction are the same as used for the An alternative solution for circular or
gravity-type quay walls discussed in Chap- square footings is given in Dawson (1983).
ter 5. Furthermore, an accurate prediction Carrier and Christian (1973) developed a
of settlement requires a thorough knowl- solution for the prediction of vertical dis-
edge of the soil's properties, in particular, placements of a rigid, concentrically loaded
its time-dependent consolidation character- circular foundation placed on a foundation
istics, and drained and undrained elasticity with a modulus of elasticity, E, that in-
modulus (Young's modulus, E), as well as creases linearly with depth. Solutions for
subsurface variations across the foundation other base geometries andjor where the
site. soil is anisotropic and exhibits a layered
Settlements caused by cyclic loading are profile are summarized by Poulos and Davis
difficult to predict due to the generation of (1974).
As discussed in Chapter 4, the long-term
excess pore pressures in the foundation soils
settlement of the structure depends on the
and subsequent reduction in soil stiffness.
soil type and its time-dependent consolida-
Young et al. (1975) suggest that the predic-
tion characteristics. In the case of granular
tion of vertical, horizontal, and rocking
soils, with their rapid dissipation of pore
movements of a gravity-type structure may
pressures, long-term settlement is very
be estimated by the method described by
small and is usually ignored. On the other
Poulos and Davis (197 4), which applies to hand, consolidation settlement in cohesive
isotropic and homogenous foundation mate- soils is gradual because excess pore pres-
rial. Poulos and Davis (1979) suggest that sure dissipates very slowly in a high im-
the settlement of a circular structure with a permeable medium.
rigid foundation base resting on the seafloor Where the foundation soil is comprised of
comprised of isotropic and homogeneous several individual layers of cohesive soils,
material may be computed by the total consolidation settlement is esti-
mated as the sum of the settlement com-
puted for each individual soil layer.
(8-14) For settlement analysis, a thick layer of
homogeneous soil is usually divided in
914 Offshore Deep Water Terminals

thinner layers. Where the added vertical -Energy Ports of the Future?" Proceedings
effective stress varies greatly through these ASCE Specialty Conference PORTS '80,
layers, the stress at the layer's midpoint is N orfolk, VA
used for the settlement analysis. Settle- DAWSON, T. H., 1983. Offshore Structural Engi-
ment of individual soillayers can be evalu- neering. Prentice-Hall, Englewood Cliffs, NJ.
ated by the method given in Chapter 5 or EAU, 1990, Recommendations of the Committee
computed by any other credible method. for Waterfront Structures. Ernst & John,
Berlin.
REFERENCES EMERY, K. 0., 1969. "The Continental Shelves
with Biographical Sketch." Scientific Ameri-
AMERICAN PETROLEUM INSTITUTE, 1984. "Plan- can, Voi. 221, No. 42.
ning, Designing and Constructing Fixed Off-
shore Platforms." API RP 2A FERGUSON, N. J., 1981. "Planning, Layout and
Design of Bulk Terminals for Large Ships."
ANDRESEN, A, BERRE, T., KLEVEN, A and LUNUE, Proceedings 25th PIANC Congress, Voi. 3,
T., 1979. "Procedures Used to Obtain Soil
S II, Edinburgh.
Parameters for Foundation Engineering in the
North Sea," Marine Geotechnique, Voi. 3, FOCHT, J. A, Jr. and KOCH, K. J., 1973. "Rational
No. 3. Analysis of the Lateral Performance of Off-
shore Pile Groups." Proceedings 5th Annual
BISHOP, A W. and HENKEL, D. J., 1962.
Offshore Technology Conference, Voi. 2,
Measurement of Soil Properties in the Triax-
Paper No. 1896, Houston, TX.
ial Test, 2nd ed. Edward Arnold Ltd., London.
BoGARD, J. D. and MATLOCK, H., 1977. "A Com- Foss, I., DALHBERG, R., and KVALSTAD, T., 1978.
puter Program for the Analysis of Beam- "Foundation Design of Gravity Structures
Columns Under Static Axial and Lateral with Respect to Failure in Cyclic Loading."
Loads." Proceedings 9th Annual Offshore Proceedings 10th Annuai Offshore Technol-
Technology Conference, Paper No. 2953, Voi. ogy Conference, Voi. 1, Paper No. 3114,
3 Houston, TX. Houston, TX.
BRATTELAND, E., 1974. "A Survey on Acceptable GAYTHWAITE, J. W., 1990. Design of Marine
Ship Movements in Harbours." The Dock and Facilities. Van Nostrand Reinhold, New York.
Harbour Authority, Voi. LV, No. 647. GEISE, J. M. and KOLK, H. J., 1983. "The Use of
BRUNN, P., 1989. Port Engineering, 4th ed. Gulf Submersibles for Geotechnical Investiga-
Publishing Co., Houston, TX. tions." Proceedings of International Confer-
BSI, 1975. Method of Test for Soils for Civil ence JUBTECH '83, London.
Engineering Purpose, Standard BSI377. GERWICK, B. C., 1986. Construction of Offshore
British Standard Institution, London. Structures. John Wiley & Sons, New York.
CARRIER, W. D. and CHRISTIAN, J. T., 1973. "Rigid HANSEN, J. B., 1970. A Reuised and Extended
Circular Plate Resting on a Non-Homogenous Formula for Bearing Capacity, Geoteknisk In-
Elastic Haif-Space." Geotechnique, Voi. 23, stitute Bulletin No. 28. Danish Geotechnical
No.1. Institute. Copenhagen.
CHANEY, R. A., 1991. "Sampling and Preparation HEADLAND, J. R., 1995. "Offshore Moorings."
of Marine Sediments." Foundation Engineer- Marine Structures Engineering: Specialized
ing Handbook, 2nd ed., Fang, H-Y (ed.). Van Applications, Tsinker, G. P. (ed.) Chapman &
N ostrand Reinhoid, New York. Hali, New York.
CHANEY, R. C. and DEMARS, K. R., 1991. "Off- HUSAIN, S. I., 1983. "Design and Full Scale Test-
shore Structure Foundations." Foundation ing of Elastomeric Bridge Bearings for Trestie
Engineering Handbook, 2nd ed., Fang, H-Y and Jetty Trestie of Super Tanker Offshore
(ed.) Van Nostrand Reinhold, New York. Terminal." ASCE, Proceedings Specialty Con-
CHESSON, E., 1980. "Offshore Artificial Islands ference PORTS '83, New Orleans, LA.
Offshore Deep Water Terminals 915

LAMBE, J. W., 1951. Soil Testing for Engineers. SKEMPTON, A. W. and SowA, V. A, 1963. "The
John Wiley & Sons, New York. Behavior of Saturated Clays During
LEMAHAUTE, B. L., 1977. "Wave Agitation Crite- Sampling and Testing." Geotechnique, Vol. 13,
ria for Harbors." Proceedings ASCE Specialty No. 4.
Conference PORTS '77, Long Beach, CA SLINN, P. J. B., 1979. "Effect of Ship Movement
LOWE, J., III and ZACCHEO, P. F., 1991. "Sub- on Container Handling Rates." The Dock &
surface Explorations and Sampling." Founda- Harbour Authority, August.
tion Engineering Handbook, 2nd ed., Fang, SoROS, P., KoMAN, B., and LUBETKIN, B. M., 1986.
H-Y (ed.). Van Nostrand Reinhold, New York. "lnnovative Bulk Ports in Latin America."
MARLOWE, L. H., 1995. "Computer Analysis of The Dock & Harbour Authority Vol. LXVII,
TerminaljShip Mooring." ASCE Proceedings No. 785.
Specialty Conference, PORTS '95, Tampa, FL. SUGIN, L., 1983. "Conventional versus Offshore
MATLOCK, H. and REESE, L. C., 1960. "Gener- Loading and Unloading." Indonesian Coalj
alized Solutions for Laterally Loaded Piles." Liguite Dev. and U.S. Technical Symposium,
ASCE Journal of the Soil Mechanics and J akarta, Indonesia.
Foundation Division, Vol. 86, No. SM5.
SULLIVAN, R. A, WRIGHT, S. J., and SANNER, D.
McCLELLAND, B., 1972. Techniques Used in Soil W. F., 1980. "Evaluation of Design Parame-
Sampling at Sea. Offshore, Vol. 32, No. 3. ters from Laboratory Tests." Offshore Site In-
NAVAL FACILITIES ENGINEERING COMMAND, 1986. vestigation: Proceedings of a Conference held
"Fixed Moorings," Design Manual 26.4, March, 1979, Graham & Trotman, London.
Alexandria, VA TERMPOL Code, 2nd ed., 1983. Canadian Coast
NooRANY, I. and SEED, H. B., 1965. "In situ Guard, Ottawa.
Strength Characteristics of Soft Clay." ASCE
THORENSEN, C. A, 1988. Port Design. Tapir Pub-
Journal Soil Mechanics and Foundation
lishers, Trondheim, Norway.
Division, Vol. 91.
TSINKER, G. P., 1995. Marine Structures Engi-
OCIMF, 1978. Guidelines and Recommendations
for Safe Mooring of Large Ships at Piers and neering: Specialized Applications. Chapman
& Hali, New York.
Sea Islands. Witherby & Sons, London.
OCIMF, 1992. Mooring Equipment Guidelines. VRIJLING, J. K. and 00STING, J., 1985. "Prob-
Oil Companies International Marine Forum. abilistic Design of Flexible Dolphins." The
Dock & Harbour Authority, September.
PIANC, 1984. "Report of the International Com-
mission for lmproving the Design of Fender WINTERKORN, H. F. and FANG, H-Y, 1991. "Soil
Systems," Supplement to Bulletin No. 45. Technology and Engineering Properties of
PIANC, 1995. "Criteria for Movements ofMoored Soils." Foundation Engineering Handbook,
Ships in Harbours," Supplement to Bulletin 2nd ed. Fang, H-F (ed.). Van Nostrand
Reinhold, New York.
No. 88.
PouLOs, H. G. and DAVIS, E. H., 1974. Elastic WRIGHT, J., CHILD, T. J., REBEL, J. M., and
Solutions for Soil and Rock Mechanics. John ROBERTON, J., 1987. "Port Bonython Jetty,
Wiley & Sons, New York. South Australia." Proceedings Institute Civil
Engineering, Part 1.
QUINN, A DeF., 1972. Design and Construction
of Ports and Marine Structures, 2nd ed. YARON, S. L. and SHIMONI, J., 1982. "The Hadera
McGraw-Hill Book Co., New York. Offshore Coal Unloading Terminal: A New
REESE, L. C. and SULLIVAN, W. R., 1977. Docu- Approach to Unloading of Bulk in the Open
mentation of Computer Program COM624- Sea." Proceedings of Offshore Technology
Ports 1 and Il Analysis of Stresses and De- Conference, Houston, TX.
flections for Laterally Loaded Piles, Including YOUNDALE, J. E. and SHRIVASTAVA, S. M., 1986.
Generation of P-Y Curves. University of "Design and Construction of Marine Facilities
Texas, Austin. for Ridley Island Coal Terminal." Proceedings
916 Offshore Deep Water Terminals

ASCE Specialty Conference PORTS '86, tures." Proceedings 7th Annual Offshore
Oakland, CA. Technology Conference, Vol. 3, Paper No.
YOUNG, A. G., 1991. "Marine Foundation 2371, Houston, TX.
Studies." Handbook of Coastal and Ocean ZONAILO, G. W., 1992. "Bulk Commodity Termi-
Engineering, Vol. 2, Herbich, J. B. (ed.). Gulf nals-Planning for the Future; Competitive
Publishing House, Houston, TX. and EnVironmental Challenges." ASCE Pro-
YoUNG, A. G., :KRAFT, L. M., Jr., and FOCHT, J. A. ceedings Specialty Conference PORTS '92,
Jr., 1975. "Geotechnical Considerations in Seattle, WA.
Foundation Design of Offshore GraVity Struc-
9
Modernization of Existing
Marine Facilities

9.1 INTRODUCTION sign and construction of new cargo handling


terminals, and modernization of existing
A port is a dynamic system, growing and ports. At present, the "ideal" terminal is
changing as the purpose it was constructed one that has sufficient ft.exibility of being
for changes. Nowadays, port authorities and adapted to the different types of cargo han-
terminal operators are increasingly seeking dling systems without substantial changes
ways in which they can adapt existing port to its infrastructure. For this, the modern
infrastructure to meet the changing de- multipurpose marine terminal is usually
mands of their markets. From an engineer- designed for a uniformly distributed load of
ing point of view, the port is a system that 50 kN jm2 with consideration given to
comprises of miscellaneous facilities di- heavily concentrated live loads that are
rected to the economically efficient and safe dependent on the chosen cargo handling
handling of cargo. When the type of cargo or and hauling equipment, which are often
transportation mode changes, the port needs custom-made. In some instances, handling
to be modernized to be effectively adapted special cargos may result in very heavy
to the new updated cargo handling and loads; these loads, however, may be han-
hauling equipment, new types of vessels, dled at specially designated areas, where
ground transportation, and so forth. the waterfront structure is designed to sus-
During the last four decades the call for tain these loads.
more effective port operation stimulated the These term.inals and their equipment
development of new cargo handling meth- must be adequately sized. However, it must
ods and new cargo handling and hauling be kept in mind that the life of the marine
equipment, and also resulted in dramatic facility is rather long and terminals de-
changes in ship size and shape. signed today must be expected to meet de-
The technological advances in transport- mands during their lifetime which cannot
ing commodities by water transportation be foreseen. On the other hand, overdesign-
results in a new approach to planning, de- ing can be very costly. Therefore, new ter-

917
918 Modernization of Existing Marine Facilities

minal planning and design must be prop- older structures typically suffer a deteriora-
erly balanced. tion of their structural components that re-
Many of the existing ports presently in duces their ability to carry loads. Structural
operation worldwide have been built in deterioration depends on material, quality
post-World War II years. Naturally, these of maintenance, exposure to marine envi-
ports and their marine facilities have been ronment, and degree of physical damages
designed to service comparatively small sustained by the structure from ship im-
vessels and are equipped with less sophisti- pacts, operation of cargo handling and haul-
cated and lighter cargo handling equip- ing equipment, and others.
ment. At that time, ports were designed for Normally, the old structure, ifplanned to
a maximum uniformly distributed load be integrated or used in a new facility,
equal to 40 kNjm 2 • Hence, to meet today's should be rehabilitated. This should include
service requirements the load-carrying ca- an evaluation of its actual load capacity.
pacity of these facilities must be carefully This typically consists of an inspection of
reviewed in-service. Furthermore, most of the structure and an engineering evalua-
the older facilities still in service have to tion; where required, the existing struc-
struggle with problems such as those asso- tures must be structurally upgraded.
ciated with insufficient water depth in front This may include either structural repair1
ofthe quay. rehabilitation andjor the addition of new
The spectacular growth of ship sizes, es- structural components to the old structure
pecially in bulk transport, and the develop- that in combination with the existing struc-
ment of new ship types such as container, ture will create a new system able to serve
roll-onjroll-off, car-carriers, and large ferry new types of vessels andjor support new
ships make many of the existing port facili- and heavier cargo handling and hauling
ties obsolete. Furthermore, larger and more equipment. Principles of structural evalua-
powerful ships with increased draft and tion of in-service marine facilities are dis-
therefore less keel and propeller clearance cussed in detail in Tsinker (1995).
approaching the berth with no tug assis- In some instances, the performance of
tance can cause considerable scouring dam- existing soil-retaining structures (e.g., quay
age, especially if the structure is built on an walls or sheet-pile bulkheads) can be im-
erodible foundation; this can undermine the proved by reducing the lateral soil pressure
quay wall base resulting in structural dam- (exerted on them) andjor modifying a soft
age or unacceptable displacements. foundation soil to achieve better bearing
Substantial scour in front of a sheet-pile capacity or increased passive soil pressure.
wall may result in heavy irreversible over- Practica! methods used for the rehabilita-
stress of both sheet piles and the wall an- tion of distressed foundation soils and re-
choring system. For a detailed discussion on ducing lateral soil thrust on soil-retaining
this subject, the reader is referred to structures are given in Tsinker (1995).
Tsinker (1995). Last but not least, where highly powered
To avoid the adverse effects of seafloor vessels, particularly those with side
erosion on the performance of waterfront thrusters, are planned to be used, the ma-
structures, a greater depth must be created. rine structure must be protected from scour.
Basic principles of structure modernization Detailed discussions on scouring effects on
that applies to depth of water increase and seafloors produced by ship propellers and
associated relevant case histories are fur- basic scour protection methods are given in
ther discussed in this chapter. In addition Tsinker (1995).
to stability problems associated with sea- The following two sections are dedicated
floor erosion and increased loadings, the to the discussion of the basic methods used
Modernization ofExisting Marine Facilities 919

for modernization of existing waterfront similar; they basically include the use of a
facilities and some characteristic case different pressure-relieving system, instal-
histories. lation of additional anchor systems, and
strengthening of some structural and foun-
dation components.
Piled structures, such as piled platforms
9.2 MODERNIZATION OF or pressure-relieving platforms, are nor-
MOORING STRUCTURES mally modernized by extending these struc-
tures seaward where deeper water can be
As mentioned in the previous section, old created by dredging or where it exists natu-
waterfront structures usually need to be rally. The other commonly used method is
modemized in order to accommodate ves- to install a sheet-pile wall that permits
sels larger than those they were originally dredging of the seafl.oor in front of the exist-
designed to handle. In most practica! cases ing structure.
this will call for deeper water in front of the A common dilemma faced by port author-
waterfront structure. However, dredging the ities and terminal operators is whether to
seafl.oor in front of the existing wharf may reconstruct (modemize) existing structure
create a serious stability problem for the or to replace the obsolete structure with a
solid-type construction waterfront structure new one. The decision is usually made on
and reclamations behind the quay apron. the hasis of evaluation of the technical and
For example, in the case of a piled struc- economic merits of both altematives. lf the
ture, overdredging will reduce pile penetra- decision is made in favor of a new facility,
tion, thus effectively reducing the piles then the existing one can be demolished
capacity to support vertical loads and to and replaced by a new structure. Alterna-
resist horizontal loads, and in the case of a tively, the obsolete structures may be buried
gravity-type structure, it may undermine behind_or under the completely new marine
the wall by reducing its effective bearing structure.
area. In the case of sheet-pile bulkheads, In general, modernization of the existing
overdredging will reduce a sheet-pile pene- facility is aimed at increasing the depth of
tration, thus effectively reducing passive water in front of the existing wharf, in-
pressure and increasing the effective active creasing the load carrying capacity of the
pressure on a wall. The latter will result in structure, or both. The techniques used to
a greater reaction force in the bulkhead's achieve these goals vary depending on foun-
anchor system. dation geotechnical parameters, type and
These examples indicate that in most physical condition of the existing structure,
cases just simple dredging of a seafl.oor in and environmental and operational condi-
front of the wharf may not be feasible. The tions (e.g., water fl.uctuation, pressure of
solution is commonly found in modernizing ice, type of ship to be served, type of cargo
the existing structure by the addition of handling and hauling equipment, etc).
structural elements that help to mitigate Generally speaking, the physical condi-
the effects of deepening of the seafl.oor. The tion of any structure before modernization
best solution is generally the one that al- must be evaluated, and if needed, adequate
lows utilization of the existing structure as rehabilitation works must be performed.
much as possible. The subject "dock-in-service: evaluation of
Methods used for modernization of soil- load-carrying capacity, repair, and rehabili-
retaining structures, such as quay walls tation" is discussed in detail in Tsinker
and sheet-pile bulkheads, are generally very (1995).
920 Modernization of Existing Marine Facilities

9.2.1 Modernization of The former include soil grouting, use of


Gravity-Type Ouay Walls miscellaneous anchor systems, and use of
piled structures of miscellaneous construc-
The depth of water in front of the gravity- tion; the latter not only considers the use of
type quay wall can be increased by utilizing anchorages of miscellaneous design but also
one of four basic methods depicted in the construction of pressure-relieving sys-
Figure 9-1, and the ability of these walls to tem such as slabs, piled platforms, and re-
carry heavier loads can be increased by one placement of the existing backfill soil with a
of the methods illustrated in Figure 9-2. good quality granular material. Essentially,

-
b

bedrock

'---

5- ·- ~

}__ ~'
~~ 6-
7-"'Y'">~

7----t-----t-----1

Figure 9-1. Modernization of gravity-type quay walls; typical structural arrangements


designed to increase depth of water in front of the quay: (a) Underpinning; (b) base grouting; (c)
installation of large-diameter king piles and short sheet piling; (d) use of piled platform.
1-Gravity-type quay wall; 2-concrete grout; 3-ground anchor; 4-concrete deck; 5-large-
diameter (size) steel king piles; 6-steel sheet piling; 7-steel or concrete piles; 8-
slope protection; 9, 10-old level of seafloor and new dredge line; 11-protection from scour;
12-' post tensioned bar.
Modernization ofExisting Marine Facilities 921

-e 1

Figure 9-2. Modernization of gravity-type quay wall; structural modifications designed to


increase wall's load-carrying capacity (soil pressure relieving systems). (a) Use of ground (rock)
anchors; (b) use of pressure-relievingjanchor slabs; (c) replacement of the backfill soil with
granular fill; (d) use of piled pressure-relieving independent platform; (e, f) use of pressure-
relieving slab. 1-Existing quay wall; 2-new groundjrock anchor; 3-new relievingjanchor
slab supported by backfill; 4-new granular fill; 5-new excavation of existing backfill;
6-new pressure-relieving piled platform; 7-new pressure-relieving slab supported on wall and
piles; 8-new pressure-relieving slab propped against wall; 9-slip surface.

the combination of both techniques can be case of modernization of the existing water-
used as required. front structures. In the latter case, grouting
is used for soil densification in order to
9.2.1.1 soi/ Grouting increase its shear strength and reduce com-
Soil grouting is the efficient technique by pressibility, ground strengthening under
which foundation materials can be impreg- the existing structure to prevent movement
nated under pressure with grout which is during adjacent excavation or pile-driving,
allowed to set. In general, grouting is a soil strengthening to increase lateral load
process in which grout in liquid is pumped resistance of sheet-piling, underpinning of
into the voids of the soil or rock and then the existing quay walls, slope stabilization,
hardens. As a result, the soil and rock are and so forth.
densified and gain additional strength. In marine application, the soil grouting
Grouting is usually used to alleviate dif- is usually accomplished by injecting the ce-
ficult foundation problems and for remedia- ment slurry or sandjcements mortars. The
tion or maintenance of existing foundation; grout may be installed by the conventional
it has also proved to be very effective in the injection grouting method, more commonly
922 Modernization of Existing Marine Facilities

known as grouting, or by high-pressure of cement slurry with the in situ soil. The
grouting known as jet grouting. General in- properties of soilcrete depend on the native
formation on soil grouting as well as a com- soil and the way the process is applied. In
prehensive list of references is given in the United States, jet grouting is being suc-
Caron et al. (1975) and Wintercorn and cessfully implemented in both granular and
Pamukcu (1991). cohesive soils (Tarricone, 1994). For details
As shown in Figures 9-1a and 9-1b, on jet grouting, consult Wintercorn and
dredging of the seafioor in front of a gravity Pamukcu (1991) and Tsinker (1995). A vari-
wall can be accomplished by the installa- ety of practica! examples are given in Pettit
tion of concrete grout under the wall base to and Wooden (1988), Coomber (1987), Prucz
the required depth. The resulting solid et al. (1992), and Tarricone (1994).
foundation material effectively extends the
wall height and safeguards the existing 9.2.1.1.2 Grouts
structure from being undermined by dredg-
ing (or overdredging). To date, cement grouts are the only
In some instances an increase in effectivegrouts that have been used in marine appli-
height of the wall can affect its ability to cations with different grouting techniques.
The advantage of using cementitious grouts
resist the horizontal thrust of soil and over-
turning loads. In this case, the additional is that seawater can be used for the mix,
support to the wall can be obtained by us- whereas freshwater is generally mandatory
ing posttensioned ground -rock anchors or for chemical grouts. Cement grouts vary in
similar systems (Figs. 9-1b and 9-1a). terms of type andjor proportions of Port-
laud cement, pozzolan, admixtures, and
9.2.1.1.1 Jet grouting water1cement ratio.
Chemical grouts are typically used in an
Jet grouting is a rather remarkable tech- injection grouting technique. They vary con-
nique that is now gaining ground in North siderably in terms of their chemical compo-
America. Although severa! techniques of sition and end product (solids or gels). Simi-
jet grouting exist, the process of greatest lar to cement grouts, chemical grouts have
commercial application to date derives been used to grout different types of soils.
from a 1971 Japanese patent granted to The category and type of grout within this
Nakanishi. Over the past 10 years, jet category, which can be used in a specific
grouting has been successfully used in type of soil, depends primarily on its rheo-
North America in a number of diverse ap- logical behavior (the manner in which it
plications such as excavation support, deforms and fiows). The physical character-
ground-water cutoff barrier, scour protec- istics that determine rheological behavior of
tion around bridge piers, and slope stabi- a grout are viscosity, rigidity, and granular
lization. The most common and valuable content. Some grouts are nongranular and
application of jet grouting is to underpin liquid (nonrigid) and maintain the same
existing foundations. viscosity with time until mass polymeriza-
Jet grouting can strengthen soft soils or tion takes place, and because of the water-
create load-bearing structural members by like nature of these grouts, they are well
forming "soilcrete" (soil-cement mixture) suited for grouting sands and silts. For
elements in horizontal, vertical, and inter- detailed information on grouts, the reader
locked rows. The soilcrete column is formed is referred to Caron et al. (1975) and
by complete, hydraulically induced mixing Wintercorn and Pamukcu (1991).
Modemization of Existing Marine Facilities 923

9.2.1.2 use of Miscellaneous a backfill soil. Soil thrust against the wall
Piled Systems for Quay can also be reduced by replacement of a
Wa/1 Modernization poor backfill material with a good quality
granular fiii. (See Fig. 9-2).
Piled systems of miscellaneous designs
are used to increase the depth of water in
front of a gravity-type quay wall. Two most 9.2.1.4 Soi/ Rep/acement
used examples are depicted in Figures 9-1c
The reduction in soil thrust against a
and 9-1d. In the case depicted in Figure wall due to replacement of a poor soil by a
9-1c, the large-diameter (section) king piles,
quality one is basically attributed to smaller
in combination with conventional sheet piles
values of the coeffi.cient of lateral pressure
driven between king piles, provide lateral f rom the new fiii . F or examp1e, rep1acement
support for the original seafloor dredged to of a silty sand with quality rockfill will
the desired level on a seaward side of the
reduce soil lateral pressure by about a fac-
piling. The bottom part between the new tor of 2 (Tsinker, 1995). To be effective, the
sheeting and the existing wall is usually ·
new· fiii must extend far enough beyond the
protected from erosion caused by strong ship slip surface (Fig. 9 _ 2c); if left within a po-
propeller-induced currents. Naturally, these t ent"1aIly uns t a ble s l"1p we d ge, the new fiii
king piles must also be out of reach from a may promote a heavier than original lateral
vessel's bulbous bow and must be strong
thrust against the wall.
enough to resist impacts by the ice floes or
occasional flotsam where relevant. U se of good quality granular material for
The more conventional solution to the soil pressure reduction is most efficient
where this material is available locally. In
problem is construction of a piled platform
general, for a better performance the new
in front of the existing structure as de- fiii s h ould b e pace 1 d from th e e d ge of exca-
picted in Figure 9-1d. The construction se- vation toward the wall. This will allow for
quence in this case is as follows. The seafloor
maximum mobilization of shear forces
is dredged first and the vertical piles only within the fill material.
or vertical piles in combination with batter
piles are installed, as required. Next, the
slope protection system is installed, and 9.2.1.5 use of Pressure-Relieving
finally, the deck is constructed. The new Slabs and Piled
structure can be either independent of the P/atforms for Quay Wa/1
existing one or joined to it in one way or Modernization
another. Naturally, the newly dredged slope
These structures are illustrated in Fig-
must be located a safe distance from the
ures 9-2b through 9-2f. Pressure-relieving
face of the existing structure. Again, it must
slabs and platforms are perhaps the oldest
also be suffi.ciently protected from scour.
techniques used by engineers for reduction
of the lateral thrust against soil-retention
9.2.1.3 Pressure-Relieving structures. For example, in Russia they
Systems have been known and used for more than
Basically these systems include slabs that 130 years (Dubrova, 1959).
may be supported on piles and thus become Being placed behind a soil-retaining
independent of the structure, or completely structure these structure can effectively re-
supported on the structure, or be supported duce the lateral soil thrust against the wall
on one end on the structure while the other and increase the wall sliding and overturn-
end is supported by either piles or rests on ing stability.
924 Modernization of Existing Marine Facilities

9.2.1.5.1 Pressure-relieving s/abs pressure on the foundation (e.g., rectangu-


lar, or close to rectangular shape of bearing
Where water levels permit, for better pressure diagram, as opposed to a conven-
performance these slabs are typically in- tional trapezoidal or triangular diagram). If
stalled at a depth of approximately one- extended far enough beyond the slip sur-
fourth the wall height. They must be of face, the pressure-relieving slab will also
sufficient length to provide for the required act as an anchorage to the wall. In this
shielding effect of lateral soil pressure. latter case the slab may be broken in two or
The mininmm required length of a soil more parts joined together through hinges
pressure-relieving slab is usually consid- to reduce the value of bending moment
ered to be 0.6H for gravity walls, where H (Tsinker, 1995).
is the height of the wall from the dredge
line to the top. As noted earlier, the
pressure-relieving slab may be supported 9.2.1.5.2 Piled p/atforms
by one end on the wall while the other end These platforms placed behind the wall
is supported on back:fill material (Fig. 9-2b), is a very effective technique that is used to
piles (Fig. 9-2e), ora system that is propped reduce the soil pressure against the soil-
against the wall (Fig. 9-2f). In all these retaining structure (Fig. 9-2d); it may
cases the reduction in lateral soil thrust reduce lateral soil thrusts by up to 50%
due to shielding effects of the slab is com- and more. Several practica! examples of
bined with the stabilizing effects of the re- pressure-relieving slabs and piled platforms
action force transmitted from the slab to and calculation methods for the pressure-
the interior face of the wall. relieving systems are given in Tsinker
All of the above systems result in a (1995). They can also be found in Chapters
smaller and better distribution of bearing 4 and 5 of this text.

7 6 2

Figure 9-3. Modernization of pile structures; typical structural arrangements designed to


increase depth of water in front of the wharf: 1-Existing piled, relieving platform; 2-existing
piled platform; 3-new sheeting, which may consist of regular sheet piles or comprise
large-diameter (section) king piles and regular sheet piles driven in between; 4-new piling;
5-new concrete superstructure; 6-new concrete deck; 7-posttensioned bar; 8, 9-old level of
seafioor and new dredge line, respectively; 10-granular fill; 11-protection against scour.
Modernization of Existing Marine Facilities 925

9.2.2 Modernization of As an alternative, the structure can also


Piled Wharves be modernized by installation of a sheet-pile
wall in front of the e:xisting structure as
As pointed out earlier, these structures are indicated in Figure 9-3a. The obvious ad-
typically constructed in the form of an open vantage of this type of construction is its
piled pier and marginal wharf or a piled simplicity. Depending on the new wall
pressure-relieving structure. Details on height andjor presence of a heavy crane or
like loads, the sheeting can be comprised of
piled structures are provided in Chapter 7.
conventional sheet-piling or it can be con-
The problems associated with piled struc-
structed as a composite structure incorpo-
tures and their potential deterioration and
rating large-diameter (section) king piles
repair jrehabilitation are discussed in de-
with conventional sheet piles driven in be-
tail in Tsinker (1995).
tween. The new sheeting can have an inde-
Iflarger vessels are considered to be used
pendent anchor system or where possible,
at the modernized port, then the existing
the existing structure can be used as an-
waterfront structure(s) must be modernized
chorage. The space between this new
to meet the new requirements. These nor-
sheet-piling and the existing structure is
mally include evaluation of the present ca-
usually filled with good quality granular
pacity of existing structures and their abil-
material. The dredging of the seafloor is
ity to resist new loads associated with new, started when the new sheet-piling is com-
potentially heavier material (cargo) han- pleted, properly anchored, and the space
dling and hauling systems. Essentially, if between the new and old structures is filled.
required, the existing structure(s) must be
upgraded accordingly. Details are provided
in Tsinker (1995).
The use of larger vessels at existing 9.2.3 Modernization of
terminals usually requires the deepening Sheet-Pile Bulkheads
of the seafloor in front of the wharf struc-
ture(s). Typically, this problem is solved in In most cases, the modernization of the ex-
one of two ways: extension of the existing isting sheet-pile bulkheads is a complicated
facility to deeper water through construc- task and roquires special attention and a
tion of a new open piled platform in front of thorough knowledge of the nature of the
the existing structure (Fig. 9-3b), or con- interaction between flexible soil-retaining
structing of a sheet-pile barrier in front of structures and soil. Details are given in
existing wharf (Fig. 9-3a). Chapters 4 and 6 and in Tsinker (1995).
The structural extension typically pro- The complexity of the sheet-pile wall-soil
ceeds in a sequence as described earlier. It interaction lies in the fact that the stress-
may be constructed as an independent deflection condition that is inherited by a
structure or integrated with an existing one. flexible wall from previous loadings (over-
In this case, the reinforcement of the exist- loadings), scour or overdredging effects, ma-
ing structure is exposed and extended to terial deterioration, or due to other reasons
the new structure as required; alterna- is practically irreversible.
tively, dowels may be installed at the face If the wall strength must be increased or
of the existing structure into predrilled the seafloor needs to be dredged in front of
holes and extended into the new structure. a sheet-pile wall, one of the methods illus-
Naturally, the new composite structure trated in Figure 9-4 can be considered.
should be analyzed as a single integrated These methods include the installation
piled system. of an additional anchor system(s) to re-
926 Modernization of Existing Marine Facilities

---\- -- --
3 1
1
/'-11
1
13 /~+cJ>/2

d f

9 / 10
Il

,_ li

l----1f--7
__ 12
L_
-
13

14

F~gure 9-4. Modernization of anchored sheet-pile bulkhead; typical structural arrangements


designed to increase depth of water in front of the wall. (a) Use of ground (rock) anchors; (b)
installation of additional tieback; (c) replacement of existing backfill by quality granular
material; (d) adding pressure-relieving system; (e) installation of new sheet-piling with partial
use of some components from the old system; (f) use of pressure-relieving (anchor) slabs.
1-Existing components of sheet-pile bulkhead: sheeting and anchor system; 2-new ground
(rock) anchor; 3-new tie-rod; 4-new piles driven in front of existing anchor wall; 5-quality
granular fill; 6-excavation; 7-new piles; 8-new pressure-relieving superstructure; 9-added
piece of anchor rod; 10-new pressure-relieving (anchor) slab; 11-slip surface; 12, 13-old level
of seafloor and new dredge line, respectively; 14-new sheeting.

duce the wall span (Figs. 9-4a and 9-4b), designed to reduced the effective span of
soil replacement (Fig. 9-4c), use of soil the existing sheet-piling, should be in-
pressure-relieving systems (Figs. 9-4d and stalled below the existing system and there-
9-4f), and the installation of brand new fore underwater (see Figs. 9-4a and 9-4b).
sheeting with partial or full utilization of This necessitates the use of special under-
components from the existing wall (Fig. water techniques or installation of a
9-4e). portable cofferdam around the working area
in order to carry out the work in dry
9.2.3.1 lnstallation of Additional conditions.
Anchors Most recently, cofferdams of this type
Because the original anchors are usually have been used in the United Kingdom and
placed very close to the mean water level, in Canada (Tsinker, 1995). To be effective,
the additional anchors, which are usually the new anchors must be tensioned in order
Modernization ofExisting Marine Facilities 927

to prevent or limit defiection of the sheet 3. The borehole is drilled through the fill by
pile under the design live load. Because employing a casing pipe (diameter of
anchor tensioning cannot significantly re- 100-150 mm depending on the diameter
duce the overall defiection of a sheet-pile ofthe tie-rod). The boring process is aided
wall, it also cannot reduce the stresses al- by a special electronic sensor system that
ensures an extremely accurately guided
ready inherited by this wall from previously
boring process. Cutting a hole through
applied loads.
the sheet-piling is allowed only if the sur-
Additional anchors are usually designed vey indicates that the deviation of the
either in the form of a ground (rock) anchor borehole from a designed position is
drilled through the sheeting into the back- within an acceptable range. After cutting
fill, or in the form of a conventional tieback the hole through the sheeting the boring
rod secured at either the existing (some- machine is shifted to a new position.
times strengthened) anchorages (e.g., con-
4. The new tie-rod is installed and secured
crete deadman, anchor or plates, or piled by divers at the new wale, and the rod is
structures of miscellaneous construction), or tensioned at the anchor block location.
at a brand new structure located some dis- The magnitude of the tension load ap-
tance behind the existing ones. plied is predetermined by design compu-
If the existing anchor block or anchor tation. Tensioning of the lower tie-rod
systems cannot safely resist the combined . system alone cannot significantly affect
load from both existing and new anchor the existing tension load in the upper
rods, it must be reinforced. This can be anchor rods. However, even the slightest
done in a variety of ways; for example, piles backward movement of the sheeting in-
can be driven in front of the anchor blocks, duced by tensioning of the lower anchors
regular soil in front of anchors can be re- can change the tension load at the upper
placed with well compacted rock fill, and so tie-rods. Therefore, it is imperative to keep
tension loads at both levels under control.
on. N aturally, new tie-rods are installed
from the water side of the wall. 5. Finally, the space between the new an-
Ishiguro and Miyata (1988) published a chor and the casing pipe is grouted from
report on new technology developed in the water side upward, and the previ-
Japan for underwater installation of addi- ously excavated material is backfilled to
tional anchor tie-rods. This technology al- its original elevation. Grouting is usually
lows for installation of new tie-rods through used to provide tie-rods with corrosion
protection and to prevent the possible
an existing or new deadman and backfill
loosening of backfill material after corro-
toward sheet piles, and then at desired lo- sion of the casing pipe.
cations through sheet piles. The new rod
installation proceeds in the following se-
quence: The obvious disadvantage of installation of
the additional anchors is that it requires
1. The existing backfill is excavated just be- use of the external wale(s) for distributing
hind the existing anchor block. The width the anchor force among individual sheet
of the excavation should be large enough piles. This wale system may be easily dam-
to accommodate a boring machine. aged by the ship if not properly protected.
2. Precise survey of the sheet-piling is done On the other hand, it may also pierce holes
and the boring machine is set up in the
in the ship hall. Where an external wale(s)
exact position for the drilling operation.
To avoid deviation of the drill from both is used, the face of the wall superstructure
vertical and horizontal directions, the (capping) must be extended far enough sea-
boring machine has to be firmly fixed in ward to prevent dangerous contact between
the predetermined position. the wall and a ship.
928 Modernization of Existing Marine Facilities

As noted earlier, before attempting any stresses in the sheet-piling, it should be


modernization the existing sheet-pile bulk- moved slightly away from the fiU.
head must be repairedjrehabilitated as re- It must be noted that the system in ques-
quired. Details are given in Tsinker (1995). tion is only appropriate when the sheet-pile
system can carry some verticalload; in gen-
9.2.3.2 Soi/ Rep/acement eral, a sheet-pile bulkhead comprised of
conventional sheet piles are designed to re-
As discussed in the previous section, the sist lateral soil thrust only. However, those
effect of soil replacement on a sheet-pile that include large-diameter (section) king
bulkhead is less pronounced than it may be piles can carry substantial verticalloads in
in the case of a gravity-type wall. As pointed combination with lateral soil thrust.
out earlier, the sheeting cannot rebound at It must be remembered that pile-driving
its middle section (location of maximum in close proximity to the existing sheeting
stress) after excavation is done in the area, can be detrimental to this existing system.
as is shown in Figure 9-4c. Hence, only the Hence, if these piles are required, they can
upper part of the wall will change its curva- be installed in predrilled boreholes. Alter-
ture after excavation of the indicated part natively, low-displacement piles (e.g., H-
of the existing fill is done. Some stress re- piles or open-ended pipe piles) or bored piles
duction in sheet-piling can be achieved by
can be used. For a discussion on this sub-
distressing the anchor rod to allow for some
ject, the reader is referred to Chapter 6. A
slight wall movement away from the fill.
relevant practica! example is discussed by
Practically, this may be accomplished by
Barker and Chattaway (1990).
loosening the nut at the anchor block or by
using the excavated turnbuckles.
Porter (1986) reports a successful sheet- 9.2.3.3.2 Pressure-relieving s/abs
pile bulkhead repair by using lightweight
When used in sheet-pile bulkheads, the
expanded shale for partial replacement of
slabs have to have a minimum length equal
the existing fill.
to no less than 0.7H, where H is the height
from the dredge line to the wall top. This is
9.2.3.3 use of Pressure-Relieving required to ensure that this slab will inter-
Systems sect a slip surface behind the wall and,
In a manner similar to gravity-type walls, therefore, will have support from a stable
these structures can be either piled plat- portion of the backfill at its rear end. These
forms or slabs of miscellaneous designs slabs are hinged to the sheet-pile system
(Figs. 9-4d and 9-4f). and therefore transmit a substantial verti-
calload to the sheeting.
Hence, the new structural system that
9.2.3.3.1 Piled platforms incorporates the pressure-relieving system
In the case of sheet-pile bulkheads these (e.g., platform or slab) must be reviewed for
platforms can be installed next to the wall the combined effect of a lateral soil thrust
or incorporated into the sheet-pile system. smaller than in the original design com-
The latter is depicted in Figure 9-4. In this bined with a substantial verticalload.
case, the pressure-relieving slab is sup- Naturally, sheet-pile penetration into the
ported on new bearing piles and in part on foundation soil must be adequate to resist
the sheet piling. Anchorage for the new sys- the vertical design load. Design methods
tem is stiU provided by the existing anchor and some practica! examples are given in
system. Again, to relieve the existing Tsinker (1995).
Modernization ofExisting Marine Facilities 929

9.2.3.3.3 New sheet-piling that the tie-rods, with the exception of 1-1.5
m immediately adjacent to the wall and
If the existing sheet-piling has deterio- concrete anchor blocks, could be reused.
rated to the extent that it cannot be eco- The terms of reference for dock modern-
nomically restored to its original capacity, a ization called for an increase in the wall
new sheet-piling system, in lieu of the old height by 1.2 m. The detailed engineering
one, is usually considered. Dilapidation of evaluation of the existing structure brought
the structure in the marine environment is designers to the conclusion that the new
most severe in the splash zone. This is usu- sheeting installed in front of the existing
ally the area where a great deal of steel is wall in combination with the existing one
lost to corrosion and affects not only the below the mean water level, and the exist-
sheet-piling but also the wale that is a vital ing anchor system extended seaward by
structural component of a sheet-pile bulk- about 2.0 m could successfully resist the
head. new increased live loads and the heavier
The properly engineered modernization soil active pressure.
of a badly deteriorated sheet-pile bulkhead
must consider the maximum use of the ex-
isting material that is still in good con-
dition. This is schematically depicted in 9.2.4 New Wall Construction
Figure 9-4e. The practica! example of this
kind of sheet-pile bulkhead restoration is In some instances, port modernization re-
discussed by Tsinker (1995). In this latter quires not only a deepening of the existing
case, the steel sheet-pile bulkhead, 12.7 m wharf but also an expansion of its working
high, anchored by steel tie-rods, and se- area. In this case, if the available water
cured at the concrete anchor block, was area is sufficiently large, then the new
badly deteriorated, basically in the splash waterfront structure is constructed some
zone. During underwater inspection of this distances ahead of the existing one; the new
structure it was found that the sheet piles working area is created by filling the space
below the minimum mean water level were between the new and the old structures
in good condition. It was also concluded with an appropriate fiU (Fig. 9-5a and 9-6).

Figure 9-5. U se of new structures for modernization of an existing wharf: (a) new quay placed
ahead of the old structure; (b) construction of new structure behind the existing wall.
1-Existing wall; 2-new structure; 3-dredged bottom; 4-new backfill; 5-excavation for new
wall construction.
930 Modernization of Existing Marine Facilities

6 6
- -':..... - - 1--+--t--+-t---i
7 7 4

Figure 9-6. Modernization of obsolete marine facilities: (a) by installation of new piled
structure; (b) by installation of new sheet-piling system. 1-Existing wharf; 2-new pressure-
relieving platform; 3-new sheet-piling system; 4-anchor system: ground (rock) anchors or
the like; 5-granular fiii; 6, 7-old level of seafioor and new dredge line, respectively.

On the other hand, if the available water gravity-type massive wall it was decided to
area is limited or needs to be enlarged to implement a very high-pressure grouting
allow the operation of larger vessels, then technique (VHP) for reinforcing the founda-
the new structures can be constructed just tion soils. (Ponnet et al., 1992). The princi-
behind the existing one (Fig. 9-5b). In this ple of this technique is very similar to the
case the existing structure may be used as jet grouting method that was discussed
a cofferdam or temporary working platform earlier. The VHP process is carried out in
for new construction. The new structures the following sequence:
are usually built by employing the "quiet"
techniques (e.g., slurry walls, boring piles, 1. A borehole is drilled through the wall and
ground or rock anchors, etc.). This is re- underlain foundation soil to the required
quired to avoid potential damages to the old depth.
structures. After the structure is built, the
2. A grout pipe is installed through the hole.
old one is dismantled and the seafloor is
dredged to the required depth. 3. A cement grout is injected horizontally
under very high pressure through the
nozzle(s) located at the bottom part of the
grout pipe while the pipe is rotating and
9.3 MODERNIZATION OF lifting.
WATERFRONT STRUCTURES:
CHARACTERISTIC EXAMPLES Due to very high pressure (up to 800 bars),
the grout jet completely destroys the natu-
ral matrix of the soil which, in the process,
mixes with the injected grout. After harden-
9.3.1 Gravity-Type Ouay Walls
ing, a homogeneous soilcrete column is ob-
t ained. The compressive strength of the
9.3.1.1 Port-of-Antwerp- Pilot
cured soil grout mix is about 10-15 N jmm2 •
Project
The column can be reinforced by a steel bar
As a part of a massive modernization of inserted into the column before hardening,
the Port-of-Antwerp the seafloor in front of provided that the obtained soil- grout mix
some existing marine facilities had to be contains no large and hard inclusions (e.g.,
substantially dredged. In the case of a stones, large size debris, etc.).
Modernization of Existing Marine Facilities 931

VHP piles can be installed next to each stallation of ground anchors angled from
other, thus forming a continuous wall. To just above mean sea level at 3 ro center-to-
confirm the effectiveness of the VHP method center toward the fiU. The service capacity
for the previously mentioned quay wall of these anchors is 650 kN. To ensure relia-
modernization, a decision was roade to first bility and long-term performance, the an-
carry out a pilot project for which berth No. chors have double corrosion protection. The
170 at the Third Harbor dock was selected new wall stability was verified for the soil
(Fig. 9-7). and water lateral thrust and for a uni-
The quay wall there, 14.0 ro high from formly distributed surcharge load of 60
the seafloor level, was constructed from two kNjm 2 .
materials; the bottom part was built from Construction was executed in the follow-
solid unreinforced concrete, and the upper ing steps:
part was of a brickwork construction. Be-
fore modernization, the depth of water at 1. Boreholes were drilled through the wall
the berth face was 11.7 ro; the project speci- and their accuracy verified by inclinome-
fied deepening it by 3.0 ro. The totallength ters connected to a computer.
of modernized wall was 40 ro. 2. Ground anchors were installed at 35° to
The double row of soilcrete columns that horizontal and tensioned. Each anchor
have been installed at the wall face have an was tested to 1.5 times the design load.
inclination of 4° toward the sea and extends 3. A double row of longitudinal soilcrete
down to about 3.5 ro below the new dredged columns were installed at the wall face; a
bottom level. Additionally, transverse walls compressive resistance of these columns
of 10-15 Njmm 2 was achieved. Forty-
formed from similar soilcrete columns were
millimeter-diameter steel bars were in-
installed at 6.0 ro center-to-center along the serted through the wall into the soilcrete
wall. This wall (diaphragm) provided addi- columns and grouted within the body of
tional support to the wall and helps to re- the wall. This helped to reinforce both
sist shear loads. parts of the wall (e.g., mass concrete and
Additionally, the new wall sliding and masonry) and integrated the existing wall
overturning stability was enhanced by in- with the new foundation.

Figure 9-7. Modernization of gravity quay wall at Port-of-Antwerp-pilot


project: (a) Existing wall; (b) modernized wall. 1-Existing quay wall; 2-
modernized wall; 3-longitudinal double-column soilcrete wall; 4-traversal at
6.0-m center-to-center one-column soilcrete wall; 5-post tensioned ground
anchor at 3.0 m center-to-center; 6-blasted offtoe; 7, 8-old and new bottoms,
respectively.
932 Modernization of Existing Marine Facilities

4. Transverse soilcrete diaphragms were this grout from spreading beyond the wall's
also installed. face, the impermeable screen was installed.
5. The wall toe was blasted off. This was The wall modernization process pro-
accomplished by explosives installed into ceeded in the following sequence:
predrilled splitting system.
6. Finally, the blasted-off toe was removed 1. First, the impermeable screen, immedi-
and the bottom was dredged to the re- ately in front of the toe of the wall, was
quired depth. installed. In was extended vertically
through the rubble mattress and quay
apron to ensure that the material from
9.3.1.2 Port-of-Larnaca, the subsequent grouting of the rubble
Cyprus- Deepening mattress would not permeate into the ma-
of the Existing terial tobe dredged later. The grout was a
Oravity Quay Wa/1 50 : 50 mix of bentonite and cement pro-
ducing a strength ofonly 1-2 Njmm 2 . It
This is discussed by Hallett et al. (1992). was injected into two overlapping rows of
The wall in question (Fig. 9-8a) is of mass holes drilled in the wall apron and rubble
concrete blockwork construction, built using foundation to several meters below the
the inclined slice technique. The wall is design dredged level (Fig. 9-Sb). A fixed
founded on 2-3-m-thick rubble foundation volume of grout was pumped into each
hole, and each panel length of 6 m was
and has a massive concrete superstructure.
inspected by a diver to ensure that group
The front face of the wall is inclined at emerged at the surface. A drilling guide
1 : 20 from the vertical. The depth of water template was used to ensure accuracy of
in front of the wall was 10 m, and by the drilling.
terms of reference, it was to be increased by 2. Next, the rubble mattress was grouted
2m. (Fig. 9-Sc). For this, the wall was drilled
The natural foundation material was vertically and cement grout was injected
comprised of dense to loose fine sands with from a level nominally at the bottom of
some silts and clays, and occasionally traces the rubble mattress up. It continued until
of organic matter. Some cemented sand- the diver noticed that the grout flowed
stone is also present. This natural stratum out above the impermeable screen, previ-
ously installed, at the toe of the wall.
is underlain by a mari stratum some 4-6 m
In the process it was found that fine
below the underside of the wall rubble sand had filled the lower part of the rub-
foundation. ble mattress, especially at the front of the
An engineering evaluation of the existing wall; it was assumed that this fine mate-
structure was carried out and indicated that rial may inhibit the flow of grout. In prac-
significantly increased toe loads, resulting tice, however, this fine material did not
from the water depth increase, could be inhibit, but quite to the contrary, encour-
transferred directly to the bedrock stratum aged the formation of the grouted
by underpinning the wall in one way or columns. It was accepted that grouting of
the open rubble, coupled with column for-
another. In the final analysis the very high
mation in the areas filled with sand, would
pressure (VHP) grouting technique, similar be structurally adequate and .sound.
to that described in the previous section,
3. Mattress solidification was followed by in-
was used to create the interlocking soilcrete stallation of soilcrete columns extending
columns to underpin the wall. However, to from bedrock to the underside of the
make these columns effective, it was con- solidified rubble mattress (Fig. 9-Sd). Ac-
cluded that the rubble mattress must be cess to the foundation soil located be-
solidified by cement grout first. To prevent tween the bedrock and mattress was
b ~
.!

2
.!! .!
1.4 6.3
M.L.W. .-
-;:.--
1
M
cd

~
~
o
p..
M
cd 8;:;·
·.
~
o
.. 1 1
::s
? 1 1 s,
1 1
_, tr:1
1 ~.
1
[ .' ,.
"'c:
~
bedroct 7 s::
1:0
;J.
~
"%j
~
~
o-.
CD
Figure 9-8. Port-of-Larnaka, Cyprus-deepening the existing quay wall: (a) Existing quay wall; (b) installation of bentonitej
cement grout; (c) rubble mattress cement-grouted behind bentonitej cement screen; (d) completely modernized wall; (e) plan of
"'
soilcrete longitudinal and transverse walls and rock anchors (not to se al). 1-Existing blockwork quay wall; 2-rubble mattress;
3-apron; 4-bentonitej cement screen; 5-cement-grouted rubble mattress; 6-soilcrete column; 7-50-mm-diameter rock anchor; ~
~
8-outline of the wall base. ~
934 Modernization of Existing Marine Facilities

achieved by drilling through the wall and As noted by Hallet et al. (1992), the con-
wall mattress. The holes were generally struction process was carefully monitored
drilled at 0.7 m centers with the objective and a number of field tests performed. Ad-
that the complete soilcrete column would ditionally, the contractor was able to derive
have a diameter of approximately 1 m much confirmatory information from the
and therefore overlap to form continuous operations of drilling and grouting in the
wall(s). The soilcrete columns were formed form of resistance to process, grout use, and
by ejecting cement grout at pressure of
so on. For example, grouting pressure was
about 400 bars, similar to that described
varied from a maximum of 400 bars to 200
in the previous section. The column diam-
eter was controlled by the speed of rota-
bars near the seabed and in the rubble
tion and the speed of extraction of the foundations; the speed of rotation and lift
grout pipe and by jet pressure. The col- during column formation were also con-
umn formation was continued upward trolled by roethods available to the
through the rubble mattress and 1 m contractor.
above the underside of the wall. Finally, a Finally, it should be pointed out that this
reinforcing bar, 22 mm in diameter, was project involved over 10 miles of drilling
placed in each column extending up into through concrete, 5 miles of VHP columns,
the body of the wall; subsequently, the 3 roiles of grout screen columns, and 4 roiles
hole was grouted to surface level. The of rock anchors. The project was devised,
design value of compressive strength of tendered, and launched on site in 7 weeks.
soilcrete (8 Njmm2 ) was achieved. In It was sufficiently completed the receive the
practice, the real strength of soilcrete first large vessel about 10 weeks later, and
columns is somewhat greater than 8 totally completed in about 6 months.
Njmm 2 •
The layout of soilcrete longitudinal and
transverse walls are shown in Figure 9.3.1.3 Port-of-Antwerp-
9-8e. Renovation of the North
4. Finally, 50-mm-diameter rock anchors ouay of Third Harbor
were installed at the rear face of the wall. Dock
They were installed through the rear soil- This is discussed by Thibant et al. (1992).
crete columns in transverse walls and A master plan adopted by the Port Author-
through two separate columns located be-
ity for renovation of the Port-of-Antwerp
tween the transverse walls as indicated
considered massive modernization of water-
in Figure 9-8e. These rock anchors were
ways and port infrastructure. It included
extended 7 m into the sound marl stra-
tum and grouted there. deepening the seafloor in front of the major
wharves presently in operation, which al-
5. The quay modernization was completed lows mooring of Panamax-size ships having
by dredging the seafloor to the designed drafts of 12.20 ro.
depth. Dredging in the vicinity of the wall The North Quay of the Third Harbor
was accomplished by using a large back-
Dock is among the major facilities of the
hoe dredger mounted on a jacked-up
port that underwent substantial modern-
barge. The rest of the harbor basin was
dredged by a trailing suction hopper ization. This quay is 1140 ro long and its
dredger. mooring wall is of a gravity-type construc-
Last but not least, the wall was fur- tion (Fig. 9-9). The total height of this wall
nished with the fender system designed is 16.3 ro; the lower part (7.5 ro high) is of
to absorb energy of a berthing ship and to mass concrete construction and the upper
provide sufficient clearance of the ship's part (8.7 ro high) consists ofbrickwork. The
hull at the toe of the wall. original depth of water in front of the
Modernization of Existing Marine Facilities 935

started thereafter. The concept of wall mod-


ernization included the driving of large-
diameter king piles, the installation of con-
ventional sheet piles between the king piles,
and the construction of a concrete deck to
bridge the space between king piles and the
existing wall in order to provide a platform
to support the gantry crane, single train
track, and paved road. The required depth
ofwater of14.7 m was obtained by dredging
the seafloor in front of the new sheeting to
the required elevation. The seafloor be-
tween the sheeting and the wall was pro-
Figure 9-9. Port-of-Antwerp-modernization of tected by a fascial mattress. Finally, ground
N orth Quay of Third Harbor Dock. 1-Existing wall, anchors were installed to resist the horizon-
the bottom part of the wall is of mass concrete con- tal pull created by mooring forces. A typical
struction and its upper part is of masonry construc- cross section of a modernized quay wall is
tion; 2-new concrete-filled, open-ended steel pipe
illustrated in Figure 9-9.
piles, d = 1016 mm and L = 24.9 m; 3-steel sheet-
piling; 4-prefabricated concrete slabs 10.63 X 2.86 X The third phase-renovation of the re-
0.2 m; 5-prefabricated concrete panels 2.6 X 2.8 X 0.5 maining 311 m of the wall-will be carried
m; 6-cast-in-situ concrete deck; 7-sand ballast; out after execution of the second phase of
8-bottom protection; 9-ground anchor; 10-existing wall modernization.
bottom; 11-new dredged seafioor.
Existing quav wa/1
North Quay is 11.5 m. In order to provide As mentioned earlier, the wall is of a
sufficient underkeel clearance for new de- massive composite construction. Wall sta-
sign vessels, it must be deepened down to bility is derived from its mass, which en-
14.7m. sures against sliding and overturning sta-
New design criteria for wall moderniza- bility. Because the new concrete deck is
tion speci:fied the following load conditions: joined with the existing wall by steel dow-
els, it can transfer horizontal thrusts to-
• Uniformly distributed surcharge load of 60 ward the land and in the opposite direction;
kNjm 2 the former is due to ship impact load and
• Crane railload of 320 kNjm on both rails the latter is generated by mooring forces.
of the crane track
To resist the former load, the masonry part
• Train track load of 50 kNjm
of the wall was reinforced by steel rebars
• Fender reaction force of 1.1 MN
• Mooring force of 1.0 MN
(diameter 20 mm) drilled into the wall to
the depth of 3.5 m. To increase the effec-
tiveness of these rebars they were installed
In order to keep interruptions in dock oper-
at 40° to 45° to vertical. The horizontal
ations to the possible minimum and at the
component of mooring load is resisted by
same time carry out modernization work
rock anchors.
with maximum efficiency, the whole process
of quay renovation was split in three execu-
tion phases; the renovation of the first part
King piles and sheet pi/es
ofthe 369-m-long wall (:first phase) began in Steel open-ended piles, 1016 mm in di-
1990 and completed in 1992. The second ameter, 24.9 m long, with a wall thickness
phase, the renovation of the next 460 m 1 of 14 mm, were installed at approximately
936 Modernization of Existing Marine Facilities

2.8-m centers in front of the existing wall, The deck .was completed with granular
as indicated in Figure 9-9. These piles were ballast, crane, and train tracks. The tracks
installed by a powerful hydraulic vibrator rest on concrete beams that project up from
and driven several meters to their final set the deck. Similar to the existing wall, the
by a diesel hammer. 1\fter installation, the deck structure was built in units each 22.96
soil plug was cleaned out of the pile's inte- m long. Each deck unit is equipped with one
rior and the interior was filled with con- 1000-kN-capacity bollard and two pieces of
crete. Protection of these piles from corro- high-energy absorption rubber fenders
sion was achieved by adding 3 mm of thick- mounted on the front panels.
ness to the pile wall. A set of three conven-
tional sheet piles, 9 m long with a total Rock anchors
width of approximately 1.8 m, were then Every deck unit is provided with four
driven between the king piles. ground anchors that are designed to resist
the horizontal component of the mooring
concrete deck load. These anchors are drilled at a 35°
angle to the horizontal into the wall backfill
This has been constructed as a composite material and beyond into a stable stratum.
structure comprised of precast concrete Each anchor is designed to sustain a work-
components and cast-in-situ concrete. The ing load of 57.5 tonnes and has sufficient
deck was constructed in the following se- protection from corrosion.
quence. First, precast concrete panels
10.62 X 2.86 m, 0.2 m, thick and weighing Bottom protection between the
15.0 tonnes each were installed to bridge new sheeting and the existing
the space between the king piles and the wa/1
wall. Next, precast front (fender) panels This protection was provided by fascine
2.6 X 2.8 m, 0.5 m thick, and weighing 9.5 mattresses in which conventional stone
tonnes each were installed and joined with riprap has been substituted for selected
previously installed flat horizontal panels. concrete debris (weight 5-80 kg) obtained
The front panels were carefully aligned with from the demolished quay wall. The joints
the help of special bolts. This resulted in a between the mattress and the existing wall,
perfectly straight quay wall face. Thus, all and the new sheet-piling were sealed and
concrete surfaces exposed to the marine en- reinforced by the installation of tremie
vironment were constructed from good concrete.
quality prefabricated concrete components.
For better performance of precast con- 9.3.1.4 Port-of-Vancouver,
crete components in the marine environ- British Columbia-
ment, steel fibers 35 kgjm 3 were added to Modernization of
the concrete mix. Subsequently, the 15-cm- Existing Grain Shipping
thick layer of concrete was installed on top Jetty
of the precast horizontal panels; this re- As discussed by J ensen and Siemons
sulted in the formation of a 35-cm-thick (1986) and Birkeland (1989) the original
composite slab able to withstand loads cre- grain shipping jetty at Port-of-Vancouver
ated by the weight of steel reinforcement, was constructed in 1924 and repaired and
the fresh concrete, the construction equip- reconstructed in 1971. It was constructed in
ment used for construction of the rest of the a form of a pier comprised of four concrete
deck. Once complete, the concrete works dolphins linked with each other and a
were carried out. causeway by the railway bridges. The re-
Modernization of Existing Marine Facilities 937

pair included rehabilitation work on all four loss of supporting material, both resulting
original concrete caissons and replacement in a reduction, or complete loss of sliding
of the original timber railway bridges be- and overturning stability of the structure.
tween the caissons with concrete box gird- Also, ali of the above may result in a
ers supported on caissons and concrete pile drastic increase in bearing pressure under
bents. A generallayout of the jetty, its ele- the crib base that may collapse the crib
vation, and typical cross section are de- bottom slab under one or more rows of cells,
picted in Figure 9-10. which ultimately may lead to a complete
The original jetty can accommodate structural failure. To prevent this from hap-
30,000-DWT vessels; it was recently up- pening it was decided to grout the granular
graded to handle ships up to 70,000 DWT. material under two caissons as is shown in
For this its length and depth, one side of Figure 9-10a. Cement grout was chosen by
the pier had to be increased. Modifications the contractor for the job and two trial
to the ship loading equipment in the gallery grouting sessions with subsequent testing
as well as new loading spouts were also were performed. As a result, it was con-
required. cluded that a measure should be adopted to
The main problem with dock modifica- restrict the grout travel beyond the cais-
tion was the need to deepen the existing sons. It was also decided that for better
water level on one side of the jetty from results the well-grouted zone was to be ex-
10.1 m to 13.7 m. This required a consider- tended from the base of the caisson to the
able amount of rock drilling and blasting in bottom of the granular material, and from
close proximity to the structure, as well as the face of the caisson and back under the
dredging some amount of granular mate- caisson a minimum of 4.5 m. Subsequently,
rial. The most severely expected adverse steel sheet-piling at the caisson face and to
effects of these works were the loss of some distance around its sides was in-
ground support near the crib's edge and stalled; it was intended to be temporary

./'
(
50000dwt
l
. . __ -- ---- --==~ :-7___ ----------
~

Figure 9-10. Port-of-Vancouver-modernization of grain-shipping jetty to accommodate a


70,000-DWT ship. (a) Elevation; (b) plan; (c) typical cross section through caisson. 1-New piled
dolphin; 2-e.xisting concrete caisson 17.1 X 15.3 in cross section and 15.0 m high; 3-new box
girder train bridge; 4-approach jetty; 5-grouted granular soil; 6-till underlaying granular
soil; 7 -sandstone and conglomerate; 8-e.xisting bottom; 9-dredged seafloor.
938 Modernization of Existing Marine Facilities

only and tobe removed when the dredging 9.3.1.5 Port-of-Providence,


operation was completed. Postgrouting in- Rhode /sland-
spection revealed that the sheet piles at the Deepening the Municipal
caisson base were about 1.2 m out from the Wharf by Underpinning
caisson face; the acceptable distance, how-
As discussed by Calabretta and Sullivan
ever, was estimated tobe 0.6 m. This was
(1980) and Pierce and Calabretta (1986) this
required in order to prevent direct contact
wharf is comprised of approximately 1100
between the ship hull and any projecting
m of a gravity-type continuous quay wall.
elements located beyond the face ofthe 55% The wall is a classic example of a turn-of-
compressed fender units.
the-century granite masonry construction
The steel sheet piles therefore had to be similar to those constructed throughout the
removed and the grouted material project- world and described in Chapter 5. It has
ing beyond 0.6 m had to be cut back. The been used as a terminal for break bulk,
excess material was "chiseled" off with a bulk, and LPG cargos. In the mid-1970s the
pair of steel sheet piles welded together and city of Providence undertook a major port
reinforced by welded-on high-strength steel modernization program to update its facili-
plates; at two locations the overgrouted ma- ties in order to accommodate modern trends
terial was removed by drilli:hg and blasting. in maritime commerce and provide ex-
All ofthe above, however, resulted in a very panded services for larger, deeper draft
rough face of grouted material which, in ships. In order to accommodate these ves-
general, consisted of relatively thick well- sels, the bottom of the berths, in general,
grouted hard layers interspersed with thin- had to be dredged from approximately 9.9
ner weakly cemented sandy layers. m to 12.0 m at a mean low water level
At the corner of one caisson, the founda- (MLW); at berth No. 3, the dredged depth
tion material was missing over a distance of was reduced to 10.7 m. This dredge, how-
approximately 2.4 m measured from the ever, would result in an undermining ofthe
corner. After a review of different alterna- existing wall and the disturbance of the
tives of how to better improve the quality of soil-bearing stratum. In this sense, the wall
the grouted bases, the contractor selected to at berth No. 3 represented the greatest con-
reinstall the steel sheet-piling, this time cern. This wall, which was constructed in
vertical and flush with the face of the cais- the late 1800s, is of considerably smaller
son and wrapped around the caisson cor- geometry than at other berths. Wall mod-
ners. Sheet piles were cut off just below the ernization basically included installation of
fenders and bolted to the caisson face. Then steel sheet piles at the wall toe and con-
the space between the sheeting and the struction of the underpinning system be-
caisson was filled with tremie concrete. At neath the wall base by drilling through the
final observation the caissons behavior show wall and installing small-diameter grouted
no displacement either horizontally or piles (Fig. 9-11). This procedure is quite
vertically. common in Europe (Bray and Tatham,
In conclusion, it should be mentioned that 1992); however, it is seldom used in North
a very similar strengthening exercise of a America.
granular base under a gravity-type dolphin The wall in question consists of 17
(caisson) has been successfully carried out courses of interlocking granite blocks that
during rehabilitation of a major bulk mate- are about 0.8 m thick. The wall's base is
rial loading facility in Newfoundland, placed on an inboard slope of approxi-
Canada (Tsinker, 1994). mately, 1.5: 12 and a similar slope is ob-
Modernization of Existing Marine Facilities 939

Detailed engineering evaluation of the


wall was carried out and a construction
technology was developed and eventually
successfully implemented. It included in-
stallation of a steel sheet-pile curtain wall
at the toe of the wall, dredging the trench to
expose the wall toe, installation of grout
anchors to connect the sheet-piling to the
wall, installation of a concrete plug to seal
the space between the sheeting and the
wall, dredging the seafloor in front of the
wall to the required depth, and, :finally,
dredging the bottom in front of the sheet-
piling by an additional 1.2 m and installa-
tion of a riprap plug into the excavated
void.
As the work progressed, the preestab-
w s ..,:;-7 lished survey program for monitoring wall
movement revealed considerable outboard
Figure 9-11. Port-of-Providence, Rhode movement of the wall. This movement was
Island-Modernization of Municipal WharfNo. 3.
1-Existing granite blockwork wall; 2-old fender
apparently caused by a disturbance of the
piles; 3-existed ground surface; 4-temporary (inter- foundation soil under the toe of the wall
mediate) rubble berm; 5-new steel sheet-piling; and the effects of vibration caused by the
6-scour protection; 7-steel dowel; 8-concrete plug; percussion drills; the wall movements
9-grout-filled stitching hole; 10-underpinning piles; stopped when construction was completed.
11-miscellaneous fill. [From Pierce and Calabretta
(1986).)
The total horizontal and vertical move-
ments reached approximately 30 and 5 cm,
respectively.
At the end of the construction period, the
served for the wall face. The base of the wall movements gradually subsided as the
wall is approximately 7.3 m wide, and its new pile underpinning system assumed
toe protrudes outboard approximately 0.6 m the foundation loads. In the process of wall
from the face of the wall. Borings taken rehabilitation prior to dredging, the con-
through the wall indicated numerous voids tractor installed a temporary gravei and
in the granite courses. The foundation soil crushed stone berm at the wall's toe, which
includes a thin layer of sand overlaying a was in place until the foundation work was
consolidated layer of gray ("Providence") completed.
varved silts whose thickness varies from 1.5 The underpinning grout piles were in-
to 3.0 m. Below this silt is a stratum of very stalled by drilling 15-cm- and 20.5-cm-
dense silty sand and gravei (glacial till) diameter holes vertically and sloped on
underlaid by gray stone bedrock. The 1 : 2.5 batter through the granite courses
bedrock elevation from MLW varied from and foundation soil to suitable foundation
15.2 to 24.4 m. It should be noted that the material; steel reinforcing was placed into
underside of the toe of the wall was esti- the predrilled holes and grout was injected.
mated et al. 9.9 m from MLW, hence, the Simultaneously, the casing was gradually
required depth of dredging 10.7 m would withdrawn to form the piles. The pile capac-
undermine the wall by 0.8 m. ity was confirmed by conducting field load
940 Modernization of Existing Marine Facilities

tests on special piles installed through and mize, overdredging. The latter often is the
behind the wall. The initial construction cause of wall distress and instability result-
specification called for the piles to be sock- ing in structural damages and excessive
eted approximately 1.5 m into the underlay- deflections.
ing bedrock. Load tests, however, indicated Essentially, excavation below the level of
that the piles could provide suitable founda- the toe reduces the bearing capacity under
tion without reaching the bedrock. As a the toe as well as the wall resistance to
result, pile tip elevations were raised, re- sliding and overturning. Successive dredg-
sulting in considerable project savings. ing operations may slowly reduce the level
The design of the piles performance was of soil at the wall face, thus reducing the
based on the assumption that the wall chance of overdredging. The effect of this
would perform as a monolithic unit. It was, mode of dredging, however, may not neces-
therefore, important that pile reinforcing sarily be obvious for a number of reasons;
extend into the wall and that the voids in for example, the bottom-level investigation
the wall be consolidated during the grout did not produce sufficiently accurate data
installation process. Grout was installed at (e.g., echo sounding can give a false reading
ambient hydrostatic pressure and ali pre- close to a vertical wall), the presence of silt
cautions were taken to prevent potential obscures the level of the firm material, and
formation of voids or discontinuities in the so forth.
pile due to a sudden grout pressure drop If bedrock bottom is excavated by use of
which could have resulted in negative pres- explosives, then the rock under the wall's
sure at the injection head. Grout injection base can be fragmented by adjacent blast-
within the wall was essential; however, it ing, resulting in degradation of the founda-
proved to be somewhat difficult due to the tion. Hence, the amount of explosives used
presence of large voids encountered be- for rock excavation must be carefully cali-
tween granite courses. The presence of these brated to avoid the aforementioned effects.
voids resulted in a loss of grout through the
wall face. To prevent grout loss, the outside
joints between courses were sealed by divers
with packing material that consisted of a 9.3.2 Modification of Piled
mixture of cement, bentonite, and metal coai-Loading Pier No. 6
fibers. The material was made in the form at Norfolk, Virginia
of balls prepared on the surface and deliv-
ered underwater to divers for wall packing. This pier is located on the east side of the
Additionally, vertical "stitching'' holes Elizabeth River, at Lambert Point, Norfolk
were drilled and grouted close to the wall Harbor. It is owned and operated by the
face. The line of these holes drilled at ap- Norfolk Southern Corporation and is one of
proximately 2.75 m center-to-center helped the world's largest and fastest coal-loading
to consolidate the body of the wall and also facilities. According to Prucz et al. (1992) its
helped to seal the wall face. "Stitching'' holes design capacity is 16,000 tonnes per hour,
were drilled through the cap stone to a with a maximum capacity of 20,000 tonnes
depth of approximately 11.0 m and grouted per hour. The north side of the pier is used
with a thickened gravity-injected grout as a berth for two large vessels and its
mixture. south side is utilized as a lay-up berth. The
In conclusion, it is worthy to note that pier supports two traveling loaders than
the dredging in front of any wall, and the can serve two vessels at one time, or con-
gravity wall in particular, must be carried centrate on a single vessel. The pier was
out cautiously to avoid, or at least mini- constructed in 1962; its total length is ap-
Modernization of Existing Marine Facilities 941

proximately 500 m and the width is about the 16.8-m depth of water, the projected
25 m. The pier is built in a form of a cast- average ship size for the year 2030 is
in-place concrete deck supported on pre- 115,000 DWT.
stressed concrete piles 610 X 610 mm in The pier capacity evaluation criteria con-
cross section, which are vertical and bat- sidered a design vessel of 150,000 DWT,
tered in both transverse and longitudinal 30-60% loaded, approaching the pier with
directions. The typical bottom profile and the velocity normal to berth of 10-25 cmjs.
pier cross section are illustrated in Figure Also, the current velocity of 20 cmjs and
9-12. the operating wind velocity of 65 kmjh were
From 1987 to 1988, the U.S. Army Corps taken into account. A maximum wind veloc-
of Engineers increased the operating depth ity of 160 kmjh (extreme load condition)
of the Norfolk Harbor Channel from ap- with allowable stresses increased by 33.3%
proximately 13.7 m to 16.8 m. AB a result, was also considered.
the pier operator decided to use larger ves- A second design vessel of 225,000 DWT,
sels with increased draft. Essentially, this representative of a large shallow draft de-
required the deepening at the pier face to sign, was also considered to address the
about 17.7 m. The original pier was de- potential accommodation of occasionally
signed to accommodate ships ranging in size large ships. Normally, the vessels calling at
from 25,000 DWT to almost 200,000 DWT. pier No. 6 are positioned along the pier's
The largest ships, those above 80,000 DWT, side with the aid of tugs; therefore, the
require an average maximum loaded draft berthing velocity is well under control.
of 13.7 m and greater. Hence, the vessels The pier capacity evaluations model used
larger than 80,000 DWT arriving at the for static structural analysis included verti-
pier were able to take on a partialload. For cal and lateral rigidity of the deck, the axial
stiffness characteristics of piles, and the
lateral and axial stiffness characteristics of
the pile-soil system; the elastic properties
of the fenders were also included.
In order to determine the dynamic loads
in the piles due to ship impact load, the
ship-pier system was modeled with a two
degree of freedom system, and a step-by-
step direct integration procedure was used
to analyze the nonlinear ship-pier dynamic
coupling effects. ·
Several alternative structural solutions
were developed and analyzed, and in each
' jj4
,_ case, the pile loads obtained were checked
against the piles' structural capacity and
CL.·,..O

the foundation soil capacity. The ultimate


l pile-soil capacities were computed based on
commonly accepted formulations. The pier
analysis indicated that in order to ensure
pier structural integrity, for a dredged depth
Figure 9-12. Coal-loading pier No. 6 at Norfolk Har- 17.7 m, the existing fender system should
hor, Virginia: Typical cross section. 1-Existing pier;
2-original ground surface; 3-soilcrete wall; be replaced with a new higher-capacity,
4-dredged bottom; 5-new fender system. [From lower-reaction-force fender unit, and the
Prucz et al. (1992).]. original soil elevation under the pier should
942 Modernization of Existing Marine Facilities

be maintained in order to provide adequate substantial modernization. The study con-


penetration to the battered piles. Also, the ducted by the Port Authorities concluded
end of the pier should be reinforced. For that about 7 km of existing quay walls had
this, numerous structural alternatives were to be renovated. This included deepening
investigated to ensure soil retention under the bottom at these walls from 8.75 m to
the pier structure; among them were the 13.0 m. The Port Artur quay with its 2000-
use of sheet-piling, precast retaining walls, m-long berthage line was the first one tobe
and installation of the grout-injected inter- modified according to the established se-
locked soilcrete columns. quence of port modernization (Simeon et al.,
The latter proved to be the most econom- 1990).
ica! solution and the soilcrete cutoff wall The existing quay wall that was built in
was constructed on both sides of the pier the 1950s replaced an old,. deteriorated and
and at its head. The interlocked rows of a obsolete gravity-type quay wall. The exist-
soilcrete column were keyed into the two ing quay wall was built in the form of
exterior rows of piles. The columns were pressure-relieving platform supported on
constructed by coring through the deck and concrete batter piles and also included steel
injecting two columns simultaneously using sheet-piling. The structure was estimated
the high-pressure grouting technology as as being in sound condition, but not suit-
discussed in Section 9.3.1. Core samples able for deepening of the bottom in front of
taken from the columns confirmed the ade- it by approximately 4.0 m, as required; the
quate strength of the soil-retention system.
dredging of this amount would certainly
Finally, the bottom was dredged to the re-
undermine the wall's stability. On the other
quired depth.
hand, the wall condition suggested that it
could be fully utilized as a useful compo-
nent of a new structure. This resulted in
9.3.3 use of Piled structures the solution illustrated in Figure 9-13. The
and Sheet-Pile Walls for essential component of the new structure is
the composite sheet-pile wall installed in
Modernization of Existing
front of the existing quay. This wall in-
structures cludes the steel H-piles (HZ) and intermedi-
ary sheet piles (ZH) driven in between the
The following are descriptions of severa!
H-piles.13
characteristic case histories which demon-
Both HZ and ZH piles are slightly bat-
strate how piles and sheet piles can be used
tered toward the new existing wall. Sheet
effectively for modernization of the existing
piles are driven to the existing bottom level
piled structures and quay walls of solid
and have the ability to retain soil behind it
construction.
after the bottom is dredged to the required
depth. It also provides support for the con-
9.3.3.1 Port-of-Ghent, Belgium- crete slab that is installed by the tremie
Modernization of the technique in between the new and existing
Port Artur Quay sheet-piling to protect the soil surface from
The effectiveness of the operation at potential erosion. The king (HZ) piles have
Port-of-Ghent strongly depends on the im- been installed to most of their penetration
provements that were made to the access by a vibrator and have been driven to their
channel in the 1970s. The channel opened final set by diesel hammer; because the
access for deep draft vessels to the existing sheet piles do not carry any vertical load,
port facilities, which, in turn, required a they were installed by vibrator only.
Modernization of Existing Marine Facilities 943

HZ Klno pile
RH lnlerlocll

Figure 9-13. Port-of-Ghent, Belgium-Modernization ofthe Port Artur Quay.


(a) Typical cross section; (b) detail of new composite front wall system. Note: HZ
and ZH are respectively Arbed (Luxembourg) H-piles and sheet piles. [After
Simoen et al. (1990).].

To achieve the required precisiOn, the The live design load on the new structure
king piles were driven through a special was kept the same as in the old design. This
template. Once the composite wall was in- load included crane and train loads and
stalled, a concrete superstructure was con- uniformly distributed surcharge loads 3
structed to integrate both new and existing tjm 2 between the crane rails and 10 tjm 2
structures. For better results, the upper 0.5 in the stockpile area.
m of the head beam of the existing struc- The quay modernization started in 1978
ture was removed and both new and old and was successfully completed in 1983.
structures were pressed against each other In conclusion, it should be mentioned that
by a system of posttensioned anchor bolts. the use of the composite sheet-piling simi-
The prefabricated front concrete panels lar to that described in this case history in
were installed before new concrete was 1980 through the 1990s has become almost
poured; these panels were used as a part of standard in Belgium and other European
the formwork system. countries.
The new land-side crane raii foundation
was constructed on top of the existing plat- 9.3.3.2 Port-of-Hamburg,
form. The harbor-side crane raii is placed oermany- Reconstruction
directly against the new wall. Because king of the Shipyard
piles are driven into a very dense sand Fitting-Out Wharf
layer and do not settle much under the
vertical load, only small shear forces are The existing piled quay wall at this
developed at the interface between the new shipyard was seriously damaged during
and existing concrete superstructures. This World War II, obsolete, and out of service
shear force is resisted by frictional forces (Hofmann, 1989). It was decided to modern-
created by tensioning the aforementioned ize this structure in order to use it for ship
anchor bolts. repair and construction of new vessels. To
944 Modernization of Existing Marine Facilities

serve this purpose, the bottom in front of in Bray and Taham (1992). One of them is
the existing wall was to be deepened by 6.0 shown in Figure 9-14b. In this example,
m; as is seen from Figure 9-14a, this was the sheet-piling is installed in front of the
impossible because bottom dredging to the existing wall and is anchored by ground
required depth would certainly undermine anchors drilled through the existing struc-
the structure. The best solution to the prob- tures.
lem was found in the construction of a new Mazurkiewicz and Zylicz (1990) discuss a
composite-type sheet-pile wall installed just case history in which the concrete sheet-pile
in front of the existing quay. This wall was bulkhead has been constructed in front of
comprised of king bearing piles and conven- the deteriorated timber relieving platform
tional steel sheet piles driven between the in order to increase the water depth by 1.0
king piles. The new wall was designed as m. In this case, the sheet-piling was an-
part of a new pressure relieving platform chored by steel tie-rods secured at a con-
that enveloped the existing wall (Fig. crete deadman placed just behind the
9-14a). Because the bottom in front of the existing structure.
existing wall was littered with surface and
underground debris, the latter needed to be 9.3.3.3 Port-of-Seattle,
removed before commencing any pile- washington-
driving activities. Dredging in front of the Modernization of
existing wall was regarded as unsafe and container Terminal No. 5
therefore not feasible, because it may lead The demand for greater terminal area,
to wall failure. coupled with changes in vessel size, more
Subsequently, the contractor carried out stringent seismic requirements, and eco-
the piling works in the following sequence. nomic pressures for greater terminal effi-
First, bearing king H-piles and batter an- ciency necessitated the renovation of the
chor H-piles were driven and connected with container terminal No. 5 at Port-of-Seattle
each other through provisional steel walling. (Ritchie and Watson, 1983). Among other
This was followed by installation of a short, modifications, this required substantial
6 m long, sheet pile driven between the modernization of the existing piled platform
back flanges of the king piles; the latter with a sheet-pile bulkhead located at the
enabled dredging out the miscellaneous ob- platform land side.
structions on the bottom. The platform is elevated 5.6 m above the
Mter removing the obstacles by dredging mean lower low water (MLLW) and the
a shallow trench, the intermediate sheet depth of water at the berth face is 12.2 m.
piles were driven between the front flanges The underdeck slope is 1.5 : 1 and the un-
of king piles. Wall renovation was com- supported height of the sheet-pile bulkhead
pleted with the installation of bored con- is approximately 6.1 m. Inspection of the
crete piles just behind the existing wall, structure has revealed some damages to
filling the space between the new sheeting the underdeck slope. It was concluded that
and the old wall with granular material, the strength and stability of the sheet-pile
and the construction cast-in-situ pressure- bulkhead is adequate to resist the lateral
relieving platform supported on the front loads such as soil pressures and associated
and rear bearing piles. The wall's lateral live loads, especially under earthquake con-
stability was provided by batter piles. ditions; this bulkhead was apparently
Finally, the bottom was dredged to the re- creeping toward the harbor.
quired depth. The other reason for wharf modification
A variety of similar solutions used for was the replacement of the existing cranes
renovation of existing quay walls are given . with new ones with considerably heavier
Modernization ofExisting Marine Facilities 945

Figure 9-14. Use of anchored sheet-pile walls and pile·s for


modemization of existing structures. (a) Modernization of obso-
lete shipyard outfitting berth [from Hofmann (1989)]; (b) recon-
struction works at Penros Quay, Cork. U.K. [from Bray and
Tatham (1992)]; (c) Port-of-Seattle, Washington-modemization
of container terminal No. 5 [after Ritchie and Watson (1983).]
1-Existing steel sheet-piling; 2-existing cast-in-place concrete
capping; 3-existing prestressed concrete panels; 4-existing
octagonal prestressed concrete piles, 420 mm in diameter;
5-existing cast-in-place concrete wall; 6-existing cast-in-place
concrete pile cap; 7-existing fender system; 8-new concrete-
filled steel pipe piles 460 mm in diameter; 9-new cast-in-place
concrete slab with new utility trench covered with steel plate;
10-new octagonal prestressed concrete piles 420 mm in dia-
meter; 11-new prestressed concrete panels 380 mm thick;
12-existing ballast and asphalt concrete pavement.
946 Modernization of Existing Marine Facilities

'''- .._ .....e ' - r· 2


- 112
c
;
.,
'
9
5
6

101 - ~
. -
1\ ~
\ 3
r-----7
. . (- Jl,~
a MLLW

1 ij\\ ·~
l]p\)..l,...
4

Figure 9-14. Continued

wheel loads; the maximum design loads for provide information on the acquired bend-
the new cranes are approximately 38.6 ing stresses that result from pile pulling
tonnes for the operating condition and 47.6 back as discussed. As pointed out in
tonnes for the stowed boom condition. The Chapter 7, these stresses are sometimes
analysis indicated that these new cranes substantial.
will significantly overload the front row of To resist new and heavier loads produced
piles, located beneath the waterside crane by new cranes, the existing deck panels,
raii; hence, this row of piles required some located at both edges of the platform, are
reinforcement. reinforced by new reinforced concrete slabs
The proposed modifications to the wharf installed as depicted in Figure 9-14(c). The
structure are illustrated in Figure 9-14(c). joint between existing panels and the new
The designers were basically concemed with slabs is made through dowels installed in
unloading the existing sheet piling and pro- the existing panels. This helps to create a
viding the additional piling to increase the composite structural condition.
capacity of the existing front row of piles.
The first goal is achieved by excavating
9.3.3.4 Naval Homeport at the
some backfill material from behind the ex-
Naval Air station,
isting sheet-piling to a depth of approxi-
Pensacola, Florida-
mately 4.1 m. The excavation slope 1.75: 1
Modernization of
is protected by the riprap, and, the space Existing Waterfront
between the slope's edge and the existing structures
platform is bridged by the 380-mm-thick
prestressed concrete panels. This new deck For many years, the Naval Air Station at
is supported on new octagonal prestressed Pensacola, Florida was the homeport for
batter piles, 420 mm in diameter, installed some smaller U.S. Navy aircraft carriers. It
at the land side of the structure and on the needed to be modernized when the super-
existing platform at the other side. The ad- carrier USS Kitty Hawk was selected to be
ditional concrete-filled steel pipe piles are homeported there. For this new use, the
driven by the large mounted pile driver just whole facility, waterfront structures in-
at the edge of existing frontal panel. When cluded, required some rehabilitation andjor
those piles reach the required capacity they modemization (Smith et al., 1992). The lat-
are cut off, pulled beneath the deck struc- ter included bottom dredging by approxi-
ture, filled with concrete, and completed mately 2.0 m to ensure the water depth in
with a cast-in-place concrete cap. Unfortu- front of the mooring structures of approxi-
nately, Ritchie and Watson (1983) do not mately 13.3 m.
Modemization of Existing Marine Facilities 94 7

The existing waterfront structure (Pier was built outboard of this bulkhead. This
303) consisted of a 240-m pile-supported allowed the bay bottom to be dredged to an
concrete deck marginal wharf. The width of elevation of -11.3 m. This series of projects
the deck of this structure ranged from 6.2 m formed the total existing Pier 303 system.
along the northern 193 m of the wharf to Mter the field inspection of the existing
20.1 m along the 47 m length of its south- structures, geotechnical investigation, and
ern part. Extending from the southern end engineering evaluation of load-carrying ca-
of Pier 303 was a 85-m-long finger pier and pacity of the existing structures, the follow-
two piled mooring dolphins. These dolphins ing was concluded:
have been linked with each other and the
finger pier via catwalks. Historically, the • The existing deck should be demolished.
southernmost portion of Pier 303 (with dol- • A new sheet-pile wall seaward of the exist-
phins) was constructed in the 1960s. This ing bulkhead should be constructed; this
was designed to replace the original piled will prevent soil erosion through the exist-
granite blockwork quay wall built in the ing corroded steel sheet-piling. Subse-
mid-1850s. quently, the space between the new and
The. elevation of the top of this quay wall the existing sheet-pile walls should be
was about 1.8 m above the mean sea level filled with granular material to preclude
and the adjacent dredge elevation was ap- continuing loss of soil.
proximately -6.1 m. A new quay was built • Existing concrete piles can be incorporated
into the new waterfront system.
in the form of a pile-supported concrete pier
that allowed for the bay bottom to be
dredged to approximately -11.3 m out- Several alternative solutions for modern-
board of the pier face. ization of Pier 303 have been investigated.
The northern part of the existing Pier These included utilization of conventional
303 was constructed in four different pro- anchored sheet-pile bulkhead, a piled re-
jects. The original structure was built in lieving platform, and an open piled struc-
1918, with a major addition and dredging ture. In final analysis, two factors con-
occurring in 1940. The original construction trolled the accepted solution to structures
consisted of a concrete gravity retaining modernization: the required depth of water
wall supported on a timber platform, the equal to 13.9 m and the presence of a soft
outboard of which was a concrete cutoff clay stratum in the vicinity of the dredge
wall. This wall extended from the bottom of line.
a gravity wall at elevation zero to a tip (toe) The piled relieving platform appeared to
depth of approximate elevation -16.8 m. In represent the most reliable and economica!
1941, a steel sheet-pile bulkhead was in- solution (Fig. 9-15). In the new structure,
stalled outboard of the gravity wall and the existing concrete piles were supple-
concrete cutoff wall and the bottom was mented with new, additional concrete piles
dredged to an elevation of -9.1 m. This wall installed between the existing pile bents,
was anchored with steel tie-rods secured at and new sheet-piling comprised of PZ40
a deadman system; this system imposed sheets with light cover plates both of Grade
loading on the gravity wall and the timber 50 steel installed in a typical relieving plat-
relieving platform system. form fashion. The granular backfill was in-
Hence, in effect, the new sheet-pile bulk- stalled through special openings left in the
head was supported by the wall built in deck; these holes were filled with concrete
1918. In the early 1960s, the steel-pile bulk- after the fill was installed.
head was encapsulated with concrete and a On the northmost part of the new con-
new pile-supported concrete deck structure struction, lateral stability of the structure
948 Modernization of Existing Marine Facilities

EL. +10.5'

EL. 0.0 CMSLI

T
1

EL. -43.5'

T
E.
VAIIES

;-l:_7
~8
1' EL. +10.5'

~ EL. 0.0 CMSL)

Ţ2
~

12
6

Figure 9-15. Naval homeport at the Naval Air Station,


Pensacola, Florida-modernization of Pier 303; typical cross
sections. (a) Northern part; (b) southern part. 1-Existing
timber relieving platform; 2-existing granite quay wall;
3-existing steel sheet-piling; 4-existing concrete sheet-
piling; 5-new concrete piling; 6-new steel sheet-
piling; 7-:first phase of new relieving platform construc-
tion; 8-utility trench; 9-second phase of new relieving
platform construction; 10-new concrete apron; 11-
granular backfill; 12-new anchor system.

was provided by a row of batter piles sup- hind the new structure. During pile instal-
plemented by a conventional tieback sys- lation, their capacity was monitored by us-
tem. The lateral soil thrust for this new ing pile-driving analyses. This, as well as
section was determined taking into account field load tests, helped to establish reliable
the dowel effect provided by timber piles embedment of these piles. Modernization of
of the existing relieving platfonn just be- Pier 303 was completed by the construction
Modernization of Existing Marine Facilities 949

O ~ IOifl
-=-=o

Figure 9-16. Port-of-Lisbon, Portugal-Modernization of the Alcantara Wharf;


typical cross section. 1-Existing wall; 2-new quay wall; 3-fill. [From Da Silva and
Brioso Pina De Jesus (1986).]

at its southern end ofthree new piled moor- cent example of this kind of dock modem-
ing dolphins that were required for han- ization is the O'Swaldkai Terminal, Port-
dling the USS Kitty Hawk mooring hawsers. of-Hamburg (Hofmann, 1989). There a deep
water wharf was built on land just behind
the old piled gravity-type wall. The soil be-
hind the wall was excavated to a certain
9.3.4 construction of Brand depth to ensure that the numerous under-
New Structures ground obstacles to pile-driving were re-
moved. After the new wall was completed,
In some instances, port modernization re- the old one was dismantled and the bottom
quires not only deepening of the wharf but in front of the new wall was dredged to the
also expansion of its working area. In this required depth.
case, a brand new waterfront facility needs
to be constructed some distance from the
existing structure. One practica! example is REFERENCES
shown in Figure 9-16. It represents a typi- BARKER, J. E. and CHATI'AWAY, J., 1990. "Deepen-
cal cross section of modernized Alcantara ing and Developing Quay Walls in Felixstowe
wharf in Port-of-Lisbon, Portugal (DaSilva Port. Examples of Special Design and Con-
and Brioso Pina De J esus, 1986). In this struction Techniques in Difficult Situations."
case, the new quay wall was constructed 80 PIANC, Proceedings of 27th International
m away from the existing gravity-type Congress, Osaka, Japan.
structure. The new construction provided BIRKELAND, 0., 1989. "Trends in the Design
both a 5-m increase in water depth as well of Marine Terminals: The Canadian Experi-
as substantially increased working area. ence." PIANC Bulletin No. 67.
Similar case histories are found in Hof- BRAY, R. N. and TATHAM, P. F. B., 1992. Old
mann (1989) and many other publications. Waterfront Walls. Management, Maintenance
In most cases involving construction of a and Rehabilitation. E & FN Spon., London.
new quay, the new structure is placed some CALABRETI'A, V. V. and SULLIVAN, T., 1980. "New
distance ahead of, or sometimes very close Foundation System for Granite Seawall."
to, the existing structure. In some in- ASCE Proceedings Specialty Conference
stances, however, the new structure is con- PORTS '80, Norfolk, VA.
structed just behind the existing structure CARON, C:, CATI'IN, P., and HERBST, T. F., 1975.
(Fig. 9-5b), in which case the latter may "Injections." Foundation Engineering Hand-
serve as a cofferdam or working platform book, Wintercorn, H. F. and H.-Y. Fang (eds.).
for new construction works. The most re- Van Nostrand Reinhold Co., New York.
950 Modernization of Existing Marine Facilities

CooMBER, D., 1987. "Restoration of Quay Wall." PONNET, L., VAN DER EECKEN, DEDEYNE; R., and
Civil Engineering, January jFebruary. THIBAUT, W., 1992. "Deepening of Gravity
DA SILVA, J. M. and BRIOSO PINA DE JEsus, J. A., Quay Walls by Means of the Very High Pres-
1986. "Le Port De Lisbonne. Deux Ouvrages sure Grouting Technique. A Pilot Project in
Maritimes Avec Utilisation D'Elements Pre- the Port of Antwerp." Proceedings 10th Inter-
fabriques an Beton." PIANC Bulletin No. 54. national Harbour Congress, Antwerp.
DUBROVA, G. A., 1959. Design of Economically PORTER, D. L., 1986. "Immovative Repairs to
Efficient Hydrotechnical Structures." River Sheet Pile Structures." ASCE Proceedings
Transport Publishing House, Moscow (in Specialty Conference PORTS '86, Oakland,
Russian). California.
HALLETI, D., PRITCHARD, D., and MELETION, M., RITCHIE, W. D. and WATSON, W. W., 1983. "Mod-
1992. "Innovative Underpinning Technique ernization of a Container Terminal." ASCE
Permitting Alongside Deepening at Existing Proceedings Specialty Conference PORTS '83,
Gravity Quay Walls." Proceedings 10th Inter- New Orleans, LA.
national Congress, Antwerp. SIMOEN, R., GUNST, D., SERRAS, J., MORTIER, P.,
VANDENBOSSCHE, D., DE REGGE, J., GoOSSENS,
HOFMANN, K.-F., 1989. "New Methods for the
D., and BERNARD, A., 1990. "Achievements of
Construction of Quay Walls, in the Port-of-
Some Port Renovation Projects Planned, De-
Hamburg." PIANC Bulletin No. 64.
signed and Realized in Belgium." PIANC Pro-
ISHIGURO, K. and MIYATA, Y., 1988. "Rein- ceedings 27th Congress S-2, Subject 4, Osaka.
forcement of an Anchored Sheet Pile Wall
SMITH, G. W., EVANS, C. H., and KNOTT, M. A.,
with Additional Lower Tie-Rods." PIANC 1992. "Naval Homeport Facilities at Pen-
Bulletin No. 61. sacola, Florida, and Mobile, Alabama." ASCE
JENSEN, J. V. and SIEMONS, A. E. L., 1986. Proceedings Specialty Conference PORTS '92,
"Overdredging at Existing Gravity Structure." Seattle, WA
ASCE Proceedings Specialty Conference TARICONE, P., 1994. "Jet Grouting Gaines." ASCE
PORTS '86, Oakland, CA Civil Engineering, Vol. 64, No. 12.
MAzURKIEWICZ, B. K. and ZYLICZ, A., 1990. "Some THIBANT, W., HIMPE, J., and VAN CELST, E., 1992.
Problems of Developing Countries." PIANC "Renovation of the North Quay of the Third
Proceedings of 27th Congress, S.II-5, Osaka. Harbour Dock in the Port-of-Antwerp." Pro-
PETTIT, P. and WOODEN, C., 1988. "Jet Grouting: ceedings 10th International Harbour Con-
The Pace Quickens." ASCE Civil Engineering, gress, Antwerp.
Vol. 58, No. 8. TSINKER, G. P., 1994. "Assessment ofDock Struc-
PIERCE, F. C. and CALABRETTA, V. V., 1986. "Re- tures." Unpublished Report, Acres Interna-
habilitation of Gravity Wall by Under- tional Limited.
pinning." ASCE Proceedings Specialty Con- TSINKER, G. P., 1995. Marine Structures Engi-
ference PORTS '86, Oakland, CA. neering: Specialized Applications. Chapman
PRucz, Z., MARTIN, B. T., and RICHSTEIN, J. L., & Hall, New York.
1992. "Modification to Coal Pier 6 Made Nec- WINTERCORN, H. F. and PAMUKCU, S., 1991. "Soil
essary by a Deeper Channel." ASCE Pro- Stabilization and Grouting." Foundation En-
ceedings Specialty Conference PORTS '92, gineering Handbook, 2nd ed., Fang, H.-Y. (ed.).
Seattle, WA. Van Nostrand Reinhold, New York.
10
Breakwater Design
<By s. Takahashil

Breakwaters are constructed to provide a waves, even in deep waters. Typical com-
calm basin protected from waves harbor or posite breakwater consists of a rubble foun-
for ships and to protect port facilities. They dation and a vertical wall of miscellaneous
are also used to protect the port area from designs. This type of construction is usually
intrusion of littoral drift. In fact, for ports classified as the vertical breakwater.
open to rough seas, breakwaters play a key Floated-in concrete caissons are most often
role in port operations. used for construction of the upright section
Because sea waves have enormous power, of the composite breakwater. This type of
the construction of breakwaters designed to construction proved to be very stable in
mitigate such power is not easily accom- rough deep seas. This has led to construc-
plished and the history of breakwaters tion of this type of breakwater worldwide.
records a great deal of damages and fail- In this chapter, different types of break-
ures sustained by the breakwaters. On the water and their historical development as
other hand, maritime technology has pro- well as advantages and disadvantages are
gressed a great deal, especially since 1945, described. Also, failures of the breakwaters
and this has gradually made it possible to are discussed to demonstrate the critical
construct breakwaters with a high stability aspects of their stability. Finally, different
against waves. analytical methods used for vertical and
There are two basic types of break- rubble-mound breakwaters design are dis-
waters: rubble-mound and composite-type cussed. The detailed design methodology for
breakwaters. A typical rubble-mound the rubble-mound breakwaters is beyond
breakwater has a rubble mound and an the scope of this work; it can be found in
armor layer that usually consists of heavy many textbooks on the subject. This work
rocks or shape-designed concrete blocks. U se is focused basically on the design of the
of shape-designed blocks enabled the vertical-type breakwater.
modern-day rubble-mound breakwaters to The author extends his deep gratitude to
effectively resist the destructive power of many authors and organizations from whom

951
952 Breakwater Design

the valuable information has been drawn 10-1 illustrates basic structural concepts of
and used in this chapter, especially to Y. Ito different types of breakwaters.
(1969), K. Horikawa (1978), Y. Goda (1985),
K. Tanimoto et al. (1987), H. F. Burcharth 10.1.1.1 Sloping or
(1993), P. Bruun (1985), M. A. Losada Rubble-Mound
(1990), Port and Harbour Research Insti- Breakwater
tute, Overseas Coastal Area Development
Institute of Japan (OCDI) (1991), CIRJA The sloping or rubble-mound type of
(special publication 83, CUR Report 154, breakwater basically consists of a rubble
1991), and the U.S. Army Corps of Engi- mound as is shown in Figure 10-1A. Note
neers (1984). that the typical rubble mound is defined as
a mound of random-shaped and random-
placed stones protected with a cover layer
of selected stones and specially shaped con-
10.1 HISTORIC DEVELOPMENT OF crete armor units (armor units in primary
BREAKWATERS cover layer may be placed in orderly
manner or dumped at random). The most
typical sloping-type breakwater is one with
randomly placed stones (Fig. 10-1A-a). The
10.1.1 structural Types multilayered rubble-mound breakwater
(Fig. 10-1A-b) ensures better stability and
Many structural types of breakwaters are reduced wave transmission. This type of
used worldwide. As indicated in Table 10-1, mound breakwater usually has a core corn-
all breakwaters can be classified as: sloping posed from quarry run. The stability of the
or mound, vertical and special types. Figure armor layer of the multilayered breakwater

Table 10-1. Structural types ofbreakwaters

Sloping (mound) type Rubble-mound breakwaters


Rubble-mound breakwaters (multilayer)
Rubble-mound breakwaters armored with
blocks
Concrete block breakwaters
Reshaping rubble-mound breakwaters
(berm breakwaters)
Reefbreakwaters (submerged breakwaters)
Vertical (upright) type Monolith concrete breakwaters
Composite type Block masonry breakwaters
Horizontally composite type Cellular block breakwaters
Concrete caisson breakwaters
New caisson breakwaters
Special (nongravity) type Curtain wall breakwaters
Steel pile breakwaters
Horizontal plate breakwaters
Floating breakwaters
Pneumatic breakwaters
Hydraulic breakwaters
Breakwater Design 953

sion is not reduced much by this type of


(a)~ sz breakwater, its simple construction proce-
dure and relatively high permeability are
advantageous features. Recently, reef-type
breakwaters or submerged breakwaters
(Fig. 10-1A-e) have been constructed for
coastal protection; these structures do not
create a so-called "visual pollution" and
therefore are attractive from the environ-
mental point of view. Reshaping break-
waters (Fig. 10-1A-f) utilize the basic con-
cept of the dynamic equilibrium between
the slope of the rubble stone and wave
action; that is, the rubble-mound forms
the S-shaped slope to stabilize itself against
wave actions. This type of breakwater has a
large berm on exposed to the wave slope
(d) which is ultimately reshaped by the wave
action. Reshaping breakwater is also termed
the berm breakwater or dynamically stable
breakwater. It should be noted that this
(e) concept is not new. Siene ancient times, the
rubble-mound breakwaters have been re-
shaped naturally by wave action and by
subsequent repairs.
(f)
10.1.1.2 vertical- and
composite-Type
Breakwaters
Figure 10-lA. Sloping-type breakwaters: (a)
The original concept of the vertical
rubble-mound breakwater; (b) multilayer rubble-
mound breakwater; (c)multilayer rubble-mound break- breakwater was to refl.ect waves. Figure
water armored with concrete blocks; (d) block-mound 10-1B illustrates four types of vertical
breakwater; (e) submerged breakwater; (0 reshaping breakwaters that are founded on mounds of
(berm) breakwater. different heights. The basic type of vertical
wall breakwater is shown in Figure 10-1B-
a, whereas the other three are composite
can be enhanced by using concrete blocks of breakwaters with a rubble-mound founda-
miscellaneous geometry; wave transmission tion. These are defined as the low-mound
can be reduced by using a concrete super- (Fig. 10-1B-b) and high-mound composite
structure (wave screen or wave wall), which breakwaters (Fig. 10-1B-d). By convention,
can also function as an access road to the the high-mound composite breakwater has
breakwater (Fig. 10-1A-c). a mound that is higher than the low water
Breakwaters comprised of only concrete level (LWL). The low-mound breakwater
blocks (Fig. 10-1A-d) have been con- does not cause wave breaking on the mound,
structed as detached breakwaters used for whereas the high-mound breakwater does.
coastal protection. Although wave transmis- However, the high-mound composite-type
954 Breakwater Design

HWL
L.W L.~_;z.. 1 (o) (o)

(b)

(d)

Figure 10-1C. Horizontally composite breakwaters:


(a) low rubble foundation; (b) relatively high rubble
Figure 10-1B. Vertical-type breakwaters: (a) verti- foundation; (c) high rubble foundation; (d) inner core.
cal wall breakwater; (b) composite breakwater (low
mound); (c) composite breakwater; (d) composite
breakwater (high mound). or large rocks have been used in the past
for strengthening the vertical wall break-
waters that have sustained damages from
breakwaters could be unstable due to wave waves; these blocks or rocks have been
generated impulsive pressure and scouring placed in front of vertical wall to dissipate
caused by the breaking waves. For this rea- the wave energy and to reduce the wave
son, composite-type breakwaters with a low force, especially that from breaking waves.
mound are more practical. The composite In modern construction practice, shape-
breakwater with a relatively high mound designed concrete blocks such as tetrapods
(Fig. 10-1B-c) although lower than LWL and others are used more often.
occasionally generates impulsive wave pres- The horizontally composite breakwaters
sure due to wave breaking. are very similar to rubble-mound breakwa-
Sometimes concrete blocks are placed in ters annored with concrete blocks. Figure
front of the breakwater to reduce both the 10- 1C shows how the cross section of hori-
wave refiection and the breaking-wave force zontally composite breakwater varies with
on the vertical wall. This type of construc- mound height; as the mound height in-
tion is termed a composite breakwater cov- creases, the horizontally composed break-
ered with wave-energy-dissipating concrete water becomes very similar to the rubble-
blocks; it is also termed the horizontally mound breakwater. This is particularly true
composite breakwater (Fig. 10-1C). Such for ones with core stones placed in front of
breakwaters are not new. Concrete blocks the vertical wall (Fig. 10-1C-d) that is
Breakwater Design 955

structurally nearly the same as the rubble- Col~

~ll~:-
mound breakwater. Both, however, are con-
ceptually different, because the concrete
blocks of the rubble-mound breakwater act &Sltr - -
as the armor for the rubble foundation,
whereas the concrete blocks of the horizon-
tally composite breakwater function to re-
duce the wave force and size of the refl.ected
waves. In general, the horizontally compos-
ite breakwaters are considered to be the
improved version of the vertical ones.
Figure 10-1D illustrates several types of
composite breakwater having different
types of the upright section. The upright
wall made from block masonry, in which
many different methods were applied to
strengthen the interlocking between the
blocks, was most popular in the 1800s (Fig.
10-1D-b). Sometimes, concrete cellular
blocks have been used to form the upright
wall of vertical breakwaters (Fig. 10-1D-c).
However, the invention of the fl.oated-in
concrete caissons (Fig. 10-1D-d), made
breakwaters more reliable, and many
breakwaters that comprised these caissons
were constructed worldwide. Subsequently,
caisson-type breakwaters have been im-
proved by using a sloping superstructure or
a perforated wall (Figs. 10-1D-e and
10-1D-f). The rubble-mound foundation of
composite breakwaters is vital for stability
of the upright section; it must safely carry
the caisson's weight, be stable against wave Figure 10-ID. Composite breakwaters: (a) monolith
concrete block; (b) block masonry; (c) cellar block; (d)
forces acting against both the upright sec- upright caisson; (e) sloping top caisson; (0 perforated
tions and the foundation itself, and must be wall caisson.
protected against scouring by currents and
waves action. 10-1E; the structures shown in this figure
are of a nongravity nature.
10.1.1.3 Special Types of The most common special-type breakwa-
Breakwater ters occasionally used are the piled struc-
tures ofmiscellaneous designs, fl.oating, and
Special-types of breakwater are those pneumatic breakwaters. Use ofthese break-
employing some kind of special feature. Al- waters is limited to special conditions. For
though they are not commonly used, their example, the curtain-wall breakwater (Fig.
history is long; in fact, some were con- 10-1E-a) is commonly used as a secondary
structed in ancient times. Some special-type breakwater to protect small-craft harbors
breakwaters are illustrated in Figure located within protected harbors and the
956 Breakwater Design

(o) 10.1.2 conditions for


Breakwater Selection
V

Breakwaters are normally selected taking


into account the following:

(b) • Layout of breakwater(s)


• Environmental conditions
• Utilization conditions
• Execution conditions
• Costs of construction
• Construction terms

-m
(el • Importance of breakwater(s)
• Available construction materials
• Maintenance

Also, the breakwater's effects on the sur-


rounding topography and the coastal line
due to wave reflection and on the quality of
water are taken into consideration for the
selection of the type of breakwater.

!el
10.1.3 comparison of Sloping-
SI
····=:.:...::·_:·-~.:...~ ···' 1
··::
...
.. and vertical-Type
6 Breakwaters
Figure 10-lE. Special breakwaters: (a) curtain wall Each type of breakwater has advantages
breakwater; (b) sheet-pile breakwater; (c) horizontal
plate breakwater (jacket type); (d) :fioating breakwater; and disadvantages (Larras, 1961). As sum-
(e) pneumatic breakwater. marized by Lamberti and Franco (1994),
the advantages of using the caisson-type
breakwater (composite breakwater) in com-
sheet-pile wall or continuous row of pile parison with a rubble-mound breakwater
(Fig. 10-1E-b) is used to break relatively armored with concrete blocks are as follows:
small waves. A horizontal plate breakwater
(Fig. 10-1E-c) can reflect and break waves, 1. A smaller body which requires a smaller
and as shown in the latter figure, is some- quantity of material needed for the break-
times supported on a steel jacket. The water construction. This is a major ad-
vantage which makes the breakwater
floating breakwater (Fig. 10-1E-d) is very
construction more economica!, especially
useful for installation in deep water, but its in deep water. In addition, a smaller-body
effectiveness is limited to breaking small, breakwater covers less seabed area and
relatively short waves. The pneumatic therefore has smaller impact on a seabed
breakwater (Fig. 10-1E-e) breaks waves due marine life and also increases the usable
to water circulation that is induced by the water area.
air bubble flow; it is considered to be effec- 2. The vertical-type (composite) breakwater
tive for breaking small waves having short requires less maintenance in comparison
lengths. with a rubble-mound breakwater because
Breakwater Design 957

the armor blocks of rubble-mound break- body, which reduces the harm to the envi-
waters require relatively frequent main- ronment.
tenance. 4. Creation of a natural reef. The slope of
3. Rapid construction, reduction of failure the rubble-mound breakwater provides a
during construction, and smaller environ- suitable place for sea life (e.g., fish
mental impact during construction. The habitat).
vertical-type (composite) breakwater can
be constructed rapidly and is fully stabi- It should be noted that some of the disad-
lized once the caissons are filled. In com-
vantages of composite breakwaters can be
parison, the rubble-mound breakwater
needs a longer postconstruction period for
improved by using horizontally composite
stabilization of its inner layers that may breakwaters or perforated wall caissons.
also be subjected to damage during con-
struction. Furthermore, the vertical-type
breakwater requires relatively small 10.1.4 Historical Development
quantities of quarry run material to be
dumped into the sea. This produces less
of Breakwaters
turbidity and lesser environmental dam-
age to the surrounding area associated The following is a brief review of the history
with the construction activities. of breakwater development (lto, 1969).
4. Potential for reuse of the dredged mate-
rial, easy dismantling procedure, and 10.1.4.1 Historical Breakwaters
fewer underwater obstacles left after re-
moval of the upright section are also con-
10_1.4.1.1 Breakwater5 in ancient
sidered to be advantages of vertical-type times
breakwaters. In addition, the use of a Breakwaters constructed in ancient times
vertical-type breakwater may be the only were simple mounds made from stones.
choice at locations where availability of
However, as early as 200 B.C. a stone ma-
the rubble stones is limited.
sonry breakwater was constructed in
Alexandria, Egypt. Figure 10-2 shows a
On the other hand, the advantages of the rubble-mound breakwater located in
rubble-mound breakwaters are summarized Chivitavecchia, Italy. It was constructed by
as follows: the Roman Emperor Trajanus (A.D. 53-117)
and is recognized as being the oldest known
1. Use of natural material. The latter is a
existing rubble-mound breakwater. This
major advantage of the rubble-mound
breakwater because this can reduce mate-
breakwater reached its equilibrium slope
rial costs, especially when a large supply after a long history of damages and subse-
of rubble stones is readily available. quent repairs.
2. Use of smaller construction equipment.
The construction of the rubble-mound 10.1.4.12 Breakwater5 built in the
breakwaters can be carried out from land 17005 through 18005
and does not usually require large-scale
floating construction equipment such as
The age of modern breakwaters is consid-
work barges and heavy-lift cranes. ered as starting in the late 1700s and early
3. Less environmental impact due to smaller
1800s, subsequent to the beginning of the
reflected waves and more water exchange industrial revolution in Europe. The break-
within the harbor; the wave's energy is waters built in Cherbourg, Plymouth, and
absorbed by the rubble mound and the Dover are considered to be the pioneering
long-period waves such as tidal waves are developments of the modern-day break-
transmitted through the breakwater's waters.
958 Breakwater Design

Breakwater at Cherbourg. The con- structure is shown in Figure 10-4a. The


struction of a bay-mouth breakwater at crown elevation of this breakwater was at
Port-of-Cherbourg, France began in 1781. +3 m and the exposed slope was 1: 3; in
This breakwater was initially built as a 1841 the crown was elevated to +6 m in
rock-filled, 50-m-wide, cone-shaped crib. order to reduce frequency of wave overtop-
However, large cones failed soon after in- ping.
stallation, and in 1789 the breakwater was The cross section of the breakwater has
redesigned and changed to a rubble-mound been changed after it suffered various dam-
breakwater built around and above the ages and subsequent repairs. In 1824, its
failed cones. The exposed slope of the new exposed slope was leveled to 1 : 5 and stone
structure initially was 1: 3; however, after pitching was added above the LWL. The
subsequent frequent damages and repairs, breakwater's cross section in 1841 is shown
it was leveled out at a 1 : 8 slope. The upper in Figure 10-4b. At this time it had a berm
part of this breakwater, above the low near the LWL and a width of approximately
water level (LWL), suffered most frequent 110 m. This breakwater continued to re-
damages, and in 1830 a vertical wall was quire a great amount of additional stones
constructed above this level. Hence, it was even after the work done in 1841. In 1921,
perhaps the first high-mound composite its slope reached 1 : 12, which is close to the
breakwater ever built. Changes that occur equilibrium slope. Dedicated maintenance
in the breakwater's cross section during its has ensured the breakwater's existence at
lifetime are shown in Figure 10-3. the present time.
Rubble-Mound Breakwater at Vertical Wall Breakwater at Dover,
Plymouth. The original breakwater at United Kingdom. It was constructed in
Port-of-Plymouth, United Kingdom was 1847 (Fig. 10-5). The construction condi-
built in 1812. It actually was a copy of tions there were extremely difficult; con-
a rubble-mound breakwater built at struction proceeded very slowly and the
Cherbourg. The cross section of the original breakwater eventually was completed at a

Figure 10-2. Rubble-mound breakwater in Chivitavecchia, ltaly.

Figure 10-3. Cherbourg Breakwater; France: stages ofmodernization.


Breakwater Design 959

(al

+3.0

( bl

-2-";j~-
...L.H~-~~:_Ji.iJ!i§i'i'iliiil~~~ :'il s:~~:
/h

Figure 10-4. Plymouth Breakwater, United Kingdom: (a) original


construction (1812); (b) modernized cross section (1841).

+ 8.8 St. Catherine

"\1
~
&:· .:...... ·.~ + 5. 7
-='f==-----l-r41Irl·... : ··... : ... = \ - - - - - . - : : . . : - ' - -
r;;;..

±0

13.7

Figure 10-5. Dover breakwater, United Kingdom.

high cost. This, however, eventually paid off Alderney (-20.0m)


because its commissioning of the breakwa-
ter performed well and experienced only
slight damages. A half a century later, the
speed of construction was significantly in-
creased when a similar vertical wall break-
water was built in the adjacent area.

::11
10.1.4.2 composite Breakwaters
ffrom High to Low Peterhead
MoundJ ·~5~,0
Examples of this type of construction are
given in Figure 10-6. Many high-mound
composite breakwaters were built after con-
Figure 10-6. Change in height of mound from high
struction ofthe Cherbourg composite break- to low in breakwaters constructed at different loca-
water. For example, in the United Kingdom tions in the United Kingdom.
960 Breakwater Design

composite breakwaters were built at St. 1"5.0


Catherine and Alderney.
As noted earlier, the wave action on the
rubble mound may cause scouring of the
mound and affect the stability of the verti-
cal wall. To avoid this type of distress, the
scouring area is usually covered with large
rocks or concrete blocks; alternatively, the
wall may be placed at a lower level. For
example, for better stability, the breakwa- Figure 10-7. Algiers north breakwater, Algeria.
ter at Alderney (Fig. 10-6b) was changed
from a high mound to a low mound; the 10.1.4.3.1 Breakwaters in Algeria
river-mouth breakwater in Tyne, United
Kingdom was initially changed from a The historic Port-of-Algiers is in exist-
high-mound to a low-mound composite ence since the 16th century. The port's orig-
breakwater, and in the 1890s it was finally inal breakwater was of a rubble-mound
converted to a vertical breakwater with no construction and required continuous
rubble foundation at all. The breakwater at maintenance. In 1833 French engineer
Peterhead (Fig. 10-6d) is a very low-mound Poirel carried out reinforcement work using
2-3-m 3 rocks for breakwater protection, but
composite breakwater.
these rocks proved to be unstable. The
Composite-type breakwaters similar to
breakwater was later successfully rein-
that illustrated in Figure 10-6 have been
forced by installation of 20-m 3 rectangular
constructed elsewhere in the world and
concrete blocks.
specifically in the ports of then British
The cross section of the north breakwater
colonies (e.g., Karachi, Colombo, and
at Port-of-Algiers as existed in 1840 is de-
Madras).
picted in Figure 10-7. At that time its cross
section was very similar to that used in
10.1.4.3 Rubble-Mound modern breakwaters. Its rubble-mound core
Breakwater Armored was protected with 15-m3 concrete armor
with Blocks blocks. These blocks were placed with a
slope of 1 : 1 and saved a lot of stones used
Parallel with the development of the for the rubble mound in comparison to the
composite breakwater, the rubble-mound Plymouth type of rubble-mound break-
breakwater underwent a very impressive water.
development due to the invention of con- From the middle to the end of the 19th
crete blocks. The primitive cement that ap- century the rubble-mound breakwaters ar-
peared around 3000 B.C. was significantly mored with concrete blocks were built at
improved in the 18th and 19th centuries, several other ports in Algeria (e.g., Algers,
and the major improvement in concrete Oran, Philippeville, etc.). During their ser-
technology occurred in 1824 when J. Aspdin vice life, these breakwaters sustained dam-
invented Portland cement. The advance- ages from time to time that required some
ment in concrete technology resulted in de- repair work. These damages were at-
velopment of concrete armor blocks used for tributed basically to steep slope, insufficient
construction of rubble-mound breakwaters. weight of concrete blocks, insufficient depth
The following examples illustrate the devel- of the armor layer, and rough placement of
opments in this area. armor blocks.
Breakwater Design 961

1859

1869

-
(Domogel 1865 +4.5

~5::1 ~~-·~~~
1869 1869 +5.0

Extension

Figure 10-8. Breakwater at Oran, Algeria: recon-


struction stages.

Figure 10-9. Breakwater at Phillippeville, Algeria:


reconstruction stages.

Figure 10-8 illustrates changes in the


cross section of the breakwater at Oran,
which was severely damaged by a heavy
storm in 1869 because its armor layer ·was
not extended deep enough under the sur- Philippeville. This structure has experi-
face of the water. Even though the depth of enced some damages and subsequent repair
the armor layer was later extended to - 9.5 work even during construction. This even-
m in the improved cross section, the break- tually led to the construction of a stable
water again experienced some damages. cross section. Figure 10-9 illustrates the
Finally, the Marseille-type cross section chronology of the breakwater modi:fications
was adopted when the breakwater was that took place from 1860 through 1890. As
extended. Marseille-type breakwaters are is seen from Figure 10-9, the breakwater's
described later in Section 10.4.3.2. stability was improved by incorporation of
Figure 10-9 shows changes in the cross the large superstructure and the heavy
section of the breakwater built at armor blocks in its cross section.
962 Breakwater Design

10.1.4.3.2 Marseille-type breakwater. The major drawback of the


breakwater Marseille-type breakwater is that concrete
rectangular blocks tend to be very heavy
Expansion of the outer Port-of-Marseille,
and are difficult to handle.
France started in 1845, and for protection
of the port water area, both vertical and
rubble-mound breakwaters were con- 10.1.4.3.3 Shape-designed
structed there. The rubble-mound break- concrete blocks
water is illustrated in Figure 10-10. It
The invention of shaped concrete block
includes the following special structural
for use in breakwaters and shore protection
features:
structures revolutionized the construction
of breakwaters. Some representative
• The stones of the rubble core vary in shape-designed concrete blocks that are
weight, with light stones placed in the used in modern port and harbor engineer-
inner core. ing are discussed later in this chapter and
• The armor layer comprised of rectangular
were illustrated in Chapter 2. Tetrapods,
concrete blocks is extended to a sufficient
depth below the LWL and the armor layer
introduced by P. Danel in 1949, greatly im-
above sea level has a gentle slope that proved the rubble-mound breakwaters and
effectively dissipates wave energy. The su- enhanced their performance.
perstructure is placed some distance away Figure 10-11 illustrates difference be-
from the water, with most of it being cov- tween the Marseille-type rubble mound-
ered with armor blocks. breakwater armored with 45-t concrete
• The slope of the lower level is relatively blocks and a rubble-mound breakwater ar-
steep. mored with 25-t tetrapods built in Port-of-
• The armor blocks are installed carefully. Sefi, Morocco. Both alternatives have been
considered for construction; however, the
The Marseille-type breakwater was not only latter proved to be more economica!. U se of
popular for use in the Mediterranean but tetrapods resulted in a 70% reduction in
also in other geographical areas. Many the amount of concrete and a 5% reduction
breakwaters built worldwide copied the of stones. This breakwater performed very
cross section of the Marseille breakwater, well during the heavy 1957 storm that pro-
and they are ali called the Marseille-type duced 9-m-high waves.

Figure 10-10. Marseille breakwater, France.


Breakwater Design 963

(a) head of the St. Vincenzo breakwater (Fig.


Concrete + 9.5 10-12b). This breakwater had a steep
stepped wall. Obviously if the step-wall be-
comes very steep, it turns the breakwater
into the vertical wall composite breakwater.
One noteworthy example of composite
breakwater with a very steep step wall is a
detached (island) breakwater constructed in
1900s in Naples (Fig. 10-12c).

( b)
10.1.4.4.2 cvctopean blocks and
caissons
The stability of the vertical wall was in-
creased by the use of large heavy blocks.
For example, heavy, 330-t Cyclopean con-
crete blocks have been used for construction
of the vertical wall of the composite break-
water at Catania, Italy (Fig. 10-12e). Note:
Figure 10-11. Marseille-type breakwater, con- the word Cyclopean comes from Cyclops,
structed at Safi, Morocco: (a) protection with concrete who, according to Greek mythology, was a
blocks (45 tonnes); (b) protection with tetrpods (25
tonnes). giant with a single eye in the middle of his
forehead.
The use of Cyclopean blocks in Italy for
10.1.4.4 step-Type and construction of breakwaters infiuenced the
composite Breakwaters design of breakwaters in the other parts of
the Mediterranean. For example, Cyclopean
10.1.4.4.1 Step-type and
blocks have been employed for construction
composite breakwaters of the Mustafa breakwater in Algiers in
in ltatv 1923; Sainfiou used an interlocking Cyclo-
The step-type rubble-mound breakwater pean block in his design of the composite
was developed in Italy, where the rubble- breakwater for the outer breakwater in
mound breakwater having a step-type ar- Marseille (Fig. 10-13), with each Cyclopean
mor layer was designed by Parodi and con- block weighing 450 tonnes. This particular
structed as the Galliera breakwater in design, however, has not been adopted for
Genoa (Fig. 10-12a). This step-type armor construction of otherwise similar types of
layer was considered to be more stable be- composite breakwater that has been con-
cause of the interlocking network of uni- structed in Marseille. Figure 10-14 illus-
formly piled concrete blocks. Subsequently, trates changes made in the cross section of
many breakwaters of this type have been this breakwater in the period from 1930 to
built in the 1880s and 1890s. However, they 1953. The interlocking network of this
were not as successful as the original one. breakwater was further reinforced as a de-
The Galliera breakwater also suffered se- sign improvement.
vere damage in 1989. One of the causes of The vertical wall of a composite breakwa-
the Galliera breakwater damage was the ter can be constructed from caissons of mis-
differential settlement of the rubble mound. cellaneous designs. U se of large caissons
In Naples, a step-type breakwater sys- increases the strength of the wall and its
tem was adopted for construction of the stability. Walker proposed using caissons
964 Breakwater Design

Catania
(el .. Cyclopean

Naples· Granili
(d)

+7.0
Cellular

Naples (detachedl
(c l
Composite l
2iL
type

Figure 10-12. Changes from step type to composite type in breakwaters constructed at
different locations in ltaly.

1930~32 + 6.0

Figure 10-13. Cyclopean block breakwater designed


by Sainfiou.
1933"-35

for breakwater construction as early as the


1840s, and in 1886, Kinipple proposed that
the concrete caisson reinforced by iron
members be used in breakwater construc-
tion. Metal caissons were used in Bilbao,
Spain in 1894 and they were later adopted
for breakwater construction in several other
ports. Concrete caissons were used for
breakwater construction in Barcelona,
Spain and other ports in Europe. In the
United States, reinforced concrete caissons
were employed in 1901 for breakwater con-
Figure 10-14. Cyclopean block breakwater at Mar-
struction on the Great Lakes. In Japan, the seille, France: Original structure and subsequent mod-
reinforced concrete caissons were used for ernizations.
:a
Breakwater Design 965

Wove-dissipoting
Holyheod
Mod:"
JJiJ
+13.5

:tJ?J'
Jj 12 +5.5 tO

~ Fih
Ymuiden Mormugoo Colombo +3.5

.Ali.
+4.2

+9.2
Buffolo Civitovecchio
+3.8

-75

Figure 10-15. Breakwaters with wave-dissipating blocks.

+ 5.5 breakwaters armored with concrete blocks.


Although the wave-dissipating concrete
blocks were usually placed to reinforce the
damaged breakwater in some breakwaters,
they were incorporated into the initial
design.
Figure 10-16 shows the Agha breakwa-
ter constructed in Algiers, which has a wave
Figure 10-16. Wave screen at Agha breakwater, Al-
screen (i.e., a vertical wall that reduces wave
geria.
transmission through the breakwater). This
breakwater and a composite breakwater
the first time in 1907 in Kobe. U se of the with wave-dissipating blocks are almost
reinforced concrete floated-in caissons pro- identica!. However, based on its design con-
moted further development of composite cept, this type of breakwater is considered
breakwaters worldwide. to be a rubble-mound breakwater having a
large wave crown (screen).
10.1.4.4.3 wave-dissipating blocks
The performance of the composite break- 10.1.4.5 Revival of the
waters is enhanced greatly by placing the Rubble-Mound
wave-dissipating blocks in front of the ver- Breakwater
tical wall. Several examples are illustrated
in Figure 10-15. The original wave- The development of breakwaters, which
dissipating blocks have been of rectangular started with the construction of a gentle
geometry, similar to those used for the ar- rubble-mound breakwater, led to the con-
mor layer in the rubble-mound breakwa- struction of the low-mound composite
ters. Therefore, the cross section of these breakwaters worldwide. However, low-
breakwaters looks similar to rubble-mound mound breakwaters suffered from various
966 Breakwater Design

types of damage, and in Europe, damaged


composite breakwaters were converted into
rubble-mound breakwaters. The following
are some characteristic examples.

10.1.4.5.1 Failure of the Catania


breakwater
The composite breakwater built at
(b)
Catania, Italy failed during construction be-
tween 1930 and 1931. The failure was
caused by insufficient interlocking of the
Cyclopean blocks. The breakwater was sub-
sequently converted into a Marseille-type
rubble-mound breakwater in steps as indi-
cated in Figure 10-17.

10.1.4.5.2 Failure ofthe Leixoes


breakwater
The original breakwater at Leixoes,
Portugal was one of Marseille-type rubble-
mound construction. During its lifetime this
(d} +8.5 +

~
breakwater was redesigned and rebuilt sev-
eral times. The history of this breakwater
reconstruction is illustrated in Figure
10-18. The breakwater which replaced the
original one in 1932 was a composite-type
breakwater. It failed during construction
between 1934 and 1936 and was eventually Figure 10-17. Reconstruction ofrubble-mound
breakwater at Catania, ltaly: (a) original section of
replaced by a rubble-mound breakwater breakwaters; (b) damaged breakwater section (1933);
comprised of large concrete blocks. (c, d) subsequently redesigned and then reinforced
sections.
10.1.4.6 Rubble-Mound
Breaklvaters;Recent
Developments
10.1.4.6.1 Rubble-mound and manuals [e.g., the U.S. Army Corps of
breakwaters armored Engineers' (1984), Shore Protection Man-
with shape-designed ual, in which the Hudson formula was in-
concrete blocks troduced as the standard design method for
Recent developments in rubble-mound armor layer design]. In the past 10-15 years
breakwaters are largely based on using the design ofrubble-mound breakwaters has
shape-designed concrete blocks. Many suc- been substantially enhanced by use of
cessful rubble-mound breakwaters have computer-assisted numerica! analysis and
been constructed using armor layers com- sophisticated physical modeling technology.
prised of such blocks. The modern design This resulted in the design and construc-
methods for rubble mound breakwaters are tion of economica! rubble-mound break-
presented and discussed in several books waters. One example of improved design of
Breakwater Design 967

Old ( 1884"-' 92) +11.2

Executed

Figure 10-18. Reconstruction of rubble-mound breakwa-


ter at Leixoes, Portugal.

(o)
Wove Woll

16-20tStone
9-20t 112-Gt
Core Material ~ -- 4 -25.0m

··.·.·.. ·.·.·.·.··:.·.
. ·· ... : .. ·_::
·.'

(b) Two Loyers 90r Blocks


Bottom Loyer"Robloc'
Top Loyer "Antifer·

Figure 10-19. Sines, Portugal breakwater: (a) as constructed; (b)


as rebuilt after damage. [From Bruun (1985).]
968 Breakwater Design

such structures is the breakwater con- This breakwater has a large berm in its
structed at Sines, Portugal. front part. The quarry stones that have been
Figure 10-19 shows the typical cross sec- used for breakwater construction are not
tion of the Sines breakwater. This is a typi- very large. Such design allows for some
cal rubble-mound breakwater protected berm deformation which eventually ends up
from wave attack by the shape-designed forming an equilibrium slope. Berm break-
concrete blocks. Note that the cross section waters like this have been built worldwide.
of this breakwater is quite small even Many studies have been carried out on berm
though the water depth there exceeds 30 m breakwaters (Willis et al., 1987; Baird and
and the design significant wave height ex- Hali, 1984; Fournier et al., 1990; Burcharth
ceeds 10 m. The shape-designed concrete and Brejnegaard-Nielsen, 1986; Burcharth
block Dolos has been employed there. This and Frigaard, 1988). Note that the berm
block has high interlocking strength which breakwater resembles older rubble-mound
enables a more economica! breakwater con- breakwaters (e.g., the Plymouth break-
struction by reducing the amount of re- water).
quired materials.
Despite its robust design, this break- 10.1.4.7 Recent Deve/opments in
water suffered serious damage in 1978. The composite Breakwaters
breakdown of Dolos blocks is thought to be
one of the main causes of breakwater fail- Figure 10-21 illustrates one of the first
ure; Dolos blocks are relatively weak al- modern breakwaters designed by Hiroi and
though their interlocking strength is high. built in 1897 at Port-of-Otaru, J apan. Many
The redesigned cross section of the re- breakwaters constructed in Europe around
built Sines breakwater has a rather flat this time were either rubble-mound break-
slope and the armor layer was made from waters or composite breakwaters with block
the low interlocking blocks (Fig. 10-19b). masonry vertical wall. The technology in-
The new cross section very closely resem- troduced in Japan was related primarily to
bles the 19th-century rubble-mound break- the composite breakwaters. It was subse-
waters armored with concrete blocks. quently developed into the currently used
After the failure of the Sines breakwater composite breakwaters with vertical wall
and severa! more breakwaters of similar comprised from concrete floated-in caissons.
design, a major effort of breakwater design- In Italy and other Mediterranean countries,
ers was directed at improving the design caisson-type breakwaters were being devel-
method of the rubble-mound breakwater oped gradually based on the technology
and experimental techniques. New im- available at the end of the 19th century.
proved design methods have been devel- The development of composite breakwaters
oped. The latter are summarized in re- following 1945 was rapid due to advance-
cently published books and manuals [e.g., ments in concrete technology, development
CIRIAjCUR (1991)]. The modern break- of sophisticated design procedures for con-
water design methods usually include van crete structures, and significant progress
der Meer's new formula for designing the made in marine sea construction technol-
armor layer (van der Meer, 1988b, 1991). ogy. Eniployment of large working vessels
and high-capacity floating cranes helped
10.1.4.6.2 Berm breakwaters significantly reduce the time required for
placing the caisson, the rubble mound, and
Figure 10-20a shows the cross section of the heavy armor blocks.
a breakwater built in Racine, Michigan. The current status of composite break-
Breakwater Design 969

water technology is summarized as follows breakwater and the design method for the
(Tanimoto and Takahashi, 1994): breakwater's upright section.

10.1.4.7.1 Design method of 1.0.1.4. 7.2 Horizontallv composite


conventional composite · breakwaters
breakwaters
The composite breakwater that includes
The design technique for composite wave-dissipating blocks is an improved ver-
breakwaters is nearly established. It in- sion of the conventional composite break-
cludes the analytical procedures for water and is now being constructed,
determining the wave forces acting on the frequently, especially in breaker zone.

Col
Loke Side Horbor Side

(b)
Loke Side Horbor Side

Figure 10-20. Berm breakwater at Racine, Michigan: (a) stages of deformation;


(b) original cross section. [From Montgomery et al. (1987).]

o
727

12 tf

unit: m
Figure 10-21. Breakwater at Otaru, Japan.
970 Breakwater Design

10.1.4.7.3 New caisson-tvpe the wave force increases the caisson stabil-
breakwaters ity. The particular advantage of this type of
breakwater is that the wave force vector
Many new types of breakwater have been
passes through the center of the circle,
invented in order to mitigate the drawbacks
thereby increasing the caisson's overturn-
associated with the conventional composite
ing stability. This type of construction is
breakwaters.
A new caisson-type breakwater with a expected to exhibit high-efficiency perfor-
relatively shallow
seaward perforated wall has been invented mance in areas with
by Jarlan (1961) and, subsequently, many water and yet high waves. The practica!
breakwaters of this type have been con- example of this type of breakwater is de-
structed. Another innovative type of break- picted in Figures 10-23b and 10-23c.
water includes the incorporation of a sloped Figure 10-24 illustrates a cross section
front wall (e.g., the sloping top, trapezoidal, of the bay mouth composite breakwater
and semicircular caissons). Figure 10-22 constructed in Kamaishi Bay, Japan .. The
illustrates a perforated wall breakwater maximum depth ofwater at the bay's mouth
with vertical slits, constructed at Port-of- is 63 m, which makes this breakwater the
Yobuko, Japan. In this structure the waves world's deepest. The breakwater is com-
can enter and leave the wave chamber dis- prised of a substantial rubble mound, and
sipating their energy in the process. Fur-- an upright section made up from the con-
thermore, this breakwater has a pair of side crete floated-in semitrapezoidal caissons.
openings that allow the water to pass The trapezoidal geometry of the lower part
through, thus providing for efficient water of the caisson is required for better caisson
exchange. The wave-dissipating caissons are stability. This also helps to reduce the ec-
usually employed where waves are rela- centric load acting on the rubble mound.
tively small. Hence, the forces acting on the The caisson's upper part has a wave-
caisson's members are also relatively small. dissipating system consisting of double-
This type of construction meets the need for horizontal-slit walls. In general, the cais-
low reflectivity and high seawater exchange sons of trapezoidal geometry suit the deep
efficiency. water sites.
Figure 10-23a illustrates the conceptual Figure 10-25 illustrates the conceptual
design of a semicircular caisson breakwater design of the dual-cylinder caisson-type
recently developed in Japan. In this break- breakwater and several practica! examples
water the vertical downward component of of this type of construction built at several

Harbor Side 13.0


15 10.0

.-::.: :'·::··:. _: .'· .. :·


. .. . . :. :. ':.

unit: m
Figure 10-22. Perforated wall caisson breakwater at Yobuko, Japan.
Breakwater Design 971

(a) (b )

(c)

Figure 10-23. Semicircular caisson breakwater at


Miyazaki, Japan: (a) artistic impression; (b) breakwa-
ter sections at prefabrication yard; (c) installed at
construction site.

19 ·0 Harbor Side
+3.0

-60.0

Figure 10-24. Deep water breakwater at Kamaishi, Japan.


locations in Japan. Similar to the previous for the dual-cylinder caisson breakwater is
example, this type of construction also has almost fully established, and the successful
a deep water application and is able to demonstration test of this structure has
dissipate energy of the large waves. This been carried out at Sakaiminato, Japan
breakwater caisson consists of inner and (Tanimoto et al., 1992).
outer cylinders. The outer cylinder wall is a In addition to the above-described cais-
kind of shell stmcture that is able to with- son breakwaters, the new systems, such as
stand forces generated by large waves. The curved-slit caisson breakwater and others,
upper part of the outer cylinder consists of have been constructed (Okada et al., 1990).
a perforated wall. The space between the In Sakata Port, Japan a wave-power-
inner and outer cylinders constitute the extracting caisson breakwater was built.
wave chamber that forms the wave- This project represents the unique concept
dissipating structure. The design procedure in the breakwater development and demon-
972 Breakwater Design

(a)

Figure 10-26. Promenade breakwater at


(b )
Wakayama, Japan.

strates the feasibility of converting the wave


power (Takahashi et al., 1992a).
Public access to breakwaters is usually
prohibited because of the potential danger.
However, some of the breakwaters have
been designed for public access. For exam-
ple, the breakwater in Brighton Marina,
United Kingdom was designed to have a
(c)
pedestrian promenade deck constructed on
top of the breakwater. Figure 10-26 shows
a promenade breakwater constructed in
Wakayama Marina City, Japan. This struc-
ture was specifically designed with the pur-
pose of enhancing the amenity and the local
landscape. Because the future construction
of such breakwaters is expected to grow,
the safety of operation of these breakwaters
must be of paramount importance (Endo
and Takahashi, 1994). ·
(d )

10.1.4.8 Recent Failure of


Seoword Stde Offshore Vertical
Breakwaters
Recent developments in design methods
- 250 and construction technology has resulted in
unit · m
more reliable designs and construction of
vertical and composite-type breakwaters.
Figure 10-25. Dual-cylinder caisson breakwater: (a) However, not ali recently constructed
artistic impression; (b) installation of the breakwater
unit at the Sakai-Minato Port breakwater, Japan; (c)
breakwaters happened to be stormproof.
breakwater constructed at the Nagashima Port, Japan; The breakwaters are commonly designed to
(d) breakwater at Shibayama, Japan-typical cross withstand waves having a return period of
section. 50 years. However, a high probability exists
Breakwater Design 973

that higher waves than the design ones This breakwater is comprised of rein-
could attack the breakwater. Consequently, forced concrete caissons 21 X 24 X 16 m
the probabilistic approach to determining (L X B X H) founded on rubble foundation
the design wave height can be more reli- composed from random 30-300-kg stones.
able. It is expected to be introduced in the The rubble-mound base is protected by con-
near future to enable quantitative evalua- crete armor blocks weighing 8 tonnes. On
tion of the breakwater failure probability February 16, 1991 during a storm, the
during the design stage (Burcharth, 1989; largest ever recorded wave, 9.94 m high,
Takayama and Fujii, 1991). attacked northern Japan. Naturally, this
The following case history illustrates the wave exceeded the design wave. Subse-
most recent failure of the offshore breakwa- quent to the storm, the breakwater suffered
ter constructed at the Mutsu-Ogawara Port four types of damage that are typical for
that occurred in February 1991 (Hitachi, breakwaters of composite construction. They
1994). It illustrates some problems associ- are as follows:
ated with the current design practice. The
Mutsu-Ogawara Port is located in the (a) Large-scale scouring in front ofthe break-
water
northern part of J apan and faces the Pacific
(b) Meandering sliding at the northern end
Ocean. Figure 10-27 shows the plan of the
(c) Scattering of wave-dissipating concrete
breakwater. From this plan it is seen that
blocks and caisson failure at the concaved
the two wings of the breakwater form a section due to impulsive breaking pres-
concaved corner portion which is covered sures
with the wave-dissipating blocks designed (d) Scouring underneath the caisson at the
to reduce the wave pressure there; that is, southern breakwater head
the concave part of the breakwater is a
horizontally composite type. Scouring that occurs in front of the break-

~ Oomoged Port
N -= <

.· ...
. . . .

Figure 10-27. Plan view ofMustu-Ogawara Port, Japan. [From Hitachi (1994).]
974 Breakwater Design

Coisson No.46 ~
L
m
8
m
H
21.0 X 24.0 X 16.
5":1!
Armor Stones I,OOOkg
____________:l~9- --- ___ b:;J

Figure 10-28. Mutsu-Ogawara Port, Japan: meandering sliding of caissons at the


northern end of the breakwater. [From Hitachi (1994).]

Sea Side

Concave Section 1-8 Section 150m

Seo Side Horbor Side


Foot Protection locks

Figure 10-29. Mutsu-Ogawara Port, Japan: scattering of artificial


blocks and damage to the breakwater caisson. [From Hitachi (1994).]
Breakwater Design 975

water has resulted in removal of seafloor However, the previous storm which oc-
sand up to 1-2 m. The latter caused settle- curred in 1990 had scattered the armor
ment and deformation of the rubble mound blocks, which were then further scattered
toe; however, no direct damage to the main by the recent storm. The transition portion
caisson body has been observed. Two deteri- of the breakwater was later extended to-
orative consequences of the seafloor erosion ward caisson No. 7, which slid about 10 m
should be pointed out: from its original location and had its upper
wall completely destroyed. Caisson No. 8
• The breakwater was designed to resist slid only about 1 m. This failure was obvi-
forces generated by the breaking wave ously caused by the impulsive wave pres-
which was limited by the original water sures resulting from the insufficient block
depth. Increase in the water depth due to covering of this caisson.
seafloor scour resulted in increased depth The damage sustained by the breakwater
of water and consequently in the wave at this section also demonstrated the weak-
height. ness of the concrete blocks (50-t tetrapods)
• Due to settlement and deformation of the used at the transition from the conven-
rubble mound, the interlocking of armor tional to the horizontally composite type of
concrete blocks in the concaved section construction (as at the breakwater head
perhaps was loosened, which may be one section). In contrast, the concrete blocks at
of the reasons that these blocks were scat-
the breakwater trunk held firmly even
tered there.
though the actual wave was much higher
than the design wave.
Seventeen caissons having a total length of The foot protection blocks at the south-
360.4 m slid from 0.14 to 4.95 m, forming a ern breakwater head were scattered and
plane view similar to that of a meandering the rubble stones there were washed away
river (Fig. 10-28). The caissons slid because from underneath the caisson ·(Fig. 10-30).
the actual waves exceeded the design wave
height and, therefore, imposed heavier than
design loads on the caissons. Forming a
meandering shape in the breakwater align-
ment is typical in a sliding mode of failure
that is caused by the refracted waves pro-
duced at the breakwater head; this is dis-
cussed later in this chapter.
If the composite-type breakwater is com-
prised ofboth the conventional and horizon-
tally composite type of construction, then
the breakwater caissons in their transition
part from the conventional composite type
to the block covered type (horizontally com- A-A' Section B-a· Section

posite type) must be reinforced to ensure No.38

safety. Insufficiently protected by wave- SOcm-lm

dissipating concrete blocks, vertical walls I00-300kg


may be subjected to excessive impulsive
pressures. This is why caisson No. 8 (Fig.
Figure 10-30. Mutsu-Ogawara Port, Japan: scour-
10-29) was designed to withstand larger ing underneath the breakwater caisson. [From Hitachi
wave forces than its neighboring caissons. (1994).]
976 Breakwater Design

This damage was probably caused by waves


coming from a westerly direction, where the
caisson edge on the harbor side acted as an
edged corner against the waves; the armor waters to low-mound ones, as well as from
blocks of the rubble mound at this edge has rubble-mound breakwaters to composite
breakwaters.
a high probability of suffering from scatter-
ing and scouring, which, in the worst case, (b) Historically, the breakwater development
results in caisson tilting andjor sliding. was strongly affected by the progress
made in design and construction tech-
nologies.
10.1.4.9 summary of Breakwater
(c) The failure of the new types of break-
History waters has always resulted in returning
Table 10-2 summarizes the history of to the old breakwater designs.
breakwaters, especially those built in the
19th and 20th centuries. Important aspects It can be concluded that the lessons learned
to note are as follows: from the breakwater design, construction,
operation, and especially failures, in combi-
(a) The general trend of breakwater develop- nation with new information obtained from
ment is from gentle slope breakwaters to the recent extensive investigations, demon-
vertical ones (i.e., from flat-slope rubble- strated that both sloping and vertical types
mound breakwaters to steep slope ones) of breakwater can be designed with a high
and from high-mound composite break- degree of confidence and reliability.

Table 10-2. Summary ofhistorical development ofbreakwaters.

Type 1800 1850 1900 1950 2000

/li
Upright Dover (1847)

' '
(Low) j Low Mound __l_,_ Cellular Block
!! . /
H1gh Mound '
Cyclopean Block
Caisson New Caissons
Composite

(High)
i
i'
1 /
/
Cherbourg
(1830)
/

!
1

!' Failures
Return to
"
(Wave Dissipating)
Horizontally
Composite type

(Steep slope) A Uniforn::tep cirave screen) Dolos


/ j / Placement Tetrapod / J.lJ
Rubble Failures
mound Cherbourg Concrete Block : / Return to
(1781)---...;......_ .....--"' "-.._ ! Mild Slope

(Mild slope)
!
:
Plymouth
(1812)
"""' Marseille type
1
t .
Berm Breakwater
1 :
1

Cement 1824 :Tetrapod 1949


Reinforced concrete 1867? !Model Experiment 1930's
Note ilribarren(1938), Hudson(1958), van der Meer
:Hirol (1919) Sainflou (1928) (1988)
iGoda(1973)1CCE (1950) '
iWorld War II (1939-1945)
Breakwater Design 977

10.2 DESIGN OF CONVENTIONAL The rubble foundation is underlined by vinyl


VERTICAL BREAKWATERS sheets which are used to prevent the sand
of the seafl.oor from erosion by wave action.
The rubble foundation is extended 8.0 m
seaward by placing the 1.5-m-thick gravei
10.2.1 Examples of mat. This mat is used for additional protec-
conventional Vertical tion against scouring. On the harbor side,
Breakwaters the armor stones are underlined by a vinyl
mat. The water depth above the rubble
10.2.1.1 Caison-Type Breakwater mound including the armor layer is 10 m at
the LWL, and the berm width of the rubble
Figure 10-31 shows a typical cross
mound is 12.8 m.
section of a conventional caisson-type
This particular breakwater is designed to
breakwater. The upright section of this
withstand the design wave of Hv = 11.6 m
breakwater consists of a concrete caisson
(cross section 21.5 m X 27 m) installed on a (significant wave height Hs = H 113 = 6.66
3.5-m-thick rubble foundation with a 4.0- m), significant wave period T8 = T 113 = 13
m-thick concrete capping crown. The depth s, and wave approach angle 8 = 22° at a
of the caisson is 11.0 m and the maximum water depth h = 21 m. The design wave at
height of the concrete crown is 4.6 m. The the breakwater site was evaluated using a
caisson is divided into 5 X 6 cells by 20-cm- wave transformation calculation, with wave
thick inner walls and 40-cm-thick outer pressures on the caisson being evaluated by
walls. These cells are filled with sand and the extended Goda pressure formula. Dis-
capped with concrete. Subsequently, the cussion on Goda's formula is provided fur-
concrete superstructure (crown) is cast on ther in Section 10.2.3.5.
top of the caisson.
Foot protection blocks are placed in front 10.2.1.2 Block Masonry
of and behind the caisson to prevent the Breakwater
scour of the rubble foundation while shape-
designed concrete blocks are installed to act Figure 10-32 illustrates a typical cross
as the armor layer of the rubble foundation. section of a block masonry breakwater. This

Seoword Side
Horbor Side

Figure 10-31. Noshiro Port, Japan: typical cross section of a vertical-type


breakwater. [From Kataoka and Saida (1986).]
978 Breakwater Design

is a secondary breakwater placed behind the transmitted waves behind the breakwa-
the principal offshore breakwater. Hence, ter. Wave reflection is sometimes a problem
the design significant wave is small (i.e., because it creates additional agitation. Min-
H 113 = 1.8 m and T 113 = 14 s). The break- imization of wave transmission is especially
water structure is comprised of four basic important in breakwater design because the
parts: rubble foundation (mattress), block- principal function of the breakwater is to
work wall, cast-in-situ concrete cap, and prevent wave propagation from occurring,
armor stones for protection of a rubble bed- thereby creating a calm water area behind
ding. The rubble foundation is made by ex- the breakwater.
cavating the sand bed; the cross section of The amount of wave reflection and trans-
the blockwork wall includes three concrete mission are usually determined by the
interlocking blocks capped by the cast-in- wave-reflection coefficient K R and wave-
situ, 2.1-m-thick concrete crown (super- transmission coefficient Kr, which are de-
structure). fined as follows:
KR =HRIHI
(10-1)
Kr=HrfHI
10.2.2 wave Transmission and
Reflection by Vertical where H1 is the incident wave height, HR
is the reflected wave height, and Hr is the
Walls transmitted wave height, all of which usu-
ally correspond to the significant wave.
When waves act on breakwaters, some en-
ergy of the incident waves is dissipated and
some of the remaining energy is reflected
10.2.2.1 wave rransmission
and generates reflected waves in front of Transmitted waves are caused by wave
the vertical-type breakwaters; the rest of transmission through the structure and
the wave energy is transmitted and yields overtopping. The transmission coefficients

Figure 10-32. Akasaki Port, Japan: typical cross section of a


block masonry breakwater. [From Kataoka and Saida (1986).]
Break:water Design 979

by both causes are respectively denoted as


Krt and Kr 0 , with the total transmission
coefficient Kr being expressed as

KT = (K2Tt + K2To )o.s (10-2)

The transmitted waves created by the


overtopping waves have a complicated form
with high-frequency components. They are o
produced by waves generated at the lee,
which result from the impact of the fall -0.1
of the overtopping mass. Therefore, in
general, the transmitted wave height and - 0.2 .___.J...__...L_._..L.._ _L_....l

period are different from those of incident o 0.2 0.4 0.6 0.8 1.0
waves; that is, the wave period of transmit- dlh
ted waves is generally smaller. Figure 10-33. Nomograph for determining the fac-
Another interesting phenomenon is that tor {3 = f(djh).
transmitted irregular waves change charac-
teristics as they propagate over a long dis-
diagram in Figure 10-33 and h' is the dis-
tance; for example, the distributions of wave
tance from the design water level to the
height and period vary with the distance
bottom of the caisson.
away from the breakwater.
Although Eq. (10-3) is based on regular
Wave transmission by vertical wall
wave tests, the relations are still applicable
breakwaters is mainly by overtopping;
to the transmission coefficient of irregular
therefore, the ratio ofthe breakwater's crest
waves with a significant wave height. Most
height h c to the incident wave height H 1 is
breakwaters in Japan are designed with a
the principal parameter governing the
relati~e crest height hc!H113 = 0.6, where
wave-transmission coefficient. Based on
H 113 1s the design significant wave height.
regular wave tests, Goda and Kakizaki
The transmission coefficient calculated from
(1966) proposed the following equations for
Eq. (10-3) is then equal to about 0.2 for the
the determination of the transmission co-
typical conditions of djh = 0.6 and h' jh =
efficient for vertical breakwaters:
0.7. Figure 10-34 represents the transmis-
For f3 - a< hjH1 < a- f3 sion coefficient for vertical wall breakwaters
obtained from Eq. (10-3).

10.2.2.2 wave Ref/ection


Waves that usually exist in front of the
vertical breakwaters are standing waves,
(10-3)
reflected by the wall. The reflection coeffi-
cient of vertical wall breakwaters is gener-
For hjH1 ;::: a - f3 ally high, although less than 1.0, usually
due to effects of the rubble-mound founda-
tion andjor wave overtopping. The wave
reflection coefficient, K R, is considerably re-
duced when breaking waves act on the
breakwaters. Figure 10-35 shows the re-
where a = 2.2 and f3 is obtained from the sults from two series of experiments using
980 Breakwater Design

1.0
~ Hr 1- B--j fc HI

'"" ~
dlh=0.7
d!h=0.5
~--.0.8
~ ~~ dlh=0.3
c 7Ji'Wi<W)' ~ NWR//

-~0.6 ~'
-
'Qj
o
"' ~ ""'-
dlh=O

u
c:0.4
~ ~~
.Q
CI)
CI)

-~0.2
~"'
~
~
~"" r---...
c:
o ~ t:--_'-...... 1---
~ ~
o-2.5 -20 -1.5 -1.0 -0.5 o 0.5 1.0 1.5 2.0
Re 1otive Crest El evot ion, hc 1 H1

Figure 10-34. Transmission coefficient for a vertical breakwater. [From Goda (1969).]

various wave conditions which are repre- height of the upright sections to the water
sented by the incident significant wave depth, hcfh, is changed in the second
height H 113 and the wavelength L 113 corre- series.
sponding to the significant wave period T 113 The x axis in Figure 10-35 is taken as
(Tanimoto et al., 1987). In the first series, the ratio of the incident significant wave
the relative thickness of the rubble-mound height to the depth in front of the upright
foundation to the water depth, djh, is pri- sections above the armor layer, H 113 jd,
marily changed, whereas the relative crest which is considered to be a principal factor

1.0 r----r-.,.---,--.--.,--.---~~----l Series 1 hei h =0.23 8 - 0.478


- · - · - { d 1 h =0.350- 0.762
hiL11 3 =0.0964~0.241

c:
~ 0.61----+---==~-==-'...___,;:~,;..,_,.;:._.::=;".......,_ţ.:,.
:;:
Q;
o
u
-~ 0.41-----t----+---+---l---.l....----L----1---~

tt ')-VL
0 Series [

~ o2 L_jf-o-~-:~-30+--i- ~_-+:-ho_~-~-+d/-~-~4-=04·- ~-:~- -_-L


__ _ _j

; 2 ~
1:50 "1:100
0.102 --o-- g ct ----

0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6


H113l d

Figure 10-35. Reflection coefficients of vertical breakwaters. [From Tani-


moto et al. (1987).]
Breakwater Design 981

representing the degree of wave breaking. u wave frequency (u = 2TTjT, where T is


=

The average relation from Series I data is the wave period)


indicated, as are the upper and lower limits t = time
of Series II data obtained using different h = water depth
values of the relative water depth, hjL 113 . w 0 =specific weight ofwater; w 0 = p0 g, where
Test results illustrated in Figure 10-35 in- p 0 is the specific density and g is the
dicate that K R tends to decrease with in- gravity acceleration.
creases in H 113 jd. Another important pa-
rameter is the relative crest height to the Equation (10-4) indicates that the nondi-
incident significant wave height, hc/H113 , mensional value of positive peak pressure
because it strongly influences breakwater (nondimensional pressure intensity
overtopping. Series II results refl.ects this, pjw 0 H) is 1.0 at the water surface and
although the crest height is expressed in decreases along the vertical wall down to
the dimensionless form using h instead the bottom, being equal to 1jcosh kh there.
of H 113 . Obviously, for shallow-water waves, the
wave pressure is constant along the vertical
wall in the direction of water depth. This
10.2.3 wave Forces on vertical equation also implies that the longer waves
Walls produce the larger wave force on the verti-
cal wall when the water depth and wave
10.2.3.1 standing-Wave Pressure height are the same. It is important to un-
derstand Eq. (10-4) because it expresses
10.2.3.1.1 Sma/1-amp/itude-wave the basic concept of the wave pressure on a
theorv vertical wall.
The wave pressure on a vertical wall
(Fig. 10-36) can be estimated by small- 10.2.3.1.2 Non/inear wave theorv
amplitude-wave theory for a standing wave; Figure 10-37 shows the time history of
that is, wave pressure on a vertical wall obtained

p(z) = w0H (
cosh k(h +
cosh kh
z)) cos ut (10-4)
from model tests. In these tests, the wave is

H1 = 42.8cm
where
T= 3.04s
p(z) = wave pressure along the vertical wall
p h = 104.4cm
z = vertical axis along the vertical wall (z = Woff1
- h at the bottom) 1.0
H = wave height
o~~~~~L--L-----L~------L--
k = wave number (k = 2TTjL, where Lis the
1.0
wave length)
0~==~~~~---L-~======~
z 1.0

~=~ c:.:.:.----:::::.> <


z =o --- . ~
o~~~~L-~-------L--~----~
1.0

P(Z)

-h " ' = = " ' - - - - - - - - - - - -


Figure 10-37. Typical sample from model tests
Figure 10-36. Standing wave on a vertical wall. showing standing-wave pressures.
982 Breakwater Design

not sinusoidal and has double humps at the opposite occurs, which is a unique
positive pressure and a relatively flat pro- feature of wave pressure in deep water.
:file at negative pressure due to nonlinear-
ity. Goda and Kakizak.i (1966) calculated 10.2.3.2 Limitations of wave
the pro:file of nonlinear wave pressures us- Pressure Calculations
ing a fourth-order perturbation method, as Although small- and finite-amplitude
illustrated in Figure 10-38. These pro:files wave theory can directly evaluate wave
show the nondimensional averaged wave pressure on a simple vertical wall, the wave
forces at the positive peak (wave crest) and pressure on the vertical wall of breakwaters
negative peak (wave trough), or Pmaxfw 0 Hh is not easily determined due to following
and Pminfw 0 Hh, respectively. Interesting reasons:
features of the nonlinear wave pressures
are as follows: (i) Wave Breaking and Sea Bottom Slope.
Wave breaking has not been included
(i) Clearly, the average pressure intensities in wave theories; however, the design
vary .with respect to the wave height wave for breakwaters is usually a
and, in some instances, exceed values ob- breaking wave. In particular, when the
tained by the small-amplitude-wave sea bottom slope is steep, impulsive
theory, especially when hjL is small. pressures on a vertical wall will be
generated due to wave breaking.
(ii) In general, the average pressure inten- (ii) Rubble mound. A rubble foundation
sity at the wave crests increases first, that is usually used for composite
then decreases, and finally increases breakwaters increases the wave pres-
again with the increase in wave height. sure intensity, which, in the worst
(iii) As commonly accepted, the absolute cases, generates impulsive pressure by
value of the pressure intensity at wave causing wave breaking on the mound.
trough is usually smaller than that at (iii) Overtopping. The crown of vertical
the wave crest. However, when the rela- breakwaters is usually lower than the
tive water depth hfL is larger than 0.25, crest elevation of the design wave;

':;. c
.....~--~--+-"r';"Ji:J:
~ ~----+~---j-gl--o-~:--:1--~-~-e-+---+---+-;~
(?-~- h/L=0.05
·' 1

s h -
:: ~
1.5

.2 ~
h ' -· 006 11 •
• o..~
nE 'mln
.._ 5 ~~~=~saos 1
.,- -1.0 1----t- / - -- OJO 1.0 .,
~ -0.9- ~r- - ---- OJ 2 -0.9 ~
~!>! -0.8 l--f--l---jf----jf----jr-Jt r-- ___ 0 J6 -l 0.8 ~~
I..L:J-~;;;;;~2:r-~~~-t..Q~.~I4d--
3 -o.7- 3
o.6 :. :
--=::r-- ____o.ta
:lJ. :
.)!!;V~ a1

-o.6t~ht~l.·~o~.2!5 :f~n,~L~~o~.z~~~~Ţ_~== F==t~f~f~-jo.2o- 0.5


n. -0.5 - 03 _.- .L L o'7'r:9/
--"-- ___ , -- 0.25 n.
<Cw

e ?:5i'j
o~

z%'
t= -0.3 - . - :=::::::;:;;--
~ -0.4~~:~5~~~;~~§~~ g~~§~t~~;~~~~~~~03 -0.3 w
-0.4 ~
~ =--::::;:::~ 2-0
(3 -02
w .
z -0.1,--
r--- - --
--:::: - -
~
-
~-
-0.1
02 Cii
o
tl.
0 ~ 1 11111.1 0
0.10 0.08 0.06 0.04 0.02 o 0.02 0.04 0.06 0.08 0.1 o
Wove Steepness, HIL

Figure 10-38. Forces generated by a nonlinear standing wave. [From Goda and
Kakizaki (1966).]
Breakwater Design 983

therefore, the breakwaters are nor- chest and stomach severely slap against the
mally overtopped. This usually reduceswater. Such a collision between the water
the wave pressure intensity, especially
surface and the body produces a strong im-
when the wave pressure is impulsive. pulsive water pressure on the body. Impul-
sive wave pressure due to a breaking wave
Due to (i)-(iii), the wave pressures on verti- is similar to this phonomenon (i.e., the colii-
cal wall breakwaters are detennined in most sion of the wave surface with the vertical
practica! cases from physical model tests or wall). Impulsive wave pressure is one of the
by empirica! fm:mulas fonnulated from most important problems in the design of
model tests. the vertical breakwater; it is regarded as an
Another problem in the calculation of inherent disadvantage. On the other hand,
wave forces concems the irregularity of if the nature and the generation conditions
waves. Until recently, the wave pressure on of this pressure are understood, it could be
vertical breakwaters has been evaluated by avoided.
using the significant wave height, which Figure 10-40 shows the waves impacting
usually produces a smaller value than the against the model of a vertical wall placed
actual wave pressure could be. It is on a horizontal bottom extended by a 1 : 10
presently popular, however, to employ the slope (Takahashi et al., 1992b and 1994).
maximum wave height when calculating the
design wave pressure, assuming that the
highest wave produces the highest wave
force. It should be realized, however, that
waves of equal height could generate forces
of different magnitudes due to their differ-
ent profiles. At present, this phenomenon is
not clearly understood and requires further
investigation.
The wave direction is another important
parameter in wave pressure calculations. If
the wave is nonbreaking, the wave pressure
in oblique seas can be calculated relatively
easily. However, the effect ofwave direction
becomes especially pertinent when the wave
is breaking; nonnally, this is evaluated ex- Figure 10-39. Example of a "belly-flop."
perimentally. In addition to the wave direc-
tion, the plane configuration of breakwater
causes complex phenomena associated with (a)
wave actions, which are discussed in
Section 10.2.3.7. T=3s
E
u
10.2.3.3 /mpulsive Pressure and
o-
o
Dynamic Response of a
Caisson
10.2.3.3.1 Jmpulsive-wave pressure
Figure 10-40. Infiuence ofberm width (B1 ) on wave
Figure 10- 39 shows a photograph of a pressure profile: (a) laboratory setup; (b) variation of
boy doing a "belly-:flop." Because his body is wave profiles due to variation in berm size, B 1 . [From
almost parallel to the water surface, his Takahashi et al. (1992b) and (1994).]
984 Breakwater Design

Figure 10-40. Continued


Breakwater Design 985

The depth of water in all model tests is amount of trapped air increases and the
h = 30.0 cm. Although the offshore wave pressure intensity decreases. In fact, for a
profile in aU tests is the same at T = 3 s wave impacting the wall after breaking (B 1
and H = 35.1 cm, the development of the = 200 cm), the impulsive pressure intensity
breaking condition for each wave is differ- is significantly reduced.
ent; it is largely dependent on the length of It should be noted that two types of im-
the horizontal part of the bottom, B 1. Fig- pulsive pressures exist depending on the air
ure 10-41 illustrates the corresponding layer: impulsive pressure occurring without
wave pressure profiles at the still water air entrapment is termed Wagner-type
level. From this figure it is seen that when pressure, whereas that with air entrapment
the wave is not breaking (B 1 = O), the wave is known as Bagnold-type pressure. The
pressure is not very impulsive. When the most severe impulsive pressure takes place
wave hits the wall just before breaking (B 1 in the transitional region between both
= 25 cm), it produces severe pressure on aforementioned types of impulsive pressure.
the wall without entrapping the air layer;
WAGNER-TYPE PRESSURE. As is shown in
under this condition, the wave pressure is Figure 10-42, a wedge hitting the water
increased by about 10w 0 H, where w 0 is the surface receives impulsive water pressure,
specific weight of water. When the wave which is Wagner-type pressure. Karman
starts breaking (B 1 = 50 cm), a small air (1929) investigated this type of pressure
layer is entrapped and compressed against when he studied the water pressure acting
the wall. This compression generates an im-
pulsive pressure equal to about 10w 0 H. As
the wave breaking is progressing, the 7r
(a) 2VNcotf3
vNt ~
Woves ( T= 3s ) Woter Depth h = 30cm
H=.35.1 Bottom Slope tona= l/10
1 . t .,
B,=Ocm~

(b)

o 2 3
1 (s l

Figure 10-41. Wave pressure profiles due variation


in berm size, B 1 , as depicted in Figure 10-40. [From Figure 10-42. Wedge impacting the surface ofwater
Takahashi et al. (1992) and (1994).] (Wagner-type pressure).
986 Breakwater Design

on the float of a seaplane during landing. (a)


He found that the force on the wedge in-
creases with decreasing attacking angle {3
or in the angle between the wedge and
water surface. Hence, the impulsive force is
at its maximum when the attacking angle
is zero; under these conditions, the impul-
sive force is equal to p 0 cw VN, where p 0 is
the specific density ofwater, cw is the speed
of sound in water, and VN is the decent
speed of the wedge.
Wagner (1932) investigated the same (b)
phenomenon and defined the pressure dis-

~k.,
tribution as a function of the attacking
angle and decent speed of the wedge (Fig.
10-42). He found that the impulsive pres-
sure reaches a high peak at the intersection
of the water surface and wedge surface,
with the resultant pressure distribution ex- Figure 10-43. Air compression with some leakage
tending as the submerged area of the wedge due to breaking-wave impact (Bagnold-type pressure).
extends. Therefore, the time history of
Wagner-type pressure is characterized by wave pressure that travels at cw and acts
a sudden rise and exponential decay in simultaneously with air pressure on the
pressure. vertical wall is felt some distance away from
Wagner-type pressure was recently the vertical wall.
termed "flip through" due to the occurrence The peak value of the Bagnold pressure
of a pile-up effect of water at the intersec- increases with the increase in the wave-
tion point (Fig. 10-42) (Cooker and front speed or with a decrease of the air
layer thickness. It is a common misconcep-
Peregrine, 1990). Wagner also pointed out
tion that the air layer contributes to induc-
this effect, especially for the case when the
ing a high impulsive pressure similar to
attacking angle is small. Goda (1967) called
that occurring in the explosion process. In
this type of pressure "single peak pressure"
actuality, the air layer acts as a cushion
and Lundgren (1969) and Horikawa and
that is reducing the impact force.
Noguch (1970) called it "ventilated shock."
Results of the most recent studies on the
BAGNOLD-'I'YPE PRESSURE. Bagnold-type impulsive pressures have been reported by
pressure is very popular among the break- Arami and Hattori (1990), Partensky (1988),
water designers for representing impulsive Kirkgoz (1991), Oumeraci et al. (1991),
wave pressures. Bagnold (1939) discussed Oumeraci and Kortenhaus (1992) and Chan
the effects of the air compression pressure (1994).
on the wall, and Mitsuyasu (1966) devel-
oped the pressure calculation method that 10.2.3.3.2 ca/cu/ation model of
includes damping factors due to some air impu/sive pressure
escape (Fig. 10-43). The latter is character- Figure 10-44 illustrates three impulsive
ized by a damping oscillation of air pres- wave pressure models developed by
sure, and sometimes only a single pressure Takahashi et al. (1983, 1992b, 1994). Ac-
peak occurs when the air escape is substan- cording to these investigators, when the at-
tially large. Note that the effects of the tacking angle {3 of the wave front is larger
Breakwater Design 987

Wagner Type Transition Type Bognold Type


VN VN V,.,

f3 > 8 8> f3 >o 0>(3


Figure 10-44. Three basic types ofimpulsive pressure. [From Takahashi et al. (1994).]

than the curvature angle 8 of the wave H = wave height


front, the Wagner-type pressure acts on the p 0 = atmospheric pressure
wall; when {3 is negative, the Bagnold-type h' = water depth at the bottom of wall
pressure arises; and when {3 is between hc = crest elevation of wall
these values, a transition-type pressure g = gravity acceleration
acts, generating an impulsive pressure sim- w 0 = specific weight of water
ilar to the Bagnold type. The wave pressure
can be evaluated using {3, the curvature From Eq. (10-5) it is obvious that intensity
angle 8, wave-front height l, and impacting
of the impulsive pressure increases as the
speed VN. wave front height increases and the amount
Tak.ahashi et al. (1983, 1992b, 1994)
of entrapped air decreases. For example:
found that the maximum average wave
when h = 22.5 m, h' = 17.5 m, hc = 5 m,
pressure intensity (p jw 0 H) that appears
H = 8 m, ka = 0.06, and k 1 = 0.5, the maxi-
in the transition region and its duration
mum average pressure intensity is equal to
time, T, can be approximated as follows:
about 8 w 0 H and the duration time T =
0.2 s.

10.2.3.3.3 tmpu/sive pressure due


to a steep sea-bottom
(10-6) slope
The wave that is traveling across a steep
where sea bottom generates impulsive pressure by
ka = air thickness coeffi.cient; ka is related to {3 forming a vertical wave front when it
and 8, and its minimum value is approxi- breaks. Mitsuyasu (1961) carried out series
mately on the order of 0.01-0.1. of the laboratory tests aimed at measure-
k 1 = impulsive height coeffi.cient; k 1 is the ratio ments of the wave pressure on a vertical
of the wave-front height l to the wave wall that is placed on a steep slope. Figure
height H; theoretically, it ranges from O to 10-45 illustrates some typical results of
1, although it is usually used as 0.4-0.9 nondimensional pressure as a function of
k m = added mass correction factor; for practica! the ratio of the water depth at the wall to
calculations k m is usually assumed equal the wave height in deep water, hjH0 • In
to 0.83 these model tests, h varied from 5 to 18 cm,
'Y = specific heat ratio; 'Y = 1.4 H 0 was equal to about 10 cm, the wave
h = water depth period varied from 1.02 to 2.57 s, and the
988 Breakwater Design

slope was equal to 1 : 15. When the test (ii) When H 0 jL 0 is a constant, the pressure
results depicted in Figure 10-45 are evalu- intensity peaks at the critica! water
ated together with those obtained at slopes depth, hM, which is dependent on the
at 1 : 30 and 1 : 50, the following conclusion slope angle tan a and H 0 jL 0 , that is,
can be drawn:

Ho = (0.59- 3.2tan a) ( H
hM 0
(i) The smaller H 0 jL 0 value and steeper Lo )
(10-7)
slope result ·m larger nond.imensional
pressure intensity. These factors obvi-
ously affect the mode of wave breaking.
Figure 10-46 illustrates different types (iii) When h is deep enough, nonbreaking
of wave breaking as functions of wave wave pressure appears, and when it is a
steepness and angle of seafioor slope. little larger than the critica! depth h M
Note that in the region of a spilling [i.e., (1.0-1.2)h M ], the Wagner-type im-
breaker, no significant impulsive pres- pulsive pressure occurs. When h is
sure occurs, whereas in the regions of slightly less than hM [i.e., (0.7-l.O)hM],
plunging or surging breakers, impulsive the Bagnold-type pressure appears.
pressures are generated. Clearly, the im-
pulsive pressures are dependent on the
projection of the wave front that is hit- Based on these results, impulsive
ting the vertical wall. The breaker-type pressure may occur under the following
wave can also be defined by the so-called conditions:
surf-similarity parameter; Iribarren
number (Ir = (tan a)(H0 jL 0 )(-l/ 2 ),
with a plunging breaker occurring when tan a
I = > 0.46 (10-8)
Ir > 0.46 (Battjes, 1974). r (Ho/Lo)0.5

Ho: Wove height in deepwoter h


0.7 < - < 1.2 (10-9)
11 Lo : Wove length in deepwoter hM
SLOPE 1/15

SYM BOL lb!Lo 0.07


o 0.0095 o
.....

7
• 0.019 )0.06
o 0.032 ~
t::. 0.067 Spilling wove

1/
~0.05

~"' ./
~t-- h
r/
1/
~0.04 - / f----
-....:: 1---
Cl
.5

1/
::l0.03
"'a.
c: Plunging wove
"'
!!!0.02
lfl
!:
~0.01
~urgi~1'-
o
0.4 o 0.02 0.04 0.06 0.08 0.10 0.12
h/Ho Seo bottom slope ton a

Figure 10-45. lmpulsive pressure due to wave Figure 10-46. Breaker types due to sea-bottom slope
breaking on a steep slope. [From Mitsuyasu (1961).) and wave steepness.
Breakwater Design 989

10.2.3.3.4 tmpulsive pressure due W0 c threshold weight in water against


=
to a high and 1 or Jong sliding (determined experimen-
mound tally for a given wave condition)
U0 and Pa = the total uplift force and the total
When wave forces act an an upright horizontal wave force, respec-
section having a high and/ ar long rubble- tively, calculated by the Goda for-
mound foundation, the wave pressure be- mula.
comes strongly impulsive. Figure 10-47
illustrates the example of variation in the Hence, p is not the actual wave pressure
wave pressure intensity, p, averaged over but rather the quasi-static wave pressure
the total height ofthe upright section, where intensity which allows examination of the
the horizontal axis represents the berm breakwater stability against sliding; p
width of the rubble-mound foundation rela- actually represents the shear force at the
tive ta the water depth, and the vertical bottom of the breakwaters upright section
axis represents the thickness of the rubble- including the dynamic response effect.
mound foundation, including the armor The variation of the pressure intensity
layer, relative to the water depth. (pjw 0 H) in Figure 10-47 is shown by the
Data for p are obtained from wall sliding equal value contour lines for the wave con-
tests carried out in a wave tank. In these dition in which hjL = 0.0712 and Hjh =
tests, different sizes of the rubble-mound 0.719. This example illustrates how strongly
foundation have been employed. The follow- the wave pressure intensity is infiuenced by
ing relation is used to obtain p from the both the height of the rubble-mound foun-
sliding tests results (Tanimoto et al., 1981): dation and the berm width, and also that
the breaking wave forces become very large
.u.Woc when the rubble-mound foundation is too
p(h' + hc) = U p (10-10)
1 + .u al a high andjor too wide. Under these wave
conditions, the horizontal wave pressure
where
(quasi-static pressure) exceeds 2.5w 0 H and
,u = coefficient of friction between the reaches a maximum when (h- d)jh = 0.6
wall and the rubble foundation and BMIL = 0.12.
The impulsive pressure coefficient, a 1 ,
1. o ,---.---.-.-------,--:-:-=-===--:-:-:-:-=-o=--.
h1L=0.0712, Hlh = .719
was derived from the aforementioned test
results; a 1 can be used ta determine the
critica! level of impulsive pressure, which
includes the effect of dynamic response of
the vertical wall and is used together with
the extended Goda formula that is de-
scribed further in Section 10.2.3.4.
0.3 f----1--+-+--+---1-+--+-t--+---1
0.2 1----+--+-+--t---1-+---+-t---t----i
10.2.3.3.5 Dynamic response of
vertical wa/1
0.1 1---+--t-t---+---t-+--+-t---t--1
o L_~~-L--L~-~~--L--L~ Although the impulsive pressure acts di-
o 0.5 1.0 1.5 2.0 2.5 30 3.5 4.0 4.5 5.0
BNI h
rectly on the vertical wall, the total upright
o 0.1 0.2 0.3 section of the breakwater dynamically re-
sponds ta this pressure as a part of an
Figure 10-47. Horizontal wave pressure intensity
elastic system that includes the rubble
obtained from wall sliding tests. [From Tanimoto et al. mound and a seafioor soil. This results in
(1981).] the significantly reduced shear force that
990 Breakwater Design

may cause the upright section sliding Another point of interest is that the cais-
(Hayashi and lmai, 1964; Horikawa et al., son moves significantly at a relatively low
1975; lto et al., 1966; Shimosako and Taka- frequency. The shear force is reduced due to
hashi, 1994). inertia force, and it peaks at the negative
The practica! example of the vertical wall peak of the inertia force. This is defined as
dynamic response to concentrated load is dynamic response of the caisson. From this
illustrated in Figure 10-48. The 22.5-m- numerical example it follows that the slid-
high, 20.0-m-wide wall, as shown in this ing force is expected to be less than 3 w 0 H,
figure, is placed on a 5.0-m-thick rubble as is demonstrated by Goda (1973a). An-
mound. The wall was analyzed by using the other conclusion is that the sliding force is
finite element method (FEM) based on Biot's approximately proportional to the momen-
equations. The wave force FH, inertia force tum ofthe impact force.
mxc, shear force F8 , and horizontal dis-
placement xc are indicated along with val- 10.2.3.3.6 Mitigation of the Effects
ues of the impulsive component PP and of the Jmputsive
static component Ps of the input force FH, Pressure on a vertical
and the duration time r of the impact. In Wa/1
these analyses the intensity of PP is 800 PREVENTION OF DANGEROUS CONDITIONS.
tfjm, or approximately 5w 0 H which corre- The best defense of the vertical-type break-
sponds to wave height 8.0 m. It should be water from the effects of the excessive im-
pointed .out that greatly reduced Fs peaks pulsive pressures that is defined by the con-
at about 300 tonne-force per meter, being ditions expressed in Eqs. (10-8) and (10-9)
only 40% of PP. or by the impulsive pressure coefficient de-
scribed in Section 10.2.3.5 is prevention of
such conditions from occurring. This in-
(o)
cludes the following measures:

(i) Where possible, the breakwater should be


situated on a gentle-slope sea bottom, or
at sufficiently deep locations.
(ii) The caisson's mound must have its height
and length optimized; it is recommended
that the water depth d in front of the
caisson should be deeper than 0.6h, where
h is the depth of water just in front of the
breakwater structure.
800..---r---.---.:..::'b-r-l- - - - , - - - - - , 0.4

Pp •BOOtflm
Ps •16011/m
0.3
x.
The incident wave angle () to the break-
T•0.24s !ml water's alignment is also an important
0.2
factor that affects the impulsive pressure.
0.1 In fact, if () is greater than 30°, then the im-
pulsive pressure on a vertical breakwater is
almost nonexistent (Takahashi et al., 1992b,
30f-----t-----+-----l----1
2 (5 l 3 1994).

Figure 10-48. Calculation results ofbreakwater dy-


ADOPTION OF ALTERNATIVE BREA.KWATER
namic response. [From Shimosako and Takahashi, 8TRUCTURES. lf the high and large rubble
(1994).] foundation is unavoidable and the sea bot-
Breakwater Design 991

tom is steep, then in order to prevent for- Horbor


mation of the high-magnitude impulsive Si de
pressure acting on the vertical wall, the
other structural types of breakwater should
be used. One such alternative structure is a
horizontally composite-type breakwater; for
example, a composite breakwater having its
front covered with wave-dissipating con-
crete blocks. This design results in a com-
pletely different wave profile impacting the
breakwater; in this case, the impulsive
Figure 10-49. Simplified Sainflou wave pressure di-
pressure is remarkably reduced by the cov- agram.
ering blocks. The design of horizontally
composite breakwaters is discussed in
Section 10.4. whereas the pressure at the wave trough is
Impulsive pressure may also be substan- determined from the following formula-
tially reduced by using perforated wall cais- tions:
sons, which are also called wave-dissipating
caissons. The perforated front wall of these p~ = w 0 (H- h 0 ) (10-13)
caissons in combination with the wave
chamber located just behind this wall can w0 H
(10-14)
significantly reduce the impulsive pressure P2 = p 2 = cosh(21ThjL)
acting on the vertical wall. The design of
perforated wall caissons is discussed in Sec- (10-15)
tion 10.3. The sloping top caissons can also
reduce the destructive effects of the wave
impact on the structure.
The Sainflou formula properly describes
the standing-wave pressure and has been
10.2.3.4 wave Pressure Formulas used ali over the world for many years. It
10.2.3.4.1 Sainflou Formula was also used in J apan prior to the develop-
ment of the Goda formula. In Eqs. (10-11)
In 1928, Sainflou introduced a wave
through (10-15), H is the design wave
pressure formula for standing waves that
height; it is recommended that the maxi-
was based on trochoidal wave theory. At
mum wave height be applied for the design
that time his formula improved the existing
wave.
wave pressure formula proposed by Benezit
(1923). The Sainflou formula simplified the
wave pressure theory and provided the wave
10.2.3.4.2 Hiroi formula
pressure distributions at the wave crest and Hiroi (1919) developed a wave pressure
trough (Fig. 10-49). Accordingly, the wave formula using analogy of the hydrodynamic
pressure at wave crest is determined by the pressure. His formula is based on field
following equations: measurements obtained by Stevenson-type
pressure gauges (Stevenson, 1886) and is
applied to breaking waves in relatively
shallow seas. The pressure distribution is
assumed to be uniform along the vertical
wall face as is shown in Figure 10-50. The
· wave pressure p at the wave crest is ob-
992 Breakwater Design

10.2.3.4.4 Nagai formula


Nagai and Kurata (1974) developed a
comprehensive wave pressure formula
which recognizes the infiuence of water
depth d in front of the upright section rela-
tive to the site's water depth h and the
berm width B M.

10.2.3.5 Extended Goda Formula


In 1973, Goda used results of his own
theoretical and laboratory studies pub-
lished in 1972 to develop a comprehensive
formula for calculation of the design wave
forces acting on vertical wall breakwaters.
This formula has been modified later to
Figure 10-50. Wave pressure diagram in accor-
dance with the Hiroi pressure formula. account for the effect of the oblique waves.
It was successfully used to design the verti-
cal breakwaters built in J apan.
tained from The original formula (Goda, 1973b) has
many advantageous features, with the main
p = 1.5w 0 H (10-16)
ones being as follows:
where H is the design wave height. The 1. It can be used for all wave cond.itions (i.e.,
wave crest elevation is assumed to be at the both for stand.ing and breaking waves).
height 77* = 1.25H above the still-water 2. The design wave that is included in Goda
level. formula is the maximum wave height;
The Hiroi formula is very simple, yet it it can be evaluated by the provided
reasonably accurately describes the design diagrams andjor equations.
pressure developed by breaking waves; it 3. It is partially based on nonlinear wave
was used for many years in Japan until theory and can represent ·wave pressure
development of the Goda formula. In the characteristics by considering two pres-
Hiroi formula, the design wave is usually sure components: the breaking and slowly
varying pressure components. Hence, it
assumed to be H 113 . However, the differ-
can be extended relatively easily for ap-
ence between the maximum wave height plication to different types of vertical wall
and the significant wave height, H 113 , is breakwaters.
small in shallow seas. In practica! terms, 4. The Goda formula clarifies the concept of
the water depth was used by Hiroi instead the uplift pressure on the wall bottom; in
of the wave height when he designed the this formula, the buoyancy of the vertical
breakwater built in Hakodate. wall in stiU water and the uplift pressure
due to wave action are defined separately.
10.2.3.4.3 Minikin formula The d.istribution of the uplift pressure has
a triangular shape.
Minikin (1950) proposed a breaking-wave
pressure formula that includes impulsive Subsequently, the Goda formula was ex-
pressures. This formula, however, cannot tended to include the following parameters:
properly describe changes in the impulsive
pressures due to shape of the rubble 1. The incident wave direction (Tanimoto et
foundation. al., 1976)
Breakwater Design 993

2. Modification factors applicable to other (10-19)


types of vertical walls
(10-20)
3. The impulsive pressure coefficient
(Takahashi et al., 1994)
in which
In the extended Goda formula, the wave
pressure acting along the vertical wall is
assumed to have a trapezoidal distribution
both above and below the still-water level, (47rh/Lv) ) 2
whereas the uplift pressure acting on the a 1 = 0.6 + 0.5 ( sinh( 4 7rh/Lv) (10-22)
bottom of the upright section is assumed to
have a triangular distribution as is shown a* = max{a 2 , a 1 } (10-23)
in Figure 10-51. The buoyancy is calculated
using the displacement volume of the up-
right section in stiU water at the design
a2 = min{( 1-hb d) (Hvfd)z'
3
~}
Hv
(10-24)

water level. As indicated in Figure 10-51, h


denotes the water depth in front of the
breakwater, d is the depth above the armor
aa = 1 - ( :, ) { 1 - cosh(2~h!Lv) } (10-25)
layer of the rubble-mound foundation, h' is
h*
the distance from the design water level to a4 = 1 - _c (10-26)
the bottom of the upright section, and hc is TJ*
the elevation of the breakwater above the
design water level. The theoretical eleva-
tion at which the wave pressure could be
where
exerted, 'TJ*, and the representative wave
pressure intensities p 1 , p 3 , and p 4 in a (} = angle between the direction of
generalized form are obtained from the fol- wave approach and a line nor-
lowing formulations: mal to the breakwater
Av A2 , and A3 = modification factors dependent
TJ* = 0.75(1 +cos (})A 1 Hv (10-17) on the structural type
p1 = 0.5(1 + cos(}) Hv, Lv = design wave height and length,
respectively
X(A1 a 1 + A2 a*cos 2e)w 0 Hv (10-18) a1 = impulsive pressure coefficient

se a harbor ·
si de si de

Figure 10-51. Wave pressure diagram in accordance with


the Goda pressure formula.
994 Breakwater Design

w0 = specific weight of sea water 10.2.3.5.2 tmpu/sive pressure


(wo = Pog) coefficient a 1
Hb = offshore water depth ata dis-
tance five times the significant The coefficient a 1 was obtained by reana-
wave height, H 113 lyzing results of comprehensive wall sliding
min{a, b} = smaller value of a or b in Eq. tests carried out by Takahashi et al. (1994).
(10-24) It has a nondimensional value and repre-
sents the impulsive pressure component,
max{a, b} = larger value of a and b in Eq.
(10-23) which should be regarded as an additional
effect of the slowly varying pressure compo-
nent. This coefficient was introduced be-
10.2.3.5.1 Pressure component cause the coefficient a 2 in Goda's formula
and Pressure does not estimate accurately the effect of
Coefficients ( a 1 , a 2 , the dynamic (impulsive) wave pressure and
and al) therefore the effective (quasi-static) pres-
Figure 10-52 illustrates the transition of sure due to impulsive pressure under all
wave pressure from nonbreaking to impul- practical site conditions. Numerical values
sive pressure, where the pressure compo- of a 1 can be obtained from the diagram
nent is indicated by coefficients av a 2 , and presented in Figure 10-53 in which a 1 is
a 1 : a 1 represents the slowly varying pres- expressed as a function of a 10 and an. The
sure components and a 2 is the breaking factor a 10 represents the effect of wave
pressure component; a 1 represents the im- height on the mound and the factor an
pulsive pressure component, which includes represents the effect of mound shape (shown
the dynamic response effect on wall sliding. by contour lines in Fig. 10-53).
The coefficient a 1 varies from O to 1.1 as
the relative depth decreases, and coefficient
(10-28)
a 2 increases as djhb decreases, although it
peaks and then decreases as d j h b de-
Hjd when H :S 2d
creases; a 2 ranges from O to 1.0. The mean- alo = {
2.0 when H > 2d
ing and value of coefficient a 1 is described
in more detail in the following section. (10-29)

-1.0 -1.0
(a ) Stand ing Pressure ( b )Breoking Pressure (c) Impulsive Pressure

Figure 10-52. Transition ofwave pressure.


Breakwater Design 995

10.2.3.5.3 Modification factors (A 1 ,


A2 , and A3 )
Because Goda's formula was originally
proposed to describe loads on the ordinary
vertical-type breakwaters, the factors A1 ,
A2 , and A3 are taken equal to unity. The
modi:fication factor A1 represents the reduc-
d
tion or increase of the wave's slowly varying
h pressure component; factor A2 represents
changes in the breaking pressure compo-
nent (dynamic pressure component or im-
pulsive component); and factor A3 repre-
sents changes in the uplift pressure. These
modi:fication factors are explained further
in Section 10.3 and 10.4 for different types
of vertical breakwater.
O L-___L_ _L__ _L___J_ ____L_ _L__ _L____JI.O
o O. 1 0.2 fM 0.3 0.4
10.2.3.5.4 Design wave height
Figure 10-53. lmpulsive pressure coefficient. [From .A13 pointed out earlier, the design wave
Takahashi et al. (1994).] height and length are those of the highest
wave in the design sea state. The height of
the design wave is taken equal to H D =
Hmax = H 11250 = 1.8H113 seaward of the
surf zone, or within the surf zone as the
when 82 >O largest breaking wave height, Hb, of aran-
dom breaking wave at water depth h b.
(10-30)
The design wave height can be obtained
when 811 :-:;O from the following empirica! formulas:
(10-31)
when 811 >O
KsHo
_ { 4.98 22 when 822 :-:; O
82 - 3 o~ 22 w h en o~ 22 > o
(10-32) { when hjL 0 ~ 0.2
H 11 3 = min{( {3 0 H 0 + {3 1 h), f3maxH0, K.Hb}
when hjL 0 < 0.2
811 = 0.93( B;: - 0.12) + 0.36( h ~d - 0.6)
(10-34)

822 = -0.36( B;: - 0.12) + 0.93( h ~d - 0.6) 1.8K.H0


(10-33) when hjL 0 ~ 0.2
Hmax = Hl/250 min{(f3J Hb + f3i h), (10-35)
The value of the coefficient a 1 reaches a f3!axHb, 1.8K.Hb}
maximum of 2 at BM!L = 0.12, djh = 0.4, when hjL 0 < 0.2
and
Hjd > 2; when djh > 0.7, a 1 is always
close to zero and is less than a 2 • It should where min{a, b, c} represents smaller value
be noted the impulsive pressure signi:fi- of a, b, and c. Other parameters such as /3 0
cantly decreases when the approach angle and f36 can be determined from formula-
of incidence wave, e' is oblique. tions given in Table 10-3. The nonlinear
996 Breakwater Design

Table 10-3. Coefficients for approximate estimation of wave height within the surf zone

Coefficients for H 113 Coefficients for Hmax


{3 0 = 0.028(H0 /L 0 )- 0·38 exp(20tanl. 5 8) 136 = 0.052(H0/L 0 )- 0 ·38 exp(20tanl. 5 8)
{3 1 = 0.52exp(4.2tan 8) f3i = 0.63exp(3.8tan 8)
l3max = max{0.92,0.32(H0 /L 0 )- 0 ·29 exp(2.4tan 8)} {3:,.x = max{l.65, 0.53(H0/L 0 )- 0 ·29 exp(2.4tan 8)}

Source: From Goda (1985).

h!Lo
3.0
OI o 15 02 0.3 04 0.6 0.8 1.0
.0~

~~~~m ffffll 1'111111 H#litfl 1111111111


\ 0>,..(0
1 le
0.9

O.ţ,.

1~. ·:.-~1111 III 1


~/-",1L ~. 1
r2Si"O Ha!Lo =
'i:
2.0 1\. ilo', 0.005 1 ' '
~
Q.l
_1{!-f •\ 1 !
Qj H(,/Lo=O.OOO i
8 1'- :::,.0 0.001 r'-1.: ' . .OI l
O'
1 . ~
'
.E 0.002 \
1
o 1.5 c-...; O.O?I 1
o
.s:::
(f) .'
::-..
::::--; t=:..: ".:;. Op4
r-
1.0 -
0.004 0.006 0.008 0.01 0.015 0.02 0.03 0.04 0.06 0.08 0.1
Relative Woter Depth, h ILo

Figure 10-54. Diagram of nonlinear wave shoaling. [From Goda (1985).]

shoaling coefficient K s can be obtained from 10.2.3.5.5 wave direction


Figure 10-54, and Hb is the equivalent
As shown in Figure 10-55, the wave an-
deep water wave which is hypothetically
gle (} is the one between the direction of
assumed to include the effect of wave trans-
wave approach and the line normal to the
formation due to wave diffraction and re-
breakwater alignment. Starting from its
fraction. Equations (10-34) and (10-35) are
used solely for evaluating the effect of
wave transformation due to shoaling and
breaking.
It should be noted that the design water
depth is not the site's water depth h, but
the water depth h b at a distance equal to
5H113 seaward of the vertical breakwater
face. By including hb in the design wave
analysis, its height will increase, especially
when the sea bottom is steep. Subse-
quently, the design wave pressure will also
/
increase to .a certain extent due to effects of
the sea-bottom slope. Figure 10-55. Angle ofwave approach.
Breakwater Design 997

principal direction, the wave direction friction coefficient, ,u, between concrete
should be rotated toward the line normal to base and rubble stones is usually taken
the breakwater alignment by up to 15°. This equal to 0.6.
adjustment is made to compensate for both W0 = weight of the upright section per unit
the uncertainty in estimating the wave length in stiU water
direction and the waves' directional U = total uplift force per unit length
spreading. P = total horizontal wave force per unit length
calculated by Eqs. (10-17) through (10-27).
10.2.3.6 stability of the vertical
Wall The safety factor against overtuming of the
vertical wall section, SF0 , is determined
10.2.3.6.1 Sliding and overturning from
The breakwater's upright section (verti-
cal gravity-type wall) must be stable against (10-37)
sliding and overtuming (Fig. 10-56). Its
safety factors against sliding and overtum- where
ing must be greater than 1.2. In most cases,
the wall sliding stability is more critical t = horizontal distance between the center of
gravity and the heel of the upright section
than overturning, especially when the
M u = uplift moment about the heel of the up-
breakwater's crown is relatively low. The
right section
safety factor against sliding due to wave
action, SF8 , is determined from MP =moment due to the horizontal pressure at
the wall base
,u(W0 - U)
SF8 = - - p - - (10-36) 10.2.3.6.2 Bearing capacity of the
rubble foundation
where To fully determine the stability of a verti-
,u = coefficient of (static) friction between the cal gravity-type wall, the bearing capacity
upright section and rubble mound. The of the rubble mound and the underlying
subsoil must be investigated. In this
procedure, a trapezoidal or triangular
(a) distribution of bearing pressure on rubble
foundation is usually assumed. The bearing
pressure on the rubble mound and the un-
derlaying subsoil is determined according
to the procedures described in detail in
Chapter 5.
The maximum allowable bearing pres-
sure on rubble mound, sometimes called the
"toe pressure," is usually taken equal to
(b)
40-50 tonne-force per square meter (ap-
proximately 400-500 kNjm2 ). Sometimes
the upper limit of this pressure was raised
up to 60 tfjm 2 (approximately 600 kNjm 2 ).
Kobayashi et al. (1987) proposed a new
method for calculating the bearing capacity
Figure 10-56. Sliding (a) and overturning (b) mode of the gravity-type structures that are rest-
of failure of upright section. [From Goda (1985).] ing on a rubble mound. In this method, the
998 Breakwater Design

authors use the simplified Bishop method of (Tanimoto and Ojima, 1983). The friction
circular slip failure for analysis of the rub- coefficient between flat concrete blocks is
ble-mound foundation. usually taken to be 0.5, and the effects of
For these analyses, Kobayashi et al. interlocking between blocks, if any, should
(1987) recommend the value of rubble cohe- be considered.
sion coefficient c = 2 tfjm 2 (approximately
20kNjm2 ) and the angle of shear resist- 10.2.3.7 Other Problems Related
ance <P = 35°. These values have been ob- to wave Forces
tained by the latter investigators from the 10.2.3.7.1 Concave Section
triaxial tests carried out on large samples.
It should be noted that these tests have As pointed out earlier, the vertical wall
been carried out on materials that are rou- breakwaters reflect waves, and therefore the
tinely used in Japan for harbor construc- wave reflection and diffraction from this
tion. The safety factor in this method must type of breakwater should be taken into
not be less than 1.0. account when computing the wave forces.
The wave forces strongly depend on break-
water alignment.
10.2.3.6.3 stability of the block
Figure 10-58 illustrates the calculated
masonrv wa/1
distribution of the ratio of the wave height
Stability of the block masonry wall can to the incident wave, Kd, in front of the
be determined using the extended Goda vertical breakwater which has the concave
pressure formula. The masonry wall stabil- alignment with respect to the incident
ity should be examined at each level of waves. From this figure it is seen that due
the blocks taking into account the quasi- to wave reflection from one of the breakwa-
hydrostatic pressure that may exist be- ter lines, the wave height along the vertical
tween the adjacent blocks due to wave ac- wall is not simply double for standing
tion. This hydrostatic pressure acts as the waves; it is obvious that the value of Kd
uplift force on the upper block. In the exam- exceeds 2.0. The amplification factor of the
ple shown in Figure 10-57 this pressure incident wave to each part of the breakwa-
acts on a second from the bottom block. The ter can be obtained by Kd/2. It should be
pressure distribution diagram is usually as- pointed out that the amplification factor is
sumed as triangular with maximum ordi- affected by standing-wave breaking; based
nate equal to horizontal pressure (p 8 ) at on a series of experiments, its maximum
the joint between two adjacent blocks value is recommended as 1.4.

Figure 10-57. Loads acting on block masonry wall.


Breakwater Design 999

increases near the breakwater head. If the


wave height that attack the breakwater ex-
ceeds the height of the design wave, it may
result in sliding of the breakwater's upright
section. If the breakwater is composed of
separate caissons or similar units, then un-
der the above condition its alignment may
subsequently take the form of the meander.

10.2.3.7.3 Long caisson


When a relatively long caisson is erected
in an oblique sea, the wave crest moves
Figure 10-58. Wave height distribution along con- along the vertical wall, as shown in Figure
caved breakwater alignment. [From Kobune and Osato 10-60. In this case, the total wave force can
(1976).] be averaged along the breakwater align-
ment. The solid line in Figure 10-60 shows
the averaging factor 88 for long caisson as a
10.2.3.7.2 Meandering effect function of the wave approach angle and
Another problem related to the breakwa- the relative length of the caisson. It is cal-
ter alignment is the so-called meandering culated using small-amplitude-wave theory.
effect (Ito and Tanimoto, 1972). It was dis- Although 88 is small when 8 is large
cussed earlier in Section 10.1.4.8. Figure and the caisson's relative length is long, the
10-59 illustrates the calculated value of the caisson's stability against horizontal rota-
ratio Ka for a single-line detached break- tion could be critica!. The interrupted line
water 200 m long and wave length L 113 = in Figure 10-60 shows the limit of 88 when
92.3 m; the wave height fluctuates along the caisson stability against rotation is
the breakwater's length and significantly about 0.8.

3.2~------~----~~,_-------,,--------,

8=60°

Le= 200m
0.81---

0.41---

OL-------~5~o~----~,~oo=-------7.,5~o~----~2oo
x (m)

Figure 10-59. Wave height distribution along detached one-line


breakwater. [From Goda and Yoshimura (1971).]
1000 Breakwater Design

/si L
8=45~·r----o,.~2--~0T.4~--o~.rG____o=·~s~~~~-o~---~~-2~--~~~.4
8=30o0r--=r-~r-~~-=Or.8~~~~~~~~~~.6~~~~2.0
8=15°0r---~--~~--~~~~--~~--~L_~~---

Figure 10-60. Averaging of wave force by a long caisson. [From


Takabashi et al. (1990b).]

10.2.3. 7.4 coefficient of friction comparison with the high-crest wall. It


The various recommended values of co- should be noted that the Goda wave pres-
efficient of friction p, are given in Table sure formula .includes the overtopping effect
10-4. These values are based on measure- because in his experiments he dealt with
ments carried out on models and proto- the models of the ordinary relatively low-
types. It must be remembered, however, crown breakwaters.
that breakwater failure may occur due to Mizuno et al. (1993) found that the high
sliding in other places of the breakwater walls that do not allow for wave overtop-
structure; for example, between the rubble ping creates a larger wave force than that
stones or between the rubble mound and determined by the Goda formula. The latter
the foundation soil. investigators also found that the impulsive
pressure acts on breakwater even when a
10.2.3.7.5 High- and low-crest spilling-type breaking wave hits the high
breakwaters vertical wall (Fig. 10-61a). This occurs be-
cause the spilling breaking wave that is
Wave overtopping is allowed for ordinary riding on the upper part of the approaching
breakwaters. Obviously, the wave force on wave hits the vertical wall if it is high
the low-crest vertical wall is smaller in
enough, whereas it spills over the wall if it
is of ordinary height (Fig. 10-61b). Hence,
Table 10-4. Various coefficients of friction used for the wave pressure on a wall with a very low
breakwater design crest is significantly reduced due to over-
Concrete on concrete 0.5 topping.
Concrete on rock base 0.5 Nakada et al. (1984) conducted a series
Concrete on rubble stone 0.6 of experiments to investigate the effect of
Rubble stone on rubble stone 0.8 the crest height and proposed the modifica-
Breakwater Design 1001

(a) effects on armor protection strongly de-


pends on the approach angle. The practica!
example is shown in Figure 10-62. This
example illustrates the movements of the
armor stones as a function of the wave
approach angle, e. From this figure, it is
seen that for the given condition (H113 and
T 113 ), the armor stones are scattered from
the mound's slope down when incident
waves approach the breakwater at e = o
(Fig. 10-62a). Note the heavy scattering of
stones around the breakwater head. When
e = 45° (Fig. 10-62b), the armor stones not
only scatter down the mound's slope but
( b)
they are also scatter on the fiat berm near
the caisson. Under these conditions, the
stones also scatter around the breakwater
downstream head. Where e = 60° (Fig.
10-62c), the armor stones on the slope do
not move, although they significantly scat-
ter on the fiat berm, especially near the
caisson.

Figure 10-61. Wave pattern hitting a wall: (a) high


crown wall; (b) overtopping a low crown wall.

tion factors for the low-crest height vertical


walls.
Ti/3=2.92s (b)
Hu3=175cm

10.2.4 Design of Rubble-Mound ,..rrMi·


f.ti~:r_-:-_;-_J'-_i:j
,· ,-, ""T.Crl-r ,-.;L.

Foundation ~lil'l2112011911BLI7116JI5114113112111110191
:- -:-~-~

10.2.4.1 Armor for Rubble


Foundation
10.2.4.1.1 wave force on armors <el

The rubble-mound foundation under the


vertical-type wall breakwater should be
protected from potential of scour due to
action of waves and currents. This is usu-
ally accomplished by covering the mound
with armor stones or concrete blocks of mis- Figure 10-62. Scattering of armor stones of rubble
cellaneous designs. The character of waves foundation. [From Kimura et al. (1994).]
1002 Breakwater Design

It is postulated that the stability of ar- Figure 10-64 shows typical distribution
mor stones or blocks is governed mainly by of the peak water partide velocity around
the water partide velocity induced by waves the breakwater head. In this figure it is
(i.e., by the drag and uplift forces produced seen that high velocity occurs around the
by the water partide velocity). In the edge of the breakwater upright section.
example depicted in Figure 10-62 in areas
where armor stones are moved, the water 10.2.4.1.2 Weight of armor
velocity is high, which supports the above To withstand velocity-induced forces, the
hypothesis. armor stones or concrete blocks should be
Figure 10....:63 shows a hydrograph of the heavy enough. Their required weight can be
water partide velocity at the breakwater evaluated by the Isbash formula for stones
trunk. From this figure, it is seen that for embedded in the bottom of a sloped chan-
the given conditions when (} = O, the water nel. This formula relates the stable weight
partide only moves perpendicular to the of armor stones to the water partide veloc-
breakwater alignment and the velocity is ity as follows:
almost zero near the vertical wall, being
largest at the node of the dapotis (standing
wave). As (} increases, the water velocity
component parallel to the breakwater align-
ment increases, with the velocity near the where
vertical wall also significantly increasing. W = required weight of armor stane
'Yr = specific weight of armor stane ( 'Yr = Prg)
y = Ishbash number; y = 1.2 for embedded
T=2.92s, H=lO.Ocm stones and y = 0.86 for stones placed on a
flat bottom

e __
oo
45°---
600 -------

Figure 10-63. Water partide velocity hydrograph Figure 10-64. W ater partide velocity distribution
near vertical wall. [From Kimura et al. (1994).] around breakwater head. [From Kimura et al. (1994).]
Breakwater Design 1003

sr = specific gravity of stone where


U = water partide velocity on the stone
g = acceleration of gravity (10-41)
a = bottom slope 2kh'
(10-42)
K1 = sinh2kh'
Brebner and Donnelly (1962) proposed the
method to directly determine the necessary (K2)B = max{a 8 sin26cos 2 (kBM cos 6),
weight of the armor unit from the wave (10-43)
height. In their method, the stable weight
of the armor unit, W, can be obtained from where
the following formulation: max{a, b} = denotes the maximum values of a
and b
h' = water depth above the rubble-
(10-39)
mound foundation,
K = 27TjL'

where L' = length of the wave that corresponds


the significant wave period at the
'Yr = specific weight of the armor unit depth h'
H 113 = design significant wave height BM = berm width
N 8 = stability coefficient a 8 = 0.45 is a correction factor obtained
from the wave tank experiments
'fhe coefficient N 8 depends on variables such
as the shape of the armor unit, the manner Equation (10-40) is also used for the evalu-
of placement, shape of the rubble-mound ation of the stability of the rubble-mound
foundation, and wave parameters (e.g., armor layer installed in the breakwater
height, period, and direction). Tanimoto et head. In this case, the term (K 2 )r is used
al. (1982) proposed a formula for the deter- instead of ( K 2 )B to represent the water par-
mination of the stability coefficient, N 8 , for tide velocity at the breakwater head. Hence,
two layers of quarry stones. This formula is
based on some analytical considerations K = Kl(K2)T (10-44)
and the results of random wave tests.
Takahashi et al. (1990a) modified the ( K2 )r = tC a;r 2) (10-45)
Tanimoto's formula in a way that made it
applicable for situations where the oblique where 'T is the ratio of the water partide
incident waves can be a problem. According velocity at the breakwater head to that of
to these workers the modified coefficient velocity incident wave; 'T = 1.4 is usually
N 8 can be obtained from the following used for a wave angle less than 45°.
The sample calculations of the required
expression.
weight of armor stones are illustrated in
Figure 10-65. The shape of the breakwater
1K~/3K) ( ~~
Ns = max{ 1.8, 1.3( 3 )} is indicated and the necessary weight of
armor stones for different wave conditions
and water depths in front of the vertical
+ 18 exp [ -1.5 ( (1 K~/: )' )( :,;,)] wall are calculated. From this figure it is
seen that under e = 60°, the weight of the
armor stones is the largest, whereas that
(10-40)
for e = 30° is the smallest. It is obvious that
1004 Breakwater Design

16.0 the wall. Figure 10-66 shows the failure


+4.0 unit:m
modes related to such foot protection. Where
some blocks are removed, the erosion of
rubble mound takes place near the foot of
the wall. Also, through-wash (rapid current
through the rubble mound) may cause
scouring of the underlaying seabed. The foot
protection concrete blocks are installed to
prevent the direct intrusion of wave pres-
sure into the rubble mound and subsequent
pressure-induced current in the mound.
These blocks work as a fi.lter and also pro-
vide additional weight to the rubble mound.
The critica} force on a foot protection
block is that due to differential pressure
between the upper and lower faces of the
block. The absolute value of the wave pres-
sure under the block is experimentally
found to be 5-40% less than that on the
upper side. The differential pressure can be
reduced and the stability of the block in-
creased by introduction of the holes in the
blocks. The block perforation ratia of 10% is
recommended by Tanimoto et al. (1982). It
should be noted that the holes should be
small enough to prevent the small stones
and gravei from being washed away from
the mound.
The thickness of perforated block, t ', with
a 10% hole ratia can be obtained from the
5 10 diagram given in Figure 10-67. Sizes of
Hu• (mJ
some typical foot protection blocks are sum-
Figure 10-65. Weight of armor stones for rubble marized in Table 10-5. A typical block, 1.2
foundation as a function of significant wave height m thick, is illustrated in Figure 10-68.
(H113 ) and wave approaching angle. [From Kimura et It should be noted that concrete foot pro-
al. (1994).]
tection blocks also act as armor blocks for
the rubble mound as discussed earlier, es-
weight of armor stones increases as the
wave period increases or the water depth in
front of the wall decreases. When the re-
quired weight is large, then armor concrete
blocks must be used to cover the rubble
mound.
.. . . . .
.....
10.2.4.2 Foot Protection . ..

Foot of the upright wall is usually pro- Figure 10-66. Rubble-mound and seafloor failure
tected by concrete blocks placed in front of due to damage to toe protection blocks.
Breakwater Design 1005

l
Table 10-5. Specification of foot protection blocks
ţ '
t- ~ r--.._.~-
t- ...... t--
Head
- - - Trunk. t' (m) l (m) X b (m) X t' (m) W(t)
0.3
t- ....... "" ~0.8 2.5 X 1.5 X 0.8 6.2
-
-
- "" ~ ~ 1.0 3.0 X 2.5 X 1.0 15.6

"'\
t/H - ~ 1.2 4.0 X 2.5 X 1.2 24.8
~ 1.4 5.0 X 2.5 X 1.4 37.0
~
,-- ......
- ~ 1.6 5.0 X 2.5 X 1.6 42.3
-
0.2 - ~ 1.8 5.0 X 2.5 X 1.8 47.6
-
\ \ ~2.0 5.0 X 2.5 X 2.0 52.9
\
r- \ 1\ ~2.2 5.0 X 2.5 X 2.2 58.2

\\
t-
t-
f-
f-
f-
0.1 t b L
r- 40 .
\~\
r- 4.0x 2.5x 1.2
--
-
-
~
\\
0.0 t-1 1 1 1 1 1 1
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
d!h
Figure 10-68. Example oftypical foot protection
Figure 10-67. Thickness offoot protection blocks at concrete block.
breakwater's head and trunk. [From Ushijima et al.
(1988).]

ward mound toe area was significantly


pecially in oblique seas and at the break- scoured, which deepened this area by about
water head. Therefore, they must be stable 2-3 m. Despite the substantial erosion,
against the velocity-induced force. however, the caisson remained intact and
Furthermore, care should be taken to its foot protection blocks and armor blocks
prevent the occurrence of scouring under- of the rubble mound functioned as designed.
neath the rubble foundation; a too thin rub- Funakoshi (1994) stated that sea:floor
ble mound may cause this type of failure to scouring in the toe area can be effec-
occur due to severe wave actions. A geotex- tively prevented by installation of wave-
tile sheet is sometimes placed on the sand dissipating blocks. The latter also may have
bed to prevent this type of bed erosion from a favorable effect on restoration of the
occurring. eroded seafloor to its original level. The
typical cause of toe area scouring is strong
wave-induced velocity; these velocities can
10.2.5 Rubble-Mound Toe be further increased by the effect of the
Protection Against reflected waves. The scouring mechanism is
scouring very complicated and has not yet been prop-
erly explained. Generally, two types of toe
Figure 10-69a illustrates the scouring scouring exist: local scouring and large-scale
damage at the toe of the rubble mound that seabed material movement.
occurred at the composite breakwater con- Local scouring by standing waves was
structed on a sandy seabed (Funakoshi, investigated by Xie (1981), lrie et al. (1986),
1994). In this case, the entire breakwater Oumeraci (1994), and others. lrie et al.
structure slightly settled due to the weight found the occurrence of N- and L-type
of the vertical wall. In addition, the sea- (node-loop) scours, in which the former type
1006 Breakwater Design

of scour is due te movements of a "bed load" severe scouring is expected. The scour pro-
that includes relatively coarse sand parti- tection methods include use of gravei blan-
des which cause accretion at the node of a kets, geotextile, or asphalt mats. These
standing wave, whereas the later type of methods can limit scouring of the sea bot-
scour is due to suspension of relatively fine tom to some extent; however, they cannot
sand particles that causes erosion at the guarantee full protection from the scour.
node and accretion at the loop. L-type scour Hence, at locations where a scour condition
appears to be predominant in prototype seas can be expected, the scour monitoring and
where erosion at the node is usually found. breakwater repair jmaintenance program
Typical definitions for N- and L-type (node- should be in place.
loop) scour are depicted in Figure 10-69b.
Unfortunately, scour is usually unavoid-
able where vertical breakwaters are built 10.3 DESIGN OF NEW TYPES OF
on a sandy sea bottom. However, scour is VERTICAL BREAKWATER
not a fatal problem where protection fea-
tures are provided to the rubble mound. These include mainly perforated and in-
Scour protection elements should be always clined wall breakwaters of miscellaneous
included in the breakwater structure where designs.

Measured
--------- Oct. 1980
Harbor Side Seaward Side -·-·-Aug.l982
--- .. -Jul. 1985
---Aug.l990
±0.0 ( l Oct. 1974
24.5
it

Rubble moond

Figure 10-69. (a) North breakwater at Sakata Port, Japan; scour around rubble-mound toe.
[From Funakoshi (1994).] (b) Typical definition for N- and L-type (node-loop) scour as related
to effects of standing waves.
Breakwater Design 1007

10.3.1 Perforated Wall which waves can enter and leave the wave
Breakwater chamber; in the process, the wave energy
dissipates through generating eddies. The
Sometimes perforated wall breakwaters are larger difference in water levels inside and
referred to as "wave-dissipating upright outside the wave chamber creates better
breakwaters." The :first perforated wall conditions for dissipation of the wave
breakwater was proposed by Jarlan (1961). energy.
This structure was comprised of a perfo- The size of the wave chamber strongly
depends on the length of the design wave,
rated front wall and a wave chamber lo-
and, in general, the performance of the per-
cated just behind this wall constructed in
forated wall breakwater is typically most
1966 in Comeau Bay, Canada. In Japan,
efficient when the width of the wave cham-
the :first perforated wall breakwater was
ber constitutes approximately 10-20% of
constructed in 1970 at Takamatsu Port. It
the wavelength. This is discussed later in
should be noted, however, that a year ear-
this section.
lier, a quay wall composed from perforated
The wave chamber normally has a bot-
wall concrete caissons was built in Kobe
tom slab, and in order to avoid the impul-
Port. The perforated wall breakwaters are
sive uplift pressures generated inside the
usually composed from concrete floated-in
wave chamber, the ceiling slab is usually
caissons, having the perforated front wall
not used. However, to effectively utilize
and the wave chamber.
space on the top ofthe breakwater, a ceiling
Due to their high wave-absorbing capac-
slab is installed sometimes to act as a
ity and high stability against waves, this waterfront promenade. In this case, ade-
type of caisson has been used increasingly quate reinforcement in this slab should be
worldwide for construction of seawalls and provided.
breakwaters. Although initially the perfo- Perforated wall breakwaters are distin-
rated wall caissons were intended for use in guished from each other mainly by the type
relatively calm seas, at present they are of perforation. At present, the most popular
employed in rough open seas. one is the vertical-silt wall breakwater;
Figure 10-70 illustrates the concept of however horizontal-slit walls, circular-hole
the perforated wall breakwater. As is seen walls, and curved-slit walls are also popu-
from this :figure the upright section of the lar. To widen the range of wave period for
breakwater has a perforated wall through wave energy dissipation, more complex
shapes of perforated walls have been used
(e.g., double-slit walls or dual-cylinder
breakwaters).
As stated earlier, the upright section of a
perforated wall breakwater is usually corn-
posed offloated-in caissons; however, it m~y
also be constructed from block masonry, m
which case the shape-designed (perforated)
concrete blocks are placed to form a perfo-
rated wall. This is an innovative and now
frequently used design, especially in waters
that are not too rough. The slit wall caisson
Figure 10-70. Conceptual sketch of the perforated and perforation block masonry are dis-
wall breakwater. cussed in the following sections.
1008 Breakwater Design

10.3.1.1 verticai-Siit Wa/1 H,/L


composed of Floated-in m o 0.009-0.010
Caissons <D 0.018 - 0.029

Figure 10-71 illustrates a typical cross


section of a vertical-slit wall breakwater.
0.6
This structure was constructed in a rela-
tively deep bay (h = 38 m). The vertical
wall of this breakwater is composed of con- 0.4
crete floated-in caissons that are designed
to withstand a moderate design wave (H113
0:2
= 4.8 m, T 113 = 10.4 s). The caisson has a
1-m-thick slit wall and a 7-m-wide wave
chamber. The perforation ratio of the front 0.0 L - . . - - ' - - - - - L - - - ' - - - - - ' - - - - - 1
wall e = 0.135 and the wave chamber 0.0 0.1 0.2 0.3 0.4 0.5
depth from low water level d 1 = 4.5 m (q = l;(
d jh = 0.28). No armor stones and foot
1 1
Figure 10-72. Reflection coefficient KT of a perfo-
protection blocks are installed because of rated wall breakwater. [From Tanimoto et al. (1982).]
the relatively small wave height and sub-
stantial depth of water (h = 16 m).1

by Tanimoto and Yoshimoto. There, l is the 1

width of wave chainber and L is the wave- 1

10.3.1.1.1 wave ref/ection and length at the wave chamber depth d 1 • As it


transmission is seen from Figure 10-72 the reflection
Tanimoto and Yoshimoto (1982) carried coefficient, KR, is reduced to less than 0.4
out a series of experiments to investigate in the range l jL = 0.1-0.3, with the
1 1

the phenomenon of wave reflection by the smallest value around 0.18. Based on re-
perforated wall caisson. Figure 10-72 shows sults from numerous model tests carried
the reflection coefficient versus relative out to study the K R of the perforated wall
width of the wave chamber l 1L obtained1 1
, caissons, it is possible to conclude that, in

~H.WL.+ 3.3
~ L.WL. ! 0.0
Sand

'.
·,?
·'2.

Rubble Stones

Figure 10-71. Breakwater at Nagasaki Port, Japan; typical cross section through a vertical-slit
caisson. [From Kataoka and Saida (1986).]
Breakwater Design 1009

general, to effectively dissipate wave en- porated in the design of the perforated wall;
ergy, the wave chamber width, l', should be the "target wave" is one which should be
equal to approximately 10-20% of the dissipated by the breakwater; it is usually
wavelength. not a heavy storm wave but a rather ordi-
Figure 10-73 shows theoretical predic- nary one. Due to a better wave-absorbing
tions of the refl.ection coefficient in which performance, the wave-transmission coef-
K R is expres sed as a function of the perfo- ficient of the perforated wall breakwaters,
ration ratio 8, the water depth ratio q = Kr, is a1so reduced in comparison to the
d' jh' for the wave chamber in which performance of the conventiona1 vertica1
l' jL = 0.154 and ljL = 0.166, and wave wall breakwaters. However, because the de-
parameter ratio HjL = 0.04. As the water sign wave is usually longer than the "target
depth in the wave chamber decreases, the wave," the reduction of wave transmission
optimum perforation ratio increases with is not so significant for the design wave.
the minimum refl.ection coefficient also Furthermore, the fl.ow rate of waves over-
tending to increase. This has been sup- topping the perforated wall breakwater is
ported by results of many model tests. In considerably smaller than that of the con-
practice, the perforation ratio 8 is usually ventional vertical wa1l breakwater.
used in the range 15-30%.
The refl.ection from the perforated wall 10.3.1.1.2 wave forces on
can also be numerically estimated from the- perforated breakwaters
ories proposed by Terrett et al. (1968), The wave forces on different components
Richey and Solitt (1970), Kondo and Takeda of a perforated breakwater reach their peaks
(1983), Al1sop and McBride (1994), and at different phases of wave-structure inter-
Kakuno (1994). It should be pointed out action; therefore, the wave pressure distri-
that perforated walls of different shapes bution at each of these phases must be
have almost identica! fundamental charac- evaluated. In fact, in the case of a perfo-
teristics of wave refl.ection. rated breakwater, the peak load that may
It should be also noted that "the target cause this breakwater to slide or overturn
waves for wave absorption" must be incor- does not necessarily occur when the wave

E
Figure 10-73. Reflection coefficient KR versus the opening ratia € and wave
chamber depth. [From Tanimoto et al. (1982).]
1010 Breakwater Design

crest is located just in front of the perfo- curs at the crest-Ilb phase. However, when
rated wall. The most critica! phases of the the perforated wall breakwater is used for
wave-breakwater interaction are discussed wave dissipation in rough seas, where it
in this and the following sections. can be subjected to highly impulsive waves,
The forces that act upon each structural the peak sliding or overturning force does
component of the perforated breakwater not always occur at crest-Ilb.
peak at six basic wave phases: crest-I,
Design Pressure Distribution at Positive
crest-Ila, crest-Ilb, trough-I, trough-II, and
Peaks. In the design of perforated wall
trough-III as is shown in Figure 10-74.
caissons, the positive pressure distributions
These six phases are defined as follows.
at three phases (e.g., crest-I, crest-Ila, and
crest-Ilb) should be evaluated. They can be
Crest-I: the phase during which the wave calculated based on the Goda pressure for-
forces on the front wall (both perforated and mula with modification factors A1 , A2 , and
the lower imperm.eable part) reach their posi- A3 , where A1 is the modification factor of
tive peak.
the slowly varying wave pressure compo-
Crest-lla: the phase during which the wave nent, A2 is the impulsive wave pressure
force that acts on the wave chamber rear wall component of the horizontal pressure, and
reaches the impulsive peak. This phase, how-
A3 is the uplift pressure component Pu as
ever, is not always distinct ifthe wave force is
not impulsive.
discussed in Section 10.2.3.5.
For the calculation of the impulsive pres-
Crest-IIb: the phase during which the wave
sure component for the wave chamber rear
force on the wave chamber rear wall reaches
a lower peak following the impulsive peak.
wall by the Goda formula, the coefficient a]
is used instead of a1 ; the coefficient a] is
Trough-1: the phase during which the wave
obtained by making the following substitu-
force on the front wall reaches its negative
tions: d is replaced by d' (wave chamber
peak.
depth), L is replaced by L' (wavelength at
Trough-11: the phase during which the wa-
d'), and BM is replaced by BM' = l - (d -
ter level in front of the caisson is the lowest.
d '), where l is the width of the wave cham-
Trough-111: the phase during which the ber. Note, if B~ is negative, then B~ =O
water level in the wave chamber is the lowest. should be used.
Figure 10-75 shows the typical wave
When the wave force is not severe, the peak pressure distribution at crest-Ha, where the
sliding andjor overturning force usually oc- subscript on A indicates specific locations
· on the perforated wall breakwater; for ex-
ample S stands for the slit wall, L for the
impermeable front wall, R for the wave
chamber rear wall, M for the wave cham-
ber bottom slab, and U for the uplift force.
The modification factors (A 1 , A2 , and A3 )
used in the extended Goda formula for a
typical perforated vertical wall breakwater
are given in Table 10-6. Note that the ap-
plication of the modification factors reduces
both the impulsive pressure and the slowly
varying pressure components.

Figure 10-74. Typical phases of wave-perforated Design Pressure Distribution at Negative


wall breakwater interaction that are used for wave Peaks. The wave pressure from the nega-
pressure analysis. [From Takahashi et al. (1991).] tive peaks can be determined using the wa-
Breakwater Design 1011

>..R

Crest J[a
/
/ / ..... - - - - - - .,..,---t--r-r-r--r-...--,r-T-r--r-,.-,-.--r-..-,~---- ---
/ ./'
..,.,...-"""" ..........
/ ""

L~;;~~~~----------~---~~---J-----~~-----------------
Figure 10-75. Wave pressure distribution at crest-IIa phase (see Fig. 10-74) for a
perforated wall breakwater.

Table 10-6. Modification factors for various phases ofwave-perforated breakwater interaction

Crest-I Crest-IIa Crest-IIb

Slit wall Asl 0.85 0.7 0.3


As2 0.4 (a* ;;;:; 0.75) o o
0.3/a*(a* > 0.75)
lmpermeable ÂLl 1 0.75 0.65
front wall ÂL2 0.4 (a*;;;:; 0.5) o o
0.2/a*(a• > 0.5)
Wave chamber ÂR1 o 20lj3L' (ljL' ;;;:; 0.15) 1.4 (Hv/h;;;:; 0.1)
rearwall 1.0 (ljL' > 0.15) 1.6 - 2Hv/h (0.1 < Hv/h < 0.3)
1.0 (Hv!h;;;;; 0.3)
ÂR2 o 0.56 (a* ;;;:; 25/28) o
0.5ja* (a*> 25/28)
Wave chamber ÂM1 o 20lj3L' (ljL' ;;;:; 0.15) 1.4 (Hv/h;;;:; 0.1)
bottom slab 1.0 {ljL' > 0.15) 1.6- 2Hv/h (0.1 < Hn/h < 0.3)
1.0 (Hv!h;;;;; 0.3)
ĂM2 o o o
Uplift force Âu1 1 0.75 0.65
Âu2 o o o
Source: From Takahashi et al. (1991).

ter level in front of the perforated wall and elevation of the wave chamber's rear wall.
that in the wave chamber. Figure 10-76 Although negative wave pressure is not a
shows distribution of the wave pressure at critical factor that affects stability of the
trough-I, trough-II, and trough-III. In this breakwater under normal design condi-
figure, h 01 denotes the crest elevation ofthe tions, it may become critical when the
perforated wall and h 02 denotes the crest breakwater is located in deep water. Hence,
1012 Breakwater Design

in the latter case, the negative wave pres- 0.8 to 0.9 and the breakwater stability is
sure should be calculated using more accu- critica! at phase crest-I when Hjh < 0.2,
rate finite-amplitude standing-wave theory whereas at phase crest-Ila, the stability is
(Fig. 10-38). critica! when Hjh > 0.3.
Perforated Wall Breakwater Versus Con- When the rubble mound is comparatively
ventional Vertical Wall Breakwater. The thicker (djh = 0.5), Fcc!Fca is equal to
wave forces acting on a perforated wall about O. 7 and the breakwater's stability is
breakwater are generally smaller than those critica! at phase crest-IIa for Hjh > 0.3,
acting on conventional vertical wall break- regardless of the value of d'. The latter is
water. This is illustrated in Figure 10-77. noteworthy because it indicates that perfo-
In this figure, the symbols Fcc and Fca rated wall breakwater has better stability
denote the respective weights of the perfo- against wave forces than the conventional
rated and conventional breakwaters that one under both the low- and high-mound
need to resist similar wave force deter- conditions.
mined by the Goda formula. Note that depth
ofwater, d, and depth ofwave chamber, d', Wave Uplift Pressure on the Ceiling Slab
vary for the perforated wall breakwater. of the Wave Chamber. If the wave cham-
When the rubble mound is not high ber is covered with a ceiling slab, then as
(djh = 0.7) and d' jd = 1.0, the ratio of mentioned earlier this slab may be exposed
Fcc!Fca is approximately 0.8, except when to quite heavy uplift forces. The magnitude
Hjh < 0.3, and for all Hjh values, Fcc is of these forces depends on the ceiling slab
at its maximum (i.e., the sliding stability is clearance from the still-water level. This
critica!) at crest-Ilb. When djh = 0.7 and pressure is caused by the compression of
d' jd = 0.5, the Fcc!Fca ratio ranges from the entrapped air in the wave chamber and

min {0.35H ,(2hcl+hczl/3}

... ~ .....
/<6:~:~;;
--..-1!- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -·

[TiOU9hJIT}
_________..L..:::=:::::==-~
/

_/,/ -min { 0.6woH, 0.35wod'}


--""~~----------------------------------------
Figure 10-76. Wave pressure distribution of a negative peak.
Breakwater Design 1013

T=9s o Crest-I
l!L=0.063 • Crest-lla
<t Crest-llb

dlh dld
0.5
0.7
0.7
1.0
0.5
__...__
0.5
0.5 -------
1.0 _____...,. ____
0.5

0.2 0.4 0.6 0.8


H 1 h
Figure 10-77. Weight of a perforated wall breakwater versus conventional
vertical wall breakwater; practica! example.

is usually the Bagnold-type pressure. This


pressure typically occurs due to a lag in
phases between the water surface in the
wave chamber and that in front of the per-
forated wall (Fig. 10-78).
The pressure on a ceiling slab can be
evaluated using the upward speed of the
water surface in the wave chamber, u 0 , and
the thickness of the entrapped air, D
(Takahashi et al., 1985). For example, in
the case of the vertical-slit perforated
breakwater built in Tarumi, Japan, the de-
sign uplift pressure is equal to 2 w 0 H for
the design wave of H 113 = 4.6 m (Hn = 8.3
m) and T 113 = 8 s. It should be noted that Figure 10-78. Air compression in wave chamber.
in model tests of this breakwater, the air
pressure corresponding to the design wave
was measured as more than 4w 0 H. How- as "perforation blocks." The example of such
ever, it was reduced to 2w 0 H when the wall is depicted in Figure 10-79; it illus-
scale effect of air compression was taken trates a perforation block masonry break-
into account. water comprised of three blocks of special
design. For better stability, ali blocks are
keyed with each other as well as with a
10.3.1.2 Perforation Block
large bottom block and a large crown block.
Masonry Wall
The design wave height for this par-
This type of perforated wall comprises ticular breakwater is H 113 = 2.25 m and
special masonry blocks usually referred to T 113 = 6 s.
1014 Breakwater Design

The block masonry breakwaters are nor- overtopping occurs in comparison with solid
mally employed in a relatively sheltered vertical wall breakwaters.
area in which the design significant wave
height is less than 3 m. Several different 10.3.1.2.2 wave forces
types of perforation blocks are shown in
Because the block's shape is very com-
Figure 10-80. The perforation blocks are
plex, it makes it difficult to determine the
also used for construction of the quay walls
wave force on each component and ulti-
and the shore protection structures.
mately on the breakwater structure. The
impulsive uplift pressure that may exist
under the crown block especially compli-
10.3.1.2.1 wave ref/ection and cates the determination of the wave force.
transmission Therefore, model tests are usually carried
out to determine the realistic wave forces
Because the perforation block masonry on the wall, as well as on its particular
wall is also a perforated wall, it absorbs structural components. For the preliminary
wave energy and its wave-reflection coeffi- design, the wave pressure distribution can
cient, KR, is dependent on the ratio ljL, in be determined by the extended Goda for-
which l is the length of the wave chamber mula with modification factors A1 and A3 =
and L is the length of the wave. Figure 0.8 to 1.0, and A2 = O. The stability of the
10-81 represents a plot of the results of perforation block masonry wall as well as
several tests that have been carried out on stability of the individual blocks that are
perforation block masonry walls. From this included in this wall are examined in a
figure it is seen that K R tends to be rather similar manner that is used in the case of
small when ljL = 0.05 to 0.25. conventional blockwork wall (consult Sec-
Wave transmission of the perforation tion 10.2.3.6.3 and Chapter 5). It should be
block masonry breakwater is similar to that pointed out that the uplift pressure on
of the perforated wall breakwaters; reduc- blocks located above the still-water level
tion in the transmission coefficient Kr due and on the crown block is very impulsive
to perforation blocks is not significant, al- and could reach a value that is two to three
though a clear reduction in the breakwater times that calculated by the Goda formula .

....._____ _ _--, +5.10


szH.H.W.U-4.30 +4.10

szHWL.+ 2.30
+1.30
szl.WUO.OO

(unit:m)
Figure 10-79. Higashi·Hazu Port, Japan; typical cross section of a perforation block
masonry breakwater. [From Kataoka and Saida (1986).]
Breakwater Design 1015

ANG - LOCK CROSS-HOLLOW OI AER I G LOO

PERFORCELL PILOCK T 1 NE WAROCK

Figure 10-80. Perforation blocks.

10
wave forces. In addition, the downward force
K.
.!J:i- > o~ acting on the slope cancel at least part of
06 Bloch
)( A
the uplift pressure, thereby increasing the
breakwater stability.
...
B
06 c
E
The sloping-top breakwater has been
04
o G used for many years, with the oldest struc-
H
ture ofthis type constructed at Naples Port,
~

o 1
02 <> J Italy in 1906. Another one was built in the
middle of the 1960s at Hanstholm Harbor,
00
Denmark, where the overturning moment
OI 02 03 0 4
l/L
and the total horizontal force acting on a
breakwater were reduced to about one-half
Figure 10-81. Refiection coefficient of a perforation the values of the vertical-type breakwater
block masonry breakwater.
(Juhl, 1994; Ligteringen, 1994). Sloping-top
caissons have also been constructed in deep
water regions in Taiwan, China, and Libya,
10.3.2 lnclined Walls where wave conditions are severe (Kuo,
1994; Xie, 1994).
10.3.2.1 Sloping-Top Breakwaters In Japan, breakwaters of this type have
been built at Niigata, Miyazaki, and
Breakwater construction in open deep Hitachinaka ports and one is currently
waters has become more prevalent in recent under construction at Naha Port. This
years. In deep water locations, the large latter breakwater is covered with wave-
waves generate tremendous forces acting on dissipating blocks, thus being a combina-
a breakwater, and the sloping-top-type tion of the sloping-top and the wave-
breakwater has been found suitable for this dissipating block breakwater (Sato et al.,
application. Sloping-top breakwaters have a 1992). The typical cross section of the
superstructure that is sloped to reduce the breakwater constructed at Niigata Port is
1016 Breakwater Design

Horbor Side
8 L H
220x 18.0xl6.5

H.W.L. ... ~2.0 +0.5


50.0 48.0

Figure 10-82. Niigata Port, Japan; typical cross section of a sloping-top breakwater. [From
Kataoka and Saida (1986).]

shown in Figure 10-82. This structure is of the sloping breakwater must be higher
designed to dissipate waves having H 1i 3 = than that of the ordinary vertical wall
8.0m, Hmax = 14.3 m, T 113 = 14 s, and breakwater to obtain the same transmis-
() = 21°. sion coefficient, Kr.
The slope of the breakwater is normally
located above sea level due to construction 10.3.2.1.1 wave transmission and
reasons. However, if the sloping part is reftection
extended below the still-water level, the
capability of the breakwater to resist wave Figure 10-83 shows tests results which
attack is increased. have been carried out to investigate the
The sloping-top caisson breakwater is wave transmission coefficient, Kr, for the
very stable, yet the overtopping wave is six profiles of sloping-top breakwaters
significantly large. Hence, the crown height shown in Figure 10-84 (Takahashi et

Kr
Sloping Top Caissons

Type-1 - - •
2 - - - ....
3---.
4 -------- o
5---6
6----- o

0.1

o.o~~-L~__L__L~--L-~-L--L-~-L~~~

0.0 1.0 2.0 3.0

Figure 10-83. Transmission coefficient for sloping-top breakwaters. [From


Takahashi et al. (1994).]
Breakwater Design 1017

Q<;t"
oo.q: 00
type 1 ~ o type2 type3
~~------~~~~

N
o:)
type4 1()
type6

(unit: cm)
Figure 10-84. Typical cross sections of sloping-top breakwaters used in model
tests. [From Takahashi et al. (1994).]

al., 1994). In this :figure, breakwater type either a is smaller andjor the value of de
1 is a standard sloping-top breakwater, is negative and larger. Taking such factors
having a slope that starts from the still- into account, KT for a sloping-top caisson
water level (de= 0): the type 2 structure is can be formulated as follows:
a semisloping-top breakwater, having a
slope starting from above the water level; when ~ + ~s - 2.2 < hc/Hl/ 3 < 2.2 -
types 3 through 6 are semisubmerged (~ + ~s)
sloping-top breakwaters having a sub-
merged slope that enhances the breakwater
stability, creating a substantial downward KT- { 0.25[1- mn( .~.l( ;,:, + ~ + ~.) r
2) 0.5
component of the wave force.

+0.01(-h-)
In these tests, the crest elevation above 1- h'
still-water level, hc, was 32 cm in the case (10-46)
ofbreakwater types 1 through 3, and 16 cm
in the case of breakwater types 4 through when hc/H113 ~ 2.2 - ( ~ + ~8 )
16. The type 5 model had the most deeply
submerged slope (27.6 cm). The slope angle
(a) in ali models but type 6 was equal to KT = 0.1( 1- : )
45°; the latter model had a= 56°. The
where
depth of water, h, in ali experiments was
104.4 cm. ~s = - 0.3[(hc - 2dc)f(H113 tan a)) 0 ·5
From these experiments it is obvious that ~=factor obtained from Figure (10-33)
the sloping-top breakwater has a relatively h = depth of water
large transmission coefficient, KT, com- h' = depth of the caisson
pared to ordinary vertical wall breakwa- hc = crest elevation
ter~, and that KT becomes larger when de= elevation ofthe downward edge ofthe slope;
1018 Breakwater Design

note that when the downward edge of the Goda Formula for vertical breakwaters.
slope is submerged, de is negative Takahashi et al. (1994) further modified the
a = slope angle Goda formula to account for force variation
due to wave height. Figure 10-86 shows the
It should be noted that Kr values obtained design wave pressure distribution on the
from Eq. (10-46) are in good agreement sloping-top breakwater. From this figure, it
with the test results given in Figure 10-84. is seen that wave pressure distribution is
The crown height hc of the standard practically the same as that postulated by
sloping-top breakwater is usually taken the Goda formula.
equal to H 113 , whereas in conventional
Wave Force on the Slope. The horizontal
vertical-type breakwaters, hc is usually
component FsH and the vertical downward
considered tobe equal to 0.6H113 .
component F8 v of wave force on a sloped
The wave-reflection coefficients, K R, ob-
part are obtained from the following:
tained from the aforementioned model tests
as a function of H 1 jh (Fig. 10-85) indicate (10-47)
that wave reflection by the sloping top
breakwaters is smaller than that by the F8 v = FP cos a = X8 LF1 sin a cos a (10-48)
conventional vertical wall breakwater. This
is basically ·attributed to wave overtopping where
and the eddies that are generated at the F1 = horizontal wave force on the equally high
lower edge of the slope. vertical wall
a = slope angle
10.3.2.1.2 wave forces XsL = modification factor, the modification fac-
tor AsL depends on HjL and a and can
In Japan, the design method for sloping- be expressed as
top breakwaters was proposed by Morihira
and Kunita (1979), who used the extended AsL = min{max{l.O, -23(HjL)tan- 2a

+0.46tan- 2a
+sin - 2a}, sin - 2a} (10-49)

and, therefore, XsL sin 2a is equal to


~ti 6

ilw
• e
0111.
:i:
1&<>•
·sO.o

~{·
îjţl
p
KR Jf ... ,.P?Jt
(])
~ - 23(HjL) cos 2a
8
0.6 A<t Il>
rf! + 0.46 cos 2a + 1.0}, 1.0} (10-50)
h/L 023 0.14 0099
Typel • o o
2
3
.A.


ţi

ţf "'
D
4 <) <D e
5 ~ & >4.
6 [J IIl 8

O.OL---~-----L-----L----~--~
0.0 0.1 0.2 0.3 0.4 0.5
Hrlh

Figure 10-85. Reflection coefficient for sloping-top Figure 10-86. Design wave force acting on a slop-
breakwaters. [From Takahashi et al. (1994).] ing-top breakwater.
Breakwater Design 1019

In practica! calculations of FsH, the effect types of the sloping-top breakwater and are
of XsL is as follows: shown in Figure 10-84. In Figure 10-87, Fc
and Fca respectively are weights of the
(i) When HjL is relatively small, FsH = F 1 sloping-top breakwater model and the verti-
(ii) When HjL is large, FsH = F 1 sin2a cal one that are required to resist the wave
(iii) When HjL is between cases (i) and (ii), attack having a H = 61.9 cm and T = 3.5 s.
FsH decreases with the increase in HjL. As is seen in Figure 10-87, among break-
water types 1 through 3, the latter has the
Note: When HjL is large and a= 45°, then smallest Fc value and, in general, the
FsH is equal to one-halfthe horizontal wave sloping-top breakwater stability is en-
force on an equivalent portion of a vertical hanced with decrease in either de or a.
wall, F 1 . Also, the required weight of the sloping top
Wave Force on the Vertical Section. The breakwater is approximately 60-80% of
wave forces on a vertical portion of a slop- that of the conventional vertical wall break-
ing top breakwater, F v, and the total uplift water that is needed to resist an identica!
force, Fu, can be calculated from the Goda wave force.
formula, as follows:
10.3.2.2 Trapezoidal-Type
Fv = >..vF2 (10-51) Breakwaters · ·
Fu= 0.5puB (10-52)
Trapezoidal-type breakwaters belong to
where p u is the uplift pressure at the the category of the vertical breakwater; in
breakwater wall front toe and F 2 is the general, however, they are more stable
wave force on the vertical part of the slop- against waves attack than the conventional
ing breakwater, both calculated by the Goda vertical wall structures. Tanimoto and
formula. The modification coefficient Av is Kimura (1985) carried out a series of model
obtained from the following: tests to investigate wave forces on a trape-
zoidal breakwater. They have found that
A.v = min{l.O, max{1.1, 1.1 + lldc/L} the wave pressure on a trapezoidal break-
-5.0HjL} (10-53) water can be determined by the Goda for-
mula with wave pressure distribution as
where de is the height of the slope. The indicated in Figure 10-88. Similar to
coefficient Av indicates reduction in wave
pressure due to wave rapid upward velocity
induced by the slope.
The value of wave forces calculated by 1.0
O Colulat ional
the proposed method are in good agree- • Experimental
ment with results of the model tests. This 1)
'l'"
suggests that the proposed method can
evaluate reasonably well the effects of 0 .5 1-
1

structural variations on stability of the


sloping-top breakwaters.
o
10.3.2.1.3 Stability. of S/oping-Top Type 1 2 3 4 5 6
Caissons
Figure 10-87. Required weight of sloping-top break-
The data presented in Figure 10-87 are water in comparison ta conventional vertical wall
based on test results carried out on six breakwater. [From Takahashi et al. (1994).]
1020 Breakwater Design

sloping-top breakwaters, the vertical com- This, of course, is due to greater width of
ponent of the wave pressure contributes to the wall base.
better stability of trapezoidal breakwaters.
The Tanimoto and Kimura experiments
demonstrated that the uplift wave pressure
on the bottom of the upright section is re- 10.4 DESIGN OF HORIZONTALLV
duced due to upward water-particle velocity COMPOSITE BREAKWATERS
enhanced by the breakwater slope. These
investigators suggested that the modifica- A typical cross section of a horizontally
tion factor ..\3 for the uplift pressure in the composite breakwater covered with con-
Goda formula can be obtained from crete blocks is shown in Figure 10-89. This
breakwater is composed of a 14.5-m-wide
floated-in concrete caisson protected with
32-t concrete blocks placed at a slope of
1: 4/3. These blocks (tetrapods) dissipate
where ld = h' tan a' (a' is the slope angle wave energy and reduce the breaking-wave
of the front wall). Equation (10-54) is appli- force. The top level of the block section is at
cable to a trapezoidal caisson with inclina- the same elevation as the caisson crown,
tion up to 20° and ld < 0.1L, where Lis the and the width of the block section is twice
wavelength. the block height. This breakwater has been
Another advantage of trapezoidal geome- designed to withstand design breaking
try of the upright section is a smaller toe waves having H 113 = 6.01 m, Hmax = 9.96
pressure applied on a rubber foundation. m, and T 113 = 14 s.

Figure 10-88. Wave pressure distribution on a trapezoidal breakwater. [From


Tanimoto et al. (1985).]
Breakwater Design 1021

145 2
+5.0 +

.
g L.W.U0.43
~--

Tetropod32
3.03.0

Figure 10-89. Typical cross section of a horizontally composite breakwater. [From


Kataoka et al. (1986).]

10.4.1 wave Transmission and 0.4.----.--.----,---.,-----r--,


Reflection Kr b.li..vo•0.05-0.07 1
b. "2 (Il,+ bol o
b.ILII>
o 05-0 07
b•ILI/3•0.07-0.IO~~ o:o1-o:1o
• 0.10-0.15
10.4.1.1 Transmission coefficient o hh (h/h.•0.6)

The transmission coefficient of a horizon-


tally composite breakwater, KT, to some
extent is in:fluenced by the width of the
berm made up by covering concrete blocks
relative to the length of the incident wave
and by the principal parameter hc/H113 •
Tanimoto et al. (1987) carried out inves-
tigation of the horizontally composite Figure 10-90. Wave transmission coefficient Kr of
breakwater exposed to irregular waves. In horizontally composite breakwaters. [From Tanimoto
the process, the heights of significant waves et al. (1987).]
have been measured. Results of these inves-
tigations are presented in Figure 10-90 in fluenced by the relative covering width, B.,
the form KT = f(hJH 113 ) for different ra- which is computed from the following for-
tios of befL 113 , where be = 0.5(bc + bb); be mulation (Tanimoto et al., 1987):
is defined as the equivalent covering width,
and bc and bb are respectively the width of B =b - ( - -
cot a )
the berm and the base of the concrete block • O h + hc
installation in front of the caisson.
ABis seen from Figure 10-90, the value X [Jhc ( cosh 2 27T(h + z) jL) z dz
of KT ranges from 0.10 to 0.16 for a relative -h' cosh 2 27ThjL
crest height of 0.6, which is smaller than
corresponding values for the ordinary verti- +0.5h~] (10-55)
cal wall breakwater; obviously, the value of
KT decreases as the relative width, b.,
increases. where
b0 = covering width at the still-water level
10.4.1.2 Reflection coefficient a = slope angle of the protective structure mea-
sured from horizontal
The reflection coefficient, K R, of horizon- z = upward distance away from the still-water
tally composite breakwaters is strongly in- level
1022 Breakwater Design

I.O~-~--r--,-::::=r;;=:::;::::=:f=;:::::::::;, (a) - - - Without Blocks


1
{ h,/h•0.21 ~0.40 1 ...§.._ h lh
B.1,L111 • 0.087-0098 L11s o-0.21o2-o.41a+oE
1 \ 1 ao&- aoa o o
0.8 ( h.l h•0.26 - 0.08- 0.10 • • ----------- With Blocks
B./ Lvs•0.079 0.10- 0.12 Gl .&.
.12- 0.14 a e A
D 0 o .14-0.16 IJ
0.61---~---''--'0"---,.,"'-
" 0 1 l ~jl_ L <>·o·--o..J_
o 16- o 18 ...

0.4
• ~~ ..: :-r.t~~~l----"
v.,-·"';,"J"...:..f=+~ ,.,. rli~
~i~d- ~•A<>
1'
{ h./h•02B-o40/
&/L"s•O.II •• ""1 ~ .,..
-•
. \
n=
A
02~--+--~--+-~~---~~-~ '\.rh.lh •02B-0.40
'j·'L,"·O.I4-f"'5 p
0o~-~o~.~~-~o~.2-~0~3~-0~.4~-~oL.5-~o~

Figure 10-91. Reflection coeffi.cient KR of horizon-


tally composite breakwaters. [From Tanimoto et al.
(1987).]

Results of the model tests carried out by


Tanimoto et al. (1987) are depicted in Fig-
ure 10-91. In this figure, KR is represented Figure 10-92. Change in wave pressure profile by
as a function of H 113 jh, hcfh, and BJL 113 • wave-dissipating blocks: (a) strongly breaking wave;
(b) breaking wave; (c) nonbreaking wave.
The results of the experiment indicate that
value of K R clearly decreases with an in-
crease in the BJL 113 value.

1.2 h 1 L"• • 0.073-0.108

10.4.2 wave and Block Load 1.0 ---------------


..... ·...: .. _
.-:-.~ -~.- ....
on a vertical Wall ...
0.81----------='----.......::-.:--:..:o....__ ------

0.6
10.4.2.1 wave Force 0.4 bo/L11s
• 0.046
The concrete blocks in front of the up- 0.2 • 0.053
right wall act to reduce the wave force on
the wall. Figure 10-92 illustrates the time
history of wave pressure obtained from a
model test, where the pressure profile with Figure 10-93. Reduction of wave force on the verti-
and without wave-dissipating blocks is indi- cal wall due to the presence of wave-dissipating blocks.
[From Takahashi et al. (1990b).]
cated. These results demonstrate that the
pressure component due to wave breaking
is significantly reduced by the blocks,
whereas the slowly varying wave pressure height. In fact, when the wave height is
component only slightly changes. close to the water depth, the wave force
Figure 10-93 shows the ratio A1 of the ratio A1 is reduced to less than 0.8. Based
wave force acting on the vertical wall pro- on these results, it is possible to conclude
tected by blocks to that without blocks as a that the design wave force on a vertical wall
function of H jh, where H is the wave covered with concrete blocks can be evalu-
height and h is the water depth. The ratio ated by the Goda pressure formula after
A1 tends to decrease with increases in wave incorporating in it the following modifica-
Breakwater Design 1023

tion factors (Takahashi et al., 1990b): deep waters exposed to relatively small
waves, the block force is usually very small
Ă 1 = ,\3 = 1.0 for Hjh < 0.3 in comparison with a wave force and can
,\1 = .,\3 =1.2- ~(Hjh) for 0.3 :s; Hjh<0 .6 therefore be conveniently neglected.
,\1 = ,\3 = 0.8 for Hjh ~ 0.6
,\2 =o (10-56)
10.4.3 stability of
When A1 , A2 , and A3 are applied to the wave~Dissipating
ordinary block section, it becomes apparent concrete Blocks
that the wave force is significantly reduced
by the covering blocks. lf the covering width 10.4.3.1 8/ock section
is enlarged, the wave force is reduced even
further (lnagaki et al., 1987), and the strong Many type of shape-designed, wave-
impulsive pressures are generated if the dissipating concrete blocks have been used
covering is insufficient (Kougami and in horizontally composite breakwater, some
Tokikawa, 1970). of which are described later in Section 10.5
and illustrated in Figure 10-103. The
10.4.2.2 8/ock force weight of a block usually varies from sev-
era! tonnes to up to 80 tonnes. Most types of
When concrete blocks are installed in wave-dissipating blocks are fabricated from
front of the vertical wall, the wall acts as a plane concrete; however, high interlocking
retaining wall supporting them. The resul~ blocks such as Dolos blocks are reinforced
tant block force on a wall can be significant, with steel bars.
especially in deep water. However, this force The wave-dissipating blocks should com-
is greatly reduced by movements of the up- pletely cover the vertical wall to prevent
right section that is attributed to the wave the impulsive wave-breaking pressure on
force (Fig. 10-94). Consequently, with the the wall. The top level of the blocks should
exception of the breakwaters constructed in therefore be at the same elevation with the
vertical wall crown, whereas the width of
(a) Before Wove Action
the block section at the top should be at
least as twice as much as the height of the
block. The slope of the block section, a , is
usually steep (i.e., cot a = 1.34 to 1.5) and
the gentle slope (cot a > 2) is seldom
employed.

10.4.3.2 weight of the 8/ocks


( b) Dur ing Wove Act ion Although the wave action on the wave-
dissipating blocks of a horizontally compos-
ite breakwater is slightly different from that
on the armor blocks of a rubble-mound
breakwater, the Hudson formula can be em-
ployed to evaluate the required weight of
the concrete blocks. Recently, van der Meer
(1988b) modified the Hudson formula by
Figure 10-94. Wall displacement by wave action. including parameters in it, such as the
1024 Breakwater Design

damage rate and number of waves (details Consequently, to obtain the most cost-
are provided in Section 10.5.2). effective solution it is prudent to carry
The following should be considered in the out the model. This can help to deter-
block section design. mine the optimum shape of the block sec-
tion, including the shape and mass of the
concrete block. For the same purpose, the
1. A large wave height requires employment results of the previously conducted exper-
of a heavy-weight block strength, which, iments in which the identica! conditions
as it follows from the recent spectacular have been employed can also be consid-
failure of the breakwater at Sines, ered. Tanimoto et al. (1985) and Kajima
Portugal, is of paramount importance. To (1994) carried out comprehensive studies
avoid breakwater failure due to breakage on the stability of wave-dissipating blocks
of the shape-designed heavy concrete used for horizontally composite break-
blocks, the slope of block section can be waters.
made flatter and the block's specific den-
sity can be changed as required. Details
are provided in Section 10.5. Also, it must
be realized that the wave-dissipating
blocks at the breakwater head are usually
10.5 Design of Rubble-Mound
less stable compared to those at the Breakwaters
breakwater trunk.
2. In practice, the required weight of the Two representative rubble-mound break-
wave-dissipating concrete blocks is less waters are illustrated in Figures 10-95 and
than that calculated by the Hudson or 10-96.
van der Meer formulas. This is because The Bilbao, Spain breakwater depicted
the stability of such blocks placed in front in Figure 10-95 is an example of a modern
of the vertical wall is inherently higher large rubble-mound breakwater built in
than the stability of similar armor blocks deep water (Burcharth and Frigaard, 1995).
installed on a rubble slope. This is basi-
It has a core made of a quarry run which is
cally attributed to the higher permeabil-
covered by the armor layers of rocks and
ity of the block section that is placed
against vertical wall, but an additional shape-designed concrete blocks; a large
factor is that both the Hudson and van wave screen (wave wall) is constructed on
der Meer formulas are developed for cal- the top of the rubble mound.
culating the weight of armor blocks Many rubble-mound breakwaters have
intended to protect the rubble mounds. been constructed in rather shallow water

r
Ali b}acks are parallelepiped b]ocks 1:1:1.25
Measures in m
29.00
+14.00
_:sL

Figure 10-95. Rubble-mound breakwater in deep water. [From Bureharth and Frigaard (1995).]
Breakwater Design 1025

Seoword Side Horbor Side

{unit: ml

Figure 10-96. Yokosuka Port, Japan; Rubble-mound breakwater built in


shallow water. [From Kataoka and Saida (1986).]

areas. Figure 10-96 illustrates the cross sion and refl.ection (Barends, 1988; van der
section of one such breakwater projected Meer, 1988a; and others). Wave breaking
seaward normally to the shoreline. It has a and penetration also affect the breakwater's
core of a quarry run, which is protected stability.
from the wave attack by 5-tonne tetrapod
blocks. 10.5.1.1 wave Transmission
Wave transmission by the rubble-mound
breakwater has been extensively studied in
10.5.1 wave Transmission and the past (Saville, 1963; Numata, 1976;
Reflection Tanaka, 1976; Seelig, 1980; Allsop, 1983;
Sakamoto et al., 1984; Powell and Allsop,
Wave action on the rubble-mound breakwa- 1985) At present, results of many investiga-
ter is very different from that on a vertical tions are available to the breakwater de-
wall breakwater; therefore, its wave trans- signer to determine the wave transmission
mission and refl.ection characteristics are coeffi.cient, KT. Madsen and White (1979)
different. Wave breaking on a slope affects developed the analytical solution for evalu-
the wave transmission and refl.ection, and ation of KT. Some of these studies will be
under this condition, the wave transmission reviewed later in order to better under-
and refl.ection parameters are determined stand the basic characteristics of wave
mainly by the slope angle a and the wave transmission by the rubble-mound break-
steepness s 0 m = H 113 jL 0 . These character- water.
istics can be expressed by a function of the Figure 10-97 shows a diagram of wave
surf-similarityparameter Ir = tanajsg!, as transmission over a rubble-mound break-
is shown in Figure 10-99. Extensive studies
1.0
have been carried out to investigate wave O. 1--
Kr 1 1

breaking on the slope, including wave run- ~?5


07
r-- .......___
0.8
1
0.1 t-- :- r-.... '\.
up and overtopping (TAW, 1974; Bottin et r-- .......... ~ '\ .6- ~.
8

~
02 1--

" '"
.......... ţ2-
al., 1976; Kamphuis and Mohamed, 1978; a4 0 ;..:::::-
0.8 ......._
Gunbak, 1979; Ahrens, 1981; Owen, 1980, B!Lo•
~
~ .~ ~
~ .'\..'\ ro\""'
1982; Jensen, 1984; Allsop et al., 1985;
Bradbury et al., 1988; Aminti and Franco, ~ ~,,
1'---
-
~ . ::>.. ...........

1988; van der Meer and Angremond, 1991). -3.0 -2.0 -1.0 o~ 1.0 2.0
Wave penetration into the breakwater is
another special feature ofthe rubble-mound Figure 10-97. Wave transmission diagram. [From
breakwater which affects wave transmis- Tanaka (1976).]
1026 Breakwater Design

water developed by Tanaka (1976), in which tions proposed by Seelig (1980): where
Kr varies depending on crown width B and
crest height he. From this diagram, it fol- K T = ( Kro
2 2 )0.5
+ Krp (10-57)
lows that when the crown elevation is rela-
tively high (hcfH113 > 1.0), the wave trans- (10-58)
mission is small, being governed by wave
penetration into the breakwater body. When in which Ce= 0.51- O.llBj(h + hc). In
the crown is close to the stiU water level, these formulations R is the run-up height
the wave transmission occurs through over- given in Figure 10-99 as a function of the
topping the breakwater, although, in gen- surf-similarity parameter I, and Kro and
eral, it is reduced by refiection and break- Krp are the transmission coef:ficients due to
ing. A large crown width enhances wave overtopping and due to wave penetration
breaking and reduces wave transmission. It into the breakwater body, respectively. For
should be noted that Kr for this type of example, when tan a = 1/1.5, H 113 jL 0 =
rubble-mound breakwater shows similar 0.04, B j(h + he) = 0.5, and hjH113 = 0.6,
tendencies to that of a vertical wall break- then Ce = 0.455, I, = 3.33, RjH113 = 0.9,
water as is shown in Figure 10-34. and Kro = 0.15. Here, L 0 is the wave-
Figure 10-98 shows tests results ob- length in deep water. When the wave pene-
tained by Allsop (1983) for a stone-armored tration into the breakwater body is in-
rubble-mound breakwater, where the effect signi:ficant, the wave transmission coef:fi-
of wave steepness s m = H / L and h c on the cient Kr = Kro·
transmission coef:ficient is apparent (i.e., Wave transmission over a sloping break-
waves with a smaller sm have a larger Kr). water with concrete armor blocks is primar-
The wave transmission Kr can be evalu- ily determined by the amount of wave pene-
ated using the following empirica! formula- tration. To evaluate Kr, Sakamoto et al.
(1984) proposed the following empirica} ex-
pressions:

(10-59)

t~,--- 2.0
RIPRJPIAH~ENS s1o MC~AR~EY 19k -
- -- --
1.8
RUBBLE · MOUND BREAKWATER
1.6 t- IGUNBAK,I979
hclh Sm
.- .-·
"""
------
.. 0.23}
1.4
/V' .).··
.--
• 0.39 0.04
.-· .-·-· f-·
• 0.56 1.2 / /

.. 0.39}
<l:l:î:t.o /
--- . 1-'
• 0.56 0.005-0.02 ll::i:l:
0.8
1_, ;..~
><;: lOS ARMORED BREAKWATE~-
'·:'/ / liTTIN, CHATiM,AND CARVER,
0.6
::/_,~. 1 1 19761

0.4
!IV 'R~~~~O';l~~sN:feiBREAKWATERS

0.2
11
'}'
005 0.10 0.15 0.20 0.25 o
o 23456789
.!k.. (~)0.5 lr•..!Qn..g_
H". 2..- ,Jii7f;

Figure 10-98. Wave transmission coefficient Kr of Figure 10-99. Wave run-up on a rubble-mound
a rubble-mound breakwater. [From Allsop (1983).] breakwater and riprap. [From Seelig (1980).]
Breakwater Design 1027

KT2 = 1.8(B.!L - 0.6)(hjH1/3 - 0.25) + 0.04 KR


(10-61) 1.00 Trend for smooth, lmpermeable slope

0.80
(10-62) Rip-rap(GUnbak)
0.60

0.40 Dolos(Wallingford)
where
0.20
B. = width of the breakwater at the still-water
level 1.0 20 3.0 4.0 50 6.0 7.0 8.0
lr
d B = height of the block
!fiB = coefficient that takes into account the shape Figure 10-100. Variation in wave refl.ection coeffi-
ofthe block cient KR. [From Losada (1980).]

!fiB = 1/1/1- 1/Ja)/1/12 (10-63) greatly reduced when wave overtopping


does occur.
1/11 =dVA (10-64)

1/12 = d~jV (10-65)


10.5.2 Design of Armor Layer
10.5.2.1 wave Force on Armor
where
Layer and Armor-Layer
A and V = surface area and volume of the block, Stability
respectively
1/13 = porosity of block section The armor layer is a very important com-
ponent of the rubble-mound breakwater;
For a tetrapod block, 1/11 = 2.69, 1/12 = 0.278, that is, if it is damaged, then the inner body
1/13 = 0.5, and 1/JB = 4.84. For example, if of the breakwater may be easily eroded by
h = 6 m, B. = 11.1 m, dB = 1.79 m, H 113 = the direct wave action.
32 m, and L = 73.6 m, then KA = 34.5 and, lf the slope is attacked by waves, then
therefore, Kr = 0.26. the resulting water movements inside the
breakwater body may cause displacement of
10.5.1.2 wave Reflection the rubble stones on the slope. Figure
10-101 illustrates effects of wave uprush
Many studies have been carried out on and downrush on a rubble-mound slope.
the wave-reflection coefficient KR (Losada Damage to the armor layer due to this
and Gimenez-Curto, 1980; Allsop, 1990; van movement of water usually occurs near the
der Meer, 1991, and others). Coefficient KR still-water level; thus, the stability of the
of a rubble-mound breakwater is generally armor layer near this level should be given
small because attacking waves are usually
effectively dissipated by this type of break-
water. However, as the slope angle in-
creases and the wave steepness decreases,
K R increases. Furthermore, if roughness of
the slope is small, then K R is large. Figure
10-100 shows KR as a function ofthe surf-
similarity parameter, Ir for several slope
types in which no overtopping occurs: It Figure 10-101. Wave action on a rubble-mound
should be noted that coefficient K R is breakwater.
1028 Breakwater Design

special attention in the design of the and Pn can be obtained from the following:
rubble-mound breakwşter.
The forces acting on the armor stones are Ph = CnpogHAh = Cnp 0 gHkh(W jp,g) 213
shown in Figure 10-102. There, Ph is the (10-70)
horizontal wave force, Pn is the uplift wave
force, W' is the weight of the stone, and FP
is the frictional force that resists the two
wave forces; FP and W' are expressed as
where
(10-66) H = wave height
Cn and CL = drag and uplift coeffi.cients, respec-
(10-67) tively
kh and kn = coeffi.cients that consider shape of
where f.Lp is the coefficient of friction and the stone
Sr = p,/ Po is the specific gravity of armor
stone, Pr is the mass density of armor stone, Hence, the required weight ofthe stone, W,
and p 0 is the mass density of water. The is determined from the following:
armor stone stability is determined from
the following: W = PrgH 3 !N.3 (Sr - 1)3 (10-72)

in which the parameter N. is called the


(10-68) "stability number"; N. accounts for effects
of stone shape and wave conditions and is
therefore, determined from

W' > (Ph + f.LpPn)/( f.Lp cos a± sin a) (10-69) N.3 = ( f.LpknCL ± khCn) 3 / ( f.Lp cos a± sin a) 3
(10-73)
Forces Ph and Pn are mainly due to the
Equation (10-72) can be rewritten using
drag and uplift forces generated by waves;
both of these forces are proportional to the stone nominal diameter Dn as follows:
square of the water-particle velocity; it is Hj(t:.Dn) = N. (10-74)
assumed to be proportional to (gH) 0·5 • Be-
cause the drag and uplift forces are also where
proportional to the respective projected Dn = (W/ Prg )1/3 (10-75)
areas Ah and An ofthe stone, the forces Ph
and
!:". = s,- 1 (10-76)

Equation (10-74) is very useful for under-


standing the scale of the structure in com-
parison to the wave height. Table 10-7
shows the dimensional range of coasta!
structures using parameter Hjtl.Dn, which
is equal to the stability number N •.

10.5.2.2 Armor Blocks


The invention of rectangular and shape-
Figu.re 10-102. Wave forces acting on armor stone. designed concrete blocks has promoted the
Breakwater Design 1029

Table 10-7. Classification oî coastal structures categories:


by Hji1D

Hji1D Structure 1. Shape (bulky or slender)


2. Porosity (porous or solid as one unit and
0-1 Caisson or sea walls as a section)
1-4 Stable breakwater
3-6 8-shaped and berm breakwaters 3. Interlocking (high or weak)
6-20 Rock slopesjbeaches 4. Slope inclination (steep or gentle)
15-500 Gravei beaches 5. Placement method (random or pattern
500 and up Sand beaches (during storm surges) placing (pitching), one layer or two layers)
Source: From CIRIAjCUR (1991). 6. Wave energy dissipation (high or low)

For example, the tetrapods have a rela-


development and use of the rubble-mound tively slender shape and relatively high i:n-
breakwater. These concrete blocks are now terlocking ability and, therefore, can be used
frequently used where waves are high and with a steep slope. Tetrapods are usually
the rubble stones are not heavy enough to randomly placed. This results in signifi-
safely sustain wave forces. cantly large porosity of the armor section.
Figure 10-103 illustrates some types of Thus, the armor layer roade from tetrapods
concrete blocks presently used for break- signi:ficantly dissipates the wave energy. On
water and coastal protection construction. the other hand, the Dolos blocks have a
More structural types of shape-designed higher interlocking ability than the tet-
block are found in Chapter 2. The armor rapods, but they are structurally weaker
blocks can be classi:fied in the following because of their slender shape. Strength of

ACCROPOD AKMON ANTIFER BLOCK BEEHIVE COB CUBE(modifiedl DOLOS

GAMMA-L GASSHO GROBBELAAR HEXALEG HOLLOW HOLLOW SANREN


SQUARE TETRAHEORON

SEALOCK STOLK CUBE SVEE TETRAPOD TOSKA~E TRIB AR X- BLOCK

Figure 10-103. Shape-designed concrete blocks. [Modified from Burcharth (1993).]


1030 Breakwater Design

the Dolos blocks can be increased if the Dn = 2.79 m is needed to resist the attack of
large waist ratio (r > 0.4) is employed; r is the design wave, H 113 = 7.67 m. This value
ratio of the leg width to leg length. In re- of Dn = 2. 79 m corresponds to block weight
cent years, several new bulky blocks with of 50 tf (490 kN).
increased structural strength (e.g., accro- For design waves higher than 10 m, the
pode and others) have been developed and required weight of armor block, W, becomes
successfully used. very large; hence, to reduce the weight of
Blocks like the X-block and hollow-square the armor blocks, the shape-designed blocks
block have small interlocking capability and with a high Kv value are used andjor
are quite different from those like tetrapod more a gentle slope is constructed.
or dolos. Blocks like these are normally used Recently, blocks having a high specific
for cladding the gentle rubble slopes. density, Pn have been employed; that is,
blocks made from special aggregates of
10.5.2.3 Hudson Formula heavy stones or iron ores having a Pr value
of up to 4.0 tjm 3 . If this kind of material
The evaluation of required weight of the
would be used in the above example, then
armor blocks or stones is one of the most
the nominal diameter of the block, Dn is
important aspects in the design of a
reduced to 1.63 m and its required weight
rubble-mound breakwaters. This can be
W will be only 14 tf.
done either analytically or by conducting
Recommended values for K D are given in
model tests. Also, the result of previously
Table 10-8. As indicated in this table, the
performed tests for the identical design con-
ditions can be used for the design purpose.
K D values for the breakwater head are
slightly smaller than those for the break-
Many formulas have been proposed by dif-
water trunk: This difference in K D values
ferent investigators for determination of the
reflects the greater instability of blocks lo-
weight of the armor block (Castro, 1933;
cated at the breakwater head. Also, the
lribarren, 1938; Larras, 1952; Hudson,
block stability is reduced when the break-
1958, 1959; Hedar, 1965; Price, 1979; van
water crown height is low andjor when the
der Meer, 1988b). Fundamentally, ali pro-
sea-bottom slope is steep.
posed formulations are very similar to Eq.
(10-72). The most popular one is the Hud- Note that the U.S. Army Corps of Engi-
neers (1984) Shore Protection Manual rec-
son formula, which is a modified version of
ommends the wave H 1110 as the design
the Iribarren formular. It is similar to Eq.
(10-72) in which the stability number N 8 is wave as opposed to a significant wave, H 113 .
defined as
10.5.2.4 New Formula
(10-77) Extensive efforts to revise the design
method of the rubble-mound breakwater
where the coefficient K D represents the have been recently made elsewhere in the
difference in the shape of the stone or block world, especially in Europe and North
(e.g., Kv = 8.3 for tetrapods) and a is the America. One of the most beneficia! results
slope angle. The Hudson formula is the most of this effort is a new formula, for evalua-
popular one. It was extensively used in the tion of the required weight of block or stone
past 25-30 or so years. The design wave in ăeveloped by van der Meer (1988a and
this formula is the significant wave. For 1988b). As noted earlier, this formula philo-
example, if cot a= 4j3 and Kv = 8.3, then sophically is similar to Hudson's formula.
Ns = 2.228. By assuming p 0 = 1.03 tons/ However, van der Meer developed the more
m 3 , a nominal diameter of armor stone of refined approach to calculation of the stone
Breakwater Design 1031

Table 10-8. Suggested K D values for use in determining armor unit weight

No-Damage Criteria and Minor Overtopping


Structure Trunk Structure Head
K}$ Kv Slope
Breaking N onbreaking Breaking N onbreaking cot 8
ArmorUnits na Placement Wave Wave Wave Wave

Quarrystone
Smooth rounded 2 Random 1.2c 2.4 1.1 1.9 1.5-3.0
Smooth rounded >3 Random 1.6 3.2 1.4 2.3 d
Rough angular 1 Randomb b 2.9 b 2.3 d
1.9 3.2 1.5
Rough angular 2 Random 2.0 4.0 1.6 2.8 2.0
1.3 2.3 3.0
Rough angular >3 Random 2.2 4.5 2.1 4.2 d
Rough angular 2 Special" 5.8 7.0 5.3 6.4 d
Parallelepipedr 2 Speciale 7.0-20.0 8.5-24.0
Tetrapod and 5.0 6.0 1.5
quadripod 2 Random 7.0 8.0 4.5 5.5 2.0
3.5 4.0 3.0
8.3 9.0 1.5
Trib ar 2 Random 9.0 10.0 7.8 8.5 2.0
6.0 6.5 3.0
Dolos 2 Random 15.8g 31.8g 8.0 16.0 2.0h
7.0 14.0 3.0
Modified cube 2 Random 6.5 7.5 5.0 d
Hexapod 2 Random 8.0 9.5 5.0 7.0 d
Toskane 2 Random 11.0 22.0 d
Trib ar 1 Uniform 12.0 15.0 7.5 9.5 d
Quarrystone (KRR)
Graded angular Random 2.2 2.5

an is the number of units comprising the thickness of the armor layer.


b The use of single layer of quarrystone armor units is not recommended for structures subject to breaking waves, and
only under special conditiona for structures subject to nonbreaking waves. When it is used, the stone should be carefully
placed.
c Caution: Those K n values shown in italics are unsupported by test results and are only provided for preliminary
design purposes.
d Until more information is available on the variation of K D values with slope, the use of K D should be limited to slopes
ranging from 1 : 1.5 to 1 : 3. Some armor units tested on a structure head indicate a K v-slope dependence.
• Special placement with long axis of stone placed perpendicular to structura face.
f Parallelepiped-shaped stone: long slab-like stone with the long dimension about three times the shortest dimension
(Markle and Davidson, 1979).
g Refers to no-damage criteria ( < 5% displacement, rocking, etc.); if no rocking ( < 2%) is desired, reduce Kn by 50%
(Zwamborn and Van Niekerk, 1982).
h Stability of dolosse on slopes steeper than 1 : 2 should be substantiated by site-specific model tests.
i Applicable to slopes ranging from 1: 1.5 to 1 : 5.
Source: From U.S. Army Corps of Engineers (1984).
1032 Breakwater Design

stability number, N.. The following is . a Table 10-9. Design values of S for a 2Dnso thick
discussion of N. as proposed by van der armor layer
Meer. ------------------
Initial Intermediate
Slope Damage Damage Failure
10.5.2.4.1 N5 for armor stones 1:1.5 2 3-5 8
According to van der Meer, the N. for 1:2 2 4-6 8
1:3 2 6-9 12
armor stones can be determined from the
1:4 3 8-12 17
following: 1:6 3 8-12 17
(10-78) Source: From CIRIAjCUR (1991).

where
S.WL.
N spl = 8 •7(H1/20 jH1/3 )- 1po.1s(So.2/No.1)J-o.s
r
(10-80)

Nssr = 1.4(H1j20/H1/3) - 1p-0. 13


Figure 10-104. Definition of erosion area. [From
X(S 0 ·2 jN°· 1 )(cot a) 0 ' 5 If (10-81) CIRIA/CUR (1991).]

where
Nspl = stability number for plunging waves The stone stability number, N., as defined
Nssr = stability number for surging waves by van der Meer has a much more complex
Ir = surf-similarity parameter form in comparison to Hudson's formula. In
Ir = tan ajsg! addition to a, it includes effects of N, S,
s0 m = wave steepness; s 0 m = H 113 jL 0 s 0 m, and P. It should be noted that high
L0 = wavelength in deep water; L 0 = permeability significantly increases the sta-
gTf;3/2'TT bility of the armor layer. This is basically
T 113 = significant-wave period
governed by its inner structure (Fig.
H 113 = significant-wave height
10-105).
H 1120 = average of the highest 1/20 wave
heights (see Fig. 10-106)
~(al
a = slope angle
~
.!J. = relative mass density; !J. = Prl p 0 - 1

)
Pr = mass density of stone

A
Po = mass density of water
Dnso = nominal diameter of stone; Dnso =
(MsoiPY 13 6
M 50 = 50% value of mass distribution curve

P = permeability coeffi.cient of the structures


(see Fig. 10-105) 1>1«' Nofiller
No core
S = damage level; S = AjD;50 can be ob- OAM)Aanominol dlometer of ormour stane
tained from Table 10-9 Oa~F=nominaJ diometer Of filter material
o.~·nomlnol diometer of core
A = erosion area of the cross section (Fig.
10-104) Figure 10-105. Nominal permeability factor. [From
N = number of waves (storm duration) CIRIA/CUR (1991).]
Breakwater Design 1033

subsequently proposed alternative formulas


for Ns for each of these blocks.

(a) Cubes with cot a = 1.5

N. = Hl!a/fiDn = (6.7N0°.4N- 0 ·3 + l.O)s 0!·1


(10-82)

where N 0 is the relative damage which


defines the number of displaced blocks
within width Dn in the direction ofbreak-
water alignment. The value of N 0 is ap-
proximately twice as much as the value of
S, severe damage corresponds to N 0 of
more than 1 and up.
(b) Tetrapods with cot a= 1.5

= (3.75Ng· 5N- 0 ·25 + 0.85)s 0!·2 (10-83)

where severe damage corresponds to


N 0 > 0.5.
2 3 4
(c) Accropode blocks with cot a= 1.33.
hiH~
The damage has no significant correlation
Figu~e 10-106. Wave height ratio H 1120 jH1.3 ~s a with wave number or wave steepness;
functwn of hjH0; H 0 is the equivalent deep water
wave height which includes the effects of diffraction therefore, N8 = 3. 7 for the start of damage
and refraction. (No = O) and Ns = 4.1 for failure (N0 >
0.5). Because start of acropod damage and
its failure are very close then for design
Equations (10-80) and (10-81) include
purposes Ns = 2.5 is recommended.
the wave heigl;:tt ratio H 1120 jH113 which
All values of Ns determined as described
expresses the effect of wave breaking. In
in cases (a), (b), and (c) are applicable to
the original van der Meer formula ' H 2% is
. breakwaters with a high-crown elevation.
used mstead of H 1120 , although numeri-
The stability number of low-crest breakwa-
cally they are almost the same. The ratio
ters, including submerged breakwaters, has
H1;2o/H113 is typically equal to 1.4 for non-
been studied by many investigators (Ahrens,
breaking waves and it is smaller for break-
1987; Allsop, 1983; Jensen, 1984; Givler and
ing waves (Fig. 10-106).
Sorensen, 1986; van der Meer, 1991; Losada,
1990) and the interested reader is referred
10.5.2.4.2 stabilitv number for to works of these authors.
cubes, tetrapods, and Efforts directed at establishing a reliable
accropode blocks formula for miscellaneous concrete blocks
Van der Meer carried out the model tests are still in progress; for example, Burcharth
on cube, tetrapod, and accropode blocks and and Liu (1992) proposed a new formula for
1034 Breakwater Design

Dolos blocks: ments due to the scale effect (Timco and


Mansard, 1983).
Ns = HI;a/!l.Dn = (47- 72r)cpn= 2 D~I 2 N- 0 · 1
(10-84)
10.5.3 lnner Layers. core, Toe,
and wave screen
where
D 0 =relative number of units displaced one 10.5.3.1 lnner Layer, core and
Dolos height within levels SWL ± 6.5Dn Toe
(SWL = sea-water level)
Inner layers (underlayers and filter lay-
r = Dolos waist ratio
ers) are also important for the protection of
'Pn=Z = packing density
the breakwater core made up of a quarry
run. The main purpose of inner layers is to
10.5.2.5 Structural lntegrity of reduce both the transmission through the
Armor Blocks breakwater and the construction cost by
After the failure of the Sines breakwater, using lower-cost materials for the break-
intensive research has been carried out by water core. These layers are designed to
various institutions and investigators to de- damp the wave penetration and to prevent
termine the load on various structural com- pull-out and wash-out of the fine material
ponents of armor blocks. This also included included in the breakwater core and from
direct prototype measurements (Magoon the seabed.
and Baird, 1977; Howell, 1985, 1988; Scott The design of the underlayers and filter
et al., 1986; Burger et al., 1990; Markl, layers is quite complicated and is usually
1990; Ligteringen et al., 1990; Burcharth, based on the limited information that is
1983, 1984; Burcharth and Liu, 1990, 1992). available · at present (Sowers and Sowers,
It is quite obvious that when blocks are 1970; Graauw et al., 1984; Bezuijen et al.,
broken by wave force, their interlocking ca- 1987). Figure 10-108 shows the idealized
pability is significantly reduced. This re- multilayer cross section of a rubble-mound
sults in subsequent block movements. These breakwater, where the weight of the under-
movements are accelerated because broken layer stone varies from 1/10 to 1/20 that of
blocks start hitting each other. Thus, when the upper layer. The recommended criterion
using slender blocks, especially heavy ones, for transition from one size of the stone to
the structural integrity of these blocks is another one is to ensure that the rock gra-
the factor that affects design of the armor dation will act ·as a filter as follows:
layer.
To avoid structural failure of armor D15(upper) < 5D85(under) (10-85)
blocks, more massive units should be fa-
vored where the sea condition is severe. where D 15 and D 85 are 15% and 85% ofthe
Slender blocks, however, can be used with value determined by the sieve curve, re-
some specific limitations. Burcharth and Liu spectively.
(1992) developed integrity and displace- The structure and stability of the toe
ment design diagrams (Fig. 10-107) for section is very important for stability of
Dolos blocks which indicates that these the breakwater's armor layer. Hence, the
blocks can be employed safely if the waist design of the toe section and especially
ratio is large enough. its protection from scouring by wave and
It should be noted that it is difficult to current action must be given proper atten-
reproduce block breakage in model experi- tion. Several formulas are proposed by
Breakwater Design 1035

Burcharth et al. (1995) for the evaluation of toe of the wave screen. The reason for these
toe stability. damages is easy to see by reviewing the
historical damages to high-mound compos-
10.5.3.2 wave screen ite breakwaters. There, impulsive
The wave screen or wave wall is some- breaking-wave forces that are acting on a
times installed on top _of the rubble-mound wave screen or wall are similar to that
breakwater to reduce wave overtopping; it acting on the vertical wall of composite
may be part of the concrete crown or access breakwater, thereby causing similar dam-
road built on top of the breakwater. When ages. To reduce these severe forces, the
the breakwater has a low crown height, the crown wall should be fully covered by con-
wave screen may suffer damage in the form crete blocks as discussed in Section 10.2.3.3.
of breakage, sliding, overturning, or erosion Otherwise, the wave screen or wall must
of mound rubble in front of or in back of · have the necessary structural strength and

L•gend:
------ Hydraulie stabllity limit (N.=1000, t~=0.7 ..... 6 =1 .29) carrespandlng ta nlative
number of dlsplaced unils O.
- - Tenslle strength llmlt correspondlng ta relative number of broken Oolosse B
H ~ Signiflcanf wove helghl In front of breakwater
Colos wolst ratia
S Concnte lensile strength
Relative number of broken Oolo.su
Relative number of disploced Dolosso

a) d)

c
100

80 ~·"7Tr~· .. l
100

80
.-cun

iii
a
60 60

ga 20
···;!//. .,.
~;~'~
~ ~wa
o =27.
;ga
40

20
S ::4Wpa
9= 2%
0=2X

5 10 5 10 15
H:.,.(m) H' (m)
b) •)

100 100

80 80

1 60
c
o
60

40
S=4t.tpa
9=5%
:ga 20
o
ac zo 0=5~

5 10 15
H!..,.(m)
c) 1)

100 100
r-0.~1 ..o.u ".o.:ns

}~:
80 80
?
~ 60

40

-!c 20 -! 20
"
5 10 15 5 10
H!..(m) H!...(m)

Figure 10-107. Design diagrama for structural integrity


and hydraulic stability of Dolos armor. [From Burcharth and
Liu (1992).]
1036 Breakwater Design

Seaward Leeward

Rock Size
Rock Size Loyer Grodotion ("'ol
W Prlmory Cover Loyer' 125to75 H•Wove Height
W/2ond W/15 Secondory Cover Loyer• 125to75 W=Weightof Individual Armor Unit
W/IOond W/300 FirstUnderloyer• 130to70 r = Averoge Loyer Thickness
W/200 Second Underloyer 150to50
W/4000-W/6000 Core ond Bedding Loyer 170 to30

For concrete ormor: 1 Sections m, 7, g, ( 1l, ( 2 l ond ( 6 l


2 Section ID,7; ,g ,(7)
•Sectlon ill,7,g,(8l

Figure 10-108. Idealized multilayer cross section of rubble-mound breakwater. [From U.S.
Army Corps of Engineers (1984).]

be heavy enough to provide sufficient resis- ALLSOP, N. W. H., 1990. "Reflection Performance
tance against aforementioned impulsive of Rock Armoured Slopes in Random Waves."
wave forces. Proc. of the 22nd ICCE Conf, Delft, The
In general, evaluation of the wave forces Netherlands, pp. 1460-1472.
on the wave screen is difficult, although the ALLSOP, N. W. H. and McBRIDE, N. W., 1994.
preliminary estimate of these forces can be "Reflections from Vertical Walls: The Poten-
obtained from the Goda pressure formula tial for Improvement in Vessel Safety and
for a vertical wall. Model tests have been Wave Disturbance." Proc. International
performed to determine these wave forces Workshop on Waue Barriers in Deepwaters,
Port and Harbour Research Institute,
(Tanimoto and Ojima, 1983; Bradbury et
Yokosuka,Japan,pp. 101-128.
al., 1988; Jensen, 1984; Hamilton and Hali,
1992). ALLSOP, N. w. H., HAWKES, P. 1., JACKSON, F. A
et al., 1985. "Wave Run-Up on Steep Slopes-
Model Tests Under Random Waves," Hy-
REFERENCES draulics Research, Wallingford Report SR2 .
AHRENS, J. P., 1981. "Irregular Wave Run-up on .AMINTI, P. and FRANCO, L., 1988. "Wave Overtop-
Smooth Slopes." CERC, U.S. Army Corps of ping on Rubble Mound Breakwaters." Proc. of
Engineers, Vicksburg, MS, CETA 81-17. the 21st ICCE, Malaga, Spain, pp. 770-781.
AHRENS, J. P., 1987. "Characteristics of Reef ARAMI, A and HATTORI, S., 1990. "A Model of
Breakwaters." CERC, U.S. Army Corps of En- Impulsive Pressure." Proceedings of Japanese
gineers, Vicksburg, Technical Report, 87-17. Conference on Coasta[ Engineering, Vol. 37,
ALLSOP, N. W. H., 1983. "Low-Crest Break- 629-633 (in Japanese).
waters, Studies in Random Waves." ASCE BAGNOLD, R. A, 1939. "Interim Report of Wave
Proceedings of Coasta[ Structures '83, pp. Pressure Research." Journal of the Institute
94-107. of Civil Engineers, Vol. 12, 202-226.
Breakwater Design 1037

BAIRD, W. F. and HALL, K. R., 1984. "The Design Armour." Proc. of the 20th ICCE, Taipei,
of Breakwaters Using Quarried Stones." Proc. Taiwan, pp. 1783-1796.
of the 19th ICCE, Houston, TX, pp. BURCHARTH, H. F. and FRIGAARD, P., 1987. "Sta-
2580-2591. bility of Roundheads and Trunk Erosion."
BARENDS, F. B. J., 1988. "Discussion on Paper 5." Proc. of ASCE Conf. on Berm Breakwaters,
Proceedings of the Conference on Break- Ottawa, Canada.
waters 1988, Eastbourne, UK.. BURCHARTH, H. F. and FRIGAARD, P., 1988. "On
BATIJES, J. A., 1974. "Surf Similarity," Proc. of 3-Dimensional Stability of Reshaping Break-
the 14th ICCE, Copenhagen, pp. 466-480. waters." Proc. of the 21st ICCE, Malaga,
BENEZIT, A., 1923. "Essai sur les digues mar- Spain, pp. 2284-2298.
itimes verticales." BURCHARTH, H. F. and Lru, Z., 1990. "A General
BEZUIJEN, A., KLEIN BRETILER, M., and BAKKEN, Discussion of Problems Related to the Deter-
K. I., 1987. "Design Criteria for Placed Block mination of Concrete Armour Unit Stress In-
Revetments and Granular Filters." Proc. of cluding Specific Results Related to Static and
the International Conf. on Coasta! and Port Dynamic Stresses in Dolosse." Proc. of the
Eng. in Developing Countries, Beijing. Seminaral Stresses in Concrete Armour Units,
ASCE, Vicksburg, MS.
BOTIIN, R., CHATMAN, C., and CARVER, R., 1976.
BURCHARTH, H. F. and Lru, Z., 1992. "Design of
"Waianae Small Boat Harbor, Oahu, Hawaii,
Dolos Armour Units." Proc. of the 23rd ICCE,
Design for Wave Protection," CERC, U.S.
Venice, pp. 1053-1066.
Army Corps of Engineer Waterways Experi-
ment Station, Vicksburg, MS, TR h-76-81. BURCHARTH, H. F., FRIGAARD, P., BERENGHER, J.
M., and MADIGAL, B. G., 1995. "Design of the
BRADBURY, A. R., ALLSOP, N. W. H., and STEVENS,
Ciervana Breakwaters Bilbao." Proc. of
R. V., 1988. "Hydraulic Performance of
Coasta[ Structures and Breakwaters '95. In-
Breakwater Crown Wall." Report No. SR 142, stitution of Civil Engineers, London.
HRS.
BURGER, W. W., 0UMERACI, H., and PARTENSKY,
BREBNER, A. and DONNELLY, D., 1962. "Labora- H. W., 1990. "Impact Strain Investigations on
tory Study ofRubble Foundations for Vertical Tetrapods: Results of Dry and Hydraulic
Breakwaters." Proc. 8th ICCE, New Mexico Tests." Proc. of the Seminar Stresses in Con-
City, pp. 406-429. crete Armour Units, ASCE, Vicksburg, MS.
BRUUN, P., 1985. Design and Construction of COOKER, M. J., and PEREGRINE, D. H., 1990. "A
Mound for Breakwater and Coastal Protec- Model for Breaking Wave Impact Pressures."
tion. Elsevier, Amsterdam. Proceeding of 22nd ICCE, Delft, The Nether-
BURCHARTH, H. F., 1983. "Material, Structural lands, pp. 1473-1489.
Design of Armour Units: Seminaral Rubble CASTRO, E., 1933. Diques de escollera. Revista de
Mound Breakwaters." Bull. TRITA-VBI-120, Obras Publicas.
Royal Institute of Technology, Sweden. CIRIA 1 CUR, 1991. Manual on the Use of
BURCHARTH, H. F., 1984. "Fatigue in Breakwater Rock in Coastal and Shoreline Engineering.
Armour Units." Proc. of the 19th ICCE, London.
Houston, TX, pp. 2592-2607. CHAN, E. S., 1994. "Mechanics of deep water
BURCHARTH, H. F., 1989. "Uncertainty Related to plunging-wave impacts on vertical struc-
Environmental Data and Estimated Extreme tures," J. Coastal Engineering, Voi. 22.
Event." Report of sub-group B, working group DANEL, P., 1953. "Tetrapods." ASCE, Proceed-
12, PTC 11. PIANC, Brussels. ings Fourth ICCE, Chicago, pp. 390-398.
BURCHARTH, H. F., 1993. "The Design of Break- ENDO, K. and TAKAHASI, S., 1994. "Numerically
waters." Coastal and Harbor Engineering Ref- Modeling Personal Danger due to Overtop-
erence Book, Abbott, M. B. and Price, W. A. ping Waves on a Promenade Breakwater."
(Eds.). E. & F. Spon. Proc. 24th ICCE, Kobe, pp. 1016-1029.
BURCHARTH, H. F. and BREJNEGAARD-NIELSEN, FOURNIER, C. P., SAYAO, O. J., and CALDAS, F.,
T., 1986. "The Infl.uence ofWaist Thickness of 1990. "Berm Breakwater Contamination
Dolosse on the Hydraulic Stability of Dolos Study, Sergipe Marine Terminal, Brazii."
1038 Breakwater Design

Proc. of the 22nd ICCE, Delft, The Nether- GonA, Y. and KAKIZAKI, S., 1966. "Study of Finite
lands, pp. 3194-3206. Amplitude Standing Waves and their Pres-
FUNAKOsm, H., 1994. "Survey of Long-term De- sures upon a Vertical Wall." Rept. Port and
formation of Composite Breakwaters along the Harbour Research Institute, Vol. 5, No. 10, 57
Japan Sea." Proceeding of the International p. (in Japanese); also Coastal Engineering in
Workshop on Wave Barriers in Deepwaters, Japan, Vol. 10, 1967, pp. 1-11.
PHRI, Yokosuka, Japan, pp. 239-266. GonA, Y. and YOSHIMURA, T., 1971. "Wave Force
GAILLARD, B., 1905. "Wave Action in Engineering Component for Structures of Large Diameter
Structures," Engineering News, 23 February. Isolated in Offshore." Report of Port and Har-
GIVLER, L. D. and SORENSEN, R. M., 1986. "An bour Research Institute, Vol. 10, No. 4, pp.
3-52 (in Japanese).
Investigation of the Stability of Submerged
Homogeneous Rubble Mound Structures un- GRAAUW, A F. F., VAN DER MEULEN, T., and DE
der Wave Attack." Report IHL 110-86. H.R. BYE, D., 1984. "Granular Filters: Design Cri-
IMBT Hydraulics, Lehigh Univ. Lehigh, PA teria." Journal of Waterways, Ports, Coastal
GODA, Y., 1967. Waue Forces on Structures. and Ocean Engineering, Voi. 110, No. 1.
JSCE Hydraulic Eng. Series 67-10, JSCE, GUNBAK, A R., 1979. "Rubble Mound Break-
(in Japanese). waters." Report 12/77. Division of Port and
GonA, Y., 1969. "Reanalysis of Laboratory Data Ocean Eng. Technical University of Norway,
on Wave Transmission over Breakwaters." Trondheim.
Rept. Port and Harbour Res. Inst., Vol. 8, No. liAMILTON, D. G. and HALL, K. R., 1992. "Pre-
3, pp. 3-18. liminary Analysis of the Stability of Rubble
GonA, Y., 1972. "Laboratory Investigation of Mound Breakwater Crown Wall." Proceeding
Wave Pressures Exerted Upon Vertical and of 23rd ICCE, pp. 1217-1230.
Composite Walls." Rept. Port and Harbour HAYASm, T. and IMAI, T., 1964. "Breaking Wave
Research Institute, Vol. 11, No. 2, pp. 3-45 (in Pressure and Sliding of Caisson." Proc. of
Japanese); "Experiments on the Transition Japanese Conference on Coastal Engineering,
from Nonbreaking to Post Breaking Wave Vol. 11, 184-192 (in Japanese).
Pressures." Coastal Engineering in Japan, HEDAR, P. A., 1965. "Rules for the Design of
Vol. 15, pp. 81-90. Rock-FiU Breakwaters and Revetments." XXI
GonA, Y., 1973a. "Motion of Composite Breakwa- International Navig. Congress, Section 11-1,
ter on Elastic Foundation under the Action of Stockholm, Sweden.
Impulsive Breaking Wave Pressure." Rept.
HIROI, 1., 1919. "On a Method of Estimating the
Port and Harbour Research Institute, Vol. 12,
Force of Waves." Journal of College of Engi-
No. 3, pp. 3-29 (in Japanese); or Goda, Y.,
neering, Univ. of Tokyo, Vol. 10, No. 11, pp.
1994. "Dynamic Response of Upright Break-
1-19.
water." Journal of Coastal Engineering, Vol.
22. HITACm, S., 1994. "Case Study of Breakwater
Damages-Mutsu-Ogawara Port." Proc. of
GODA, Y., 1973b. "A New Method of Wave Pres-
Intl. Workshop on Wave Barriers in Deep-
sure Calculation for the Design of Composite
waters. PHRI, Yokosuke, pp. 308-331.
Breakwater." Rept. Port and Harbour Re-
search Institute, Vol. 12, No. 3, pp. 31-70 (in HORIKAWA, K., 1978. Coastal Engineering, Uni-
Japanese); Proc. 14th ICCE, 1974, Copen- versity of Tokyo Press, Toyko.
hagen, Denmark, pp. 1702-1720. HoRIKAWA, K. and Noaucm, Y., 1970. "Relation
GonA, Y., 1985. Random Seas and Design of Between the Wave Force and Wave Profile of
Maritime Structures. University of Tokyo Breaking Wave on Vertical Wall." Proc. of
Press, Tokyo. Japanese Conference on Coastal Engineering,
GODA, Y. and JlARANAKA, S., 1967. "An Experi- Vol. 17, 177-184 (in Japanese).
ment of the Shock Pressure of Breaking HORIKAWA, K., OZAWA, Y., and TAKAHASm, K.,
Waves." Tech. Note of the Port and Harbour 1975. "Expected Sliding Distance of High
Res. Inst., No. 32, pp. 1-18 (in Japanese). Mound Composite Breakwater." Proc. of
Breakwater Design 1039

Japanese Conference on Coasta[ Engineering, International Workshop on Wave Barriers in


Vol. 19, 351-356 (in Japanese). Deepwaters, PHRI, Yokosuka, pp. 154-171.
HOWELL, G. L. (ed.), 1985. Proc. of Workshop on KAJIMA, R., 1994. "A New Method of Structurally
Measurement and Analysis of Structural Re- Resistive Design of Very Important Seawalls
sponse in Concrete Armor Units. Waterways Against Wave Action." Proc. of Intl. Workshop
Experiment Station, CERC, Vicksburg, MS. on Wave Barriers in Deepwaters, PHRI,
HOWELL, G. L., 1988. "Measurement ofForces on Yokosuka, pp. 459-518.
Dolos Armor Units at Prototype Scale." Proc. KAKuNo, S., 1994. "Status and Trends in the
of the 21st ICCE, Malaga, Spain, pp. Wave Pressure Calculations for Coasta}
2353-2369. Structures in Japan." Proc. of lntl. Workshop
HUDSON, R. Y., 1958. "Design of Quarry Stone on Wave Barriers in Deepwaters, PHRI,
Cover Layers for Rubble Mound Break- Yokosuka.
waters." Res. Rept. 2.2. WES, Vicksburg. KAMPHUIS, J. W. and MOHAMED, N., 1987. "Run-
HunsoN, R. Y., 1959. "Laboratory Investigation up of Irregular Waves on Plane Smooth
of Rubble Mound Breakwater." Journal of Slope." ASCE J. ofWaterways, Ports, Coastal
Waterways, and Harbor Div., Vol. 85, No. and Ocean Engineering, Vol. 104, No. WW2,
WW3, 93-121. pp. 135-146.
lNAGAKI, K., TSUNEI, K., HATI'ORI, M., and ENDO, KARMAN, VoN. 1929. "The Impact on Sea Plane
T., 1987. "Wave Force on a Breakwater with Floats During Landing." NACA, TN321, pp.
Concrete Block Mound of Energy Dissipating 1-8.
Type." Proceedings of Japanese Conference on
KATAOKA, S. and SAIDA, S., 1986. "Compilation of
Coasta} Engineering, Vol. 34 (in Japanese).
Breakwater Structures." Tech. Note of Port
IRIBARREN C. R., 1938. Una formula para el cal-
and Harbour Res. Inst. No. 556 (in Japanese).
cula de diques de escollera. Revista de Obras
Publicas. KlMuRA, K., TAKAHASHI, S., and TANIMOTO, K.,
1994. "Stability of Rubble Mound Founda-
IRIE, 1., KURIYAMA, Y., and ABAKAwA, H., 1986.
tions of Caisson Breakwater under Oblique
"Study on Scour in Front of Breakwaters by
Wave Attack." Proc. 24th ICCE, Kobe, Japan.
Standing Waves and Protection Method." Re-
port of Port and Harbour Research Institute, KrnKGOZ, S. M., 1991. "Impact Pressure of
Vol. 25, No. 11, 3-869 (in Japanese). Breaking Waves on Vertical and Sloping
ITO, Y., 1969. "A Treatise on Historical Develop- Walls." Ocean Engineering, Vol. 18, No. 1/2,
ment of Breakwater Design." Tech. Note of pp. 45-49.
Port and Harbour Res. Institute, No. 69 (in KOBAYASHI, M., TERASID, M., and TAKAHASHI, K.,
Japanese). 1987. "Bearing Capacity of a Rubble Mound
ITO, Y. and TANIMOTO, K., 1972 "Meandering Supporting a Gravity Structure." Rept. Port
Damages of Composite Type Breakwaters," and Harbour Research Institute, Vol. 26, No.
Tech. Note of Port and Harbour Res. Insti- 5, 215-252.
tute, No. 112 (in Japanese). KOBUNE, K and 0SATO, M., 1976. "A Study of
ITO, Y., FUJISHIMA, M., and KITATANI, T., 1966. Wave Height Distribution Along a Break-
"On the Stability of Breakwaters." Rept. Port water with a Corner." Rept. of Port and Har-
and Harbour Research Institute, Vol. 5, No. bour Research Institute, Vol. 15, No. 2 (in
14 (in Japanese); or Coastal Engineering in Japanese).
Japan, Vol. 14, 53-61 (1971). KONDO, H. and TAKEDA, H., 1983. Wave Dissipat-
JARLAN, G. E., 1961. "A Perforated Vertical ing Structures. Morikita Publishing Co. Ltd.
Breakwater." The Dock and Harbour Author- (in Japanese).
ity, Vol. 41, No. 488, 394-398. KOUGAMI, Y. and TOKIKAWA, K., 1970. "Experi-
JENSEN, O. J., 1984. "A Monograph on Rubble mental Study on the Reduction ofWave Force
Mound Breakwaters." Danish Hydraulic In- due · to Wave Dissipating ·Concrete Blocks
stitute, Horsholm, Denmark. During Construction." Proc. of Japanese Con-
JUHL, J., 1994. "Danish Experience and Recent ference on Coasta[ Engineering, Voi. 17,
Research on Vertical Breakwaters." Proc. of 205-210 (in Japanese).
1040 Breakwater Design

Kuo, C. T., 1994. "Recent Researches and MARKLE, D. G. and DAVIDSON, D. D., 1979.
Experiences on Composite Breakwaters in "Placed-Stone Stability Tests, Tillamook, Ore-
Taiwan." Proc. Intl. Workshop on Waue Barri- gon," U.S. Army Engineer Waterway Experi-
ers in Deepwaters. PHRI, Yokosuka, Japan, ment Station, TR HL-79-16, Vicksburg, Miss.
pp. 217-238. April.
LAMBERTI, A. and FRA:Nco, L., 1994. "Italian Ex- MINIKIN, R. R., 1950. Winds, Waues and Mar-
perience on Upright Breakwaters." Proc. Intl. itime Structures. Griffin, London.
Con{. on Waue Barriers in Deepwaters, PHRI,
MITSUYASU, H., 1961. "Experimental Study on
Yokosuka,Japan,pp.25-75.
Wave Force Against a Wall." Rept. Transp.
LARRAs, J., 1952. "L'equilibre sous-marin d'un Tech. Res. Inst., No. 47; or Coastal Engineer-
massif de materiaux soumis a la houle." Le ing in Japan, Vol. 5, 23-47.
Genie Civil.
LARRAs, J., 1961. Cours d'Hydrauique Maritime MITSUYASU, H., 1966. "Shock Pressure of Break-
et de Trauaux Maritimes. Dunod, Paris, pp. ing Wave." ASCE Proceedings of 10th Coastal
244-245. Engineering Conference, pp. 268-283.
LIGTERINGEN, H., 1994. "Other European Experi- MIZUNO, Y., SUGIMOTO, Y., KIMURA, K.,
ence on Deepwater Breakwaters." Proc. of YAMANAKA, K, and Knrucm, S., 1993. "Wave
Intl. Workshop on Waue Barriers in Deep- Forces on Breakwater with High Crown."
waters, PHRI, Yokosuka, Japan, pp. 199-216. Proc. of Japanese Conference on Coastal Engi-
LIGTERINGEN, H., ALTINK, H., and VAN 0RCHOT, J. neering, Vol. 40 (in Japanese).
H., 1990. "Strength of Concrete Armour Units, MONTGOMERY, R., HOFMEISTER, G. J., and
Joint lndustry Research." Proc. of a Seminar BAIRD, W. F., 1987. "lmplementation and
on Stresses in Concrete Armour Units, Performance of Berm Breakwater Design
Vicksburg, MS. at Rachine." Berm Breakwaters, ASCE,
LOSADA, M. A., 1990. "Recent Development in Washington, DC.
the Design of Mound Breakwaters." Hand- MoRIIDRA, H. and KUNITA, 0., 1979. "Model Ex-
book of Coastal and Ocean Engineering, periments on Hydraulic Characteristics of
Herbich, J. (ed.), Gulf Publishing Co., Sloped Wall Breakwater." Proc. Japanese
Houston, TX, pp. 939-1050. Conference on Coastal Engineering, Vol. 26,
LOSADA, M. A. and GIMENEZ-CVRTO, L. A., 1980. 295-298.
"Mound Breakwaters Under Wave Attack," NAGAI, S. and KURATA, K, 1974. "Investigations
Int. Seminar on Criteria for Design and Con- of Wave-Pressure Formulas due to Damages
struction of Breakwaters and Coastal Struc- of Breakwaters." Proc. 14th Intl. Conf. on
tures, Santander, Spain, Sect. II, pp. 127-238. Coastal Eng., Copenhagen, pp. 1721-1740.
LVNDGEN, H., 1969. "Wave Shock Forces," Proc.
NAKADA, K, TERAucm, K, NISHIDA, H., and
Res. on Wave Action, Delft, The Netherlands,
UMEKI, K, 1984. "Characteristic ofLow Crown
Vol. Il.
Composite Breakwaters." Proc. Japanese
MADSEN, O. S. and WmTE S., 1979. "Reflection Conference on Coastal Engineering, Vol. 31,
and Transmission Characteristics of Porous pp. 532-536 (in Japanese).
Rubble-Mound Breakwaters." MR 76-5,
Coastal Eng. Res. Center, U.S. Army Water- NVMATA, A., 1976. "Laboratory Formulation for
ways Experiment Station, Vicksburg, MS. Transmission and Reflection at Permeable
Breakwaters of Artificial Blocks," Coastal En-
MAGOON, 0. T. and BAIRD, W. F., 1977.
gineering in Japan, Vol. XIX, pp. 47-58.
"Breakage of Breakwater Armor Units." Proc.
of the Symposium on Design Rubble-Mound 0KADA, Y., WATANABE, T., SUGAWARA, T., and
Breakwaters, British Hovercraft Corporation, TANIMOTO, K, 1990. "Recent Developments of
Isle ofWight, U.K New Type Breakwaters in Japan." PIANC
MARKLE, R. G., 1990. "Crescent City Instrumen- 27th Inter. Congress, Osaka.
tal Model Dolos Study." Proc. of a Seminar 0UMERACI, H., 1994. "Scour in Front of Vertical
on Stresses in Concrete Armour Units, Breakwaters-Review of Problems." Proceed-
Vicksburg, MS. ings of International Workshop on Waue Bar-
Breakwater Design 1041

riers in Deepwaters, PHRI, Yokosuka, Japan, SATO, T., YAMAGATA, N., FURUKAWA, M.,
pp. 281-317. TAKAHASm, S., and HosoYAMADA, T., 1992.
0UMERACI, H. and KORTENHAUS, A., 1992. "Anal- "Hydraulic Characteristics of the Sloping Top
ysis of the Dynamic Response of Caisson Caisson Covered with Wave Dissipating
Breakwaters," Proc. 23rd ICCE, Venice. Blocks." Proc. Japanese Conference on Coastal
0UMERACI, H., PARTENSCKY, H. W., TAUTENHAIN, Engineering, Vol. 39, 556-560 (in Japanese).
E., and NICKELS, H., 1991. "Large-Scale Model SAVILLE, T., Jr., 1963. "Wave Run-up on Compos-
Investigations: A Contribution to the Revival ite Slopes." Proc. 6th ICCE, pp. 691-699.
of Vertical Breakwaters." Proc. Conf. on SELLING, W. N. and AHRENs, J., 1981. "Estima-
Coastal Structures and Breakwaters, I.C.E., tion of Wave Refl.ection and Energy Dissip·a-
pp. 167-180. tion Coefficients for Beaches, Revetments and
0VERSEAS COASTAL AREA DEVELOPMENT INSTI- Breakwaters." USA Army Corps of Engrs.,
TUTE OF JAPAN, 1991. Technical Standards for Coastal Eng. Res. Center, Fort Belvoir, VA.
Port and Harbour Facilities in Japan, Tokyo, SEELIG, W., 1980. "Two-Dimensional Tests of
Japan. Wave Transmission and Refl.ection Character-
OWEN, M. W., 1980. "Design of Seawalls Allow- istics of Laboratory Breakwater." Report TR
ing for Wave Overtopping." Rep. Ex 924, Hy- 80-1, CERC, WES (waterways experimental
draulics Res., Wallingford. station) U.S. Army.
OWEN, M. W., 1982. "The Hydraulic Design SELLING, W. N., 1983. "Wave Reflection from
of Seawall Profiles." Proceedings of the Coasta! Structures." Proc. of Conf. on Coasta!
Conference on Shoreline Protection, ICE, Structures '83, Arlington, VA.
Southampton, UK. SHIMOSAKO, K and TAKAHAsm, S., 1994. "De-
PARTENSCKY, H. W., 1988. "Dynamic Forces due termination of the Sliding Distance of Com-
to Waves Breaking at Vertical Coastal Struc- posite Breakwaters due to Wave Forces Inclu-
ture." Proc. of the 21st ICCE, pp. 2504-2518. sive Breaking Wave Forces." Proc. 24th ICCE,
PIERRE, R., 1953. "La jetee de Mustapha au Port Kobe,Japan,pp. 1580-1594.
d'Alger," Ann. Ponts et Chaussees, Avr.-Mai. SOWERS, G. B. and SOWERS, G. F., 1970. Intro-
PORT AND HARBOUR RESEARCH INSTITUTE, 1994. ductory Soil Mechanics and Foundations, 3rd
Proceedings of International Workshop on ed., MacMillan, New York.
Wave Barriers in Deepwaters, PHRI, Yoko- STEVENSON, T., 1886. The Design and Construc-
suka, Japan. tion of Harbours, 3rd ed. Adam & Charles
POWELL, K A. and ALLSOP, N. W. H., 1985. Black, London.
"Low-Crest Breakwaters, Hydraulic Perfor- TAKAHASHI, S. and HOSOYAMADA, S., 1994. "Hy-
mance and Stability." Wallingford, UK. Rep. drodynamic Characteristics of Sloping Top
SR57, Hydraulics Res. Caissons." Proc. of International Conference
PRICE, A. W., 1979. "Static Stability of Rubble on Hydro-technical Engineering for Port and
Mound Breakwaters." Rep. of Hydraulics Res. Harbor Construction, PHRI, Yokosaka, Japan.
Station, Wallingford, UK. TAKAHASm, S. and SIMOSAKO, K, 1990. "Reduc-
RICHEY, E. P. and SOLLITT, C. K, 1970. "Wave tion of Wave Force on a Long Caisson on
Attenuation by Porous Walled Breakwater." Vertical Breakwater and Its Stability." Tech-
Proc. ASCE, Vol. 96, No. WW3. nical Note of Port and Harbour Research In-
SAINFLOU, G., 1928. "Essai sur les digues mar- stitute, No. 685 (in Japanese).
itimes verticales," Ann. Ponts et Chaussees, TAKAHASm, S., TANIMOTO, K, and SuzUMURA, S.,
Vol. 98, No. 11, 5-48. 1983. "Generation Mechanism of Impulsive
SAKAMOTO, Y, MIYAJI, Y, UENISm, T., and TAKEDA, Pressure by Breaking Wave on a Vertical
H., 1984. "Experimental Study on Hydrody- Wall." Report of Port and Harbour Research
namic Function of Rubble Mound Break- Institute, Vol. 22, No. 4, 3-31 (in Japanese).
waters." Rept. Civil Eng. Research Inst. of TAKAHASm, S., TANIMOTO, K, and MIYANAGA, S.,
Hokkaido Development Bureau No. 82 (in 1985. "Uplift Force Due to Compression of
Japanese). Enclosed Air Layer and Their Similitude
1042 Breakwater Design

Law." Coastal Engineering in Japan, Vol. 28, TANAKA, N., 1976. "Effects of Submerged
191-206. Rubble-Mound Breakwater on Wave Attenua-
TAKAHASHI, S., KlMuRA., K., and TANIMOTO, K., tion and Shoreline Stabilization." Proc. 23rd
1990a. "Stability of Armor Units of Composite Japanese Conference on Coastal Eng. Conf.,
Breakwater Mound Against Oblique Waves." pp. 152-157 (in Japanese).
Report of Port and Harbour Research Insti- TANIMOTO, K., 1976. "Wave Forces on a
tute, Voi. 29, No. 2 (in Japanese). Composite-Type Breakwater." Proc. of the
TAKAHASHI, 8., TANIMOTO, K., and SHIMOSAKO, K., 1976 Annual Res. Presentation of Port and
1990b. "Wave and Block Forces on a Caisson Harbour Res. Inst., pp. 1-26 (in Japanese).
Covered with Wave Dissipating Blocks." Re- TANIMOTO, K., IIARANAKA, 8., and YAMAZAKI, K.,
port of Port and Harbour Research Institute, 1985. "Experimental Study ofWave Dissipat-
Vol. 29, No. 1, 54-75 (in Japanese). ing Concrete Blocks Against Irregular Waves."
TAKAHASHI, 8., SHIMOSAKO, K., and SASAKI, H., Rept. Port and Harbour Research Institute,
1991. "Experimental Study on Wave Forces Vol. 24, No. 2, 85-121 (in Japanese).
Acting Perforated Wall Caisson Breakwaters." TANIMOTO, K. and KlMuRA., K., 1985. "A Hy-
Re port of Port and Harbour Research Ins ti- draulic Experimental Study on Trapezoidal
tute, Vol. 30, No. 4, 3-34 (in Japanese); or Caisson Breakwaters." Technical Note of Port
Takahashi, S. and Shimosako, K., 1994. "Wave and Harbour Research Institute, No. 528 (in
Pressure on a Perforated Wali." Proc. of Inter- Japanese).
national Conference on Hydro-technical Engi-
TANIMOTO, K. and 0JIMA, R., 1983. "Experimen-
neering for Port and Harbour Construction,
tal Study on Wave Forces Acting on a Super-
PHRI, Yokosaka, Japan, pp. 747-764.
structure of Sloping Breakwater and on Block
TAKAHASHI, S., NAKADA, H., OHNEDA, H., and Type Composite Breakwaters." Tech. Note of
SHIKAMIRI, M., 1992a. "Wave Power Conver- Port and Harbour Research Institute, No. 450
sion by a Prototype Wave Power Extracting (in Japanese).
Caisson in Sakata Port." Proc. 23rd ICCE,
TANIMOTO, K. and YOSHIMOTO, Y., 1982. "Theo-
Venice, pp. 3440-3543.
retical and Experimental Study of Reflection
TAKAHASHI, 8., TANIMOTO, K., and SHIMOSAKO, K., Coefficient for Wave Dissipating Caisson with
1992b. "Experimental Study of lmpulsive a Permeable Front Wali." Report of Port and
Pressures on Composite Breakwaters." Re- Harbour Research Institute, Vol. 21, No. 3 (in
port of Port and Harbour Research Institute, Japanese); or Proc. of Coastal Structures and
Voi. 31, No. 5; Takahashi et al., 1994. "Dy- Breakwaters, 1995.
namic Response and Sliding of Breakwater
Caisson Against Impulsive Breaking Wave TANIMOTO, K. and TAKAHASHI, 8., 1994.
Forces." Proceeding of International Work- "Design and Construction of Caisson Break-
shop on Wave Barriers in Deepwaters, pp. waters-The Japanese Experience." Journal
362-401. o{Coastal Engineering, Vol. 22, 57-77.
TAKAHASHI, 8., TANIMOTO, K., and SHIMOSAKO, K., TANIMOTO, K., KOBUNE, K., and 0HSATA, M., 1975.
1994. "A Proposal of lmpulsive Pressure "Wave Forces on a Caisson and Stress Analy-
Coefficient for Design of Composite Break- sis of the Wali for Prototype Breakwaters."
waters." Proc. of International Conference on Tech. Note of the Port and Harbour Research
Hydro-technical Engineering for Port and Institute, No. 224 (in Japanese).
Harbor Construction, Port and Harbour Re- TANIMOTO, K., MOTO, K., ISHIZUKA, 8., and GODA,
search Institute, Yokosuka, Japan, pp. Y., 1976. "An Investigation on Design Wave
489-504. Force Formulae of Composite-Type Break-
TAKAYAMA, T. and FuJII, H., 1991. "Probabilistic waters." Proc. 23rd Japanese Conf. Coastal
Estimations of Stability of Slide for Caisson Eng., pp. 11-16 (in Japanese).
Type Breakwaters." Report of Port and Har- TANIMOTO, K., TAKAHASHI, S., and KITATANI, T.,
bour Research Institute, Vol. 30, No. 4, 35-64; 1981. "Experimental Study of Impact Break-
or Report of Port and Harbour Research Insti- ing Wave Forces on a Vertical-Wali Caisson of
tute, Vol. 31, No. 5. Composite Breakwater." Report of Port and
Breakwater Design 1043

Harbour Research Institute, Vol. 20, No. 2, VAN DER MEER, J. W., 1988b. "Stability of Cubes,
3-39 (in J apanese). Tetrapods and Accropode." Proc. of Break-
TANIMOTO, K., YAGYU, T., and GODA, Y., 1982. water '88. Eastbourne, UK.
"Irregular Wave Tests for Composite Break- VAN DER MEER, J. W., 1991. "Stability and
water Foundation." Proc. 18th ICCE, Transmission at Low-Crested Structures."
Capetown, South Africa, pp. 2144-2163. Delft Hydraulics Rep. H-45, Delft, The
TANIMOTO, K., TAKAHASm, 8., Kimura, K., 1987. Netherlands.
"Structures and Hydraulic Characteristics of VAN DER MEER, J. W. and ANGREMOND, K., 1991.
Breakwaters-The State of the Art of Break- "Wave Transmission at Low Crested Struc-
water Design in Japan." Report of Port and tures." Proc. of Conf. on Coastal Structures
Harbour Research Institute, Vol. 26, No. 5, and Breakwaters, ICE, London.
11-15. VAN DER MEER, J. W. and IIEYDRA, G., 1991.
TANIMOTO, K., ENDOH, H., and TAKAHASm, S., "Rocking Armour Units: Number, Location
1992. "Field Experiments on a Dual Cylindri- and Impact, Velocity." Coastal Engineering,
cal Caisson Breakwater." Proc. of 23rd ICCE, Voi. 15, 21-39.
Venice, pp. 1625-1638. WAGNER, H., 1932. "Uber stoss-und Gleit-
TAW, 1974. "Wave Run-up and Overtopping." vorgănge an der oberflăche von Flussig-
Technical Advisory Committee on Protection keiten," Zeishrift fur Angewandete
against Inundation. Mathematik und Mechanik, Voi. 12, No. 4,
TERRETT, F. L., OsoRIO, J. D. C., and LEAN, G. 192-215.
H., 1968. "Model Studies of a Perforated WILLIS, D: H., BAIRD, w. F., and MAGOON,
Breakwater." Proc. of llth ICCE, pp. O. T., 1987. "Berm Breakwaters." ASCE,
1104-1120. Washington, DC.
TIMCO, G. W. and MANSARD, E. P. D., 1983. "On XIE, S. L., 1981. "Scouring Patterns in Front of
the Interpretation of Rubble-Mound Break- Vertical Breakwaters and Their Influence on
water Tests." Proc. of Coastal Structures '83, the Stability of Foundation of Breakwaters."
Arlington, VA. Rept. of Department of Civil Engineering,
U.S. ARMY CORPS OF ENGINEERS, 1984. Shore Delft University of Technology, Delft, The
Protection Manual, U.S. Corps of Engineers, Netherlands.
Washington, DC. XIE, S. L., 1994. "Recent Research and Experi-
USHIJIMA, R., MIZUNO, R., and IMOTO, T., 1988. ence on Vertical and Composite Breakwaters
"Laboratory Stability Test of Foot-Protection in China." Proc. Intl. Conf. on Waue Barriers
Blocks for Upright Section of Composite in Deepwaters, PHRI, Yokosuka, Japan, pp.
Breakwaters." Rept. of Civil Engineering Re- 129-153.
search Institute, Hokkaido, Japan, Develop- ZWAMBORN, J. A. AND VAN NIEKERK, M., 1982.
ment Bureau, No. 424, pp. 1-4 (in Japanese). "Additiomil Model Tests. Dolos Packing Den-
VAN DER MEER, J. W., 1988a. "Rock Slopes and sity and Effect of Relative Block Density",
Gravei Beaches Under Wave Attack." Doc- National Research Institute for Oceanology,
toral Thesis, Delft University of Technology, Coastal Engineering and Hydraulics Division,
Delft, The N etherlands. Stellenbosch, South Africa, July.
Index

Access trestie (see marine structures) scattering of armor stones, 1001, 1002
Aids to navigation, 204, 205 stability, 1002, 1023, 1027, 1028
Anchor piles with enhanced pull-out capacity, 580-584 stability coefficient, 1003
Anchorage (see also sheet pile bulkhead; anchoring Armor stability coefficient, 1003
systems), 109-117 Atmospheric icing, 10
A-frame, 556,579,668,669
anchor (raked) pile, 557, 567, 568, 578, 579-585, Backfill material ( see backfill soil)
607,621,645-655,688 Backfill soil, 34 7
deadman (see plate) Bagnold-type impulsive wave pressure, 983-991
double tieback system, 580, 588, 606 Bedrock, (see rock)
failure, 688 Bent stiffness, 842-845
ground anchors, 557, 585-591, 675-679, 920, 921, Berm breakwater, 953, 967
926,927,929-931,933-938 Bitt, 205, 206
plate, 555, 578, 580, 618, 620, 672-674 Block masonry breakwater, 955, 959
rock anchors (see ground anchors) Block mound breakwater, 953
steel wale system, 576, 577, 674, 675 Blockwork wall
tie-rods, 554-556, 571, 572-578, 580 basic static principles, 482, 483
tiebacks (see tie-rods) block size, 476
vertical piles, 573, 671, 672 capping wall design, 484
wale (see steel wale system) concrete used for, 479
Arch-type gravity wall, 453, 454 contact pressures, 467-469, 483
Armor layer design criteria, 479, 480
armor blocks, 199, 202, 962-966, 1015, 1028-1030 design loads, 461-464
armor stones for rubble foundation, 1001-1004 design phases, 480-484
concrete blocks (see armor blocks) global stability, 476-478
damage level, 1032 loads against, 480-482
erosion area, 1032 mooring force distribution, 480
nominal diameter, 1028 rubble mattress for, 479
of the rubble foundation, 953, 954, 963-969, 1001, settlement, 473-476
1024, 1027 structural types, 399, 403-409
permeability factory, 1032 upper block design, 483

1045
1046 Index

Bollard, 205, 206 Concave layout of breakwater, 998


Breakwater at Cherbourg, 958 Concrete
Breakwater at Dover, 958 aggregates, 159
Breakwater concrete capping crown, 977 C 3 A content, 157, 158, 424
Breakwater foot protection, 977, 1004 calcium nitrate corrosion inhibitor, 158, 1424
Breakwater head, 1001, 1002 consolidation, 161, 162
Breakwater toe, 1005, 1034 corrosion, 156, 157
Breakwater wave crown, 965 cover over reinforcement, 168, 169
Breakwaters cracks, 156, 425
assembled from prefabricated components, 460, 461 curing, 162
berm breakwater, 953, 967 deterioration, 156
block masonry type, 955, 959 durability, 157
block mound type, 953 effects of microsilica fume, xxxiii, 158, 424
composite type, 199, 953-955 exothermic reaction, 160, 428
curtain wall type, 955, 956 freezejthaw effects, 159
Cyclopian blocks type, 963, 964, 978 joints, 179, 180
definition, 72, 198 light weight, 425
dual-cylinder caisson type, 970, 973 mix, 160
dynamically stable, 953
permeability, 158
failure, 971 ·
placement, 161
fl.oated-in caisson type, 199, 201, 955, 975, 977
pozzolanic additives, 157, 424
fl.oating,200,201,956
precast, 178-180, 701
history of, 952-976
prefabricated (see precast)
horizontal plate, 956
preparation, 160
horizontally composite, 954, 969, 991, 1020,
1021-1024 protection from corrosion, 157, 424
hydraulic, 203 protective membrane, 163, 164
objectives, 198, 951 quality control, 165
perforated wall, 955, 1007 required properties, 163, 164, 424
performation block masonry, 1007, 1013 slag-Portland cement requirements, 157, 424
piled, 199,200,201,956 surface finishing, 162
pneumatic, 201, 203, 956 transportation, 160
reshaping, 953 type, 155, 156
rubble-mound, 2, 199, 201,951-953, 1024-1036 unit weight, 156
sheet-pile, 200, 201, 956 waterjcement ratio, 157-158, 165,424
typicallayout, 199, 200 water used in mix, 159, 162, 163
Buoyancy and buoyant stability, 485-490 Concrete armor block
Accropode, 1033
Caisson (see concrete caisson) cubes, 1033
Calcareous sands, 804-806 dolos, 1029
Cast-iron sheet piles, 552, 553 interlocking ability, 1029
Cathodic system design, 193-195 shape-designed, 962, 966
Channel structural integrity, 1034
access, 78 Tetrapid, 1033
alignment (layout), 82-84, 96 wave energy dissipating concrete blocks, 199, 202,
cross-section, 84, 85 962,964,966
design parameters, 81, 82 Concrete caisson (see also fl.oated-in concrete caisson
depth, 85-90 wall)
design factors, 79 base grouting, 428
economy of construction, 96-98 buoyancy and buoyant stability, 485-490
layout (alignment), 82-84, 96 concrete for construction of, 424-426
operation, 79-81 constructed from prefabricated components, 411-416
protection from sedimentation, 96, 97 design loads, 461-464
two-way, 93-95 drainage system, 418
width, 90-96 field monitoring, 499
Coefficient of friction, 467, 482, 1000 global stability, 476-478
Composite type breakwaters, 953-955 installation, 426-428, 497
Index 1047

launch,401, 410,413,426,427,490-495 general trends in modern design practice, 267-269,


light weight concrete, 425 935
meandering sliding, 973, 999 ground motion-induced, 318, 319
of cylindrical configuration, 416, 419 pelletized, 271
ofunusual configurations, 414-416, 419-422 permanent, 244
settlement, 473-476 special, 244
space between adjacent (joints), 428, 429 standard highway load, 275, 276
structural arrangements, 410-421 temporary, 244
structural design, 497-499 uniformly distributed, 269-272, 720, 918
tilt ( see settlement) Design Wave, 33-37, 995
tow, 426, 427, 496, 497 Dolphin (see marine structures)
use for quay wall, 401, 409-422 Dowel effect, 683, 684, 696, 783, 862
wall thickness requirements, 425, 426 Dual-cylinder caisson type breakwater, 970, 973
with external stiffening fins, 414, 415 Dynamic response of vertical wall, 989
Concrete superstructure (see concrete deck) Dynamically stable breakwater, 953
Constructibility, 154, 154
Containers, xxvii, 271
Effect of long caisson on wave distribution, 999
Corrosion of metals, 156, 157, 187
Extended Goda formula, 993-995, 1010
Creep soils (see bulkheads on creep soils)
Crest height of upright of breakwater section, 980
Cribwork wall Facia fender panel (see fender panel)
basic static principles 464-467 Failure of composite breakwater, 972
bearing stresses, 524 Fender panel (suspended), 703, 712-722, 724
concrete, 435, 437-439, 440 Fenders
design, 522-524 buckling type, 215-222
design loads, 461-464 cell-type, 216, 222
global stability, 476-478 cold temperature effects on, 56-58, 211-213
settlement, 476-476 design principles, 226, 228
timber, 435, 438 failure, 226-228
Current-induced scour, 427, 793, 794 foam-filled, 222, 225
Currents gravity-type, 225
effect on deck orientation, 249 hollow cylindrical, 213-215
effects on navigation and port operation, 2, 16, 18 hollow rectangular, 213-215
general, 16-19, 248 kappa-type, 216, 223
littoral, 18 mechanical, 222
load, 249-251 OXYgen andjor heat effects on, 211
propeller-induced, 791, 794 pi-type, 216, 211
tidal, 17 pneurnatic,216,224
wind-stress, 17 prestressed concrete piles, 723, 724
Curtain-wall breakwater, 955, 956 purpose, 206
Cyclopean blocks type breakwater, 963, 964, 978 Raykin (sandwich-type), 225, 226
Cylindrical caisson, 416, 419 selection, 226-230
shear type, 215
Design loads solid rubber, 210-222
assumptions and considerations, 267-271 systems, 207-208
combinations, 319, 320, 244 timber, 208, 210
environmental, 244-253 used tires, 291
from fi:xed-base equipment, 282, 283 v-type, 216, 220
from forklift truck, xxvii, 272-274, 779 wooden (see timber)
from loading ramp, 268 Ferrocement (see concrete)
from mobile (truck mounted) equipment, 277, 708, Fiber reinforcement, 170-172, 936
718, 720, 779 Floated-in concrete caisson type breakwater, 955, 975,
from portal crane, 269, 279-282 977
from rail-mounted equipment, 279-282 Floated-in concrete caisson wall (see concrete caisson)
from stacking straddle carrier, 274 Floating breakwater, 200, 201, 956
from vehicular equipment, 272-279 Fog, 10, 11
general, 243 Fouling,4,7,8
1048 Index

Freshwater Oow salinity) ice Ice-caused abrasion, 53


adhesion, 4 7 Icing, 53, 54
compressive strength, 45, 46 Impulsive pressure coefficient, 989, 994
ftexural strength, 46 Impulsive wave pressure, 983-991
friction, 4 7 Bagnold-type, 985
Poisson's ratio, 47 coefficient, 989, 994
shear strength, 46, 47 due to high andjor long mound, 989
tensile strength, 46 due to steep sea bottom slope, 987
Young's modulus, 4 7 ftip through, 986
single peak pressure, 986
Goda formula, 992 transition-type pressure, 987
Goda extended formula, 993-995, 1010 ventilated shock, 986
Gravei drain protection technique against soilliquefac- Wagner-type, 985
tion,337, 338 Incident wave angle, 990
Gravity-type wall (see quay wall) Iribarian formula, 789
Ground anchors Isbash formula, 1002
definition, 585
design, 675-679 Jet grouting, 921, 922
installation, 587, 588, 920
life span, 585, 586
protection from corrosion, 588-590, 936 L-shaped wall
stressing, 590 anchor design, 510-511
types, 586, 587 basic static principles, 464-467
Grout, 922 bearing pressure distribution, 508-510
Grout injection, 178, 930-934, 936-940 cantilever,446, 505-507
construction phases, 449
contact pressures, 467-469
Harbor basin ( see port basin)
counterfort, 446, 447, 449, 450, 505, 506, 507, 508
Harbor entrance (see port entrance)
design, 506-511
High vertical wall breakwater, 1000, 1001
design loads, 461-464
High-mound breakwater, 953, 959
effective bearing area, 471
Hiroi formula, 991
ftoated-in, 450-452, 460-462
Horizontal plate breakwater, 956
Horizontally composite breakwater, 954, 969, 991, global stability, 476-478
1020, 1021-1024 prefabricated units for construction, 446-452
Hudson formula, 788, 1023, 1030 settlement, 4 73-4 76
Hydrostatic pressure, 481 with externa! anchorage, 446, 448, 506, 510, 511
with interna! anchorage, 464, 448, 449, 451, 452,
506,508-510
Ice Large-diameter cylinder wall
control, 38, 58-61
basic structural types, 430, 431
effects on marine structures, 4 7, 48, 54-56
bearing pressure, 502
effects on navigation, 38, 50-52
design loads, 461-464, 501, 502
effects on port layout design, 49, 50
double-cylinder wall, 433, 434
effects on port operation, 47-49, 52
global stability, 476-478
environmental driving forces, 294
installation, 428, 432, 436-438
formation, 39-41
joints, 428, 430
management, 38
penetration into foundation soil, 431, 432, 500
microstructure and morphology, 39
tilt,502-504
Ice build-up due to tide action, 52
Loading platform, 893, 899
Ice growth in ship track, 50-52
Low-crest vertical wall breakwater, 1000, 1001
Ice load
Low-mound breakwater, 953, 960
due to adfreeze to the structure, 299, 300
Low salinity ice (see freshwater ice)
due to ice sheet bending, 297-299
general information, 293
ice-crushing load, 295-297 Marginal piled wharf, 711, 714-722, 725-735
of thermal origin, 301, 302 Marine facility, 115
vertical on piers, 300, 301 Marine foundation, 149, 151, 152, 343-345, 890,
vertical on piles, 300, 301 901-903
Index 1049

Marine structures Monolith concrete block, 955


access trestie, 122, 123, 706, 707, 881, 899, 900 Mooring accessories, 205, 206, 899, 907, 908
bottom-fixed, 343-345 Mooring buoy system, 112
catwalk, 899-901 Mooring force, 260, 262-267, 903-906
configuration, 138, 154 Mooring lines
constructability, 153, 154 arrangement, 138, 261, 262-265
cost, 151-153 materiala, 262, 906, 907
cross-section, 138 safety factory, 265
design principles, 148-151 Mooring load (see mooring force)
design stages, 148 Mooring system, 888, 889
dolphin Morison equation, 254
breasting, 122, 124, 136, 443, 705, 706, 882-888,
893-897, 936-938 Nagai formula, 992
design, 266, 267, 903-914 Navigation-induced forces, 791, 792
flexible, 290, 705, 706, 896, 909 Negative wave pressure on perforated wall caisson,
layout, 882-888 1011
mooring, 117, 119, 122, 136, 138, 443, 882-888, Nonlinear wave pressures, 982
898, 899
warping, 116
ductile, 714, 716, 717 Observational method, 343
effects of environmental conditiona, 890 Offshore terminal, 121, 123, 414, 415, 419-421,
effects of foundation on design, 149, 151, 152, 702-710,880,881-888,890-893
343-345,890,901-903
effects of geological conditiona on design, 151, 152 Perforated wall breakwater, 955, 970, 991, 1007
elevation, 147, 148 Perforated wall caisson modification factor, 1010
environmental constraints, 153 Perforation block masonry wall breakwater, 1007, 1013
in-seismic zone, 714, 716, 717 Perforation blocks, 1013
layout, 131-138,140,881-889 Pile
loading platform, 893, 899 arrangement, 697-701, 706, 735
modernization,917-949 bearing point, 737-739
on special foundations, 147 belled,752, 753
pier, 136,137,138-141,457,459,696,702-711,781, bored, 714, 733, 744, 755
882 boxtype,732, 733,743,744
piled, 837, 890, 891 (see also piled marine struc- capping beam, 702, 703, 714, 718-721, 777-780
tures) cast-in placed (see bored)
prefabrication, 154, 701-704, 714 choice (selection), 710, 757, 758
protection from ship impact (see fenders) also 893, concrete-filled, 742, 848-853
898,896 cracking, 702, 740, 744, 749
purpose, 130, 131 cylindrical concrete, 702, 710, 714, 718-720, 746-752
redundancy, 150 damageto, 737-739
safety considerations, 345-347 deterioration, 735
specification, 154 doweleffect,683,684,696, 783,862
steel jacket type, 892-894, 898 drilled, ( see bored)
type, 138-147,890,892,893-895 H-pile, 731, 736-738
unloading platform, 893, 899 handling, 741, 744, 745
Marine terminal integration with ·concrete superstructure, 704, 705,
berth requirements, 128-130 714, 726, 777, 778
environmental considerations, 127, 128 lateral stability, 702
land requirements, 124-126, 136 load test, 768, 776
modernization, 917-949 overstressing, 607, 714, 777, 778
mooring system, 888, 889 plastic, 755, 756
offshore, 121,123,414,415,419-421,702-710,880 prestressed concrete, 706, 708, 710, 729, 730, 745,
Marseille-type breakwater, 962, 963 746, 847
Meandering sliding of caisson, 973, 999 screw, 713, 753-755
Minikin formula, 992 splice, 737, 738, 740, 741, 745, 747
Modernization of marine structures, 917-949 steel grouted (M. V.), 730
Monitoring decision method (see observational method) steel pile, 704, 710, 736-742, 848, 891
1050 Index

Pile (continued) ductile, 714


steel pipe ( see steel pile) foundation analysis, 841, 842, 853-862
structural design, 846-853 foundation stiffness, 842
tapered, 756, 757 in seismic zone, 714
timber, 712, 714, 736 modernization, 824, 825, 940-949
with enlarged end, 706, 708, 746 protection from ship impact, 722-724
wooden (see timber) superstructure, 696
Pile driving (see pile installation) Piled pier design, 838-853, 940-942
Pile driving analyzer, 812 Piled relieving platform
Pile driving hammers, 594, 595, 758, 759 area of implementation, 700, 725
Pile driving-induced instability, 785, 786 basic structural types, 698, 726-734
Pile driving rig, 594, 759-763 cast-in-situ, 777
Pile foundation analysis, 841, 842, 860-862 composite, 778-780
Pile installation construction aspects, 862-864
by.waterjet,598-605, 717,748-750,765,766 definition, 700
driving accessories, 758, 762-763 design, 840, 841, 842-845, 853-864
guiding template, 751, 761, 762 global stability, 862
installation specification, 767 joints, 840, 846
via mandrel, 760, 761, 764 length between expansion joints, 845, 846
with hammer, 758, 759 modernization,920, 924,928-930,942-949
with vibrator, 764, 765 silo effect, 700
Pile integration with concrete superstructure, 704, 705, stiffness, 839
714,726,777,778 superstructure, 725-734, 781, 782
Pile integrity test, 776 Plymouth type breakwater, 958
Pilejetting,598-605, 717,748-750,765,766 Port
Pile laterally loaded classification, 70, 71
basic theories, 820-825 definition, 69, 70
Broms' design method, 826-829 development, xxvi, xxvii
Matlock and Reese design method, 830-835 elements, 71, 73
socketed into rock, 836, 837 environmental conditions, 890
subgrade reaction design approach (see Matlock and history, xxv
Reese method) layout, 115-119
Pile load test, 768-776 location, 112, 113, 114, 880
Pile settlement modernization, 917-949
in cohesionless soil, 818, 819 modern trends, xxvii, xxix
in cohesive soil, 819, 820 protection criteria, 251, 252
Pile socketing, 724-726, 742 Port basin
Pile under static axialload anchorage area, 109-117
compressive, 806-810, 812, 815 area, 115
pull-out (see tensile) layout, 119-121
tensile, 815-817 maneuvering area, 108, 111
Pile wave-equation analysis, 811 ship stopping length, 107, 108
Pile-driving formulas, 811 size, 107
Pile-enhanced slope stability (see dowel effect) turning basin, 109, 111
Pile-soil interaction wave climate, 106, 107
basic principles, 795-802 Port entrance
effecting parameters, 795-803 cross-section, 105
laterally loaded piles, 820-837 location, 98, 103
pile settlement, 817-820 orientation, 104, 116
piles under axial static loads, 803-817 tidal currents effect, 99-103
Piled marine structure width,98, 99
area of construction, 695-697 Post-tensioned steel (see reinforcing steel)
basic structural schemes, 697-701 Prestressed concrete (see concrete)
cost, 696 Prestressed steel (see reinforcing steel)
damage, 696, 697 Probabilistic design of breakwater, 972, 973
design criteria, 837, 838 Promenade breakwater, 971, 974
design methods, 838-864 Propeller-induced current, 791, 794
Index 1051

Protection against corrosion Risk analysis, 342


cathodic protection, 189-195 Rock, 350-352
galvanic anode cathodic system, 189-191 Rock anchor (see ground anchor)
impressed-current cathodic system, 191-195 Rubble foundation bearing capacity, 997, 998
protective coatings, 188-189 Rubble foundation design, 1001, 1002
Pumped concrete Rubble mattress, 400, 416, 422, 423
aggregates, 177 Rubble-mound breakwater
equipment, 177 armor layer design, 1027-1034
placement, 177 basic types, 951-953
core, 1034
design, 1024-1036
Quaywall filter, 1034
backfill,402,408,445 inner layer, 1034
basic static principles, 464-467 underlayer, 1034
blockwork-type, 399, 403-409 wave crown, 965
caisson-type, 401, 409-422 wave screen, 965, 1035
contact pressures, 467-469, 498 wave wall, 1035
counterfort, 397, 399
design loads, 461-464 Safety considerations, 345-347
effective bearing area, 471, 910, 911 Safety factor, 150, 151, 345, 473, 783, 997
global stability, 476-478 Sainflou formula, 991
L-shaped (see L-shaped wall) Scattering of wave dissipating blocks, 973
mass concrete-type, 397, 398 Scouring at breakwater toe, 1005
merits and drawback, 892, 893 Scouring in front ofbreakwater, 973
modernization,920-924, 930-940 Scouring underneath the caisson, 973
Ravier system, 407, 408 Sea ice
adhesion, 45
rubble mattress, 400, 416, 422, 423
compressive strength, 43, 44
settlement, 473-476, 913, 914
flexural strength, 43
space between expansion joint, 402
friction coefficient, 45
stability ( see basic static principles) parameters, 41-43
superstructure, 405 Poisson's ratia, 44, 45
tilt (see settlement) shear strength, 45
Quick-release hook, 205, 206 tensile strength, 44
thermal expansion, 45
Rain, 10 Young's modulus, 44
Reinforced concrete (see concrete) Seawater, 4-6
Reinforced concrete cylinder piles, 571, 573 Seismic loads, 302-318
Reinforced concrete sheet piles, 553, 563-572, 556 Semicircular perforated caissons, 970, 971
Reinforced earth wall Sheet pile anchor wall, 575-577, 669, 671
application in marine engineering, 524-531 Sheet pile bulkhead
construction sequence (see construction underwater) anchoring systems (see also anchorage), 554-557,
construction underwater, 536-546 573-591
design principles, 532-536 anchored by raked piles, 621, 623, 645-655
longevity in marine environment, 530-532 arching behind, 346
backfill, 605-607
Reinforcing steel
built on creep soils, 655-688
anchor for prestressed steel, 166
capping, 591-593
ducts (sheaths) for prestressed steel, 165, 166
construction sequences, 347, 557
epoxy coated, 169, 170 conventional design, 628-630, 633-636
galvanized, 170 design by Danish Rules, 609, 610, 638-640
grouting, 167, 168 design by Lasebnik's method, 615, 616, 643, 644
prefabrication, 168 design by finite element, 644, 645
space between, 165 design by Russian method, 641-643
splices, 165 design by Tschebotarioffs method, 640-641
stressing, 166, 167 design criteria, 625-627
type, 165-167,426 design of anchored, 632-644
Reshaping breakwaters, 953 (see also dynamically sta- design of cantilever, 627-632
bie breakwater) effects of construction procedures, 347, 557
Riprap, 782, 788, 793 failure, 688-690
1052 Index

Sheet pile bulkhead (continued) Soil characteristics ( see soil properties)


fixed earth support method, 637, 638 Soil classifications, 349-352, 655
flexibility criteria, 612, 615-617 Soil compressibility, 256, 357
free earth support method, 633-637 Soil consolidation (see soil compressibility)
graphical design method, 630-632, 636-638 Soil densification, 348
modeling, 346 Soil dynamic pressure
modernization, 925-928 calculation method, 382-388
of composite construction, 731-733 effect of saturated backfill, 385
on creep soils, 655-688 effect of wall inertia, 386, 387
overall stability, 679-684 effect of wall movement, 387, 388
seismic design, 684-688 hydrodynamic pressures, 385, 386
Sheet pile bulkhead modernization, 925-928 Soil grouting, 921, 922, 941
Sheet pile bulkheads on creep soils, 655-688 Soillateral pressures
Sheet pile driving compaction-induced, 381
equipment, 594-596 distribution, 331
methods, 596-598 effects of ambient temperature on, 376
Sheet pile selection, 572, 573 effects of backfill freezing, 376
Sheet pile wall anchored by raked piles, 621, 623, effects of relieving platform, 857-860, 921, 923, 924
645-655 effects ofresidual built-in stresses, 367, 375
Sheet piling of composite construction, 731-733 effects of time-dependent changes in soil, 331, 332,
Sheet-pile wall breakwater, 201, 956 374
Ship effects of wall movements, 331, 342, 358, 359, 387,
allowable movements at the dock, 265, 266 388,609-625,611,612,617-622,644
classification, 73 effects of wall friction, 381, 382, 609-625, 628, 644
draft, 77-77, 86 effects ofwall flexibUity, 609-625, 644
effects on port design, 74-78, 879 reduction methods, 364-366, 368, 456-458, 466
parameters, 73, 74 Soilliquefaction, 334-342, 347
size, 75-77 Soil passive pressure
Ship impact basic theories, 376-380
calculation methods, 283-291, 903, 904 effects of soil profile on, 378
mathematical modeling of, 292, 293 effects ofuniformly distributed load on, 378
Physical scale modeling of, 292 effects of wall movements on, 611, 612
protection from (see fenders) Soil permeability, 357, 358
velocity correction factor, 291, 292 Soil plug in pile, 704, 738, 740, 742, 743
Ship-induced waves, 24-26, 790, 791 Soil pressure at rest, 380, 381
Sines breakwater, 966, 968 Soil pressure relieving platform, 364-366, 368,
Skirt,423,424,465,499, 500 456-458
Slope instability, 782-786 Soil properties, 349-358, 803, 804
Slope protection, 203, 204, 782-795 Soil shear strength, 354-356, 803
Sloping breakwater, 952 Spray, 10
Sloping-top caisson, 955, 970, 1016-1019 Squat, 86-88
Sloping-top breakwater, 1015-1019 Stability of wave-dissipating blocks, 1023
Slurry wall, 734 Standing-wave pressure, 981
Small amplitude wave theory, 981 Standoffforce, 250, 251, 253
Snow, 10, 11 Steel cylinder sheet piles, 562
Soil active pressure Steel jacket, 892-894, 898
basic theories, 359 Steel sheet piles, xxxii, 181, 182, 183, 184, 553
coefficient of, 359-362 Steel sheet-pile cell wall
effects of backfill profile, 363 basic parameters, 439, 441, 442
effects of grade load, 364-366 construction sequences, 442, 444, 445
effects of piles on, 855-857 deflection, 520, 521
effects ofresidual built-in stresses, 367, 375 design loads, 461-464, 522
effects of wall movements, 331, 342, 358, 359, design methods, 511-519
368-374 effects of concentrated load on, 522
Soil arching, 366, 367 fill, 444, 445
Soil bearing capacity, 469-473 global stability, 476-478
Index 1053

seismic design, 337, 338, 519, 520 Vessel (see ship)


superstructure design, 522 Vessel-induced waves (see ship-induced waves)
tilt ( see deflection)
typical configurations, 439, 441 Wagner-type impulsive wave pressure, 985
Steel wale system, 576, 577, 674, 675 Wall with fins, 452-456
Step-type rubble-mound breakwater, 963 Water-level variations, 8-10, 152
Structural steel Wave energy dissipating chamber, 1007
area of application in marine environment, 180 Wave energy dissipating concrete blocks, (see armor
corrosion control, 187, 188 (see also protection blocks)
against corrosion) Wave energy dissipating concrete caisson, 970, 971
corrosion in marine environment, 156, 157, 187 Wave force
fabrication, 186 impulsive pressure, 983-991
for marine construction, 186 nonlinear, 982
grade, 182-186 on a vertical wall covered with concrete blocks, 1022
materiala, 181 onan armor stone, 1028
shapes, 181 on circular pile(s), 253-259
weldability, 186 on continuous vertical wall, 259
welding procedure, 186 on large-diameter cylinders, 258, 259
Submerged breakwaters, 953 on perforated breakwater, 1009
Subsurface investigation, 333, 334, 340-342, 902, 903 on performation block masonry wall, 1015
Surf-similarity parameter, 1032 on sloping-top breakwater, 1018
Suspended fender panel, 703, 712-722, 724 standing wave pressure, 981
W ave reflection, 978
Timber W ave reflection coefficient
area of application in marine environment, 195, 196 of horizontally composite breakwater, 1021
basic properties, 196, 197 of perforation block masonry wall, 1012
preservation, 196-198, 210, 726-728 of rubble mound breakwater, 1027
Timber preservation (see timber) of sloping-top breakwater, 1018
Timber sheet piles of the perforated wall caisson, 1008
history, xxx, xxxi of vertical wall, 979
preservation, 196-198, 210, 726-728 Wave screen, 965, 1035
types, 511, 512, 558-560 W ave transmission, 978
Trapezoidal type concrete caisson, 970 W ave transmission coefficient
Trapezoidal-type breakwater, 1019 of horizontally composite breakwater, 1021
Tremie concrete of perforation block masonry wall, 1014
aggregates, 176 ofrubble mound breakwater, 1025
cement, 176 of sloping-top breakwater, 1016
formwork, 178 of vertical wall, 978
methods of placement, 172-176, 752 Wave uplift pressure on ceiling slab, 1011
plasticizers (see water reducing admixtures) Wave wall, 1035
thermal cracking, 176, 753 Wave-induced scour, 22-24
typical mix, 176 Wave-power-extracting caisson, 971
water j cement ratio, 176 Waves
waterjreducing admixtures, 176 action, 2
Tsunamis, 36 parameters, 26-30
Turning basin, 109-111 direction, 996
design, 33-37, 995
Underdeck slope, 717, 718, 719, 782-788 long-period, 21
Underwater concreting (see tremie concrete and effects on port
pumped concrete) operation, 2
Unloading platform, 893, 899 shoaling,2, 29
type, 19-22
van der Meer formula, 1030 short-period, 21
Variation in water level, 8-10, 152 breaking, 21,22,259,260
Ventilated shock due to impulsive wave pressure, 986 ship-induced, 24-26, 790, 791
Vertical type breakwater, 953, 977 theories, 31-33
Vertical wall breakwater stability, 997 vessel-induced (see ship-induced)
1054 Index

Weight of armor, 1002-1004 load,· 245-248


Wind prevailing, 12
drag, 16 rose, 12, 13
dynamic effects, 16 velocity (speed), 14, 15
gust, 14, 245 Wooden sheet piles (see timber sheet piles)

You might also like