HBRC Journal (2015) 11, 224230

Housing and Building National Research Center

HBRC Journal

http://ees.elsevier.com/hbrcj

Seismic displacement of gravity retaining walls

Kamal Mohamed Hafez Ismail Ibrahim *

Civil Engineering Dep., Suez Canal University, Egypt

Received 29 December 2013; revised 12 February 2014; accepted 11 March 2014

KEYWORDS Abstract Seismic displacement of gravity walls had been studied using conventional static methods
Gravity walls; for controlled displacement design. In this study plain strain numerical analysis is performed using
Backll; Plaxis dynamic program where prescribed displacement is applied at the bottom boundary of the
Earthquakes; soil to simulate the applied seismic load. Constrained absorbent side boundaries are introduced
Numerical analysis; to prevent any wave reection. The studied soil is chosen dense granular sand and modeled as
Displacement design; elasto-plastic material according to MohrColumn criteria while the gravity wall is assumed elastic.
Limit design By comparing the resulted seismic wall displacements calculated by numerical analysis for six
historical ground motions with that calculated by the pseudo-static method, it is found that numer-
ical seismic displacements are either equal to or greater than corresponding pseudo-static values.
Permissible seismic wall displacement calculated by AASHTO can be used for empirical estimation
of seismic displacement. It is also found that seismic wall displacement is directly proportional with
the positive angle of inclination of the back surface of the wall, soil exibility and with the earth-
quake maximum ground acceleration. Seismic wall sliding is dominant and rotation is negligible for
rigid walls when the ratio between the wall height and the foundation width is less than 1.4, while
for greater ratios the wall becomes more exible and rotation (rocking) increases till the ratio
reaches 1.8 where overturning is susceptible to take place. Cumulative seismic wall rotation
increases with dynamic time and tends to be constant at the end of earthquake.
2014 Production and hosting by Elsevier B.V. on behalf of Housing and Building National Research
Center.

Introduction stability of gravity retaining walls (Mononobe, and Matuo [1];

Limit-state analysis method based on Pseudo-static approach
is among several methods that have been used to study seismic
* Tel.: +20 1001525472.
E-mail address: kmhi123@Yahoo.com
Peer review under responsibility of Housing and Building National
Research Center.
Production and hosting by Elsevier
1687-4048 2014 Production and hosting by Elsevier B.V. on behalf of Housing and Building National Research Center.
http://dx.doi.org/10.1016/j.hbrcj.2014.03.006
Okabe [2]; Choudhury et al. [3] and Ortigosa [4]).
Pseudo-dynamic approach has the capability of considering
the dynamic nature of the earthquake loading in an approxi-
mate and simple manner compared with other methods. The
phase difference and the amplication effects within the soil
mass are considered along with the accelerations causing
inertia forces (Steedman and Zeng [5]).
Closed form solutions using elastic or viscous elastic
behavior analyzed the response of a rigid non-yielding wall
retaining a homogeneous linear elastic soil and connected to
a rigid base. For such conditions Veletsos and Younan [6]
concluded that the dynamic amplication is insignicant for
Seismic displacement of gravity retaining walls
relatively low-frequency ground motions (that is, motions less
than half of the natural frequency of the unconstrained back-
ll), which would include many or most earthquake problems.
Numerical analyses by nite element numerical model have
been developed for gravity walls found on dry sand (Al-
Homoud and Whitman [7]) using two-dimensional (2D) nite
element computer code, FLEX. Dynamic analyses in FLEX
are performed using an explicit time integration technique
(Green and Ebeling [8]). Other numerical models have been
developed using FLAC nite difference code.
Gazetas et al. [9] applied numerical models using the com-
mercial nite-element package ABAQUS for two dimensional
plane-strain conditions.
Displacement-based analysis considers the earthquake
motion vibrates with the backll soil, and the wall can easily
move from the original position due to this earthquake
motion. The methods available for displacement based analy-
sis of retaining structures during seismic conditions are based
on the early work of Newmark [10], and Kramer [11]. The
basic procedure was developed for evaluating the deformation
of an embankment dam shaken by earthquake based on the
analogy of sliding block-on-a-plane. Richards and Elms [12]
model proposed the basic Newmarks model, developed origi-
nally for evaluation of seismic slope stability and modied it
for the design of gravity retaining walls.
Some guidelines for permissible displacement based on
experience or judgment (Huang [13]) are used for the design
of retaining walls which failed during earthquakes by sliding
away from the backll or due to combined action of sliding
and rocking displacements.
The permissible horizontal displacement according to
Eurocode [14] equals 300.amax (mm), while according to
AASHTO [15] it equals 250.amax (mm), where amax is the max-
imum horizontal design acceleration.
Wu and Prakash [16] predict that permissible horizontal
displacement equals 0.02 H and the failure horizontal displace-
ment equals 0.1 H, where H is the height of retaining wall.
JRA [17] suggested that the permissible differential settle-
ment equals 0.10.2 m, and that the severe differential settle-
ment P0.2 m (damage needing long term retrot measures).
Rafnsson and Prakash [18] developed a model for a rigid
wall resting on the foundation soil and subjected to a horizon-
tal ground motion and analyzed the problem as a case of com-
bined sliding and rocking vibrations including the effect of
various important parameters such as soil stiffness in sliding,
soil stiffness in rocking, geometrical damping in sliding, geo-
metrical damping in rocking, material damping in sliding,
and material damping in rocking. The cumulative displace-
ment of retaining walls due to combined sliding and rocking
for negative back slope of walls is smaller compared to the case
of vertical face or positive back slope.
Wu [19] lists the cumulative displacements for gravity walls
4 m to 10 m high walls with a typical granular backll sub-
jected to 3 earthquakes. He found that for 8 m high wall and
base width 4.6 m when subjected to El-Centro earthquake, it
undergoes 0.135 m sliding, 0.286 m rocking with total com-
bined displacement equals 0.42 m. The later value is greater
than the permissible displacement 300 amax = 0.349 300 -
mm = 0.104 m according to the Eurocode.
Nadim [20], Nadim and Whitman [21] recommended deter-
mining the frequency ratio between fundamental ground
motion (f) and fundamental frequency of backll (f1). If the
225
ratio is less than 0.25, neglect the ground amplication, if
the ratio is approximately 0.5, increase the peak acceleration
by 2530% and if the ratio is between 0.7 and 1.0, increase
amax and V (ground velocity) by 50%.
Based on the previous literature survey the objective of the
present work is to carry out modied seismic numerical analy-
sis for gravity retaining walls and comparing the resulted dis-
placements with the corresponding values given by previous
work, also to introduce an additional analysis for both active
and limited passive earth pressure for the aim of reducing seis-
mic wall displacement to be within the permissible values.
Pseudo-static analysis of gravity retaining walls
Fig. 1 shows a sketch for a trapezoidal gravity retaining wall of
dimensions A, B and height H undergoing sliding and rocking
displacement. The ground backll is inclined by angle b. The
back surface of the wall is inclined by angle h with the vertical.
The sliding is represented by the translation from point 1 to 10
while rocking of the wall is represented by the inclination angle
i with the vertical.
MononobeOkabe developed the total static and dynamic
active coefcient of earth pressure acting on rigid retaining
walls
amax Kh g maximum ground acceleration 1
Kh and Kv are coefcients of horizontal and vertical ground
accelerations, Kv = 0 or half Kh.
Kh
w tan1 2
1  Kv
The total active static and dynamic pressure on the back
surface of the wall is PAE
1
PAE KAE H2 c1  Kv 3
2
cos2 /  h  w
KAE h qi2
sind; sin;bw
cos w cos2 h cosd h w 1 cosdhw cosbh
4
ay N  g
 
PAE cosd h  PAE sind h tan /b
tan /b  g 5
W
where: ay = the critical or yield ground acceleration causing
sliding of the wall
Fig. 1 Seismic displacement of gravity retaining wall.
226 K.M.H.I. Ibrahim

Fig. 3 Finite element mesh of gravity wall with vertical back

toward the back ll.

Table 2 Gravity wall dimensions and soil properties.

A B H b0 /b d /0 E m c
(m) (m) (m) (kN/m2) (kN/m3)
2 5 8 0 23 25 37 80,000 0.3 18.5

records earthquake data are used in the current numerical

analysis. These ASCII les contain time-series coordinates of
accelerations, but they may also contain velocity or displace-
ment series and response spectra. Table 1 shows the maximum
ground accelerations, maximum relative velocities, maximum
ground displacements, earthquake violent duration, funda-
mental frequency (ff) and date of occurrence for six historical
earthquakes.

Material and methods

Fig. 2 Range of normalized displacements using Newmark [10]
sliding block model and various equations.
Fig. 3 shows the nite element mesh for a rigid elastic concrete
gravity retaining wall 8 m high analyzed by dynamic Plaxis [22]
program. Plain strain numerical analysis is performed. The
/b = friction angle at foundation base between the wall mesh width is 100 m and 20 m depth. Horizontal and bottom
and the soil. xity is applied at side and lower boundaries. Lateral absor-
Richard and Elms [12] suggested a value for the permanent bent boundaries are at the sides for absorbing lateral waves
wall displacement and prevent their reection. The wall is subjected to the pre-
V2max a3max ay scribed displacement at the bottom boundary representing
dperm 0:087 in case of  0:3 6 the earthquake load. The foundation soil and backll are gran-
a4y amax
ular sand and are analyzed as elasto-plastic material according
where Vmax = maximum ground velocity. to MohrCoulomb. Elastic analysis is also done and its defor-
 1 mation is compared with elasto-plastic one. The wall width is
ay
3V2max amax ay 2 m at the crest and 5 m at the foundation, the height is 8 m.
dperm in case < 0:3 7 Two cases of back wall slope are studied, one case is studied
amax amax
vertical and the other is inclined with the vertical.
Fig. 2 shows the estimated displacements for different N/A
Two loading phases are applied. Phase 1 is a static elasto-
ratios (Newmark [10]).
plastic phase which represents excavation 8 m in front of the
wall. Phase 2 is a dynamic load phase and starts by resetting
Properties of applied strong motion for six historical static displacements to zero then prescribed displacement for
earthquakes the chosen earthquake is applied within the duration of
dynamic time in seconds.
SMC (Strong Motion CD-ROM) les, obtained from the U.S. The analyzed wall height, geometric dimensions and soil
Geological Survey National Strong-motion program that properties are given in Table 2.

Table 1 Dynamic property of different historical earthquakes.

Earthquake amax (m/s2) Vmax (m/s) dmax (m) t(s) ffund (c.p.s.) Date
El-Centro/Imperial Valley 3.36 0.35 0.091 80 1.5 1940
Petrolli 4.13 0.827 0.215 40 0.7
Shake 3.69 0.51 0.144 30 0.5/1.4 1906
Taft Lincoln School Tunnel/Kern County 1.81 0.17 0.062 70 0.6/1.0/2.2 1952
Topanga Canyon Fire Station/Northridge 3.22 0.15 0.039 40 0.6/1.8/4 1994
Treasure Island/Loma Prieta 1.54 0.332 0.12 20 0.5/1.8 1989
Yerba Island/Loma Prieta 0.634 0.146 0.046 30 0.8/1.8 1989
Seismic displacement of gravity retaining walls 227

0.32
Numerical wall sliding
0.28 Richard and Elms 12

Seismic wall displacement (m)

permissible disp. AASHTO 15
0.24

0.2

0.16

0.12

0.08

0.04

0
El-centro Petrolli Shake Taft Topanga Trasure Yerba

Fig. 4 Comparison between numerical, pseudo-static and permissible sliding with vertical back surface acted upon by different historical
earthquakes.
Top wall seismic displacement (m)

0.30 0.15
Mohr-Column analysis Top seismic wall displacement (m) Zero foundation depth
0.20 Elastic analysis 2 m foundation depth
0.10
0.10
0.05
0.00

-0.10 0.00

-0.20
-0.05
-0.30

-0.40 -0.10
0 10 20 30 40 50
Dynamic time (s) -0.15
0 10 20 30 40 50 60 70
Fig. 5 Top wall displacement versus dynamic time in case of Dynamic time (s)

elastic and MohrColumn analysis under the effect of Petrolli

Fig. 7 Effect of foundation depth of the gravity wall on seismic
earthquake.
displacement under applied El-Centro earthquake.

0.15 0.15
Seismic top wall displacement (m)

Inclined back wall Taft EQ.

0.10
0.10 Vertical back wall Petrolia EQ.
0.05
Top wall displacement (m)

0.05 0.00
-0.05
0.00
-0.10
-0.05 -0.15

-0.10 -0.20
-0.25
-0.15
-0.30
0 10 20 30 40 50 60 70
-0.35
Dynamic time (s)
0 10 20 30 40 50 60 70
Dynamic time (s)
Fig. 6 Effect of back wall inclination on seismic wall displace-
ment (El-Centro earthquake). Fig. 8 Effect of earthquake loading on top wall displacement.

earthquake amax = 3.36 m/s2, Vmax = 0.352 m/s, then W =

Results and discussions 67.2 t/m1, Kh = 0.144, Kv = 0, KAE = 0.4, PAE = 25.9,
ay = 1.47 m/s2 and N = 0.144 for vertical back wall surface
Conventional wall displacement and ay = 1.68 m/s2 and N = 0.168 for 20.6 positive inclined
back wall surface.
Applying the previous equations from 1 to 6 on the gravity From equation 6 and according to pseudo-static analysis,
retaining wall given in Table 2 when subjected to El-centro the base wall displacement = 10.13 cm. Other wall displace-
228 K.M.H.I. Ibrahim

0.15 0.26
Wall seismic displacement (m) Top wall disp.
0.24
0.10 Foundation wall disp. El-centro eathquake

Maximum top wall seismic

0.22

displacement (m)
0.05
0.20
0.00
0.18
-0.05
0.16

-0.10 0.14

-0.15 0.12
0 10 20 30 40 50 60 70 2.E+04 3.E+04 4.E+04 5.E+04 6.E+04 7.E+04 8.E+04

Dynamic time (s) Soil young's modulus (kN/m2)

Fig. 9 Top and foundation wall displacement distribution with Fig. 12 Effect of soil stiffness on maximum top wall
dynamic time for El-centro earthquake. displacement.

2.E-03
El-centro earthquake
Seismic wall rotation (degree)

1.E-03 El-centro earthquake

Petroli earthquake

Maximum Top wall displacement (m)

0.E+00 0.15
-1.E-03
0.14
-2.E-03
-3.E-03 0.13
-4.E-03 H
0.12
-5.E-03
-6.E-03 0.11
B
-7.E-03
0 10 20 30 40 50 60 70 0.10
Dynamic time (s)
0.09
1.2 1.3 1.4 1.5 1.6 1.7 1.8
Fig. 10 Cumulative seismic wall rotation distribution with H/B
dynamic time.
Fig. 13 Effect of H/B on maximum seismic top wall
0.15
E=80000 kN/m2 displacement.
Top wall displacement (m)

0.10 E=20000kN/m2
E=40000kN/m2
0.05

0.00
Fig. 4 shows the numerical seismic sliding for 8 m high
gravity wall due to phase 2, the pseudo-static displacement cal-
-0.05 culated by Richard and Elms [12] and the permissible displace-
-0.10 ment given by ASSHTO [15]. It is noticed that numerical
seismic wall displacements in case of Petrolli and Shake earth-
-0.15
quakes are nearly equal to those calculated by the pseudo-sta-
-0.20 tic method. In case of El-centro, Taft, Topanga, Treasure and
Yerba earthquakes the numerical seismic wall displacements
-0.25
are greater than the corresponding pseudo-static method. Per-
0 10 20 30 40 50 60 70
Dynamic time (s) missible displacements of El-centro and Taft earthquakes are
nearly equal to the corresponding numerical ones, except Pet-
Fig. 11 Effect of soil stiffness on top wall displacement under El- rolli, Shake, Treasure and Yerba earthquakes their permissible
centro earthquake. displacements are relatively smaller. Permissible displacements
only depend on maximum ground acceleration and does not
ments calculated using pseudo-static methods for other earth- consider soil properties, earthquake velocity, soil damping
quakes are shown in Fig. 4. and earthquake fundamental frequency but other methods
It is noticed that if the ground acceleration is nearly equal include.
to or less than the yield acceleration (1.68 m/s2) of the gravity Fig. 5 shows both elastic and elasto-plastic top wall seismic
wall as in case of Taft, Treasure and Yerba earthquakes, the displacement under effect of Petroli earthquake. It is noticed
resulted seismic displacement is nearly zero. that the maximum elastic displacement is 0.2 m while the elas-
to-plastic displacement is 0.33 m.
Numerical wall displacement Fig. 6 shows that for vertical back wall gravity wall the
maximum top displacement is 0.09 m while for inclined back
The static active wall displacement resulted from phase 1 is wall (20.6o ) the maximum top displacement is 0.13 m under
uniform and equals 5.3 103 m for crest and foundation effect of El-centro earthquake at the 23 second. This may be
displacement. due to additional soil mass acting on the inclined back wall.
Seismic displacement of gravity retaining walls
Fig. 7 show that 2 m foundation depth reduces the maxi-
mum top displacement from 0.133 m to 0.09 m under effect
of El-Centro earthquake. Passive resistance reduces the seismic
wall displacement.
Fig. 8 shows the top wall displacement distribution under
effect of Taft and Petrolia earthquake for inclined positive
back slope gravity wall. The maximum ground accelerations
are 1.81 and 4.13 m/s2 while the corresponding maximum
top wall displacements are -0.076 and 0.33 m at 55.6 s and
26.7 s respectively. It is noticed that Taft earthquake total
duration is 66 s and causes residual plastic top wall displace-
ment about -0.03 m while in case of Petroli earthquake the
residual plastic end displacement is about -0.19 m. The nal
permanent seismic wall displacement is directly proportional
to the maximum ground acceleration.
Fig. 9 shows the top and foundation wall displacement dis-
tribution with dynamic time for El-centro earthquake. The
maximum top wall displacement is 0.133 m occurs at 23.75 sec-
onds while the corresponding foundation displacement is
0.095 m. The difference between both displacements divided
by the wall height represents the wall seismic rotation angle
as shown in Fig. 10.
Fig. 10 shows the cumulative seismic wall rotation for both
El-centro and Petroli earthquakes. The seismic wall rotation
increases gradually with dynamic time and reaches its peak
value at end of the earthquake.
Fig. 11 shows the seismic top wall displacement with
dynamic time for three granular soils with youngs modulus
20000, 40000 and 80000 kN/m2. It is noticed that the seismic
wall displacement decreases as soil youngs modulus increase.
Fig. 12, shows that the maximum corresponding seismic
top wall displacements are 0.236, 0.166 and 0.133 m
respectively.
Fig. 13 shows the effect of increasing wall height with
respect to footing breadth. The maximum top wall displace-
ment increases with increasing the wall height. For H/b less
than 1.4 the wall behaves as a rigid body and sliding is domi-
nant while rotation is nearly zero. For H/B greater than 1.4
and less than 1.6 the wall becomes more exible and wall rota-
tion increases. For H/B equals 1.8 represents the maximum
wall seismic displacement and beyond that it is noticed that
overturning failure takes place.
Conclusion
- Some numerical seismic displacements for studied historical
ground motions are nearly equal to those calculated by the
pseudo-static method and others are relatively greater.
- Permissible seismic wall displacement calculated by
AASHTO [15] can be used for empirical displacement esti-
mation as it depends only on maximum ground accelera-
tion and does not consider the effect of soil stiffness,
earthquake fundamental frequency and dynamic time.
- Gravity retaining walls with inclined positive back slope
surface undergoes greater total sliding and rocking dis-
placements than vertical back walls due to increase in mass
and inertia horizontal force of backll.
- The cumulative seismic wall rotation increases gradually
with dynamic time and reaches maximum at the end of
earthquake.
- Increase in soil exibility increases too much the seismic
wall displacement.
229
- In the study, for rigid gravity walls having H/B less than 1.4
seismic sliding is dominant while wall rotation is nearly
zero. For greater ratios, the wall becomes more exible
and its rotation increases till overturning and failure takes
place.
Conict of interest
None declared.
References
[1] N. Mononobe, H. Matuo, On the determination of earth
pressures during earthquakes, Proc. World Engrg. Congr.,
Tokyo, Japan 1929, 9, paper (388).
[2] S. Okabe, General theory of earth pressure and seismic stability
of retaining wall and dam, J. Japan Soc. Civ. Engrs., Tokyo,
Japan 1924, 12(1).
[3] D. Choudhury, K.S. Subba Rao, S. Ghosh, Passive earth pressure
distribution under seismic condition, 15th Eng. Mechanics
Conference of ASCE, Columbia University, New York, 2002.
[4] P. Ortigosa, Seismic earth pressure including soil cohesion, the
16th International Conference on Soil Mechanics and
Geotechnical Engineering, Osaka, 2005.
[5] R.S. Steedman, X. Zeng, The inuence of phase on the
calculation of pseudo-static earth pressure on a retaining wall,
Geotechnique 40 (1) (1990) 103112.
[6] A.S. Veletsos, A.H. Younan, Dynamic modelling and response
of soil-wall systems, J. Geotech. Engrg. ASCE 120 (12) (1994B)
21552179.
[7] A.S. Al-Homoud, R.V. Whitman, Seismic analysis and design of
rigid bridge abutment considering rotation and sliding incorporating
non-linear soil behaviour, Soil Dynam. Eng. (1999) 247277.
[8] R.A. Green, R.M. Ebeling, Modelling the dynamic response of
cantilever earth-retaining walls using FLAC, Numeric. Model.
Geomech. (2003).
[9] G. Gazetas, P. Psarropoulos, I. Anastasopoulos, N. Gerolymos,
Seismic behaviour of exible retaining systems subjected to short
duration moderately-strong excitation, Soil Dynam. Earthqu.
Eng. 25 (2005) 537550.
[10] N.N. Newmark, Effect of earthquakes on dams and
embankments, Geotechnique 15 (2) (1965) 139160.
[11] S.L. Kramer, Geotechnical Earthquake Engineering, Prentice-
Hall, New Jersey, 1996, pp. 466505.
[12] R. Richard, D.G. Elms, Seismic behaviour of gravity retaining
walls, J. Geotec. Engrg. ASCE (1979) 105 (GT4).
[13] Chin-Chan Huang, Seismic displacement of soil retaining walls
situated on slope, J. Geotechn. Geo-Environ. Eng. ASCE 31 (9)
(2005) 11081117.
[14] EUROCODE 8 (European pre-standard), Design Provisions for
Earthquake Resistance of Structures-Part 5: Foundations
Retaining Structures and Geotechnical Aspects, The
Commission of the European Communities, 1994.
[15] American Association of State Highway and Transportation
Ofcials, Standard Specications for Highway Bridges,
(AASHTO) Sections 3 and 7, 2002.
[16] Y. Wu, S. Prakash, Effect of submergence on seismic displacement
of rigid walls, Earthquake Geotechnical Engineering, 1999
Balkema, Rotterdam, ISBN 90 5809 1163, 1999.
[17] JRA, Seismic Design Specications and Construction of
highway Bridges, Japan Road Association, 1996.
[18] E.A. Rafnsson, S. Prakash, Stiffness and damping parameters
for dynamic analysis of retaining walls, in: Proc. Of 2nd
International Conference on Recent Advances in Geotechnical
Earthquake Eng. and Soil Dynamics, St. Louis MO, 1991, vol.
3, pp. 19432952.
230
[19] Y. Wu, Displacement-based analysis and design of rigid
retaining walls to real earthquakes, PhD. Dissertation, Uni of
Missouri-Rolla, 1999.
[20] F. Nadim, A numerical model for evaluation of seismic
behaviour of gravity retaining walls, ScD Thesis, Department
of Civil Eng., M.I.T., Cambridge, MA, 1982.
K.M.H.I. Ibrahim
[21] F. Nadim, R.V. Whitman, Coupled sliding and tilting of gravity
retaining walls, J. Geot. Eng. Divis. ASDCE 109 (7) (1983) 915
931.
[22] Plaxis 2D Version 8, R.B.J. Brinkgreve (Ed.), Delft University
of Technology & Plaxis b.v., Netherland, P.O. Box 572, 2600
AN DELFT, Netherlands.