485ac - Viv

ATLANTIA OFFSHORE LIMITED
MEMBER OF THE IHC CALAND GROUP
VIV Application to Deepwater Risers

Steve Leverette Atlantia Offshore Limited December 2-3, 2003

Agenda
Introduction Environment Riser Response to Currents Fatigue Tendons The FIX Analysis Methods and Programs Model Tests and Data Sets Other Issues Areas of R&D Industry Experience Conclusion

Thanks to:
Thanos Moros, BP Owen Oakley, Jim Stear, Hugh Thompson, Chevron Kim Vandiver, MIT Guy Mansour, Atlantia T. Sarpkaya, Naval Postgraduate School Don Allen/Li Lee, Shell Global Solutions Rodney Masters, AIMS

Introduction
Risers and tendons Steel Catenary Riser: SCR Top Tensioned Riser: TTR Flexible Riser Free Standing Riser: FSR Highly Compliant Riser: HCR Tendons

Perspective from the industry:

The offshore industry has been made painfully aware of what may go wrong, how VIV may affect operations. The industry is having to be conservative because of uncertainties in environment and in engineering prediction tools. For water depths to 5000, we can proceed with reasonable solutions. For 10,000, we cannot afford high levels of conservatism and still afford to make money producing oil.

VIV is a resonant feedback phenomena

Engineering in uncertainty: Risks not generally in getting design numbers wrong, but in not designing for new phenomena.

Risks and Cost Impacts

Risers and Tendons represent critical structural and oil containment systems Uncertain environment new phenomena being discovered Uncertain response highly non-linear, stochastic in nature Failure has high cost

Agenda
Introduction
Environment
Riser Response to Currents Fatigue Tendons The FIX Analysis Methods and Programs Model Tests and Data Sets Other Issues Areas of R&D Industry Experience Conclusion

Environment
Geography Mode experiment Types of Ocean Currents Typical and extreme magnitudes Durations Uncertainties, unknowns Combination with other events
Ratio of environmental forces, shallow to deep




MODE Experiment
WHOI - 1970s Weather in the Oceans Fronts, temporal variation Large scale turbulence Vertical Structure

Ocean Currents
Tidal Currents (how they exist in deep water) Loop Eddy Storm (wind stress model) Inertial Slope Bottom Boundary Current (Sigsby Escarpment)

Ocean Currents loop current / eddy current

Ocean Currents Cold Core Eddy
0.000 0.0 500.0 1000.0
1.000
2.000
Cold Core Current Profile Velocity 3.000 4.000 5.000
6.000
7.000
Depth - feet
1500.0 2000.0 2500.0 3000.0 3500.0

Inertial and Slope currents

Morpeth
Hurricane Earl
Neptune
Hurricane Georges 1.8 2.2 kts in 2000

Sigsbee Escarpment
Currents identified from bottom furrows

0.000 0.0 2.000
Allegheny Current Profiles

Velocity (ft/s) 4.000 6.000 8.000
Orig. Warm Core New Warm Core Profile New Cold Core Profile
Depth - feet
Current profiles
500.0 1000.0 1500.0 2000.0 2500.0 3000.0 3500.0


Environment
Insert DW Currents presentation








Agenda
Introduction Environment
Riser Response to Currents

Fatigue Tendons The FIX Analysis Methods and Programs Model Tests and Data Sets Other Issues Areas of R&D Industry Experience Conclusion


Types of Risers: TTR - top tensioned riser SCR - steel catenary riser Flexible riser HCR - highly compliant FSR - free standing riser


Typical dimensions
4 flowline SCR 4 flowline SCR with 9 buoyancy 21 drilling riser 48 drilling riser with buoyancy 18 export SCR 120 aircans 8 tieback risers

Effects of End Conditions

Taper Joint Flexjoint Keel joint Tensioner/Gimbal Guide frames Cantilever wellhead




PRT at Full Extend, Null, Full Retract





Buoyancy, cladding, strakes, etc Touchdown point

KingKong riser with Buoyancy


Static response
Vessel Offset Increased drag due to VIV Interference analysis Touchdown trench effects


Dynamic response
Modal analysis (string, beam models) Bending vs axial modes Finite versus infinite transmission models

TLP Production Riser Lateral Motion Mode Shape

1 Mode 22 Mode 23 0.9
TLP production Riser Modes 1-2,11-12,22-23

1 Mode 11 0.9 Mode 10
Relative position, x/L
0.8
0.7
0.6
0.8
0.5

1 0.7
0.4
0.9
0.6
0.3
0.8 Mode 1
0.5
0.2
0.7
0.4
0.1
0.6
0.3
0 -1
0.5 Mode 2 0.2
-0.8
-0.6
-0.4
-0.2
0 Mode shape
0.2
0.4
0.6
0.8
0.4
0.1
0.3 -1 -0.8 -0.6 -0.4 -0.2
0 0 Mode shape 0.2 0.4 0.6 0.8 1
0.2
0.1
0 -1 -0.8 -0.6 -0.4 -0.2 0 Mode shape 0.2 0.4 0.6 0.8 1

Mode No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Frequency, Hz 0.090 0.180 0.271 0.363 0.456 0.552 0.650 0.750 0.853 0.959 1.069 1.183 1.301 1.423 1.549 1.680 1.816 1.957 2.103 2.255
Typhoon TLP Tendon Lateral Motion Natural Frequency (Pretension = 1917 kips)
2.5
2.0
Frequency, Hz
1.5
1.0
0.5
0.0 0 2 4 6 8 10 Mode Number 12 14 16 18 20

Illustration of SCR Inplane Modes

4.0E-04 3.0E-04
Curvature (1/ft)
2.0E-04 1.0E-04 0.0E+00 -1.0E-04 -2.0E-04 -3.0E-04 0 500 1000 1500 2000 2500 Mode 1 Mode 5 Mode 10
Node Number

2500
2000
LOCAL-1 AXIS
1500
LOCAL-1 AXIS
1000
500
-500
-2250 -1750 -1250
-750
-250
250
750
-500
500
1000
1500
2000
2500
Mean Position Mode No. 2
Mean Position Mode No. 32
-2250 -1750 -1250
-750
-250
250
750
LOCAL-3 AXIS ; VP=(0.00,0.00,1.00)
LOCAL-3 AXIS ; VP=(0.00,0.00,1.00)
Examples of SCR Mode Shapes
Illustration of Axial Modes Influence ATLANTIA OFFSHORE LIMITED

MEMBER OF THE IHC CALAND GROUP 8.0E-03
7.0E-03
Max Curvature (1/ft)
6.0E-03 5.0E-03 4.0E-03 3.0E-03 2.0E-03 1.0E-03 0.0E+00 0 50 100 150 200 250
Mean Position - G68_m ean_2D_XY_110 Mode No.50 - G68_m ean_2D_XY_110
In-plane Mode Number

3000 3500
LOCAL-1 AXIS
-500
0 -1000
500
1000
1500
2000
2500
-500
500
1000
1500
2000
2500
3000
LOCAL-3 AXIS ; VP=(0.00,0.00,1.00)

Steady Flow Response for an Infinite Beam with Finite Length Excitation
1 0.5
L power in = 4.5
Standing wave excitation
n o i t a t i c x e
0 -0.5 -1 -400 0.6
-300
-200
-100
100
200
300
400
e d u t i 0.4 l p m a e 0.2 s n o p s 0 e -400 r
Traveling waves Standing wave response

-300 -200 -100 0 100 200 300 400

Matterhorn Production Risers
Matterhorn production riser from main deck


Jolliet Single Piece Tendon

Steel tendons 24 in diameter.
Provided by Steve Leverette Atlantia Offshore Limited


VIV response characterization
Strouhal number, Reduced velocity Cl (AOD, Re, Vr) Pre-lockin, Lock-in, lock-out Power in, versus damping and radiation Effects of damping on response Power balance solution

VIV Characteristics
St = f st D U
Strouhal Number Reduced Velocity
Vr =
U f st D
St=0.2, Vr=5

Historical Shear7, Clift Formulation

Pre-lock-in, Lock-in, Lock-out

Dynamic Response Power Balance

Power Balance



Single-mode vs Multi-mode
In-line vs transverse


Bending and increased drag effects Spectral approach to response Uncertainties in response Typical hot spots Types of currents that are problematic Effect of increased tension (decrease mode, reduce bending)

Agenda
Introduction Environment Riser Response to Currents
Fatigue
Tendons The FIX Analysis Methods and Programs Model Tests and Data Sets Other Issues Areas of R&D Industry Experience Conclusion

Fatigue caused by VIV

Fatigue
Miners rule
D=
i =1
ni Ni
log N = log a m log log N = log a m log [ (t/tref)k] API, AWS DOE, DnV

Fatigue

Fatigue

One Year of Tension Spectra

Allegheny (Kips)2/Hz Tendon 1
1.8
100000
1.6
10000
1.4
1000
1.2
frequency (Hz)
100
1
10
0.8
1
0.6
0.1
0.4
0.2
0.01
01/01/01 00:00
02/26 12:00
04/24
06/19 12:00
08/15 Date
10/10 12:00
12/06
01/31 12:00
Confidential Property of Atlantia
orr 03/25/02 17:13:52

Safety Factors
Depends on application Typically 3 for structures Typically 10-20 for risers

Fatigue
Fracture Mechanics
Frequency Sensitivity: mode, curvature, no. cycles Ways to combine response modes
Rainflow Spectral approach to fatigue (Wirsching, Dirlik)
Sensitivity to assumptions

Agenda
Introduction Environment Riser Response to Currents Fatigue
Tendons
The FIX Analysis Methods and Programs Model Tests and Data Sets Other Issues Areas of R&D Industry Experience Conclusion

Tendons in VIV
OTC Paper
Leverette, Rijken, Thompson, Dooley

Agenda
Introduction Environment Riser Response to Currents Fatigue Tendons
The FIX
Analysis Methods and Programs Model Tests and Data Sets Other Issues Areas of R&D Industry Experience Conclusion


The FIX


Sh e ll G lo b a l So lu tio n s
Fairings for drilling riser

Fairings for TLP tendon

Fairings in moonpool

Half strakes for pipelines




Agenda
Introduction Environment Riser Response to Currents Fatigue Tendons The FIX
Analysis Methods and Programs

Model Tests and Data Sets Other Issues Areas of R&D Industry Experience Conclusion





Engineering tools, not simulators Summary of how they work Comparisons Why they vary so much CFD Wake models and other approaches

Flow diagram of Shear 7 Analysis



3 4 2 1
Figure 8: Shear7 4.3 defines a smoothed lift coefficient curve by fitting two parabolas to three points defined by four user-specified values.

VIV Program Comparison (Stride JIP - 2H) RMS and Damage

VIV Program Comparison 2 (Stride JIP - 2H) RMS and Damage

Example Shear7 Analysis




Shear7, Viva, Vivana, Vivarray Engineering tools, not simulators Summary of how they work Comparisons / Example Why they vary so much CFD Wake models and other approaches

Agenda
Introduction Environment Riser Response to Currents Fatigue Tendons The FIX Analysis Methods and Programs
Model Tests and Data Sets

Other Issues Areas of R&D Industry Experience Conclusion

The Literature
Review Papers Parkinson (1974, 1989) Sarpkaya (1979, 1995) Griffin and Ramberg (1982) Bearman (1984) Pantazopoulos (1994) Books Blevin 1990 Chen 1987 Naudascher and Rockwell 1994 Sumer and Fredsoe 1997 Au-Yang 2001

Recent Model Tests and Data Sets

Deepstar St Johns, CFD, Lake Seneca Stride (2H Allegheny and model tests) VIVA PMB HCR Lake Pend Oreille BP Exxon Shell

Agenda
Introduction Environment Riser Response to Currents Fatigue Tendons The FIX Analysis Methods and Programs Model Tests and Data Sets
Other Issues
Areas of R&D Industry Experience Conclusion

Other Issues
Wake effects, riser arrays High Re performance of fairings/strakes Issue of low mass systems, frequency independence Directionality of currents, resulting stress hot spots Directionality of response amplitude Trenching

Trenching

Agenda
Introduction Environment Riser Response to Currents Fatigue Tendons The FIX Analysis Methods and Programs Model Tests and Data Sets Other Issues
Areas of R&D
Industry Experience Conclusion

Areas of Research and Development
Empirical CFD Fairings JIPs

MIT - VIVA MIT - Vandiver JIP and Shear7 Deepstar (ARA CFD, Principia CFD, Lake Seneca, St. Johns) Several CFD proposals out


Slides from Chevron (O. Oakley)

riser undergoing hydro-elastic response; mesh and vorticity contours in the wake are shown; note the difference in the shed wake pattern between the nodes and the anti-nodes of the structure


Vortex Dislocations Visualizations

Re=100 Experiments C.H.K. Williamson (1989) Remax=100 R-DNS Simulations
Remax=1000 R-DNS Simulations

Re=100k

Re=600k

Re=4MM

Grid for cylinder

Strip theory approach to riser modeling (J. Kallinderis)


EMBER OF THE ALAND ROUP
Constant Contours of Spanwise Vorticity (non-dimensional) M IHC C G

Numerical Modeling
CFD = computational fluid dynamics, solution of Navier-Stokes equations for fluid flow
U xi i =
j
= 0 i ) 1 xi x j
U
x
(
j
j x j
(
j
) x j
DNS direct numerical solver

DES
CFD Approach:
DNS
LES
RAN S S-A k- k- SMC, ..
none
Smag . S-A ..
S-A
Turbulence Model
LES turbulence model spatial filter representation of turbulence fluctuations

CFD Spars
2dof Modeling Straked Cylinder

No chains, pipes or anodes S-A turbulence model Re = 34MM Navier-Stokes Solution: URANS - time averaging LES space averaging DES RANS in boundary layer & LES outside
RANS Grid
DES Grid

Displacement History Genesis spar, strakes only
Pressure
DES
Velocity Mag.
Pressure
RANS
041707S ES 04/17/01 07:00:01 0
-10
-20
-30
Northings (ft)
-40
-50
-60
-70
-80
-90 30
40
50
60
70 80 Eas tings (ft)
90
100
110
120

Lock-in Slope
Locked in
Lock-out Slope

Objectives and Challenges for High Mode Number, Flow-Induced Vibration Model Tests in Sheared Flow
Prof. J. Kim Vandiver MIT June 3, 2003

Principal Issues for High Mode Number VIV Response Prediction Does lock-in occur at high mode number in uniform and sheared flow? What are the statistics of single versus multi-frequency response in sheared flow? What are the statistics of in-line and crossflow response?

Principal Issues for High Mode Number VIV Response Prediction
Do hydrodynamic damping models need improvement? What fairing or strake coverage is required What is the effect of Reynolds number on St, CL damping and suppression effectiveness.

Important dimensionless groups

1. m mass ratio f D2
2. VR =
U ( x) , Reduced velocity fv D
4. What about correlation length?
3.
Rn n Reduced damping f U 2 ( x) Lin
5. s =
2 n m
rs
, and n t =
2L
Modal damping and wave attenuation

6. Re = DU ( x )
, Reynolds number

How is lock-in affected by mass ratio versus reduced velocity bandwidth?
1.
m mass ratio f D2
VR U ( x) , dVR = = lock-in bandwidth fv D VRc
2. VR =
f v = the vibration frequency

Is the response different if the wake adjusts its frequency to match the cylinder or the natural frequency adjusts to match the wake.

A/D versus Reduced velocity (based on a fixed natural frequency)

mr=0.78
VR VRc
mr=34
Castine field experiment in uniform flow(1981) Uniform flow on short lengths(75 feet)

Site layout

Cable, pipe and wire rope properties
Cable was PVC plastic tubing with accelerometers, wires, strength members and potting compound, D=1.25 inches., s.g.=1.4 Pipe was 1.631 inch steel tube with the cable pulled inside as the measuring instrument. S.g.=2.4 Wire rope was polyethylene coated oceanographic wire, 3x19 construction. D=.28 inches, s.g.=2.5

Castine pipe response, L=75 ft, D=1.63 inches, specific gravity = 2.4

Pipe lock-in and non-lock-in response

Castine cable cross-flow VIV, D=1.25 inches specific gravity = 1.4

Stable lock-in events are aided by isolated natural frequencies with no overlap.
Cable, s.g.=1.4 Ca= 0 and 1.0 Frequency Hz Pipe, s.g.=2.4
Tension

An example from Helland Hansen: Peak frequencies versus time. What is going on?
Power Spectra, Helland-Hansen, DACOS 2 (x-acc) 98-05-08 100 90 98-05-07 80 70 60 98-05-05 50 40 98-05-04 98-05-06 Date [YY-MM-DD]
From Robit Helland Hansen Report

Wire rope: L=75 ft, D=0.25 inches, specific gravity = 2.6 Steady beating around 6th mode, components 0.6 Hz apart Reduced velocity bandwidth of +-20%=+-2.6Hz

Does lock-in occur at High Mode Number ?

Increase in Uniform flow speed Increasing overlap of competing modes Reduced velocity bandwidth of +-20% would include many modes. Can just one respond in a uniform flow? What happens in shear?
Increasing Overlap with Mode Number +/- 20 % Bandwidth
30 25
Mode Number
20 15 10 5 0 0 0.5 1 1.5 2 2.5 Uniform Flow Speed (m/s)

Wire rope: L=900 ft, D=0.280 inches, specific gravity = 2.5
Nearly uniform flow on a 900 foot long sample. +- 10% flow variation along the length Lock-in at 50th mode Conclusions: High mass ratio provides stable sharp resonances. VR bandwidth allows for some spatial tolerance to flow variations. Single frequency dominance is possible in nearly uniform flows, even at high mode number.
A O The role ofLdamping in uniform flow? The reduced damping is a statement of dynamic equilibrium.
TLANTIA FFSHORE IMITED
Two forms are equal: s me - Griffin = 8 2 St2 f D2 s Sg = rs n Vandiver f U n2

A useful property is that: < C L , n > = 2 Su qn , the equation of an D hyperbola for lines of constant lift coefficient.

10
Lift Coe fficie nt in P has e with Ve locity at Lock-in Cl=2S g*A/D Cl =0.2 0.5 1.0 Cylinde r Data
D / 10 A Peak e s Response n o (A/D) p s e -1 r k 10 Cl =0.01 a e 0.02 P 0.05 0.10

10
-2
10
-2
10
-1
10
10
Re duce d Damping, S g
TLANTIA OFFSHORE LIMITED A Aflex ible cylin d er in a n o n -u n ifo rm flo w

w ith a p o w er-in reg io n o f len g th L in h as a reso n an t m o d al resp o n se g iven b y: C L ,n qn = D 2 C L ,n wU n 2 L in = R n n 2Su
In red are 3 k ey p aram eters in p red ictio n p ro g ram s. T h ey form a d im en sio n less g ro u p S u an d C L , n A verag e m o d al lift co efficien t 8 2 S t2 M n n R n n = wU 2 L in w D 2 L in
Su =

Consider a simple slab flow problem first to illustrate the example of limited power-in length.

Shroud created a still water damped region Uniform flow power-in region
Lin L U
Shroud covers this length, creates still water around the cylinder

Example Experimental Data for a Slab Flow

Tsahalis, D. T., Experimental Study of the Vortex Induced Vibrations of a Long Model Riser Exposed to Uniform and Nonuniform Steady Flow, Houston, Texas, Westhollow Research Center, 1985. Thank you to Shell Global Solutions for allowing me to publish these results.

Steel pipe, O.D. = 1.5in(3.81cm) I.D. = 1.334in(3.39cm)
Experiment Details
U= 0.33 to 6ft/s(0.1 to 1.83 m/s) in steps of 0.33ft/s(0.1m/s), modes 1 thru 3. Biaxial accelerometers at 5 points: L/8, L/4, L/2, 5L/8, AND 5L/6.

Shrouds used to simulate slab flow cases.

100% 87.5% 75.0% 62.5% 50.0% 37.5% 25.0%
Exposed length of cylinder in 7 steps


1.6
Peak response A/D, mode 2
Reduced Velocity versus Mode 2, peak A/D for various fractions of exposure length
Vr =
U fv D
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 0.0

Lin=25% Lin=50% Lin=87.5%
Delta Vr = 0.42
2.0
Lin=25% Lin=62.5% Lin=87.5%
4.0
Lin=37.5% Lin=62.5% Lin=100%
6.0
Lin=37.5% Lin=75% Lin=100%
8.0
10.0
Lin=50% Lin=75% Lin=100%
Vr, based on vibration frequency

Jump to DS 6402 presentation

Agenda
Introduction Environment Riser Response to Currents Fatigue Tendons The FIX Analysis Methods and Programs Model Tests and Data Sets Other Issues Areas of R&D
Industry Experience
Conclusion

Industry Experience (interesting bits of history)
BNOC West of Shetlands BP Scheihellion experiences Exxon Brazil Andaman Sea West Seno Risers and Tendons Auger Top Tensioned Sales Riser Allegheny SCR clashing Allegheny trenching Matterhorn (effects of cantilevered wellhead, strakes help installation, cold core re-fit) Typhoon SCR design and cold core issue Discovery of cold core/submerged current events Atlantis (and Thunder Horse) Sigsby Escarpment, high bottom currents, full straked riser 7200 WD, 12 deg top, ~10,000 for 2 export, 6 production

Agenda
Introduction Environment Riser Response to Currents Fatigue Tendons The FIX Analysis Methods and Programs Model Tests and Data Sets Other Issues Areas of R&D Industry Experience
Conclusions



Summary and Conclusions

VIV is significant design issue in deep water Highly non-linear in excitation and in damage Instability of wake, not easy to predict Industry addressing problem through many studies and investigations We can confidently design for it using high safety factors As go to ultra deep, cost of high safety factors is prohibitive, additional progress needed.

485ac - Viv

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

485ac - Viv

Uploaded by

Copyright:

Available Formats

ATLANTIA OFFSHORE LIMITED

MEMBER OF THE IHC CALAND GROUP

VIV Application to Deepwater Risers

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

Perspective from the industry:

ATLANTIA OFFSHORE LIMITED

VIV is a resonant feedback phenomena

ATLANTIA OFFSHORE LIMITED

Risks and Cost Impacts

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

Ocean Currents loop current / eddy current

ATLANTIA OFFSHORE LIMITED

Ocean Currents Cold Core Eddy

0.000 0.0 500.0 1000.0

Cold Core Current Profile Velocity 3.000 4.000 5.000

1500.0 2000.0 2500.0 3000.0 3500.0

ATLANTIA OFFSHORE LIMITED

Inertial and Slope currents

ATLANTIA OFFSHORE LIMITED

Currents identified from bottom furrows

ATLANTIA OFFSHORE LIMITED

Allegheny Current Profiles

500.0 1000.0 1500.0 2000.0 2500.0 3000.0 3500.0

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

Riser Response to Currents

ATLANTIA OFFSHORE LIMITED

Riser Response to Currents

ATLANTIA OFFSHORE LIMITED

Riser Response to Currents

ATLANTIA OFFSHORE LIMITED

Effects of End Conditions

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

PRT at Full Extend, Null, Full Retract

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

ATLANTIA OFFSHORE LIMITED

Riser Response to Currents

ATLANTIA OFFSHORE LIMITED

KingKong riser with Buoyancy