PIANC Seismic Structures

INTERNATIONAL NAVIGATION
ASSOCIATION
ASSOCIATION INTERNATIONALE
DE NAVIGATION
SEISMIC DESIGN GUIDELINES
FOR PORT STRUCTURES
Working Group No. 34 of the Maritime Navigation Commission
International Navigation Association
-

915
A.A. BALKEMA PUBLISHERS / LISSE/ ABINGDON / EXTON(PA) / TOKYO
Library of Congress Cataloging-in-Publication Data
Applied for
Cover design: Studio Jan de Boer. Amsterdam. The Netherlands.
Typesetting: Macmillan India Ltd.. Bangalore. India.
Printed by: Grafisch Produktiebedrijf Gorter. Steenwijk. The Netherlands.
CD ~ O O I Swets & Zeitlinger B.Y.. Lisse
Technical Commentary 5 is exempt from the copyright because the original material
\ \ ' J ~ produced by the US Navy for the state of California. USA. No one can exercise the
copyright.
The figures and table in Technical Commentary 6 have been reprinted from:
Handbook on liquefaction remediation of reclaimed land. Port and Harbour Research
Institute (ed.). ISBN 90 54 j() 653 O. 1997. 25 em. 324 pr..
EL"R 82.50/USS97.00/GBP5S
All rights reserved. No part ( ~ f this publication may he reproduced. stored in a retrieve!
svstrm. or transmitted in 0I1y/0l"/11 or hy means. electronic. mechanical. hy plunocopving.
recording or otherwise. without prior written permission ( ~ ( the publishers.
ISBN 90 265 1g18 8 (hardback)
Contents
PREFACE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX
Members of PIANClMarComIWorking Group 34 XI
List of Tables and Figures: Main Text XIII
Main Text
Chapter 1. INTRODUCTION .. . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 3
Chapter 2. EARTHQUAKES AND PORT STRUCTURES .
2.1 Earthquake Motion .
?? L' f: .
_._ Ique action .
2.3 Tsunamis '..
2.4 Port Structures .
2.5 Examples of Seismic Damage ~ .
Chapter 3. DESIGN PHILOSOPHY .
3.1 Performance-Based Methodology .
3.2 Reference Levels of Earthquake Motions .
3.3 Performance Evaluation .
Chapter 4. DAMAGE CRITERIA .
4.1 Gravity Quay Walls .
4.2 Sheet Pile Quay Walls .
4.3 Pile-Supported Wharves .
4.4 Cellular Quay Walls .
4.5 Quay Walls with Cranes ~ .
4.6 Breakwaters ~ .
Chapter 5. SEISMIC ANALYSIS .
5.1 Types of Analysis .
5.2 Site ResponselLiquefaction Analysis .
7
7
9
12
13
15
23
23
?-
-)
27
31
31
33
36
43
45
52
55
55
56
v r t'JAJ\lL
5.3 Analysis of Port Structures .
5.4 Inputs and Outputs of Analysis .
Technical Commentaries
TC1: EXISTING CODES Al\.TD GUIDELINES .
Tl .1 List of Seismic Design Codes and Guidelines for Port
Structures .
TI.2 Reviewed Aspects of Codes and Guidelines .
Tl.3 Seismic Design Practice for Port Structures around the World
TC2: CASE HISTORIES
T2.1 Damage to Gravity Quay Walls .
T2.2 Damage to Sheet Pile Quay Walls .
T2.3 Damage to Pile-Supported Wharves .
T2.4 Damage to Cellular Quay Walls .
T2.5 Damage to Cranes .
T2.6 Damage to Breakwaters .
TC3: EARTHQUAKE MOTION .
T3.1 Size of Earthquakes .
T3.2 Strong Ground Motion Parameters <
T3.3 Seismic Source and Travel Path Effects .
T3.4 Local Site Effects .
T3.5 Seismic Hazard and Design Earthquake Motion .
T3.6 Ground Motion Input for Seismic Analysis of Port Structures
TC4: GEOTECHNICAL CHARACTERISATION .
T4.1 Mechanical Behaviour of Soil under Cyclic Loads .
T4.2 Measurement of Soil Parameters .
T4.3 Evaluation of Pre-Failure Soil Properties through Empirical
Correlations .
T4.4 Assessment of Liquefaction Potential .
T4.5 Geotechnical Characterisation Strategy with Reference to the
Analysis Requirements .
Appendix: Interpretation (>f Geophysical TeSTS .
TC5: STRUCTURAL DESIGN ASPECTS OF PILE-DECK SYSTE!\1S .
T5.1 Design Performance and Earthquake Levels .
T5.2 Modelling Aspects .
T5.3 Methods of Analysis for Seismic Response .
T5.4 Structural Criteria for Pier and Wharf Piles .
60
66
-\'{
77
79
80
89
91
93
94
94
95
95
96
127
129
130
]34
]38
141
]51
]55 ,
]57
]68
]82
]91
204
:!08
215
217
:!]9
11J
I"',
.:s:
Contents VII
T5.5 PilelDeck Connection Details. . . . . . . . . . . . . . . . . . . . . . .. 253
T5.6 Existing Construction 262
TC6: REMEDIATION OF LIQUEFIABLE SOILS 271
T6.1 Overview....................................... 273
T6.2 Outline of Remedial Measures against Liquefaction . . . . . .. 275
T6.3 Compaction Method: Design and Installation 282
T6..+ Drainage Method: Design and Installation 284
... ...
T6.5 Premix Method: Design and Installation 287
. ... .
T6.6 Preload Method: Desizn and Use 290
...
T6.7 Design of Liquefaction Remediation 294
T6.8 Influence on Existing Structures during Soil Improvement . 299
TC7: ANALYSIS METHODS 307
T7.1 Simplified Analysis 309
T7.2 Simplified Dynamic Analysis . . . . . . . . . . . . . . . . . . .. 330
T7.3 Dynamic Analysis 355
T7.4 Input Parameters for Analysis. . . . . . . . . . . . . . . . . . . . . . .. 369
TC8: EXAMPLES OF SEISMIC PERFORMANCE EVALUATION '" 379
T8.1 Gravity Quay Wall (Grade A) 381
T8.2 Sheet Pile Quay Wall (Grade B) 397
T8.3 Sheet Pile Quay Wall (Grade S) 420
T8.4 Pile-Supported Wharf (Grade B) 423
T8.5 Pile-Supported Wharf (Grade S) 432
LIST OF SYMBOLS 441
REFERENCES 453
INDEX .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 469
Preface
Although the damaging effects of earthquakes have been known for centuries. it
is only since the mid-twentieth century that seismic provisions for port structures
have been adopted in design practice. In 1997, the International Navigation
Association (PIANC; formerly the Permanent International Association for
Navigation Congresses) formed a working group, PIANClMarComIWG34, to
focus international attention on the devastating effects of earthquakes on port
facilities. This book. entitled 'Seismic Design Guidelines for Port Structures: is
the culmination of the efforts of this working group.
This book is the first of its kind in presenting international guidelines for seis-
mic design. The provisions reflect the diverse. nature of port facilities. Although
constructed in the marine environment, the port facilities are associated with
extensive waterfront development, and provide multiple land-sea transport con-
nections. The port must accommodate small to very large vessels, as well as spe-
cial facilities for handling potentially hazardous materials and critical emergency
facilities that must be operational immediately after a devastating earthquake.
The primary goal of the working group was the development of a consistent set
of seismic design guidelines that would have broad international support. The
diverse characteristics of port structures led the working group to adopt an evolu-
tionary design strategy based on seismic response and performance requirements.
Proven simplified methods and state-of-the-art analysis procedures have been
carefully selected and integrated in the guidelines in order to provide a flexible
and consistent methodology for the seismic design of port facilities.
This book consists of a main text and eight technical commentaries. The main
text introduces the reader to basic earthquake engineering concepts and a strategy
for performance-based seismic design. The technical commentaries illustrate spe-
cific aspects of seismic analysis and design, and provide examples of various
applications of the guidelines.
The working group members, technical editors, and graphics experts are listed
below. The working group would like to express their sincere gratitude to a group
of New Zealand wharf designers/earthquake engineering experts, including Mr.
Dick Carter, Prof. Bob Park, Dr. Rob Park, Mr. Grant Pearce and Mr. Stuart
Palmer, who are not apart of the working group, for their contributions on New
Zealand design code and practice. Many colleagues in Greece, Japan and USA
X PlANe
provided constructive critical reviews of the book, including Mr. Hisao Oouchi,
Prof. Kyriazis Pitilakis, Dr. Craig Taylor, Dr. Tadahiko Yagyu, Dr. Shuji
Yamamoto, and Dr. Hiroshi Yokota. Sincere thanks are due to them for sugges-
tions that improved the quality of the book. The working group would also like to
express their sincere appreciation to the Coastal -Development Institute of
Technology, Japan, for funding the publication, to the PlANC regional sections
and organizations for sponsoring the working group meetings and activities, and
many 'of the member's organizations for partially funding activities associated
with the development of these guidelines.
The working group hopes that the recommended seismic guidelines for port
facilities will make a significant contribution towards mitigating seismic disasters
in port areas around the world.
December 15, 2000
Susumu Iai
Chairman
PlANCIMarComIWG34
Members of PIANCMarComiWorking Group 34
Chairman:
.Susumu Iai, Port and Harbour Research Institute, Japan
Secretaries:
Takahiro Sugano, Port and Harbour Research Institute, Japan
Koji Ichii, Port and Harbour Research Institute, Japan
Primary Authors:
Alberto Bernal, ByA Estudio de Ingenieria, Spain
Rafael Blazquez, Universidad Politecnica de Madrid, Spain
Hans F. Burcharth, Aalborg University, Denmark
Stephen E. Dickenson, Oregon State University, USA
John Ferritto, Consulting Engineer, USA
W.D. Liam Finn, University of British Columbia, Canada/Kagawa
University, Japan
Susumu Iai, Port and Harbour Research Institute, Japan
Koji Ichii, Port and Harbour Research Institute, Japan
Nason J. Mcflullough, Oregon State University, USA
Piet W.H. Meeuwissen, Delta Marine Consultants bv, Netherlands (from
May 1998)
Constantine D. Memos, National Technical University of Athens, Greece
MJ.N. Priestley, University of California, San Diego, USA
Francesco Silvestri, Universita della Calabria, Italy
Armando L. Simonelli, Consiglio Nazionale delle Ricerche, Italy
R. Scott Steedman, Whitby Bird & Partners Ltd., UK
Takahiro Sugano, Port and Harbour Research Institute, Japan
Contributing Working Group Members:
Steve J. Bowring, Delta Marine Consultants bv, Netherlands (through May
1998)
Valery M. Buslov, Han-Padron Associates, USA
Brad P. Erickson, TranSystems Co., USA
.x11 PlANe
D. Mandja, Ministere de I'Equipement et de I' Amenagement du
TerritoireIDIMA, Algeria
Werner Pall oks, Bundesanstalt fur Wasserbau, Germany
Filippo Vinale, Universita degli Studi di Napoli Federico II, Italy
Mentors:
Thorndike Saville, Jr., USA(through Oct. 1998)
Thomas H. Wakeman, Port Authority of New York & New Jersey, USA
(fromOct. 1998)
Technical Editors:
Elizabeth Hausler, University of California, Berkeley, USA
John Stebila, Berkeley, USA
Graphics:
Yoko Sato, Port and Harbour Research Institute, Japan
List of Tables and Figures: Main Text
Tables
Table 3.1. Acceptable level of damage in performance-based design.
Table 3.2. Performance grades S, A, Band C.
Table 3.3. Performance grade based on the importance category of port
structures.
Table 4.1. Proposed damage criteria for gravity quay walls.
Table 4.2. Proposed damage criteria for sheet pile quay walls (when an anchor is
more difficult to restore than a wall).
Table 4.3. Proposed damage criteria for pile-supported wharves.
Table 4.4. Proposed damage criteria for cellular quay walls.
Table 4.5. Proposed damage criteria for cranes.
Table 4.6. Tolerance for ordinary maintenance of container cranes.
Table 5.1. Types of analysis related with performance grades.
Table 5.2. Methods for site response analysis and liquefaction potential
assessment.
Table 5.3. Analysis methods for port structures.
Table 5.4. List of analysis methods and references.
Table 5.5. Major input parameters for analysis.
Table 5.6. Analysis output.
Table 5.7. Outputs from dynamic analysis.
Figures
Fig. 1.1.
Fig. 2.1.
Fig. 2.2.
Fig. 2.3.
Fig. 2.4.
Fig. 2.5.
Fig. 2.6.
Worldwide zoned average earthquake hazard (modified from Bea,
1997; GSHAP, 1999).
Schematic figure of propagation of seismic waves.
Peak horizontal bedrock accelerations over various return periods.
Mechanism of liquefaction.
Evidence of sand boils due to liquefaction at Hakodate Port, Japan,
during the Hokkaido-Nansei-oki earthquake of 1993.
Damage to a breakwater at Okushiri Port, Japan, during the Hokkaido-
Nansei-oki earthquake of 1993.
Typical port structures.
Fig. 3.1.
Fig. 3.2.
Fig. 3.3.
Fig. 4.1.
XIV PlANe
Fig. 2.7. Damage to a caisson quay wall at Kobe Port, Japan, during the Great
Hanshin earthquake of 1995. .
Fig. 2.8. Cross section of a caisson quay wall at Kobe Port.
Fig. 2.9. Damage to a sheet pile quay wall at Ohama Wharf, Akita Port, Japan,
during the Nihonkai-chubu earthquake of 1983.
Fig. 2.10. Cross section of a sheet pile quay wall at Ohama Wharf, Akita Port.
Fig. 2.11. Damage to a sheet pile quay wall at Shimohama Wharf, Akita Port,
Japan, during the Nihonkai-chubu earthquake of 1983.
Fig. 2.12. Cross section of a sheet pile quay wall at Shimohama Wharf, Akita
Port.
Fig. 2.13. Cross section of a pile-supported wharf at Takahama Wharf, Kobe
Port, Japan, and damage during the Great Hanshin earthquake of 1995.
Fig. 2.14. Damage to piles at Takahama Wharf, Kobe Port (after extraction for
inspection.
Fig. 2.15. Damage to a crane at Kobe Port, Japan, during the Great Hanshin
. earthquake of 1995.
(a) Overview.
(b) Close-up.
Fig. 2.16. Damage to a composite breakwater at Kobe Port, Japan, during the
Great Hanshin earthquake of 1995.
Fig. 2.17. Damage to a rubble mound breakwater at Patras Port, Greece. after a
series of earthquakes in 1984.
(a) Typical cross section.
(b) Cross section before and after the failure.
Flowchart for seismic performance evaluation.
Schematic figure of performance grades S, A, Band C.
Examples of seismic performance evaluation.
Deformation/failure modes of gravity quay wall.
(a) On firm foundation.
(b) On loose sandy foundation.
Fig. 4.2. Parameters for specifying damage criteria for gravity quay wall.
Fig. 4.3. Deformation/failure modes of sheet pile quay wall.
(a) Deformation/failure at anchor.
(b) Failure at sheet pile wall/tie-rod.
(c) Failure at embedment.
Fig. 4.4. Parameters for specifying damage criteria for sheet pile quay wall.
(a) With respect to displacements.
(b) With respect to stresses.
Fig. 4.5. Preferred sequence for yield of sheet pile quay wall.
Fig. 4.6. Deformation/failure modes of pile-supported wharf.
(a) Deformation due to inertia force at deck.
(b) Deformation due to horizontal force from retaining wall.
(c) Deformation due to lateral displacement of loose subsoil.
Fig. 4.8.
"-
Fig. 4.9.
Fig. 4.14.
Fig. 4.15.
Fig. 4.16.
LlSl Ol l a O l ( ; ~ dllU 1 15UI\o , , ~ . .
Fig. 4.7. Parameters for specifying 'damage criteria for pile-supported wharf.
Preferred sequence for yielding of pile-supported wharf.
Deformation/failure modes of cellular quay wall.
(a) On finn foundation.
Fig. 4.10. Deformation/failure modes of cellular quay wall involving cross
deformation.
Fig. 4.11. Parameters for specifying damage criteria for cellular quay wall.
(bjWith respect to stresses.
Fig. 4.12. Schematic figure of gantry crane..
Fig. 4.13. Deformation modes of gantry crane.
(a) Widening of span between the legs.
(b) Narrowing span between the legs due to rocking motion.
ec) Tilting of crane due to differential settlement of foundation.
ed) Overturning of one-hinged leg crane due to rocking/sliding.
Parameters for specifying damage criteria for crane.
Parameters for evaluation overturning of crane.
"-
Deformation/failure modes of breakwater.
(a) Caisson resting on sea bed.
(b) 'Vertically composite' caisson breakwater.
(c) 'Horizontally composite' caisson breakwater.
(d) Rubble mound breakwater.
Fig. 4.17. Examples of berthing behind breakwater.
Fig. 4.7. Parameters for specifying damage criteria for pile-supported wharf.
Fig. 4.8. Preferred sequence for yielding of pile-supported wharf.
Fig. 4.9. Deformation/failure modes of cellular quay wall.
Fig. 4.10. Deformation/failure modes of cellular quay wall involving cross
deformation.
Fig. 4.11. Parameters for specifying damage criteria for cellular quay wall.
- Fig. 4.12. Schematic figure of gantry crane.
(c) Tilting of crane due to differential settlement of foundation.
(d) Overturning of one-hinged leg crane due to rocking/sliding.
Fig. 4.14. Parameters for specifying damage criteria for crane.
Fig. 4.15. Parameters for evaluation overturning of crane.
Fig. 4.16. Deformation/failure modes of breakwater.
Fig. 4.17. Examples of berthing behind breakwater.
MAIN TEXT
CHAPTER I
Introduction
The occurrence of a large earthquake near a major city may be a rare event. but
its societal and economic impact can be so devastating that it is a matter of
national interest. The earthquake disasters in Los Angeles, USA, in 1994 (61
fatalities and -W billion US dollars in losses); Kobe, Japan, in 1995 (over 6.400
fatalities and 100 billion US dollars in losses); Kocaeli, Turkey, in 1999 (over
15,000 fatalities and 20 billion US dollars in losses); Athens, Greece, in 1999 (143
fatalities and 2 billion US dollars in losses); and Taiwan in 1999 (over 2,300 fatal-
ities and 9 billion US dollars in losses) are recent examples. Although seismicity
varies regionally as reflected in Fig. 1.1, earthquake disasters have repeatedly
occurred not only in the seismically active regions in the world but also in areas
within low seismicity regions, such as in Zones 1or 2 in the figure. Mitigating the
outcome of earthquake disasters is a matter of worldwide interest.
In order to mitigate hazards and losses due to earthquakes, seismic design
methodologies have been developed and implemented in design practice in many
regions since the early twentieth century, often in the form of codes and
standards. Most of these methodologies are based on a force-balance approach, in
which structures are designed to resist a prescribed level of seismic force
specified as a fraction of gravity. These methodologies have contributed to the
acceptable seismic performance of port structures, particularly when the earth-
quake motions are more or less within the prescribed design level. Earthquake
disasters, however, have continued to occur. These disasters are caused either
by strong earthquake motions, often in the near field of seismic source areas, or
by moderate earthquake motions in the regions where the damage due to ground
failures has not been anticipated or considered in the seismic design.
The seismic design guidelines for port structures presented in this book address
the limitations inherent if) conventional design, and establish the framework for a
new design approach. In particular, the guidelines are intended to be:
- performance-based, allowing a certain degree of damage depending on
the specific functions and response characteristics of a port structure and
probability of earthquake occurrence in the region;
- user-friendly, offering design engineers a choice of analysis methods, which
range from simple to sophisticated, for evaluating the seismic performance of
structures; and -
4 Y1ANC
EARTHQUAKE INTENSITY
ZONEO=O.OO-O.05g
ZONE1=O.OS-O.15g
ZONE2=O.1S-0.25g
ZONE3=O.2S-0.35g
ZONE4=O.3S-0.45g
ZONE5=O.4S-0.55g
':,
SEISMOTECTONIC TYPE
A=Shaliow crustal fault zones
B=Deep subduction zones
C=Mixed shallow crustal fault
and deep subduction zones
D=lntraplate zones
1
3
NOTE: Values of acceleration corresponding to a return period of475 years.
Some areas of low average seismic hazard have historically experienced major destructive
earthquakes.
Fig. 1.1. Worldwide zoned average eanhquake hazard (modified from Bea. 1997;
GSHAP, 1999).
general enough to be useful throughout the world, where the required functions
of port structures, economic and social environment, and seismic activities may
differ from region to region.
The expected users of the guidelines are design engineers, port authorities, and
specialists in earthquake engineering. The applicability of the guidelines will
reflect regional standards of practice. If a region has no seismic codes or standards
for designing port structures, the guidelines may be used as a basis to .developea
new seismic design methodology, or codes applicable to that particular region. If
a region has already developed seismic codes, standards, or established design
practice, then the guidelines may be used to supplement these design and analy-
sis procedures (see Technical Commentary I for existing codes and guidelines). It
is not the intent of the authors to claim that these guidelines should be used instead
of the existing codes or standards or established design practice in the region of
Introduction 5
interest. It is anticipated, however. that the guidelines will, with continual modifi-
cation and upgrading, be recognized as a new and useful basis for mitigating
seismic disasters in port areas. It is hoped that the guidelines may eventually
be accepted worldwide as recommended seismic design provisions.
Earthquake engineering demands background knowledge in several disci-
plines. Although this background knowledge is not a pre-requisite to understand-
ing the guidelines, readers may find it useful to have reference textbooks readily
available. Pertinent examples include Kramer (1996) on geotechnical earthquake
engineering and Tsinker (1997) on design practice for port structures.
This Main Text provides an overview of the seismic design guidelines. More
.details in the particular aspects of the seismic design guidelines can be found in
the following Technical Commentaries (TC):
TCI: Existing Codes and Guidelines
TC2: Case Histories
TC3: Earthquake Motion
TC4: Geotechnical Characterisation
TC5: Structural Design Aspects of Pile-Deck Systems
TC6: Remediation of Liquefiable Soils
TC7: Analysis Methods
TC8: Examples of Seismic Performance Evaluations
CHAPTER 2
Earthquakes and Port Structures
This chapter addresses issues fundamental to understanding seismic effects on
port structures. As illustrated in Fig. 2.1, seismic waves are generated along a
crustal fault and they propagate through upper crustal rock, travelling to the sur-
face of the bedrock at a site of interest. The ground motions then propagate
through the local soil deposits, reaching the ground surface and impacting struc-
tures. Depending on the intensity of shaking and soil conditions, liquefaction of
near-surface soils and associated ground failures may occur and could signifi-
cantly affect the port structures. If an offshore fault motion .involves vertical
tectonic displacement of the seabed, tsunamis may be generated. The engineering
aspects of these phenomena are collectively important in the evaluation of the
seismic effects on port structures.
2.1 EARTHQUAKE MOTION
(l) Bedrock Motion
The bedrock motions used for seismic analysis and design at a particular site are
characterized through seismic hazard analysis. If a specific earthquake scenario is
assumed in the seismic hazard analysis, the bedrock motion is defined detennin-
istically based on the earthquake source parameters and wave propagation effects
along the source-to-site path. Most often, however, the bedrock motion is defined
probabilistically through the seismic hazard analysis, taking into account uncer-
tainties in frequency of occurrence and location of earthquakes.
One of the key parameters in engineering design practice is the intensity of
bedrock motion defined in terms of peak ground acceleration (PGA), or in some
cases peak ground velocity (PGV). This parameter is used either by itself or to
scale relevant ground motion characteristics, including response spectra and time
histories. In the probabilistic seismic hazard analysis, the level of bedrock motion
is defined as a function of a return period, or a probability of exceedance over a
prescribed exposure time. An example is shown- in Fig. 2.2, in which PGAs at
bedrock are obtained based on geologic, tectonic and historical seismic activity
8 PlANe
}SOil Layers
/. Bedrock
Fig. 2.1. Schematic figure
of propagation of
seismic waves.
Fig. 2.2. Peak horizontal bedrock
accelerations over van-
OllS return periods.
1000
500
Return Period (years)
SanFrancisCO
/
/.
.::t.
cg
0.0
o 100
Oi 1.0-r----------------,
-
c
o
;;
cg
0.8
Q)
o
o
<t
_ 0.6
cg
-
c:
o
N
<5 0.4
:I:
"0
c
::J
e0.2
C)
data available for the region in consideration. It should be foloted that the results of
the seismic hazard analysis, such as shown in Fig. 2.2. can depend on the degree.
of knowledge about the regional tectonic environment, quality of the earthquake
data base, and on the probabilistic methods employed. For these reasons, seismic.
hazard studies are commonly updated .baseq on the most recent research. The
bedrock motion for a prescribed return period is often specified in codes and stan-
dards for a region. See Technical Commentary 3 for more details on bedrock
motion and seismic hazard analysis.
(2) Local Site Effects
The soil deposits at a particular site may significantly modify the bedrock ground
motion, changing the amplitude. frequency content, and duration. This is due to
the dynamic response characteristics of the soils, and it has been termed 'local site
Earthquakes and Port Structures
effects: Local site effects depend on the material properties of the subsoil;
stratigraphy, as well as the intensity and frequency characteristics of the bedro-
motion.
As strong ground motion propagates upwards, towards the ground surface. t
reduction in the strength and stiffness of soil deposits tends to amplify the grout
motions. Depending on their depth and properties, soft soil layers can strong
amplify particular frequencies of motion. For very soft soils, the shaking moti-
may be attenuated and large strains may develop where the imposed cyclic she
stresses approach the strength of the deposits. Care must be taken to ensure t
site response analysis is appropriate to the soil strain levels.
In engineering practice. local site effects are evaluated either by usi:
prescribed site amplification factors based on statistical analysis of existing de
or a site-specific response analysis (Kramer, 1 9 9 6 ~ ISSMGE-TC4, 1999). The s:
amplification factors are often specified in codes and standards, and used to set
the bedrock PGA or PGV to obtain the corresponding values at the groin
surface. or used to scale bedrock response spectra to define the ground surfa
response spectra. Site response analysis of horizontally layered subsoil
generally accomplished using a one-dimensional model to obtain time histories
ground surface motion. Non-linear behaviour of soil is often idealized throu,
the equivalent linear model, in which strain dependent material paramete
such as shear modulus and damping ratio are defined to idealize non-lineari
and energy dissipation imposed by design ground motions.. See Technic
Commentaries 3 and 4 for more details on site response analysis and geotechnic
characterization.
2.2 LIQUEFACTION
As saturated soil deposits are sheared rapidl y back and forth by the shakii
motion, the water pressure in the pores of the soil starts to rise. In loose saturat
cohesionless soils, the pore water pressure can rise rapidly and may reach suer
level that the particles briefly float apart and the strength and stiffness of the s(
is temporarily lost altogether. This is a condition called soil liquefaction, and it
shown diagrammatically in Fig. 2.3. The strength of soil comes about as the res'
of friction and interlocking between the soil particles. At any depth in t
ground, before the earthquake, the weight of the soil and other loads above
carried in part by forces between the soil particles and in part by the pc
water. When loose soil is shaken, ittries to densify or compact. The presence
the water, which has to drain away to allow the compaction, prevents tf
from happening immediately. As a consequence, more and more of the weig
above is transferred to the pore water and the forces between the soil particl
reduce. Ultimately, the pore water pressures may reach such a level tl;
10 PlANe
After

Contacts between the
particles are
recovered. Settlement
may occur in relation to
the volume of water
drained from the lique-
fied deposit.
During
Liquefaction
Pore water pressure
increases and the
contacts between
the sand particles
are lost.
Before

. . Water
. Sand
'. - Particle
Sand particles
maintain
with each other.
Fig. 2.3. Mechanism of liquefaction.
they cause water spouts to break through the overlying layers and the whole
weight of the overlying material is transferred to the pore water. In this condition,
the liquefied soil behaves as a heavy fluid, and large ground movements can
occur. This liquefaction condition will continue until the high pore water
pressures can drain again. and the contact between the soil particles is restored.
Some layers in the ground will densify as a result of this process, and ground
settlements will be observed. Other layers will remain in a very loose condition,
and will be prone to liquefy again in future earthquakes.
The potential for earthquake-induced liquefaction is related to the resistance
of the soil to the generation of excess pore pressures, and to the magnitude
and duration of the cyclic shear stresses/strains that the soil is subjected to.
The cyclic resistance of cohesionless soil is primarily a function of its density,
permeability, and the effective confining stresses acting on the material. Given
the influence of these factors on the results of in-situ tests such as the Standard
Penetration Test (SPT), Cone Penetration Test (CPT), and shear wave velocity
(V
s
), these geotechnical investigations have been used to characterize the

cyclic resistance of the soil. Laboratory testing methods involving the
undrained cyclic loading of high quality specimens are also used in practice to
ascertain the liquefaction resistance of the soil. See Technical Commentary 4 for
more details.
C.UllllfUut\.\..,) <UIU 1 VII <JUU\..IUl\,,:) 11
Fig. 2.4. Evidence of sand boils
due to liquefaction at
Hakodate Port, Japan.
tluring the Hokkaido-
Nansei-oki earthquake
of 1993.
In order to adequately assess the liquefaction susceptibility of a soil, both the
cyclic resistance of the material and the seismic actions on the soil by
design-level earthquake motions must be determined. The seismic actions are
represented by the intensity and duration of the earthquake motions. This
information is obtained from the site response analysis previously discussed. In
the most sophisticated analysis of soil liquefaction, its effects are taken into
account through effective stress formulation of the problem, incorporating pore
pressure generation and dissipation models into a finite element or finite differ-
en.ce computational scheme.
Dynamic pore pressures due to the earthquake are important since they lead to
the development of hydraulic gradients within the mass of soil. As a result, the
pore water begins to migrate and moves quickly to free-draining surfaces. This
aspect of liquefaction is responsible for the evidence of liquefaction' in level
ground, such as sand boils (Fig. 2.4), water springs and vent fractures. Other phe-
nomena associated to liquefaction are lateral spreading and flow failure of the
soil, which are manifestations of the existence of a liquefied sand layer at some
depth below a crest or embankment (e.g. Kramer, 1996).
12 PlANe
2.3 TSUNAMIS
-
Tsunamis are long period sea waves that are generated by vertical seafloor move-
ments. They are often associated with seismic fault ruptures, but occasionally with
submarine landslides. Although wave amplitudes may be small in the open ocean,
the wave height increases as the tsunamis approach shallower depths, occasion-
ally reaching tens of meters at the coastline. The wave height of tsunamis is also
amplified toward the end of V-shaped bays. The predominant wave period of
tsunamis ranges from five to ten minutes, when produced by earthquakes at close
proximity. Tsunamis can easily propagate long distances, such as across the
Pacific Ocean. In this case, the predominant wave period typically ranges from
forty minutes to two hours. Arrival time ranges from within five minutes for
locally generated tsunamis to one day for distant tsunamis travelling across the
Pacific Ocean. Destructive forces by tsunamis can be tremendous. An example of
damage to breakwaters is shown in Fig. 2.5, in which tsunamis due to the
Hokkaido-Nansei-oki earthquake of 1993 arrived at Okushiri Island, Japan,
within five minutes and claimed 200 lives.
Engineering approaches to mitigate tsunami-related disasters are currently
based on either issuing tsunami warnings or implementing engineering measures
at bay mouths and coastlines. For warning of distant tsunamis, the Pacific
Tsunami Warning Centre (PTWC) in Hawaii, USA, functions as the Operation
Fig. 2.5. Damage to a breakwater at Okushiri Pan. Japan. during the Hokkaido-Nansei-oki
earthquake of 1993.
. ,
Earthquakes and Port Structures 13
Centre for the Tsunami Warninz Svstern in the Pacific (TWSP). The time from
.... .
earthquake occurrence to the issuance of a warning ranges from 34 to 92 minutes.
55 minutes on average (Blackford. 1998).
For warning of tsunamis due to earthquakes in close proximity, regional
tsunami warning centres issue the warnings to local govemmentsand citizens. For
example. a highly advanced tsunami warning system has been established in
Japan. where Japan Meteorological Agency (JMA) acts as the regional tsunami
warning centre. Immediately after an earthquake. JNIA computes the hypocentre
and maznitude based on data from the seismic motion rnonitorinz network
... ...
deployed throughout Japan. and issues a tsunami warning within three minutes
after an earthquake occurrence. The warning is issued over 66 regions along the
entire Japanese coastline. Tsunami height and arrival time are the contents of the
warning. The warning is issued based on a digital catalogue compilation of the
results of computer simulations of hypothetical tsunamis assuming 100,000 fault
plane models around Japan. The height of the tsunami at the shore (H
t
) is obtained
from a height 'at an offshoresite (H,) with water depth of n; through Green's law:
H, == (ho/h
t
) l i4 Ho, with (Sekita et al., 1999). The warning is typically
issued for earthquakes in the sea with a magnitude larger than 6.5 and a hypo-
centre shallower than 60 km, and is aired to the citizens immediately through the
national TV network (NHK).
Difficulty lies in warning of tsunamis due to submarine landslides and
'tsunami earthquakes,' .the latter of which are known to be associated with a slow
seismic fault movement. Both are distinctive in that large displacement of the
seabed is induced without strong earthquake shaking. The Meiji-Sanriku, Japan,
earthquake of 1886 (magnitude 7.2) is a typical example, which claimed 22,000
lives without significant ground motion. The Nicaragua earthquake of 1992 is the
first case of a tsunami earthquake that was instrumentally monitored and deter-
mined as a tsunami earthquake.
Implementing engineering measures for mitigating tsunami disasters has
also been undertaken. For example, a composite breakwater over a length of 2 krn
at a bay mouth with a water depth of 63 m has been under construction in
Karnaishi, Japan, since 1978. This breakwater is designed to reduce the height of
the scenario tsunami based on case histories to 2.9 m at the end of the bay. In addi-
tion to the breakwater, a 4-m hizh tidal wall has also been constructed
...
along the coastline at the bay to significantly reduce the potential for a tsunami
disaster.
2.4 PORT STRUCTURES
From an engineering point of view, port structures are soil-structure systems that
consist of various combinations of structural and foundation types. Typical port
structures are shown in Fig. 2.6. This figure indicates that some port structures are
]4 PlANe
- Gnvity. soil-retaining structure.
- Sheet pile. filJ-soil foundation.
-Pile structure withocwithoutbatter piles.
-Pile foundation.
- Structureon cohmms withauxiliatyst.nlC
tures for borizontaI faceeabsorption.Some
times pII'tIylOI1-retaining.
-Column foundation.
Cellular
Pile-supported pier
- Pilestructure. oltenpartly soil-retaining
andwitha\lXlliary structuresfor anchoring.
- Pilefoundation.
Sheet PIle
- 5OJ1-retaining sheet pilesll'\1CtUre with
auxiliary SlrUClurel foc andloring.
- Sheet pile, fiJHoll foundation.
Sheet pile with latform
-Soil-retaining sheet pilestructure with
horiUllltaJ pile-supported slab.
- Sheet pile, pile. fiD-soll foundation.
-Monolithic. gravity,soi1-retaining structure.
-Foundation on rubble and soilor rock.
- Block work,gRvity, soil-retaining structure.
- Foundation on rubbleandSOll or rock.
-Monolithic. IJ'lvity, soil-retaining structure.
-FoundatiOll on rubbleandsoilor rock
- MonWitbic, aravity.soil-retailling structure.
- Folllldatioll onnabbIe and soilor rock.
. .....
Quay Walls I Piers
Caisson structure
_ _ - -Pile structure.
-Sot! sQjI -Pile foundation.
-Mound without superstructure.
-Mound on soil foundation.
Underlaver (Rock or concrete blockSl
- Mound withor withoutmonolithic crownwall
super structure.
- Mound on soil foundation.
Reshaped rubble mound breakwater (Berm
breakwater)
Reshaoed pr
WL
Multi laver rubble mound breakwater
with superstructure
Wave waU suoerstructure
(Mass orreinfo cedconcretel
Armour laver (Rock or Armour laver
concrete bIocksl concrete
Piled breakwater

.,
-- ;' {;;';
..
'. '.
.'
,,\ .......,,-
... ,- ,'I
Block workgravitystructure.
- Foundation onbeddinglayer.or rubble
mound. and soil or rock.
"Horilontally composite" caisson breakwater
Concrete
Breakwaters
caisson breakwater

prote . n .;..: :;'.fig S: fiU
- Monolithic gRvity structure.
- Foundation on rubblebedding layer and
soil or rock.
"Vertically composite" caisson breakwater
:z!t:.
- Monolithic gravitystructure.
- Foundation onhighrubblcmound.
-Monolithic gravitystructure withmoundof
blocks onone side.
- Foundation onbeddinglayerandsoil or
rock.
"Btock work" breakwater
Insitu cast
reinforced concrete
-
- Pile structure,
- Pilefoundation.
///. Z
- Monolithic gRvity structure.
- Foundation onrubbleandsoil
or rock. Sometimes pile
foundation.
Mooringand bfeasting dOlphins
Dolphins
Fig. 2.6. Typical port structures.
'mixed' and can not be fully characterized by a single structural or foundation
type. or stability mechanism. It is also difficult to produce an alternative charac-
terization based on the vulnerability of the structure to damaze bv strom! zround
. '- . -'-
motions. This is due to the relative importance of the soil-fill conditions on the
seismic performance of the structure. In fact. one of the major issues in this book
is a discussion on how to evaluate the seismic performance of a port structure
based on the specific structural and geotechnical conditions. Included in the
definition of port structures in this book are quay walls with or without cranes and
breakwaters. In particular, the quay walls discussed include those of gravity and
sheet pile types. pile-supported wharves, and other mixed structures such as
.cellular quay walls. The breakwaters being discussed are vertical face and rubble
mound types.
,., -
.... ) EXAMPLES OF SEISMIC DAMAGE
Typical examples of seismic damage to port structures are as follows.
(l) Damage to gravity quay walls
Damage to a gravity quay wall is shown in Figs. 2.7 and 2.8. This quay wall ~ a s
located in Kobe Port, Japan, and shaken by the Great Hanshin earthquake of 1995.
Fig. 2.7. Damage to a caisson quay wall at Kobe Port, Japan, during the Great Hanshin
earthquake of 1995.
Backfill Soil
"V-33.00--35.00
Unit (m)
Fig. 2.8. Cross section of a caisson quay wall at Kobe Port.
The damage involves large seaward displacement, settlement and tilt. The damage
was caused mainly by deformation in the loosely deposited foundation soil
beneath the caisson wall.
(2) Damage to sheet pile quay walls
Representative damage to a sheet pile quay wall is shown in Figs. 2.9 and 2.10.
This quay wall was located in Akita Port, Japan, and shaken by the 1983
Nihonkai-Chubu earthquake. The damage was mainly due to large pressures
applied to the sheet pile wall by the liquefied backfill sand. The sheet
pile bending moment became excessively large, resulting in the opening of
a crack in the wall. Additional examples of damage to sheet pile quay
walls are shown in Figs. 2.11 and 2.12. This quay wall was located in the
same port and subjected to ground motions of similar intensity. The anchor
was embedded in liquefiable soil: resulting in deformations in the entire
wharf as shown in these figures. These two examples indicate that the mode of
damage can differ significantly according to the geotechnical and structural
conditions of a quay wall.
(3) Damage to pile-supported wharves
Damage to a pile-supported wharf is shown in Fig. 2.13. This quay was located in
Kobe Port, Japan, and was shaken by the Great Hanshin earthquake of 1995. The
damage was caused by the large seaward movement of a dike over a liquefiable
sand layer. The bending moment in the piles became excessively large, resulting
Earthquakes and Port Structures . 17
Fig.-2.9. Damage to a sheet pile quay wall at Ohama Wharf. Akita Port, Japan. during the
Nihonkai-chubu earthquake of 1983.
"'
......--.+1.2

........
II .,.

..
f t "

ii",: i
.. ..
::.!!::
:: It
::e'1:
"8:: II t'
iL,..::
...... "..
. .tl): :
::"9-::
: :Q): :
::==::
Q...

a::
20.Om
gth
Steel Tie Rod
H'gh Tensile Stren
ssmm, Spacing2.0m
-16.0
+2.0 l:-- --- _ .._ _.__..__
L.W.L o.o:-i ":
--+- :- _--::.::..-::::-:= _=-= .._--
Fig. 2.10. Cross section of a sheet pile quay wall at Ohama Wharf. Akita Port.
in formation of plastic hinges. both at the pile-deck connection and within the
embedded portion of the pile. The piles, extracted after the earthquake for inves-
tigation, clearly exhibited signs of damage in the embedded portion as shown in
Fig. 2.14.
19 PlANe
Fig. 2.11. Damage to a sheet pile quay wall at Shimohama Wharf, Akita Port, Japan,
during the Nihonkai-chubu earthquake of 1983.
+2.0
-9.0
Unit (rn)
."
Fig. 2.12. Cross section of a sheet pile quay wall at Shimohama Wharf. Akita Port.
(4) Damage to cranes
Damage to a crane on rails is shown in Fig. 2.15. This crane was located in Kobe
Port, Japan, and shaken by the Great Hanshin earthquake of 1995. The damage
was caused not only by strong shaking, but also by the movement of the caisson
--16.7
- After Earthquake
- - - Before Earthquake.
Local Buckling
(Unit in meters)
at
pile cap
Horizontal
Displacement
1.3-1.7m
\
_
v H.W.L. +1.6
T L.W.L. +0.0
Pleistocene Clay (2)
Pleistocene Cia 1
Pleistocene Gravel (2)
Fig. 2.13. Cross section of a pile-supported wharf at Takahama Wharf, Kobe Port, Japan,
and damage during the Great Hanshin earthquake of 1995.
Fig. 2.14. Damage to piles at Takahama Wharf, Kobe Port (after extraction for
inspection).
20 PlANe
Fig. 2.15. Damage to a crane at Kobe Port, Japan, during the Great Hanshin earthquake
of 1995.
(a) Overview.
(b) Close-up.
quay wall toward the sea. resulting in formation of plastic hinges at the legs of the
crane.
(5) Damage to breakwaters
Damage to a composite breakwater with a vertical face is shown in Fig. 2.16. This
breakwater was located in Kobe Port. Japan. and shaken by the Great Hanshin
earthquake of 1995. The damage was mostly in the form of excessive settlement
.caused by significant deformation of the loose soil deposit below the breakwater.
Darnaze to a rubble mound breakwater is shown in Fiz. 2.17. The site was a
-
-southern extension of the detached breakwater of Patras Port, Greece. This
involved large settlements of the order of 3 to 4 m and was triggered by a series
of moderate earthquakes in 1984, and amplified considerably by the local geo-
technical conditions. The damage was mostly in the form of flattening of the cross
section and of intrusion of the lower mound material into the soft clay.
(6) Common features of damage to port structures
The evidence of damage to port structures previously addressed suggests that:
- most damage to port structures is often associated with significant deformation
of a soft or liquefiable soil deposit; hence, if the potential for liquefaction
Unit (m)
Clay
.- Before Earthquake
--After Earthquake
S7 LW L+Q 00
Clay
.. SZ HW.L +1.70

-0.64
: 8)0:
1000 1000 n -9.20 ', : : ..... 000 00
. I .14:00 ... ...._._.m4 l::. ..:?.. I . IT.
..::.. _.._
......z..J for Replacing Alluvial Clay Layer ....
..:.$
:
Fig. 2.16. Damage to a composite breakwater at Kobe Port. Japan, during the Great
Hanshin earthquake of 1995.
22 PIANC
v:tO.oo
Primaryarmour H rb
. a our
"
(a)
Harbour
:to.oo
V-10.O
-,
Before failure - ...... ....
..-",'" . "
, ....
.. ..
" ..
" ....
". After failure- ..
... _-- ... .,.---
.. __------------
... :::"::._------------"'*
I. 10m .1 (Unit: m)
(Q)
Fig. 2.17. Damage to a rubble mound breakwater at Patras Port, Greece, after a series of
earthquakes in 1984.
(a) Typical cross section.
(b) Cross section before and after the failure.
exists, implementing appropriate remediation measures against liquefaction
may be an effective approach to attaining significantly improved seismic per-
fonnance of port structures;
most failures of port structures, from a practical perspective, result from exces-
sive deformations, not catastrophic collapses; hence, design methods based on
displacements and ultimate stress states are desirable over conventional force-
based design methods for defining the comprehensive seismic performance of
port structures; and
most damage to port structures is the result of soil-structure interaction; hence,
design and analysis procedures should include both geotechnical and structural
conditions of port structures.
Technical Commentary 2 provides more details on case histories of seismic dam-
age to port structures.
CHAPTER 3
Design Philosophy
An evolving design philosophy for port structures in many seismically active
regions reflect the observations that:
- deformations in ground and foundation soils and the corresponding structural
deformation and stress states are key design parameters;
- conventional limit equilibrium-based methods are not well suited to evaluating
these parameters; and
some residual deformation may be acceptable.
The performance-based methodology presented in this book incorporates this
design. philosophy into practice-oriented guidelines, and provides engineers with
new tools for designing ports.
3.1 PERFORMANCE-BASED METHODOLOGY
Performance-based design is an emerging methodology, which was born from the
lessons learned fromearthquakes in the 1990s (SEAOC, 1995; lai and Ichii, 1998;
Steedman, 1998). The goal is to overcome the limitations present in conventional
seismic design. Conventional building code seismic design is based on providing
capacity to resist a design seismic force, but it does not provide information on
the performance of a structure when the limit of the force-balance is exceeded. If
we demand that limit equilibrium not be exceeded in conventional design for the
relatively high intensity ground motions associated with a very rare seismic event,
the construction/retrofitting cost will most likely be too high. If force-balance
design is based on a more frequent seismic event, then it is difficult to estimate
the seismic performance of the structure when subjected to ground motions that
are greater than those used in design.
In performance-based design, appropriate levels of design earthquake motions
must be defined andcorresponding acceptable levels of structural damage must be
clearly identified. Two levels of earthquake motions are typically used as design
reference motions, defined as follows:
Levell (Ll): the level of earthquake motions that are likely to occur during the
life-span of the structure;
24 PlANe
Level 2 (L2): the level of earthquake motions associated with infrequent rare
events, that typically involve very strong ground shaking.
The acceptable level of damage is specified according to the specific needs of
the users/owners of the facilities and may be defined on the basis of the accept-
able level of structural and operational damage given in Table 3.1. The structural
damage category in this table is directly related to the amount of work needed to
restorethe full functional capacity of the structure and is often referred to as direct
loss due to earthquakes. The operational damage category is related to the length
of time and cost associated with the restoration of full or partial serviceability.
Economic losses associated with the loss of serviceability are often referred to as
indirect losses. In addition to the fundamental functions of servicing sea transport,
the functions of port structures may include protection of human life and proper-
ty, functioning as an emergency base for transportation, and as protection from
spilling hazardous materials. If applicable, the effects on these issues should be
considered in defining the acceptable level of damage in addition to those shown
in Table 3.1.
Once the design earthquake levels and acceptable damage levels have been
properly defined, the required performance of a structure may be specified by the
appropriate performance grade S, A, B, or C defined in Table 3.2. In performance-
based design, a structure is designed to meet these performance grades.
The principal steps taken in performance-based design are shown in the flow-
chart in Fig. 3.1:
1) Select a performance grade of S, A, B, or C: This step is typically done by
referring to Tables 3.1 and 3.2 and selecting the damage level consistent with
Table 3.1. Acceptable level of damage in performance-based design.*
Level of damage -, Structural Operational
Degree I : Minor or no damage Little or no loss of
Serviceable serviceability
Degree II: Controlled damage** Short-term loss of
Repairable serviceabilirvv'v
Degree III: Extensive damage in near Long-term or .complete
Near collapse collapse loss of serviceability .
Degree IV: Complete loss of structure Complete loss of
Collapse**** serviceability
* Considerations: Protection of human life and property, functions as an emergency base
for transportation, and protection from spilling hazardous materials, if applicable. should
be considered in defining the damage criteria in addition to those shown in this table.
** With limited inelastic response and/or residual deformation.
*** Structure out of service for short to moderate time for repairs.
**** WlthoutSlgnificam effects on surroundings.
LeveI2(L2)
Design Philosophy 25
Table 3.2. Performance grades S. A. B. and C.
Performance grade Design earthquake
---------------------
Level ltl, 1)
Grade S
Grade A
Grade B
Grade C
Degree I: Serviceable
Degree II: Repairable
Degree II: Repairable
Degree Ill: Near collapse
Degree IV: Collapse
the needs of the users/owners. Another procedure for choosing a performance
grade is to base the grade on the importance of the structure. Degrees of
importance are defined in most seismic codes and standards. This procedure is
presented in Table 3.3. If applicable, a performance grade other than those of
S, A. B. or C may be introduced to meet specific needs of the users/
owners.
2) Define damage criteria: Specify the level of acceptable damage in engineering
parameters such as displacements, limit stress states, or ductility factors. The
details are addressed in Chapter 4, Damage Criteria.
3) Evaluate seismic performance of a structure: Evaluation is typically done by
comparing the response parameters from a seismic analysis of the structure
with the damage criteria. If the results of the analysis do not meet the damage
criteria. the proposed design or existing structure should be modified. Soil
improvement including remediation measures against liquefaction may be
necessary at this stage. Details of liquefaction remediation from the publica-
tion of the Port and Harbour Research Institute, Japan (1997) can be found in
Technical Commentary 6.
More details on design earthquake motion and performance evaluation are
addressed in the following sections.
3.2 REFERENCE LEVELS OF EARTHQUAKE MOTIONS
Levell earthquake motion (LI) is typically defined as motion with a probability
of exceedance of 50% during the life-span of a structure. Level 2 earthquake
motion (L2) is typically defined as motion with a probability of exceedance of
10% during the.life-span. In defining these motions, near field motion from a rare
event on an active seismic fault should also be considered, if the fault is located
nearby. If the life-span of a port structure is 50 years, the return periods for L1 and
L2 are 75 and 475. years, respectively.
In regions of low seismicity, L1 may be relatively small and ofminor engi-
neering significance. In this case, only L2 is used along with an appropriately
specified damage criteria. Here, it is assumed that satisfactory performance for L2
will implicitly ensure required performance under the anticipated L1 motion. It
26 PlANe
Earthquake level:
level 1 (l1)
level 2 (l2)
Performance grade:
S,A,B.C
Acceptable damage:
1 Serviceable
n Repairable
mNear Collapse
IV Collapse
Chapter 3: Design Philosophy
I - - - - - - - - - - - - - - - - - ~ - - - - - - - - - - - - - - I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1__ - _ ~
Analysis type :
1. Simplified analysis
2. Simplified dynamic analysis
. 3. Dynamic analysis
Input :
Earthquake motions
Geotechnical conditions
Proposed design or existing structure
Chapter 4:
Damage Criteria
Chapter 5: Seismic Analysis
------ - - - - - - - - - - - - - - ~
I
I
I
I
I
I
I
I
I
I
I
I
Damage criteria
Output:
Displacements
Stresses
(Liquefaction potential)
I
I
I
___ J
No Modification of
cross section!
soil improvement
Fig. 3.1. Flowchart for seismic performance evaluation.
may be noted that this single level approach is somewhat similar to conventional
design practice: it differs only in that a structure is designed in accordance with a
designated acceptable level of damage.
The dual level approach using both Ll and L2 attempts to: 1) ensure a speci-
fied level of safety and serviceability for L1, and 2) prescribe the level and modes
of seismic damage for L2. This dual level approach is particularly useful in
regions of moderate and high seismicity where meeting the specified damage
criteria for L2 may not be sufficient to ensure the desired degree of safety and
serviceability during LI, or meeting the performance standard for L1 is not
sufficient to ensure the specified performance standard for L2. It should be noted
here that stronger L2 excitations will not necessarily solely dictate the final
design, which may be highly influenced or even dominated by a high performance
standard for Ll.
In the remainder of this main text. the dual level approach will be the focus of
attention. The information needed for the single level approach can be easily
inferred by adopting the issues associated with L2 only.
3.3 PERFORMANCE EVALUATION
As a guide for evaluating performance criteria at a specific port, the relationship
between degree of damage and the design earthquake motion is illustrated in
Fig. 3.2. The curves in this figure form the basis for the performance evaluation
procedure. This figure is based on the specification of performance grades in
Table 3.2. The curves in Fig. 3.2 indicate the upper limits for the acceptable level
of damage over a continuously varying level of earthquake motions, including the
Table 3.3. Performance grade based on the importance category of port structures.
Performance Definition based on seismic effects on structures Suggested importance
grade category of port
structures in Japanese
code
Grade S
Grade A
Grade B
Grade C
<D Critical structures with potential for extensive
loss of human life and property upon seismic
damage
~ Key structures that are required to be service-
able for recovery from earthquake disaster
a> Critical structures that handle hazardous
materials
Critical structures that. if disrupted. devastate
economic and social activities in the earthquake
damage area
Primary structures having less serious effects
for <D through than Grade S structures. or
~ Structures that, if damaged, are difficult to
restore
Ordinary structures other than those of
Grades S, A and C
Small easily restorable structures
Special Class
Special Class or
Class A
Class A or B
Class B or C
28 PlANe
designated LI and L2 motions. Each curve in this figure is defined by two
control points corresponding to the upper limits of the level of damage for Ll and
L2 motions defined in Table 3.2. For example, the curve defining the upper limit
for Grade B should go through a point defining the upper limit for damage degree
I for LI motion, and another defining the upper limit for damage degree III for L2
motion. The shape of the curves may be approximated by line segments through
the controlling points or may be refined by referring to typical results of non-
linear seismic analysis of port structures.
The vertical coordinates of Fig. 3.2 are converted into engineering parameters
such as displacements, stresses or ductility factors specified by the damage crite-
ria discussed in Chapter 4. This conversion allows direct comparison between
required performance and seismic response of a structure. The seismic response
of a structure is evaluated through seismic analysis over LI and L2 motions and
plotted on this figure as 'seismic response curve.' As minimum requirement,
analysis should be performed for LI and L2 earthquake motions. For example, if
the structure being evaluated or designed has the seismic response curve 'a' in
Fig. 3.3, the curve is located below the upper bound curve .defining Grade A.
Thus, this design assures Grade A performance. If an alternative structural con-
figuration yields the seismic performance curve 'b' in Fig. 3.3 and a portion of the
curve exceeds the upper limit for Grade A, then this design assures only Grade B
performance.
IV : Collapse
:g, III : Near collapse
ell
E
ell
c
'0
"iii
~
-I
II : Repairable
L1 12
DesignEarthquake Motions
Fig. 3.2. Schematic figure of performance grades S, A, Band C.
Structure b
IV: Collapse
III
~ III : Near collapse
i
o
"0
l
3
1\ : Repairable
Structure a
L1 L2
DesignEarthquakeMotions
Fig. 3.3. Examples of seismic performance evaluation.
CHAPTER 4
Damaze Criteria
....
. In performance-based design, the acceptable level of damage, i.e. the damage cri-
teria, should be specified in engineering terms such as displacements, limit stress
state, and ductility/strain limit based on the function and seismic response of the
structure. The damage criteria may be established based on Table 3.1 in Chapter
3 by a group of design engineers with assistance and advice from the users and
owners of the structure/facili ty. The guidelines for establishing damage. criteria for
typical quay walls are described in Sections 4.1 through 4.4. The proposed dam-
age criteria discussed in this chapter have been established for the following
specific conditions:
1) The operational damage criteria in Table 3.1 in the previous chapter dictate the
Degree I, whereas the structural damage criteria in the same table allow for the
Degrees II through IV.
2) The quay walls considered are those where the seismic damage poses no
threats to human life, no hazardous material is handled, without cranes on rails,
and the sea space in front of the quay wall is unlimited.
Additional guidelines are offered in Section 4.5 for quay walls with cranes on
rails. Damage criteria for breakwaters are described in Section 4.6.
4.1 GRAVITY QUAY WALLS
(1) Seismic response of gravity quay walls
A gravity quay wall is made of a caisson or other gravity retaining structure
placed on the seabed. Stability against earth pressures from the backfill soil
behind the wall is maintained by the mass of the wall and the friction at the bot-
tom of the wall. For gravity quay walls on finn foundations, typical failure modes
during earthquakes are seaward displacements and tilting as shown in Fig. 4.1 (a).
For a loose backfill or natural loose sandy foundation, the failure modes involve
overall deformation of the foundation beneath the wall, resulting in large seaward
displacement, tilting and settlements (see Fig. 4.1 (bj), A wall with a relatively
small width to height ratio, typically less than about 0.75, will exhibit a predom-
inant tilting failure mode rather than horizontal displacements.
32 PlANe
7
. .,.
(a)
.
.
.
.'"
..,
v
~ ~ .
~ - .
..'
...
(b)
Fig. 4.]. Deformation/failure modes of gravity quay wall.
On the conceptual level of design, several options may be available for
increasing the stability of the gravity quay wall. An obvious measure to reduce
the earth pressures against the wall is to use backfill material with a large angle
of internal friction. Rockfill and similar materials should be preferable to loose
sand for such use. The risk of overall deformation as well as of liquefaction
would be minimized if such practice could be followed. Other possibilities for
increasing the stability of the quay wall include drawing landwards the centre of
gravity of the wall and thereby increasing the stabilizing moment, and imposing
a higher friction coefficient between the base slab and the foundation layer. In
concept, the former may be achieved by allowing for appropriate offsets of the
precast blocks along the vertical face of the structure or by filling the rear cells
of caissons with heavy materials such as metal slag. The latter has indeed been
achieved by using an asphalt or rubber mat beneath the caisson wall. The latter
may alternatively be achieved by providing a foundation slab of concrete with
a special anti-skid shape or by designing for inclined, rather than horizontal
contact surfaces at foundation and higher levels of the block wall.
Horizontal displacement and uniform vertical settlement of a gravity quay wall
may not significantly reduce the residual state of stability, and may be generally
acceptable from a structural point of view. However, tilting of the wall can
significantly reduce the residual stability and result in an unacceptable structural
stability situation. Case histories show an overturning/collapse of concrete block
type walls when tilting occurs. This type of wall needs careful consideration in .
specifying damage criteria regarding the overturning/collapse mode.
(2) Parameters for specifying damage criteria for gravity quay walls
Seismic performance of a gravity quay wall is specified based on such service-
ability considerations as safe berthing, safe operation of wheeled vehicles and
cargo handling. flooding, and level of structural damage in the form of displace-
ments and tilting (including relative displacements between the blocks of a
concrete block type quay wall). Parameters that may be used for specifying
Damage Criteria 33
damage criteria include displacements. settlements. tilting. and differential
displacements along the face line of a wall, and the deformations at apron includ-
ing settlement. differential settlement at and behind apron. and tilting (Fig. 4.2).
Damage criteria should be established by choosing and specifying appropriate
parameters from those mentioned above. More specifics are given in the next
subsection.
(3) Damage criteria for gravity, quay walls
Provided the conditions mentionedat the beginning of this chapter are applicable.
the damage criteria may be established by referring to Table 4.1. The criteria
shown in Table 4.1 are minimum requirements. Thus, in evaluating seismic
performance by referring to damage criteria based on a number of different para-
meters, the highest damage degree should be the final result of the evaluation.
4.2 SHEET PILE QUAY WALLS
( I) Seismic response of sheet pile quay walls
A sheet pile quay wall is composed of interlocking sheet piles, tie-rods, and
anchors. The wall is supported at the upper part by anchors and the lower part by
embedment in competent soil. Typical failure modes during earthquakes depend
on structural and geotechnical conditions as shown in Fig. 4.3.
Structural damage to a sheet pile quay wall is basically governed by structural
stress/strain states rather than displacements. It is important to determine the
preferred sequence and degrees of ultimate states that may occur in the composite
sheet pile quay wall system.
Tilting
Differential Settlement
behind Caisson
Horizontal Displacement
Settlement
Differential Displacement
v
Settlement of Apron
at Apron
.
.
.
..
..
..
..
Fig. 4.2. Parameters for specifying damage criteria for gravity quay wall.
34 PlANe
Table 4.1. Proposed damage criteria for gravity quay walls.
Level of damage Degree I Degree II Degree III Degree IV
Gravity Normalized residual Less than 1.5-5% 5-10% Larger than
wall horizontal 1.5%** 10%
displacement (dlH)*
Residual tilting Less than 3-5
0
5-8
0
Larger than
towards the sea 3
0
8
0
Apron Differential Less than N/A*** N/A N/A
settlement on apron 0.03-0.1 m
Differential Less than N/A N/A N/A
settlement between 0.3-0.7 m
apron and non-apron
areas
Residual tilting Less than N/A N/A N/A
towards the sea 2-3
0
* d: residual horizontal displacement at the top of the wall; H: height of gravity wall.
** Alternative criterion is proposed with respect.to differential horizontal displacemt less
than 30 em.
*** Abbreviation for not applicable.
(2) Parameters for specifying damage criteria for sheet pile quay walls
Seismic performance of a sheet pile quay wall is specified based on serviceability,
similar to that for a gravity quay wall, and in terms of structural damage regard-
ing stress states as well as displacements. Parameters for specifying damage cri-
teria are as follows (refer to Fig. 4.4).
Displacements:
- sheet pile wall and apron: refer to the parameters for a gravity quay wall:
- anchor: differential settlement. ground surface cracking at anchor. pull-out
displacement of battered pile anchor.
Stresses:
- sheet pile (above and below mudline):
- tie-rod (including joints);
- anchor.
parameters from those mentioned above.
The preferred sequence to reach ultimate states with increasing level of seismic
load should be appropriately specified for a sheet pile quay wall. If a damaged
anchor is more difficult to restore than a sheet pile wall, the appropriate sequence
may be given as follows (refer to Fig. 4.5).
1) Displacement of anchor (within the damage Degree I).
2) Yield at sheet pile wall (above mudline).
3) Yield at sheet pile wall (below mudline).
Damage Criteria 35
. - - - - - ~ ~ , . , . , . , . . . . ,
~ - - - <
y/////////--
(a)
(b)
r - - - - ~ ~ ~ ~
-
(C)
Fig. 4.3. Defonnationlfailure modes of sheet pile quay wall.
(a) Deformation/failure at anchor.
(b) Failure at sheet pile wall/tie-rod.
(c) Failure at embedment.
4) Yield at anchor.
5) Yield at tie-rod.
If a damaged sheet pile wall is more difficult to restore than an anchor, the yield
at anchor should precede the yield at sheet pile wall.
(3) Damage criteria for sheet pile quay walls
Provided the conditions mentioned at the beginning of this chapter are applicable,
the damage criteria for a sheet pile quay wall may be established by referring to
Table 4.2. The most restrictive conditions among displacements and stresses
should define the damage criteria. Structural damage to the embedded portion of
a sheet pile is generally difficult to restore, and thus necessitates higher seismic
resistance. Brittle fracture of a sheet pile wall, rupture ofa tie-rod, and the
collapse of anchor should be avoided.
36 PlANe
Stressin Tie-rod
(includingjoints)
Stressin Sheet Pile
(aboveand below
mudline)
Differential Settlement at Anchor
Ground SurfaceCraCking at Anchor
Pull-out Displacementof
Banered PileAnchor
Settlement of Apron
Differential Settlement at Apron
Tilting
HorizontalDisplacement
Settlement
~
(a) (b)
Fig. 4.4. Parameters for specifying damage criteria for sheet pile quay wall.
G)Displacement of Anchor
(Extent of Damage Degree I )
~ Y i e l d at Sheet Pile Wall ~ .
(above mudline) @YJeId at Anchor
@Yield at Tie-rod
Fig. 4.5. Preferred sequence for yield of
sheet pile quay wall.
4.3 PILE-SUPPORTED WHARVES
(1) Seismic response of pile-supported wharves
A pile-supported wharf is composed of a deck supported by a substructure con-
sisting of piles and a dike/slope. The unsupported pile length above the dike/slope
surface is variable. When rockfill suitable for construction of the dike is uneco-
nomical, a gravity or sheet pile retaining structure is also constructed 10 replace a
portion of the dike. The seismic response of pile-supported wharves is influenced
10 a great degree by complex soil-structure interaction during ground shaking.
Typical failure modes during earthquakes depend on the magnitude of the inertia
force relative to the ground displacement (Fig. 4.6). In particular, the failure
Table 4.2. Proposed damage criteria for sheet pile quay walls.
(when an anchor is more difficult to restore than a wall).
Residual Sheet Normalized residual Less than N/A N/A N/A
/. ! displacements pile wall horizontal displacement 1.5%**
(d/H) *
Residual tilting Less than 3
0
N/A N/A N/A
towards the sea
Apron Differential Less than N/A N/A N/A
settlement on apron 0.03-0.1 m
Differential settlement Less than N/A N/A N/A
between apron and 0.3-0.7 m
non-apron areas
Residual tilting , Less than N/A N/A N/A
towards the sea 2-3
0
Peak response Sheet Above mudline Elastic Plastic (less than Plastic (less than Plastic (beyond the
astresses/ pile wall the ductility the ductility factor/ ductility fuctor/strain
strains
factor/strain limit strain limit limit above mudline)
above mudline) above mudline)
Below rnudline Elastic Elastic Plastic (less than Plastic (beyond the
the ductility factor/ ductility factor/strain
tstrain limit limit below mudline)
below mudline)
Tie-rod Elastic Elastic Plastic (less than Plastic (beyond the
0
~
the ductility factor/ ductility factor/strain 3
~
strain limit for tie-rod) limit for ric-rod)
(IQ
~
Anchor Elastic Elastic Plastic (less than Plastic (beyond the
o
::I.
the ductility factor/ ductility factor/strain
,...
~
strain limit for anchor) limit for anchor)
::1.
~
*d: residual horizontal displacement at the top of the wall; H: height of sheer pile wall from mudline.
v)
--.J
**Altemative criterion is proposed with respect to differential horizontal displacement less t h ~ 1 I 1 30 cm.
38 PlANe
(a)
,..: .
.
.
(b)

'Firm Foundation

(c)
Fig. 4.6. Deformation/failure modes of pile-supported wharf.
(a) Deformation due to inertia force at deck.
(b) Deformation due to horizontal force from retaining wall.
(c) Deformation due to lateral displacement of loose subsoil.
associated with the ground displacement shown in Fig. 4.6(c) suggests that appro-
priate design consideration is required of the geotechnical aspects.
Among various structural components in the pile-supported wharves. anention
should be paid to the application of batter piles. Batter piles remain the most effi-
cient structural component for resisting lateral loads due to mooring. berthing and
crane operation. However, the batter pile-deck system results in a much more rigid
frame than one with vertical piles. As a result. large stress concentrations and
shear failures of concrete batter piles have been observed during earthquakes (e.g.
1989 Lorna Prieta earthquake, and 1994 Northridge earthquake in California,
USA). This ofpilefailure demonstrates of batter
piles_ .. __
and appropriate ductility.
When-usln'g-'the'concrete piles in the areas of high it is becoming
more common to design the pile-supported wharves with venical piles only. This
Damage Criteria 39
structural system resists the earthquake-induced lateral loads/displacements by
bending of the piles and associated moment resistance. Moderate structural
damage to vertical piles in bending may be unavoidable during strong earthquake
motions. However. preferable to
.. because the former is substantiallY
easier to repair (Buslov et al., 1996).
Innovative structural design and detailing of pile-deck connections has allowed
for the continued use' of batter piles for pile-supported wharves. The design
concepts incorporate a repairable structural 'fuse' at the pile cap. The fuse serves
as a weak link or load-limiting device that is robust enough to withstand lateral
. loads during routine port operation, but yields during strong earthquake motions.
thereby precluding damage to the wharf deck itself. The load-limiting device may
be designed as a component that relies on sliding friction (Zmuda et al., 1995), or
as a structural 'fuse-link' that yields in shear and can be easily replaced (Johnson
et al., 1998).
In the areas of high seismicity such as in Japan, it is common to use
steel piles for pile-supported wharves.
Structural to a pile-supported wharf is governed by stress/strain
states rather than displacements. It is important to determine the preferred
sequence and degrees of ultimate states to occur in the composite pile-supported
wharf system.
(2) Parameters for specifying damage criteria for pile-supported wharves
The structure of a. pile-supported wharf consists of a deck/pile system and a
dike/slope/retaining wall. The seismic performance of the dike/slope/retaining
wall may be specified based on the same criteria as for a gravity or sheet pile wall.
The effects of ground displacements, including the movement of the retaining
wall and the dike/slope below the deck, should be carefully evaluated for the per-
formance of the deck/pile system.
The seismic performance of a deck/pile system is specified based on service-
ability and structural damage. Parameters for specifying damage criteria for deck
and piles are as follows (refer to Fig. 4.7).
Displacements:
- deck and piles: settlement, tilting, differential displacements;
- apron (deck and loading/unloading area behind the deck): differential settle-
ment between deck and retaining wall, tilting, fall/fracture of bridge.
Stresses:
- piles (pile top and below dike/slope surface or mudline):
- deck (deck body, pile cap);
- bridge.
parameters from these parameters.
40 PlANe
Potential for deep seatedslope
movementand damageto piles
Items to be evaluated
for Retaining Wall: .
Referto those for
GravitylSheet PileQuaywalls
(between Deck andRetainingWall)
FallIColla e of Bridge
Horizontal Displacement
Settlement
Tilting
Tilting of Deck
+
Weaklayer
(a)
Stress in Bridge
Items to be evaluated
for Retaining Wall:
Refer to those for
Gravity/Sheet PileQuaywalls
r' co - - - - - - - - -
...,.
.
.
.
.
.
.
.
.
Pile Stress at
Embedded Part
(11)
Fig. 4.7. Parameters for specifying damage criteria for pile-supported wharf.
With respect to ease of restoration after an earthquake, the preferred sequence
to reach ultimate states with increasing levels of seismic load may be specified for
a pile-supported wharf as follows (refer to Fig. 4.8).
1) Pile cap (portion of pile embedded for a pile-deck connection).
2) Pile top (just below the pile cap - concrete beam connection).
Damage Criteria 41
at Deck
Fig. 4.8. Preferred sequence for yield-
ing of pile-supported wharf.
ing at
t
ap 1"_
....... --..
.

........

.

.
.
...
. .

. .
.
op
.
.
.
V
V
...

dded
n
Yield
PileT
g)Yield
Embe
Portio
<DYield
Pile C
3) Deck and pile below dike/slope surface or mudline (within allowable ductility
factor).
In the bridging area between the wharf deck and the retaining wall, structural
details such as a fail-safe device to prevent fall-down or an easily repairable
structure are also important, and, if applicable, a structure to absorb the dis-
placements from the retaining 'wall can be introduced. This could lead to further
development of energy absorption devices for pile-supported wharves.
(3) Damage criteria for pile-supported wharves
The criteria for the retaining wall of a pile-supported wharf may be established
by referring to those for a gravity or sheet pile quay wall. Damage criteria for the
dike/slope under the deck are less established because of the complex soil-struc-
ture interaction between the piles and the dike/slope.
Provided the conditions mentioned at the beginning of this chapter are appli-
cable, criteria for the piles and deck of a pile-supported wharf may be established
by referring to Table 4.3. The most restrictive condition among displacements
and stresses should define the damage criteria.
Structural damage to the embedded portion of a pile is generally difficult
to restore and has the potential to trigger collapse of a pile-supported
wharf. Thus, a more restrictive ductility factor should be used for the design.
Examples of ductility factors/strain limits for use in practice are found in
Ferritto (l997a), Fenitto et al. (1999) and Yokota et al. (1999). More details
can be found in Technical Commentaries 5 and 7. No case histories have
been reported of brittle fracture of steel piles during earthquakes. However,
the case histories of brittle fracture of thick steel columns during the
1995 Kobe earthquake indicate that it is still necessary to study this aspect
in steelpiles.
Table 4.3. Proposed damage criteria for pile-supported wharves.*
~
N
'ij
~
()
Peak
response
Level of damage Degree I
Residual Differential
displacements settlement between
deck and land behind
Residual tiIting
towards the sen
Piles**
Degree II
Less than 0.1-0.3 m
Less than 2-3
0
Essentially clastic
response with minor
or no residual
deformation
Degree III
N/A
N/A
Controlled limited
inelastic ductile
response and residual
deformation intending
to'keep the structure
repairable
Degree IV
N/A
N/A
Ductile response near
collapse (double
plastic hinges may
occur at only one or
limited number of
piles)
N/A
N/A
Beyond the slate
of Degree III
*Table refers only to piles & deck.
** Bending failure should precede shear failure in structural components.
DamageCriteria 43
4.4 CELLLLAR QUAYWALLS
( 1) Seismic response of cellular quay walls
A cellular quay wall is made of a steel plate or steel sheet pile cell with sand or
other till. Resistance against inertia forces and earth pressures is provided by the
fill friction at the bottom surface of the cell for a non-embedded cell (i.e. steel
plate cofferdam type), or by the resistance of the foundation subsoil at the cell
embedment. Typical failure modes during earthquakes depend on cell embedment
and geotechnical conditions "as shown in Fig. 4.9.
Structural damage to a cellular quay wall is governed by displacements as well
as stress states. It is important to determine the preferred sequence of occurrence
and degrees of ultimate states in the composite cellular quay wall system. The
1995 Kobe earthquake caused damage in a failure mode, shown in Fig. 4.10, that
included deformation along the horizontal cross section of a cell.
(2) Parameters for specifying damage criteria for cellular quay walls
Seismic performance of a cellular quay wall is specified based on serviceability,
similar to that for a gravity quay wall, and on structural damage regarding stress-
"" es as well as displacements. Parameters for specifying damage criteria are simi-
lar to those for a gravity or a sheet pile quay wall regarding displacements as
shown in Fig. 4.11. Parameters regarding stresses include stress states in cell and
cell joints.
v
7
.' . -:....:: -,;..-.:.;. . .,:.-.
v
(a)
,- -_. ---. -..;:.:. '.;..'.:.; ..,:.- ..........--
.

(b)
Fig. 4.9. Deformation/failure modes of cellular quay wall.
(a) On finn foundation.
44 PlANe
.....
A-A' Section
.
~ BB' Section
. n1failure modes of
Fig. 4.10. Deformatio ay wall involving
cellular 9
u
a1 deformation.
cross secnon
Settlement of Apron
I ent at Apron Differential Sett em ,
Tilting
Horizontal Displacement tial Settlement behind Cell
Differen I
Settlement t,{ Tilting
Differential Displacemen _
.- .... . . ~ . _ - - - -
Damage Criteria 45
(3) Damage criteria for cellular quay walls
Provided the conditions mentioned at the beginning of this chapter are applicable.
the damage criteria for a cellular quay wall may be established by referring to
Table 4.4. The most restrictive conditions among displacements and stresses
should define the damage criteria.
4.5 QUAYWALLS WITH CRANES
In order to be serviceable after an earthquake. quay walls equipped with facilities
such as cranes on rails (hereinafter referred to as cranes) or ferry facilities need to
maintain the serviceability of these facilities in addition to general purpose quay
wall serviceability. Safety and protection of human life should be guaranteed.
Consequently, the criteria for acceptable damage to quay walls equipped with
cranes should be specified based on considerations additional to those discussed
in Sections 4.1 through 4.4. For example, serviceability of quay walls with con-
veyor belts is strongly affected by differential displacements and settlements. For
ferry wharves, consideration of quay wall displacement and implementation of
fail-safe devices for passenger bridges is necessary in order to ensure passenger
safety and the safety of the passenger transport operation. Additional criteria for
quay walls with cranes, such as container cranes, are discussed in this section.
These criteria supplement those presented earlier in Sections 4.1 through 4.4.
(1) Seismic response of cranes
A crane consists of an upper structure for handling cargo, and a supporting
structure for holding in place and transporting the upper structure, as shown in
Fig. 4.12. The crane is generally made of a steel frame. The supporting structure
is either of the rigid frame type or hinged leg type, 'with one hinge at the location
A as shown in Fig. 4.12. The supporting structure rests on rails through the
wheels.
A crane at rest is fixed to rails or to a quay wall with clamps or anchors, whose
strength provides the upper limit for the crane resistance against external forces.
However, a crane in operation is not supported by clamps or anchors, and the
lateral resistance of the crane against external forces is from friction and from
the wheel flanges.
Typical failure modes during earthquakes are derailment of wheels, detachment
or pull-out of vehicle, rupture of clamps and anchors, buckling, and
overturning. As shown in Fig. 4.13(a), widening of a span between the legs due to
the deformation of the quay wall results in derailment or buckling of the legs.
Table 4.4. Proposed damage criteria for cellular quay walls.
~
0'
::g
~
o
Residual
displacements
Peak response
stresses/
strains
CellulClr
w<III
Apron
Cell
Cell joint
Normalized residual
horizontal displacement
(dllf)*
Residual tilling
towards the sea
Differential
settlement on apron
Differential settlement
between apron and
non-apron areas
Residual tilting
towards the sea
Less than
1.5%**
Less than 3
0
Less than
0.03-0.1 m
Less than
0.03-0.7 m
Less than
2-3
0
Elastic
Elastic
1.5-5%
3-5
0
N/A
N/A
N/A
Elastic
Plastic (less than
the strain limit
for cell joint)
5-10%
5-8
0
N/A
N/A
N/A
Plastic (less than
the strain limit
for cell)
Plastic (beyond
the strain limit
for cell joint)
Larger than 10%
Larger than 8
N/A
N/A
N/A
Plastic (beyond the
strain limit for cell)
Plastic (beyond the
strain limit for
cell joint)
"d: residual horizontal displacement at the top of thc wall; /I: height of deck from mudline.
**Alternative criterion is proposed with respect to diIferentiaI horizontaI displacement less than 30 em,
Damage Criteria 47
Fig. 4.12. Schematic figure
of gantry crane.
supporting
structure

+tructure
girder
.--/././ /.r
//vehicle & rail
/////////
seaside
leg
Conversely, as shown in Fig. 4.13(b), narrowing of a leg span can also occur due
to the rocking response of the crane. This is due to alternating action of the
horizontal component of resisting forces from the quay wall during rocking
type response involving uplifting of one of the legs. Derailment and detachment
of the wheel can also occur due to rocking. As shown in Fig. 4.13(c), when
differential settlement occurs on a quay wall below the crane, tilting or over-
turning of the crane may occur. If the crane has one-hinge type legs, the derail-
ment can result in tilting and overturning of the crane as shown in Fig. 4.l3(d).
Though a clamp or anchor will provide more resistance to motion under the
action of external forces, the internal stresses induced in the crane framework
will become larger in comparison to the case with no clamp, thus allowing for
rocking responses.
Crane rails are often directly supported either by a portion of a retaining wall
or by the deck of a pile-supported wharf. When the width of the gravity wall is
small, or the quay wall is a sheet pile or cellular type, a separate
gften consists of piles, is provided to support the rails. Inorder to achieve desir-
able seismic performance of quay walls with cranes, special consideration is
required for the rail foundation such as providing a dedicated and cross-tied
upper structure to support the rails.
(2) Parameters for specifying damage criteria for quay walls with cranes
Seismic performance of quay walls with cranes is specified based on the ser-
viceability requirements for the crane and the possible structural damage.
. Serviceability of the crane is specified regarding the function of the upper struc-
'ture (i.e. cargo handling) and that "of the supporting structure (i.e. transportation
and support of the upper structure). Structural damage regarding the crane is
specified not only in terms of displacements, derailment, tilting, and stress
within the crane framework but also as displacements and stresses within the rails
. and the foundation. With regard to serviceability of the crane, maintaining the
. power supply should also.be considered.
48 PlANe
(a)
(b)
(c)
(d)
(c) Tilting of crane due to differential settlement of foundation.
(d) Overturning of one-hinged leg crane due to rocking/sliding.
Parameters for specifying damage criteria are illustrated in Fig. 4.14 and are as
follows. For rails and the foundation, the parameters include rail span, rail
misalignment (including discontinuity). differential settlement (differential
levels between the rails; differential levels, inclination and vertical discontinuity
along a rail), and displacements and stresses of the rail foundation. For a crane,
the parameters include derailment of wheels, detachment of vehicle, rupture of
clamps and anchors, displacements of the crane (derailment. tilting, and
Damage Lnlt:na "::f
Overall Displacements of Crane
(Derailment, Tilting, Overturning, etc.)
Stresses in Framework
Buckling Location
Local buckling
and Gross Stability
Vehicle (Derailment)
Vehicular Mechanism (Pull-out)
Anchor/Brakes(Fracture)
~ ~ ' 7 7 / ' ; ~ W ~ ~
Foundation of Crane Rails
Displacement and Stress
Rail Span
Rail Winding
Difference in Land/Sea side rail levels
Vertical curvature of rails
inclination of rail
Fig. 4.14. Parameters for specifying damage criteria for crane.
overturning) and stresses in the frame (stress levels, location of buckling, possi-
bility of collapse).
parameters from those mentioned. More specifics are given in the next subsection.
(3) Damage criteria for quay walls with cranes
As mentioned earlier, damage criteria are established considering both service-
ability and structural damage not only to the quay wall, but also to the crane and
rail foundation. The damage criteria for the quay wall required to limit structural
damage to the crane are discussed below. Criteria for the quay wall with respect
to the rail foundation should be established separately by referring to the damage
criteria for both the crane and the quay wall.
Provided the conditions mentioned at the beginning of this chapter are appli-
cable, the damage criteria for a crane may be established by referring to Table 4.5.
The most restrictive conditions among displacements and stresses should dictate
the damage' criteria.
. The damage criteria for quay walls with cranes may be established by referring
to the following:
1) Damage Degree I of the quay wall should dictate displacements that keep the
crane supporting structure within the elastic limit. For some cases, this limit
may correspond to a widening of the leg span as large as one meter for a rigid
frame crane with a rail span of 30 m. There is no clearly defined limit to
.'
Table 4.5. Proposed damage criteria for cranes.
Level of damage Degree I
Displacements Without derailment
Peak response Upper structure Elastic
stresses/strains
Main framework Elastic
of supporting
structure
Toe Elastic
Degree II
With derailment
Elm.;tic
. Plastic (less than the
ductility factor/strain
limit for main frame-
work)
Damage to toe .
-(including pull-out ':
of vehicle, fracture
of anchor/brakes)
Degree III
Without overturning .'
Plastic (less than the
limit for upper structure)
Without collapse
Damage to toe
(including pull-out of
vehicle, fracture of
anchorlbrakes)
Degree IV
Overturning
Plastic (beyond the
limit for upper structure)
Collapse
Damageto toe
(including pull-out
of vehicle, fracture of
anchorlbrakes)
t
C
~
~
L
(
Damage Criteria 51
damage Degree I for a one-hinge leg crane because this type of crane can eas-
ily absorb the change in the leg span.
2) Damage Degree II of the quay wall should dictate displacements that may
allow the yielding of the crane supporting structure, but keep the stress state
less than the ductility factor allowable for the main framework of the support-
ing structure.
3) Damage Degree III of the quay wall should maintain the level difference
between the rails and the differential settlements and tilting of the apron with-
in the overturning limit for the crane. A hypothetical possibility is' to introduce
a measure to increase the surface friction of the-apron, and thereby increase the
overturning limit of a derailed one-hinge leg crane, or a crane with a plastic
hinge at a leg. The required friction coefficient is given by J.lb > dJH
e
, where
J..4,: coefficient of friction; d\: expansion of leg span (m), He: height of the hinge
from the apron (m) (refer to Fig. 4.15).
4) Damage Degree IV of the quay wall should be the state beyond the upper limit
of the damage Degree III.
A set of tolerances for ordinary maintenance of cranes is shown in Table 4.6
(Japan Cargo Handling Mechanization Association, 1996). This set of tolerances
Fig. 4.15. Parameters for evaluation
overturning of crane.
N: Axial load from a crane leg
Table 4.6. Tolerance for ordinary maintenance of container cranes.
Parameters Tolerance
Rail Span Lspan ( ~ < 25 m)
(25 m ~ u: ~ 40 m)
Level difference between sea and land side rails
Curving in vertical direction
Curving in horizontal direction
Inclination
Rail joint Differential displacements (vertical and horizontal)
Gap
*Relative to original layout.
10mm
15mm
~ 1 0 0 0
5 mrn per 10m
5 mrn per 10m
1/500
Imm
5mm*
52 PlANe
could be used as one of the references to establish more restrictive damage
criteria than discussed earlier.
4.6 BREAKWATERS
A breakwater is usually made of a rubble mound, a massive structure such as a
caisson, or a combination of both placed on a seabed. Stability against a horizon-
tal external load is maintained by shear resistance of rubble, friction at the bottom
of the caisson, -and with associated resistance to overturning and bearing capacity
failure. Typical failure modes expected during earthquakes are shown in Fig. 4.16.
Breakwaters are generally designed to limit wave penetration and wave
overtopping during specific design storms, and at the same time designed to resist
the related wave actions. It is unlikely that a major earthquake will occur simul-
taneously with the design sea state because the two events are typically not relat-
ed. Consequently, design storm wave action and an earthquake can be treated as
two independent load situations. Only wave actions from a moderate sea state
should be considered together with the design earthquakes. Decision on this sea
state has to be made based on the site-specific long-term statistics of the storm.
Selection of the appropriate design criteria depends on the functions of the
breakwater and the type of earthquake-induced failure modes. However, for all
breakwaters, the main criterion is the allowable settlement of the crest level
because it determines the amount of overtopping and wave transmission. For
breakwaters carrying roads and installations, additional criteria for allowable dif-
ferential settlement, tilting and displacement of superstructures and caissons are
needed.
Shaking of the breakwater may cause breakage of concrete armour units.
Criteria have been proposed with regard to maximum breakage in terms of num-
ber of broken units that may occur while the breakwater remains serviceable (e.g.
Zwambom and Phelp, 1995). The same criteria may be adopted for the earth-
quake-related damage.
Since the damage criteria for breakwaters have not been fully developed at
present, the performance grade is being described in order to indicate the relative
degree of allowable damage. The level of damage to various kinds of breakwaters,
according to the primary and secondary functions of breakwaters, is as follows:
- reduce wave penetration in basins (Grade C);
- recreational (access for people) (Grade C; but can be Grade Aor B depending
on the level of acceptable human life safety);
- provision of berthing on the port side of the breakwater, and related access
roads (Grade B) (see Fig. 4.17);
- provision of cargo handling facilities on the breakwater, including conveyor
belts (Grade B), and pipelines for oil and liquid gas (Grade A or S. depending
on the level of threat of explosion).
Damage Criteria 53
-Crest lowering due to settlement ofsubsoil
.Oifferentialsettlementandtilt ofcaissons
.Possible damage tocaissonshearkeys

Slip failures andsubsequentcrest loweringdue to liquefaction
subsoil
-. Differential settlementandtilt ofcaissons
Damage111 caisson shear keys iftheyexist
(a)
Rubble Mound
Crest lowering due to shakedawn of rubble foundation
Differential settlement ofcaissons
Caisson
.Crest lowering andlateral spreading due to settlemenl/liquefaction
of subsoil
Differential settlement ofcaissons
-Slip failures and subsequentcrest loweringdue to liquefaction of
subsoil
Differential tilt andsettlement ofcaissons
Damage 111 caiSSOll shearkeysif they exist
(b)
11-----
c..sscn I
I
I
of slender types ofconcrete armour units
.Subsequent increasedovertoppinlt and risk ofdisplacement of
caisson
(c)
.. "'. ......::::{'t:.::;
-===- 2 ;/
., ',RUllllle Mound "
-- ------------- -----
Crest loweringdue toshake downof rubblematerial
Differenrial settlement of superstructure elements
'5?

-Crest loweringandlateralspreadingdue 111
setdemenl/liquefaction of subsoil
-Differential settlement of superstructure elements
Failures due to liquefaction ofsubsoil
.Subsequent lowering ofcrest
-Possible tilt anddisplacement ofsuperstructure elements
(d)
Fig. 4.16. Deformation/failure modes of breakwater.
Damage Criteria 54
Fig. 4.17. Examples of berthing
behind breakwater.
CHAPTER 5
Seismic Analysis
Seismic analysis of port structures is accomplished in three steps that include
assessment of the regional seismicity, the geotechnical hazards, and soil-structure
interaction analysis. The first step is to define the earthquake motions at the
bedrock (see Fig. 2.1 for the relative location of the bedrock with respect to the
seismic source, ground surface and port structures). This is typically accom-
plished by seismic hazard analysis based on geologic, tectonic and historical seis-
micity data available for the region of interest. The second step involves the
following two interrelated aspects of dynamic soil response: (1) an evaluation of
local site effects for obtaining the earthquake motions at or near the ground
surface; and (2) an assessment of the liquefaction resistance of the near surface
sandy soils and the associated potential for ground failures. Once the ground
motion and geotechnical parameters have been established, then seismic analysis
of the port structure(s) can proceed.
As in all engineering disciplines, reasonable judgement is required in specify-
ing appropriate methods of analysis and design, as well as in the interpretation of
the results of the analysis procedures. This is particularly important in seismic
design, given the multidisciplinary input that is required for these evaiuations, and
the influence of this input on the final design recommendations. This chapter
provides general recommendations for the type and level of analysis required for
various port structures.
5.1 TYPES OF ANALYSIS
The objective of analysis in performance-based design is to evaluate the seismic
response of the port structure with respect to allowable limits (e.g. dis-
placement, stress, ductility/strain). Higher capability in analysis is generally
required for a higher performance grade facility. The selected analysis methods
should reflect the analytical capability required in the seismic performance
evaluation.
A variety of analysis methods are available for evaluating the local site effects,
liquefaction potential and the seismic response of port structures. These analysis
56 PlANe
methods are broadly categorized based on a level of sophistication and capability
as follows:
1) Simplified analysis: Appropriate for evaluating approximate threshold limit
for displacements and/or elastic response limit and an order-of-magnitude esti-
mate for permanent due to seismic loading. .
2) Simplified dynamic analysis: Possible to evaluate extent of displacementl
stress/ductility/strain based on assumed failure modes.
3) Dynamic analysis: Possible to evaluate both modes and the extent of
the displacementlstress/ductility/strain.
Table 5.1 shows the type of analysis that may be most appropriate for each
performance grade. The principle applied here is that the structures of higher
performance grade should be evaluated using more sophisticated methods. As
shown in the index in Table 5.1, less sophisticated methods may be allowed for
screeniJ}g or ... Wi\
excitation.
In the present state of practice, it is desirable to confirm the applicability of
methods for analysis of port structures by using suitable case histories or model
test results before accomplishing the seismic performance evaluation.
5.2 SITE RESPONSEILIQUEFACTION ANALYSIS
Site response analysis and liquefaction potential assessments are typically accom-
plished using the methods outlined in Table 5.2. Useful references that highlight
the applicability of the various modeling procedures, have been prepared by
several investigators (e.g.. Finn, 1988; Idriss, 1991).
(l) Site response analysis
In simplified analysis, local site effects are evaluated based on the thickness of the
deposits and the average stiffness to a specified depth (generally 30 m), or over
the entire deposit above the bedrock. This information is then used to establish
the site classification, leading to the use of specified site amplification factors or
site dependent response spectra. This type of procedure is common in codes and
standards.
In simplified dynamic analysis. local site effects are evaluated numerically with
models such as common equivalent linear. total stress formulations. Soil layers
are idealized as horizontal layers of infinite lateral extent (i.e. one-dimensional
(ID. These methods are used to generate time histories of acceleration, shear
stress, and shear strain at specified locations in the soil profile.
In both of these categories of analyses, the computed ground surface earth-
quake motion parameters are used as input for subsequent simplified structural
analysis as discussed in Section 5.3.
Table 5.1. Types of analysis related to performance grades.
Type of analysis Performance grade
Grade C Grade B Grade A Grade S'
Simplified analysis:
Appropriate for evaluating approximate threshold level and/or elastic limit
and order-of-magnitude displacements
Simplified dynamic analysis:
Of broader scope and more reliable. Possible to evaluate extent of
displacement/stress/ductility/strain bused on assumed failure modes
Dynamic analysis:
Most sophisricated. Possible to evaluate both failure modes and extent or
displacement/stress/ductiIity/strain
Index:
Standard/final design
Preliminary design or low level of excitations
.IIN
U ..,!t.)e""; . 'N
. . .. :.;. . . ,." ' ",' "."
'. . ,
: : ''l!. Li1I..... ,
.,,, .1 ..L.cJl ..
'.; 1",/.1..:11: t. .\..14 i ....
C/J
(\I
_.
'Jl
:;j
o'
::J
eo
'<:
Vl
Vi'
UI
-.J
Table 5.2. Methods for site response analysis and liquefaction potential assessment. **
Type of analysis Simplified analysis Simplifieddynamic analysis Dynamic analysis
VI
00
;:g
~
o
Site response Method Site category ID total stress
analysis (equivalent linear) analysis
Input Peak bedrock acceleration Time history of bedrock
parameters CPT qJSPT N-valucs earthquake motion
Stratigraphy V., GIGa-y, D-ycurves
Output of
analysis
Peak ground surface motion
(PGA, PGV)
Design response spectra
Time history of earthquake
motion at ground surface
and within the subsoil
Computed response
spectra at ground surface
ID effective stress (non-linear) analysis, or
10 total stress (equivalent linear) analysis*
For effective stress analysis:
Time history of bedrock earthquake motion
Undrained cyclic properties
For total stress analysis: the same as those
for simplified dynamic analysis
Time history of earthquake motion at
ground surface and within the subsoil
Laboratory cyclic tests and/or
Field correlation (SPT/CPT/V
s
)
+ I D total stress analysis
Field correlation
(SPT/CPT/V
s
)
Peak ground surface
acceleration (PGA)
CPT qJSPT
Stratigraphy
Method
Input
parameters
Laboratory cyclic tests and/or
Field correlation (SPT/CPT/Vs)
+ I Deffective stress analysis or
I D total stress analysis"
Time history of canhquakc For effective stress analysis: Tillie
motion at ground surface, or history of bedrock earthquake motion
time histories of shear stresses Undrained cyclic properties based
in the subsoil on laboratory cyclic tests and/or
Liquefaction resistance,
or rcyc, based on laboratory For total stress'analysis: the same as
cyclic tests and/or SPT/CPT/V
s
those for simplified dynamic anaysis
Output of Liquefaction potentia) (Fd Liquefaction potential (Fd Excess pore water pressure ratio(It/a:.
o
)
analysis Excess pore water pressure ratio Depth and time at the onset of
liquefaction
Liquefaction
potential
assessment
*If the bottomboundary of the domain in a soil-structure interaction analysis differs from the bedrock (i.e. if the bedrock level is too deep
for soil-structure interaction analysis), local site effects below the bottom boundary of the soil-structure analysis domain may he evaluated
based on IDeffective stress (non-linear) or equivalent-linear (total stress) analysis.
**CPT: cone penetration test, SPT: standard penetration test, POA, POY: peak ground acceleration and velocity, V.: shear wave velocity,
GIGo: secant shear modulus (G) over shear modulus at small strain level (Gil), D: equivalent damping factor, y: shear strain amplitude, FL:
factor of safety against liquefaction, excess pore water pressure (/I) over initial effective vertical stress r/(1:.o: shear stress ratio,
Ycy,,: cyclic shear strain amplitude.
C/)
n
_.
CLl
3
0'
::l
-<
Vl
_.
V)
VI
\0
-
,.
-- ..
60 PlANe
Dynamic analyses of soil-structure interaction (SSI) attempt to account for the
combined, or coupled, response of the structure and the foundation soils. Unlike
the simplified procedures, wherein the response of the structure is evaluated using
the free-field soil response as input, SSI analyses incorporate the behaviour of
both soil and structure in a single model. The finite element methods (FEM) or
finite difference methods (FDM) are' commonly used for advanced SSI
analyses. In this type of analysis, local site effects are often not evaluated
independently but are evaluated as a part of the soil-structure interaction analysis
of port structures. If the bottom boundary of the analysis domain for soil-
structure interaction analysis differs from the bedrock (i.e. when the bedrock
is too deep for soil-structure interaction analysis), the local site effects below the
bottom boundary of the soil-structure interaction analysis domain may be
evaluated based on ID non-linear (effective stress) or ID equivalent linear (total
stress) analysis.
In both of the simplified dynamic analysis and dynamic analysis, adequate
choice of input earthquake motion at the bedrock should be important. For details,
refer to Technical Cominentaries 3 and 7.
(2) Liquefaction potential assessment
In simplified analysis, the liquefaction potential of sandy soils are evaluated based
on standard penetration tests (SPT) or cone penetration tests (CPT) through
empirical criteria. In simplified dynamic analysis, liquefaction potential is evalu-
ated based on a comparison of computed shear stresses during the design earth-
quake and the results of cyclic laboratory tests, and/or based on SPT/CPT data.
The liquefaction potential evaluated through these categories of analysis are used
later as input for subsequent simplified deformation analysis of structures at
liquefiable sites as discussed in Section 5.3.
In dynamic analysis, liquefaction potential is often not evaluated independ-
ently, but is evaluated as part of the soil-structure interaction analysis of port
structures. Details are discussed in Section 5.3.
5.3 ANALYSIS OF PORT STRUCTURES
The method of analysis for a port structure depends on structural type. The appro-
priate method may be chosen by referring to Table 5.3. A list of readily available
methods is shown in Table 5.4. As indicated in Table 5.4, methods for analysis 0
port structures may be broadly classified into those applicable to retaining/eart'
structures, including quay walls, dikes/slopes and breakwaters, or those applies
ble to open pile/frame structures, including pile/deck system of pile-supports
wharves and cranes.
. , ~ ' #
Table 5.3. Analysis methods for port structures.
Type of analysis Simplitied analysis Simplified dynamic analysis Dynamic analysis
Structural modelling Geotechnical modelling
Gravity quay wall Empirical/pseudo-static Newmark type analysis FEM/FOM* FEM/FOM*
methods with/without soil Simplified chart based on
Sheet pile quay wall liquefaction parametric studies (see Table 5.4) Linear or Linear (Equivalent
Non-linear analysis linear) or
Pile-supported wharf Response spectrum method Pushover and response spectrum Non-linear analysis
methods 20/30**
Cellular quay Pseudo-static analysis Newmark type analysis
20/30**
Crane Response spectrum method Pushover and response
spectrum methods
Breakwater Pseudo-static analysis Newmark type analysis
*FEMIFOM: Finite element method/finite difference method.
**20/30: Two/three-dimensional analysis.
en
o_.
en
3
o'
~
;:]
e.
'<:
en _.
en
0'
0-
N
Table 5.4. List of analysis methods and references.
Type of analysis Simplified analysis Simplified dynamic analysis Dynamic analysis
"'0
-
~
o
Lysmer et al. (1915)
(equivalent Iincar/
total stress)
Iai (1998b)
(non-linear/effective
stress)
Dickenson &Yang
( 1998)
(for gravity wall,
1iquefaction
remediation
measures)
lai et al. (1999)
(for gravity
wall, with/without
liquefaction)
McCullough &
Dickenson (1998)
(for sheet pile wall,
Iiquefaction
remediation
measures)
Newmark type
analysis for
evaluating sliding
displacement
Simplified chart FEMIFDM 2D
based on parametric Linear geotechnical
studies for evaluating modeling (FLUSH)
displacement Non-linear geotechn-
ical modeling (FLIP)
1'7.2.1 (I), 1'7.2.2(1), 1'7.2.1(2)(3), T7.2.2(2), n.3.1, T7.3.2( I)(2)(4),
1'7.2.4 1'8.1.3. TR.I.4, 1'8.2.4 1'7.3.3, 1'8.1.5. 1'8.3.2
Uwabe (1983) Newmark (1965)
(without liquefaction) (for dikes)
Gazetas et al. (1990) Franklin & Chang
(for sheet pile wall) (1977)
lai (I 998a) (for dikes)
(with liquefaction) Richards & Elms
(1979)
(for gravity wall)
Whitman &
Liao (1984)
(for gravity wall)
Towhata & Islam
( 1987)
(for sheet pile wall)
Steedman (1998)
(for gravity and
sheet pile walls)
Empirical/pseudo-
static methods for
evaluating order-
of-magnitude
displacement
1'7.1.2
Pseudo-static
method for
evaluating
threshold level
1'7.1.1, 1'7.1.4,
T8.1.2.TR.2.2,
1'8.2.3
./
Ebeling &
Morrison (1992)
Tsinker (1997)
Ministry of
Transport,
Japan (1999)
Analysis
Method
Relevant
Technical
Commcn-
taries
Refercnces
Gravity quay
wall
Shcet pile
quay wall
Cellular quay
wall
. Dike/slope/
retaining
wall of pile-
supported
wharf
Breakwater
Pile and deck
of pile-
supported
wharf
Crane
Analysis
Method
Response spectrum method for
evaluating elastic limit response
Pushover and response spectrum methods
(SDOF/MDOF)
FEM/FDM 2D
Linear geotechnical
modeling (FLUSH)
Non-linear geotechn-
ical modeling (FLIP)
Relevant T7.1.3. T8.4.2
Technical
Commen-
taries
References Ferritto (1997a)
Werner (J998)
Yokota et al. (1999)
Perrino et al, (1999)
T7.2.3. T8.4.3
Ferritto (I 997a) .
Werner ( J998)
Yokota et al. (1999)
Ferritto et HI. ( 1999)
T7.3.1. T7.3.2(3).
T8.5.2
Lysmer et al. (1975)
(equivalent linear/total
stress)
lui (1998b)
(non-linear/effective
stress)
Perrino et al. (1999)
(non-I inear)
en
C'1l
Vi'
3
o'
:::s
eo
'<
VI
Vi'
0\
w
64 PlANe
(1) Methods for analysis of retaining/earth structures
a) Simplified analysis
Simplified analysis of retaining/earth structures is based on the conventional
force-balance approach, sometimes combined with statistical analyses of case
history data. The methods in this category are often those adopted in conventional
seismic design codes and standards. In simplified analysis, retaining/earth struc-
tures can be idealized as rigid blocks of soil and structural masses. Therigid block
analysis is typically applied for gravity, sheet pile and cellular quay walls,
dike/slope/retaining walls for pile-supported wharves and breakwaters.
Effects of earthquake motions in simplified analysis are represented by a peak
ground acceleration or an equivalent seismic coefficient for use in conventional
pseudo-static design procedures. These parameters are obtained from the simpli-
fied analysis of local site effects discussed in the previous section. A capacity to
resist the seismic force is evaluated based on structural and geotechnical condi-
tions, often in terms of a threshold acceleration or a threshold seismic coefficient,
beyond which the rigid blocks of soil and structural masses begin to move. When
soil liquefaction is an issue, the geometric extent of liquefaction must also be con-
sidered in the analysis.
Results of the simplified analysis are appropriate for evaluating the approxi-
mate threshold level of damage, which ensures at least the repairable state of
structural performance for Ll. Whether or not the approximate threshold level
ensures the serviceable state of structural performance for Ll depends on the
details in evaluating the design parameters for the pseudo-static method. Order-
of-magnitude displacement is also obtained by the pseudo-static method com-
bined with statistical analysis of case history data. This, however. is a crude
approximation and should be used only for the preliminary design stage or low
levels of excitation.
There is a significant difference between the design approach recommended
in these guidelines and the conventional design concept. In the conventional
design concept, especially when referring to the simplified analysis. an equiva-
lent seismic coefficient is used as an input parameter representing adequately
the ensemble of ground motions, and a factor of safety is applied to determine the
dimensions of the structure. In the proposed approach, the design is based on the
seismicperforrnance of the structure evaluated appropriately through response
analysis for a variety of input earthquake motions. The ensemble of the seismic
responses, rather than the ensemble of input motions, is used as a basisfor accom-
plishing the design in the proposed guidelines. For each response analysis, input
parameters most appropriate are those well defined in terms of applied mechan-
ics, such as a peak ground acceleration for the simplified analysis and/or an equiv-
alent parameter clearly defined in terms of peak ground acceleration.
Consequently, no factor of safety should be applied to input data used in seismic
analysis for evaluating the threshold level of the structure. This important
....
Seismic Analysis 65
distinction between conventional design and performance-based design should
also be borne in mind when interpreting the design guidelines included in the
various seismic codes. .
b) Simplified dynamic analysis
Simplified dynamic analysis is similar to simplified analysis, idealizing a struc-
ture by a sliding rigid block. In simplified dynamic analysis, displacement of the
sliding block is computed by integrating the acceleration time history that exceeds
the threshold limit for sliding over the duration until the block ceases sliding.
Effects of earthquake motions are generally represented by a set of time histories
"of earthquake motion at the base of a structure. The time histories of earthquake
motion are obtained from the simplified dynamic analysis of local site effects
discussed in the previous section. In the sliding block analysis, structural and
geotechnical conditions are represented by a threshold acceleration for sliding. A set
of empirical equations obtained from a statistical surrunary of sliding block analy-
ses is available (see references in Table 5.4). In these equations, peak ground accel-
eration and velocity are used to represent the effect of earthquake motion.
In more sophisticated analysis, structural and geotechnical conditions are
idealized through a series of parametric studies based on non-linear FEMlFDM
analyses of soil-structure systems, and the results are" compiled as simplified
charts for use in evaluating approximate displacements.
c) Dynamic analysis
As mentioned earlier, dynamic analysis is based on soil-structure interaction,
generally using FEM or FDM methods. In this category of analysis, effects of
earthquake motions are represented by a set of time histories of earthquake
motion at the base of the analysis domain chosen for the soil-structure system. A
structure is idealized either as linear or as non-linear, depending on the level of
earthquake motion relative to the elastic limit of the structure. Soil is idealized
either by equivalent linear or by an effective stress model, depending on the
expected strain level in the soil deposit during the design earthquake.
Fairly comprehensive results are obtained from soil-structure interaction analy-
sis, including failure modes of the soil-structure system and the extent of the dis-
placement/stress/strain states. Since this category of analysis is often sensitive to
a number of factors, it is especially desirable to confirm the applicability by using
a suitable case history or a suitable model test result.
(2) Analysis methods for open pile/frame structures
a) Simplified analysis
Simplified analysis of open pile/frame structures is typically done by idealizing
the pile/deck system of pile-supported wharves or the frame of cranes by a single
degree of freedom (SDOF) or multi-degree of freedom (MDOF) system. In this
66PIANC
analysis, earthquake motions are generally represented by the response spectrum.
Structural and geotechnical conditions are represented by a resonant frequency
and damping factor of the deck-pile system and/or the cranes: A ductility factor
may also be introduced. The movement of the dike/slope is generally assumed to
be negligible. Results of the SDOFIMDOF analysis are useful to evaluate approx- '.
imate limit state response of a pile-deck system or a" crane, which ensures at least
the repairable state of structural performance for LI.
b) Simplified dynamic analysis
In simplified dynamic analysis of open pile/frame structures, the SDOF or MDOF
analysis of pile-deck structure or cranes is combined with pushover analysis for
evaluating the ductility factor/strain limit. The movement of the dike/slope is
often assumed to be negligible but sometimes estimated by a sliding block type
analysis. Movement of a pile-supported deck could thereby be estimated by sum-
ming-up the dike/slope movement and structural deformation. Soil-structure
interaction effects are not taken into account, and thus there is a limitation in this
analysis. Interaction between the pile-supported wharves and cranes can be taken
into account by MDOF analysis. Displacement, ductility factor/strain, and loca-
tion of yielding or buckling in the structure are generally obtained as a result of
the analysis of this category. Failure modes with respect to sliding of retaining
walls/dikes/slopes are not evaluated but assumed and there is, thus, another limi-
tation in this type of analysis.
c) Dynamic analysis
Dynamic analysis is based on soil-structure interaction, generally using FEMI
FDM methods. Similar comments to those related to the dynamic analysis of
earth/retaining structures apply also to the open pile structures and cranes.
See Technical Commentaries 5 and 7 for further details on analysis methods.
5.4 INPUTS AND OUTPUTS OF ANALYSIS
Table 5.5 shows major input parameters required for the various types of analy-
sis. It should be noted that the reliability of the results depends not only on the
type of analysis. but also on the reliability of input parameters. It is ideal to use
input data based on thorough geotechnical investigations for more sophisticated
analysis. However, it may not be practical to demand that only the highest level
of geotechnical investigations. including the in-situ soil freezing technique for
sand sampling, be used in the design. Some of the parameters for the most sophis-
ticated methods could be evaluated based on the empirical correlation usingSPT
results, provided that this empirical correlation is calibrated through geotechnical
investigations of high level at the site of interest. See Technical Commentary 4 for
further details.
Seismic Analysis 67
Tables 5.6 and 5.7 show major outputs from the analyses. Clearly, more param-
eters of seismic performance can be evaluated by more sophisticated analysis.
These outputs are the final indices that define the ievel of damage. The perform-
ance evaluation can be completed by comparing these outputs to the damage cri-
teria discussed in Chapter 4. Examples of seismic performance evaluation are
given in Technical Commentary 8.
The displacements/ductility/strain limits in the proposed damage criteria are
intended to serve as general guides which are deemed as acceptable. However,
exceeding these values does not mean that performance is unacceptable.
Engineering judgement must be applied on a case by case basis in which
the amount of deformation/stress strain states and its spatial distribution are exam-
ined in terms of performance requirements and consequences for the port facili-
ties.
Ultimately, the economics of initial investment, risk of damage and the conse-
quences of failure of the operation must be considered. If the seismic responses
evaluated are less than those cited in the criteria, good performance can be expect-
ed. If the seismic responses are greater-than those cited in the criteria, it is up to
the engineer to demonstrate that the performance objectives can be achieved.
Table 5.5. Major input parameters for analysis.
",
'"
00
"'t1
~
o
Type of analysis Simplified analysis Simplified dynamic analysis
\
Dynamic analysis
Gravity Method Pseudo-static/empirical Newmark type method Simplified chart based on FEMIFDM
quay wall methods parametric studies
Design k.: equivalent seismic Empirical equations: OmllX: peak acceleration at Time histories of earthquake
parameters coefficient (["':IX: peak acceleration the bedrock motions at the bottom
k: threshold seismic ","lIX: peak velocity Cross section of wall boundary of analysis domain
coefficient Time history analysis: Index properties of soil Cross section of wall
(Geometrical extent of time histories of including SPT N-values For equivalent' linear
liquefiable soils relative earthquake mot ions geotechnical analysis:
to the position and lIl: threshold acceleration GIGo-r& D-rcurves .
dimensions of a wall for For non-linear geotechnical
a liquefiable site) analysis:
Input Results of site response analysis, including 1I
111ax
, HIllI liquefaction potential
Undrained cyclic properties
parameters assessment
and G, K: shear and bulk
Cross section of wall
modulus. in addition to the
Geotechnical parameters, including c, ep: cohesion and internal friction angle of
geotechnical parameters
soils; J./th 8: friction angles at bottom and back face of wall; ground water level
for pseudo-static and
simplified analyses
Sheet Method Pseudo-static/empirical Newmark type method Simplified chart based on FEMIFDM
pile quay methods parametric studies
wall
Design The same as those for The same as those for . The same as those for In addition to those for
parameters gravity quay wall. gravity quay wall gravity quay wall gravity quay wall, dimensions
Input In addition to those for In addition to those for pseudo-static method, and material parameters
parameters gravity quay wall material parameters defining limit states of for sheet pile, tic-rod and
structural parameters structures, including yield and ultimate stress states unchor
for sheet piIe, tie-rod and For linear structural analysis:
anchor. including elastic material parameters,
E, I, A, and allowable including E. I. A. yield stress
stress For non-linear structural
analysis:
yield stress, M-tP curve
Pile- Method Response spectrum Pushover and response spectrum methods FEM/rOM
supported method
wharf
Design Pile-deck structure: In addition to those for pseudo-static method, In addition to those for
parameters design response material parameters defining limit states of gravity quay wall, dimensions
spectra resistance at structures, including yield and ultimate and material parameters for
elastic limit or stress states of pile-deck structure pile-deck structure:
allowable stress level For linear structural analysis:
f3 - /kh'SUbDp
elastic material parameters,
including E, /, A, yield
- ~ 4El
stress
or p-y curve For non-linear structural
Dike/retaining wall: analysis:
the same as those for yield stress, M--tP curve. .
gravity quay wall
(/)
C10
_.
til
2.
n
~
:;:,
p
- '<'
V.
tn'
0\
\D
'nlble 5.5. Continued.
Type of analysis Simplified analysis Simplified dynamic analysis Dynamic analysis
Input Results of site response The same as above
parameters analysis and liquefaction
potential assessment
Pile-deck structure:
dimensions & material
parameters for piles
and deck, including
E, I, A, o; allowable
stress
kh.sub: coefficient of
lateral subgrade
reaction, or p-y curve
Dike/retaining wall:
relevant parameters for
gravity/sheet pile quay
walls
Cellular Method Pseudo-static method Newmark type analysis FEM/FDM
quay wall
Design For overall stability: In addition to those for pseudo-static method, In addition to those for gravity
parameters the same as those for material parameters defining limit states of quay wall, dimensions and
gravity/sheet pile quay structures, including yield and ultimate stress material parameters for cell
wall states of cell structure For linear structural analysis:
For structural elastic material parameters,
performance: including E, I, A, yield stress
elastic limit/allowable For non-linear structural
stress analysis:
yield stress, M- curve
-.'
c
'\;'
~
o
't
..
For overall stability:
the same as those for
gravity/sheet pile quay
wall
For structural performance:
. dimensions and material
parameters for cell
Crane Method Response spectrum method Pushover and response spectrum methods FEM or framework analysis
Design Design response spectra In addition to those for pseudo-static method, Dimensions and material
parameters Resistance at elastic limit material parameters defining limit states of parameters of crane
or allowable stress level structures, including yield and ultimate stress states For structural interaction
analysis of quay wall and
crane system:
The same as those for relevant
type of quay wall analysis
For crane response only:
Time history of input
earthquake motions to the
Dimensions and material
crane
parameters
Break-water Method Pseudo-static method Newmark type method FEMIFDM
Design The same as those for The same as those for gravity quay wall The same as those for gravity
parameters gravity quay wall . quay wall
Input The same as those for The same as those for gravity quay wall
parameters gravity quay Willi
C/)
~
(ii'
3
(i'
)-
:J
p)
,<r
VI
-,
VI
-.J
-
".-- -. _..._.. -_..... _. __._--------.
Table 5.6. Analysis output.
Analysis type Simplified analysis Analysis
performance
Performance
grade
Grade C
LI: Repairable
L2: Allow collapse but without
adverse effects on surroundings
Simplified dynamic analysis
Grade B
L I: Serviceable
L2: Near collapse
Dynamic analysis
Grade A/Grade S
L1: Serviceable
L2: Repairable (Grade A)/
Serviceable (Grade S)
-:J
N
~
~
o
Gravity quay wall
Sheet pile quay wall
Pile-supported wharf
Cellular quay wall
Crane
Break water
Threshold limit
Order of magnitude displacement
Threshold limit
Order or magnitude displacement
Pile-supported deck:
threshold limit
threshold limit
order of magnitude displacement
Threshold limit
Order of magnitude displacement
Threshold limit
Threshold limit
Wall displacement
Wall displacement
Stress/ducti lily
Pile-supported deck:
displacement, stress/ductility
displacement, stress/ductility
Wall displacement
Stress/ductility
Displacement
Stress/ductility
Displacement
Response/failure modes
Peak and residual displacements
Response/failure 1110des
Peak and residual displacements,
stress/ductility
stress/ductility
stress/ductility
stress/ductility
Peak and residual displacements
Table 5.7. Outputs from dynamic analysis.
Seismic Analysis 73
Structure and geotechnical
modelling
Structure modelling
Linear Non-linear
Geotechnical
modelling
Linear
(Equivalent
linear)
Non-linear
Peak response
Displacement/stresses
Failure mode due to
soil movement
Peak and residual
displacement/stresses
from soils movement
(assuming structure
remains elastic)
Failure mode of structure
Peak and residual dis-
placement/ductility
factor/stresses for struc-
tures (assuming there are
no effects from residual
displacement of soils)
Failure mode of soil-
structure systems
Peak and residual
displacement/ductility
factor/stresses including
effects from residual
displacements of soils
TECHNICAL COMMENTARIES
TCI: Existing Codes and Guidelines
T1.1 List of Seismic Design Codes and Guidelines for Port Structures
T1.2 Reviewed Aspects of Codes and Guidelines
T1.3 Seismic Design Practice for Port Structures around the World
TECHNICAL COMMENTARY 1
Existing Codes and Guidelines
This technical commentary presents a brief review of the earthquake provisions
described in the existing codes and guidelines used at the various ports in the
world. While most of the existing codes and guidelines for port structures include
earthquake provisions, the degree to which specific earthquake-resistant design
methods are addressed in these documents varies considerably. The objective of
this brief review is to present:
- a list of the primary codes and guidelines for seismic design of port structures
around the world;
- the current status of design practice around the world in light of the new
design strategy proposed in this book.
TI.1 LIST OF SEISMIC DESIGN CODES AND GUIDELINES FOR PORT
STRUCTURES
The existing codes and guidelines reviewed in this technical commentary are as
follows:
- Technical standards for Ports and Harbour Facilities in Japan' (Ministry of
Transport, Japan, 1999);
Recommendations of the Committee for Waterfront Structures Harbours and
Waterways (EAU, Germany, 1996);
ROM 0.6: Acciones y efectos Sfsmicos en las Obras Maritimas y Portuarias
(Puertos del Estado, Madrid, Spain, 2000);
US Navy seismic design guidelines (Ferritto, 1997a, b);
Seismic Guidelines for Ports, American Society of Civil Engineers -
Technical Council on Lifeline Earthquake Engineering, Ports Committee
(Werner, 1998);
California Marine Oil Terminal Standard (California State Lands
Commission, USA, under development as of May 2000);
European Prestandard, Eurocode 8-Design provisions for earthquake resist-
ance of structures (CEN, 1994):
~ Part 1-1: General rules - Seismic actions and general requirements for
structures, ENV-1998-1-I, May 1994;
80 PlANe
,. Part 5: Foundations, retaining structures and geotechnical aspects ENV
1998-5, June 1994;
- NewZealand Standards;
~ NZS 4203 (1992) "General Structural Design Loadings for Buildings"
>- NZS 3101 Part 1 (1995) "The Design of Concrete Structures"
~ NZS 3403 Part 1 (1997) "Steel Structures Standard"
~ NZS 3403 Part 2 (1997) "Commentary to the Steel Structures Standard"
>- Transit New Zealand (TNZ) Bridge Design Manual.
T1.2 REVIEWED ASPECTS OF CODES AND GUIDELINES
In order to identify the current status of design practice relative to the proposed
design strategy in this book, existing codes and guidelines for port structures
were reviewed with respect to the following aspects:
- Methodology:
>- Is the methodology a single or dual level approach?
--
>- What is the general statement on the acceptable level of damage?
Damage criteria:
~ What is the acceptable displacement/deformation for retaining structures in
design?
'"
~ What is the acceptable ductility/strain limit for pile-supported structures in
design?
- Analytical procedure:
>- What factor is adopted for specifying the effective seismic coefficient
as a fraction of peak ground acceleration for retaining structure
design?
>- What types of response spectra are used for pile-supported structures in
design?
An overview of the existing codes and guidelines reviewed with respect to the
aforementioned aspects is shown in Table T1.1. Additional specifics. to backup
the descriptions in Table T1.1, are presented below.
(1) Japanese design (Port and Harbour Research Institute, 1997: Ministry of
Transport, Japan, 1999).
The dual level approach has been adopted for structures of Special Class of
Importance. However, the single level approach is adopted for structures of Class
A, B, and C of Importance.
For structures of Special Class of Importance, the performance level is speci-
fied as follows:
For L1: Minor or no damageflittle or no loss of serviceability.
For L2: Minor or little damagellittle or short-term loss of serviceability.
rn.i )
Existing Codes and Guidelines 81
For retaining structures of Special Class of Importance. the criteria for struc-
tural damage and criteria regarding serviceability are shown in Tables Tl.2 and
T1.3.
The seismic coefficient for use in retaining structures is defined as follows for
Special Class structures.
t
k: - Q
max
( O? )
h - -- a max:S; ._g
g
( )
/ 3
I am:tll
kh =3 g (a max >0.2g)
For Class B structures (designed with importance factor of 1.0), the code-speci-
fied seismic coefficients are about 60% of those given by Eqn. (T1.1).
For a pile-supported wharf with vertical piles, analysis is performed based on
a simplified procedure and pushover method. The ductility limits for use in the
simplified procedure (see Technical Commentary 7) for Ll earthquake motion
are specified as shown in Table TIA. Pushover analysis is performed for Special
Class structures and the strain limits prescribed are:
Level I motion: equivalent elastic,
Level 2 motion: Cmax =O.44tp/D
p
for the embedded portion.
Pile-supported wharves with .vertical steel piles are designed using the
response spectra shown in Fig. T1.1 for L1 motion. These spectra have been com-
puted based on 2D soil-structure interaction analysis for typical pile-supported
wharf cross sections. For L2 motion, time history analysis should be performed
and the results should meet the ductility limits for L2 earthquake motion shown
in Table T1.5.
Comprehensive guidelines are shown on liquefaction potential assessment and
implementation of remedial measures (Port and Harbour Research Institute, 1997).
(2) Spanish design (ROMO.6, 2000)
Dual earthquake levels must be considered in the design, but only Level 2 earth-
quake motion is considered for low seismic design acceleration.
Analysis procedures specified in ROM ranges from the simpler methods for
the lower design acceleration to the more sophisticated methods for the higher
design acceleration.
In most cases, liquefaction is one of the main aspects to be considered. In addi-
tion, the codes specifies the criteria with respect to:
Gravity walls, breakwaters: overall stability, post-earthquake displacements.
Sheet pile walls, piled decks: structural behaviour.
Pile-supported wharves/piers with vertical steel piles are designed using the
ductility factors given as follows:
Wharves: Pile cap: J1 = up to 5.0 Embedded portion: J1 = 2.0
Piers: Pile cap: J1 =up to 4.0 Embedded portion: J1 =1.5
Table '1'1./. Overview ofexisting codes and guidelines for port structures.
Japanese Dual level for For Special Class:
(1997, 1999) Special Class No damage for LI;
Single level for Minor damage and
Class A, B, C lillIe or short-term
loss of serviceability
for L2
Codes/ Mcthodology
Guidelines S' I I I
, IIIg C (II' uua
level approach
Damage criteria
00
N
Analytical procedure Remarks
'i:l
.....
The factor a for Response spectra

z
specifying the
(J
effective seismic
coefficient as a
fraction of anlulg
See Eqn. (TI.I) LI: See Fig. Tl. I These codes
L2: Time history and
analysis [no guidelines
response spectra have been
specified] specifically
established
for designing
a=0.7 See Eqn. (Tl.2)
I port structures
Ductility/strain
limit for pile-
supported
wharf/pier
Simplified
procedure uses
displacement
ductility factors.
Pushover
analysis uses
strain limit
Ductility factors
for pile head
and embedded
portions.
Different for
wharf and pier
Displacement
for retaining
structures
Nut Specified
See Tables
TI.2 & TI.3
Performance level
(Acceptable
damage level)
Little or short-term
loss of serviceability
for L I: No collapse
for L2
Dual level
approach:
Seismic input
in each level
depends on the
importance
and life time
of structure
Spanish
(2000)
German
(1t)9(j)
Single level Not specified
approach.
More detailed
analysis is
required if
damage can
endanger
humane live
Not specified Not specified Not specified;
a to be taken
from national
code
Not specified
u US DUlll Method Serviceable Soil limits Ductility limits N/A Linear or
S Navy LI 75 year under LI and specified for specified for Nonlinear
A (.1997) L2 500 year repairable wharf dikes piles Dynamic
Events under L2 Analysis
ASCE- Dual level Critical structures: Site-specific Ductility limits See Eqn. (T1.1) a) response
TCLEE approach: Serviceable under based on specified for spectrum
( 1998) exposure time LI and repairable serviceability piles methods
and ground under L2 and effect on b) time history
motion are Other structures: adjacent analysis
based on the Repairable for LI structures c) pushover
importance. analysis
and design life
California Dual Method Serviceable Soil Strain limits for N/A A) Single Mode
Marine Ll 75 year under LI and deformation piles under 8) Multi-mode
Oil L2 500 year repairable under L2 limits for soil limit conditions C) Pushover
Terminals Events structure D) Inelastic
(2000) interaction Time History
Eurocode Dual level No reference to Not specified Not specified. Design ground See Eqn. (TI.3) These codes
tT1
( 1(94) approach Port Structures A guidance is acceleration refer to x
Vi'
given for the fraction for buildings
....
_.
::;)
design of { l J c s i ~ n
and other
OQ
plastic hinging is specified civil
()
0
engineenng
0-
New Zealand Dual method The Ultimate Limit Not specitied Both are Design Spectra Based on seismic
I1l
works with
til
(1992-19(7) state peak response specified are used zone, subsoil
no direct
15
spectra is given condition,
0-
reference
C)
structure period
c
10 port
_.
and ductility
0-
I1l
structures
-
factor
5'
I1l
til
00
VJ
Deeper than 7.5 m
84 PlANC
Table Tl.2. Structural damage criteria in Japanese standard.
Type of retaining walls Water depth
------------------
Shallower than 7.5 m
Horizontal displacement Horizontal displacement
oto 0.2 m 0 to 0.3 m
oto 0.3 m 0.3 to 0.5 m
_ Gravity quay wall
No repair needed for operation
Partial operation allowed
Sheet pile quay wall
No repair needed for operation
Partial operation allowed
Horizontal displacement
oto 0.2 m
0.2 to 0.5 m
Horizontal displacement
oto 0.3 m
0.3 to 1.0 m
Table T1.3. Serviceability criteria in Japanese ports.
Main body Upper limit of settlement
of retaining Upper limit of tilting
structure Upper limit of differential horizontal displacement
Apron Upper limit of differential settlement on apron
Upper limit of differential settlement between
apron and background
Upper limit of tilting
0.2 to 0.3 m
3-5
0
0.2 to 0.3 m
0.3 to 1.0m
0.3 to 0.7 m
3-5% [towards sea]
5% [towards land)
Table TIA. Ductility factor for Ll motion in the simplified analysis.
Special Class J.l = 1.0
Class A u = 1.3
Class B u = 1.6
Class C J.l = 2.3
Design response spectra are specified as follows:
For T =0 to TA: SAladesign =1 + 1.5 TITA
For T =TA to TB: SA..!adesign =2.5
For T> TB: SAladesign = KCIT
where adesign = 0.04 to 0.26g, T
A
=KC/lO, T
B
=KC/2.5
K=1.0 to 1.5 (depends on type of seismicity)
C= 1.0 to 2.0 (depends on ground conditions)
(T1.2)
(3) German design (EAU, 1996)
The German recommendations for quay walls describe a single level approach
with a conventional seismic earth pressure based on the Mononobe-Okabe equa-
tion (Mononobe, 1924; Okabe, 1924). Pressures are assumed to increase linearly
with depth, although it is accepted [R125] that tests have shown this not to be the
case. The soil-water system is assumed to move as one body. and no water hydro-
dynamic suction in front of the wall is taken into account. The requirements for
accuracy of the calculations are correspondingly more stringent when earthquake
._._ .... 1 _."'- _ , .'
Region A
1.00
Region B
1.00
10 1.0
Natural Period (s)
/.
\ I I
: ,6-
,
I ",'
@ i
;
~
! I
I
D ~
, ! ~
,
, I
I
I
I
l
!
I
i :
, I
i
II
! :
C
Ql
~
o 0.10
,!ol
E
III
'ijj
en
0.01
10 0.1 1.0
Natural Period (5)
.,
2'
: I I ' ~ I
, ,
(3)
I :
I
1 i
I
~ ~
I'
~
I
I j , ,
I I I
I
I
1
,
0.0
0.1
C
<II
'0
~
U 0.10
.!.l
~
'S
en
10 1.0
Natural Period (s)
, ,
,
I
!
(f)
"
, ,
I I ~
...<6
@
I I \
~
!
I
I
I
II'
Region 0
1.00
C
<II
:g
ii
o
o 0.10
.!ol
~
'a!
CIJ
0.01
10 0.1 1.0
Natural Period (s)
,
.. ,
, r
,
I ! I
I r. I I
l{:j\
~ B
~
III
RegionC
1.00
0.01
0.1
C
<II
:g
:::
~
U 010
g'
E
III
'a!
en
Index
CD T
g<0.1S
0.1s:iT
g<0.5s
@ O.Ss:iT
g
T
g
; Naturalperiod of the ground
Hi .
T
g
=4 r. V
si
where; Hi: thickness of l-th layer
Vsi : shear wave velocity of jth (ayec
Classification of RegionsAthrough E
based on seismic hazard.
10 1.0
Natural Period (5)
i1l
'2'
I
I
"'
ll6
!@
~
,
1 0.0
0.1
Region E
1.00
Fig. T1.1. Response spectra for LI earthquake for pile-supported wharves in Japan.
Table T1.5. Ductility factor for L2 motion in the simplified analysis.
Special Class 11 == 1.25 + 62.5(tp/D
p
) . but J1 ~ 1.25
tp == pile wall thickness
D
p
= pile diameter .
-----
-------
86 PlANe
damage can endanger human lives or cause destruction of supply facilities or the
like which are of importance to the population [R124].
(4) US Navy/California State design (Ferritto, 1997a, b; Ferritto et al., 1999)
The US Navy code describes a dual level design and a performance level that is
serviceable under Ll and repairable under L2. The damage criteria are deforma-
tion limits for wharf dikes and ductility limits for piles. The procedure requires a
linear or nonlinear dynamic analysis. California State design is similar to the US
Navy design in principle, but the ductility requirements are much more in detail
and specified by strain limits. This procedure is still under development and will
be finalized by 2001 in the form of regulatory guidance.
(5) ASCE-TCLEE guidelines (Werner, 1998)
This reference represents one of the first guidelines documents, specifically
developed for the seismic analysis and design of port structures and facilities in
North America. In addition to comprehensive treatment of seismic hazards, seis-
mic design and analysis in North American engineering practice, and guidelines
for specific port components, this document presents pertinent information on
seismic risk reduction, and emergency response and recovery at pons. It should
be noted that the ASCE-TCLEE guidelines were developed with the primary
objective of providing a framework for the establishment of improved seismic
risk evaluation and reduction procedures for ports in the United States. The rec-
ommendations outlined in the guidelines document are not to be interpreted as
codes, nor are they intended to supercede local code requirements that may be
applicable.
The ASCE-TCLEE seismic guidelines present a dual level design. consistent
with the current state of practice at major pons in the United States. The multi-
level design approach has been adopted at numerous ports in the form of
two-level, and occasionally three-level design procedures. An example of the
two-level approach as applied in the western United States is as follows:
Levell. Under this first level of design, Operating Level Earthquake (OLE)
ground motions are established, that have a 50-percent probability of exceedance
in 50 years (corresponding to an average return period of about 72 years). Under
this level of shaking, the structure is designed so that operations are not inter-
rupted and any damage that occurs will be repairable in a shan time (possibly less
than 6 months).
Level 2. Under this second level of desizn. more severe Continzencv Level
- . ..... -
Earthquake (CLE) ground motions are established that have a 10-percent proba-
bility of being exceeded in 50 years (consistent with most building codes and cor-
responding to an average return period of about 475 years). Under this level of
shaking. the structure is designed so as to undergo damage that is controlled. eco-
nomically repairable, and is not a threat to life safety.
It should be noted that the exposure times adopted for the Level I and Level 2
events in this example application. may vary regionally due to variations in the
rate of seismicity, the type of facility, and economic considerations.
The damage criteria outlined in the guideline document are presented in the
form of general performance-based recommendations. As such. the recommen-
dations address the evaluation and mitigation of liquefaction hazards and ground
failures. deformation limits for retaining structures. earth structures and other
'-
waterfront components, and ductility limits for piles. In order to evaluate these
earthquake-induced loads and associated deformations. pseudo-static methods of
analysis must often be supplemented with linear and/or nonlinear dynamic analy-
sis, the level of analytical sophistication being a function of the intensity of the
ground motions, the anticipated soil behaviour. and the complexity of the struc-
ture. General guidance on the level of analysis required for a variety of geotech-
nical and structural applications is provided (including dynamic soil-structure
interaction analyses).
(6) Eurocode design (CEN, 1994)
The methodology of the Eurocode 8 describes in general, a dual level approach
but in low seismicity zones [adc:sign ~ O.lg] and for 'well defined' structures in seis-
mic zones with small [adesign ~ O.04g], a single level approach can be sufficient.
The mentioned performance levels are:
'No collapse' requirement: retain structural integrity and a residual bearing
capacity.
. 'Damage limitation' requirement: no damage and associated limitations of use,
the costs of which would be disproportionately high compared with the cost of
the structure itself.
Damage criteria in terms of maximum displacements and/or ductility levels
are not specified. For piles, it is stated that they shall be designed to remain elas-
tic. When this is not feasible, guidance is given for the design of potential plastic
hinging and the region which it shall cover.
For the analytical procedure, the design ground acceleration adesign tends to
coincide with the actual peak acceleration for moderate to high magnitude earth-
quakes in cases of medium to long source-to-site distances, which are charac-
terised [on finn ground] by a broad and approximately uniform frequency
spectrum, while adesign will be more or less reduced relative to the actual peak for
near field, low magnitude events. The adesign corresponds to a reference period of
475 years, or as specified by the National Authority. .
For retaining structures kh = adesign IrEcg and k; = -o.5k
h
where
rEC = 2 for free gravity walls with acceptable displacements (mm) ~ 300ldesign!g
rEC = 1.5 as above with displacements (mm) ~ 200adesignlg
rEC = 1 for the rest of the retaining structures
88 PlANe
For retaining structures higher than == 10m, a refined estimate of adcsign can be
obtained' by a free field one-dimensional analysis of vertically propagating
waves.
For a linear analysis design of pile-supported structures, design spectra are
defined as:
0:<; T s r. : SA(T) =a- S.[I + ;. -( ~ -I )]
T
B
<5:.T <5:.T
c
: SA(T) = a-s- /30
q
~ [T Jk
dl
r; <5:. T <5:. TD : SA(T) = a .S.': ;
~ [0.20]a
(T1.3)
_ f30 [T
c
Jk
d l
[TDJk
C
-aS_ - .-
q T T
~ [ 0 . 2 0 ] -tx
where a: design ground acceleration; f3o: spectral acceleration amplification fac-
tor for 5% damping, q: damping correction factor with reference value q =I for
5% viscous damping.
Values of the parameters S, ks: and kd'J. are given depending on the subsoil class
specified by shear wave velocity.
0) New Zealand desisn
....
The seismic design loading for wharf structures is based on those within the N.Z.
Standard NZS 4203: 1992 'General Structural Design Loadings for Buildings'.
The requirements in this standard relate to an earthquake that has a 10% proba-
bility of being exceeded in a 50 year design life (or for a seismic event with a
return period 0{475 years). The specific material properties inclusive of damage
criteria are set out in the concrete and steel standards. The concrete and steel stan-
dards are standards that have been fully developed to cater to the science of build-
ing design in NZ. These standards do not specifically include earth retaining or
wharf quay type of structures, but they do form the basis of the design.
There is a recognition by New Zealand designers of the need to refer to NZ
bridge design and/or relevant local or overseas standards to ensure that wharf
structures are adequately detailed to address seismic cracking. variation in pile
stiffness. effects on the structures of lateral spreading and/or liquefaction of
the adjacent reclamation. pile ductility. and the diaphragm action of the deck to
.... _ '_ , _ .._, .. ... _ _e- :.,. .. :- ,. -.; . ..
transfer loads between shorter stiff piles and longer more flexible piles. The
torsional effects on the structure from variations in pile length and stiffness are
also recognised.
Both the ultimate limit state and the serviceability limit state are applicable
within the standards. Damage criteria used will depend on the form of construc-
tion and also on the material used. The design spectra are defined within the
material standard and the response spectra are based on the seismic zone, the sub-
soil description, the fundamental period of oscillation of the structure and the
structural ductility factor.
. There is scope within the New Zealand design standards to vary performance
requirements to cater for specific client requirements regarding design life. For
. example, definition of performance requirements to restrict concrete crushing
and residual crack widths after design level and ultimate level earthquakes are not
prescribed. but are recognised as an issue to be considered due to the rapid dete-
rioration of exposed reinforcing in the highly corrosive port environment once
any seismic cracking has occurred. Assumptions related to any predicted addi-
tional maintenance following a seismic event, do need to be canvassed by the
designer as part of the design process.
T1.3 SEISMIC DESIGN PRACTICE FOR PORT STRUCTURES AROUND
THE WORLD
The seismic design codes and guidelines reviewed in the previous section suggest
the following.
Since the mid-1990s, the dual level approach has been accepted as the new
common basis in seismic design. The conventional, single level approach
remains to be used for low design accelerations or structures with low seismic
capacity demands;
in the dual level approach, acceptable damage levels are typically defined as:
>- no damage or serviceable under L1 motion;
>- repairable or short-term loss of serviceability under L2 motion;
- damage criteria to specify limit conditions in engineering terms, including
wall displacement and pile ductility demand, have been under development
and further studies may be needed to establish them;
- analytical procedures adopted for design, including the factor for specifying
the effective seismic coefficient for retaining structures and the response spec-
tra used for pile-supported structures, vary depending on the region;
liquefaction is considered a primary design factor for port structures.
The current state of seismic design practice may be in a rapidly changing transi-
tional state, from the conventional design toward the performance-based design,
the latter of which is fully discussed and illustrated in this book.
90 PlANe
It must be emphasized that there is a significant difference between the design
approach recommended in these guidelines and the conventional design concept.
In the conventional design concept, especially when referring to the simplified
analysis, an equivalent seismic coefficient is used as an input parameter repre-
senting adequately the ensemble of ground motions, and a factor of safety is
applied to detemiine the dimensions and" properties of the 'structure. In the pro-
posed approach, the design is based on the seismic performance of the structure
evaluated appropriately through response analysis for a variety of input earth-
quake motions. The ensemble of the seismic responses, rather than the ensemble
of input motions, is used as a basis for accomplishing the design in the proposed
guidelines. For each response analysis, input parameters most appropriate are
those well defined in terms of applied mechanics, such as a peak ground accel-
eration for the simplified analysis and/or an equivalent parameter clearly defined
in terms of peak ground acceleration. Consequently, no factor of safety should be
applied to input data used in seismic analysis for evaluating the threshold level of
the structure. This important distinction between conventional design and per-
formance-based design should also be borne in mind when interpreting the
design guidelines included in the various seismic codes.
Te2: Case Histories
T2.1
T1.2
T2.3
T2.4
T
') -
_.)
T2.6
Damage to Gravity Quay Walls
Damage to Sheet Pile Quay Walls
Damage to Pile-Supported Wharves
Damage to Cellular Quay Walls
Damage to Cranes
Damage to Breakwaters
Case Histories
Selection of well-documented case histories of damage to port structures
(1980-1999) is presented for the events listed in Table T2.1. The sequential num-
bers 1 through 29 shown in this table may be used as indexes in referring to the
relevant case histories shown in Tables T2.2 through T2.30 in this technical com-
....
mentary. These case histories illustrate various seismic effects on port structures.
T2.1 DAMAGE TO GRAVITY QUAY WALLS
As discussed in Section 4.1 in the Main Text, a gravity quay wall is made of a
caisson or other gravity retaining structure placed on seabed. Stability against
earth pressures from the backfill soil behind the wall is maintained by the mass
of the wall and friction at the bottom of the wall. For gravity quay walls on firm
foundations, typical failure modes during earthquakes are seaward displacements
and tilting (Tables T2.3 through T2.5). For a loose backfill or natural loose sandy
foundation, the failure modes involve overall deformation of the foundation
beneath the wall, resulting in large seaward displacement, tilting and settlements
(Table T2.7). A wall with a relatively small width to height ratio, typically
less than about 0.75, exhibits a predominant tilting failure mode rather than
horizontal displacements, resulting, in the extreme case, in the collapse of the
wall (Table T2.2).
The evidence of damage to gravity quay walls (Tables T2.2 through T2.9)
suggests that:
I) most damage to gravity quay walls is often associated with significant defor-
mation of a soft or liquefiable soil deposit, and, hence, if liquefaction is an
issue, implementing appropriate remediation measures against liquefaction
may be "an effective approach to attaining significantly better seismic per-
formance;
2) most failures of gravity quay walls in practice result from excessive deforma-
tions, not catastrophic collapses, and, therefore, design methods based on
displacements and ultimate stress states are desirable for defining the
comprehensive seismic performance; and
94 PlANe
3) overturning/collapse of concrete block type walls could occur when tilting is
excessive, and this type of wall needs careful consideration in specifying
damage criteria regarding the overturning/collapse mode.
T2.2 DAMAGE TO SHEET PILE QUAY WALLS
As discussed in Section 4.2 in the Main Text, a sheet pile quay wall is composed
of interlocking sheet piles, tie-rods, and anchors. The wall is supported at the
upper part by anchors and the lower part by embedment in foundation soil.
Typical failure modes during earthquakes depend on structural and geotechnical
conditions, as shown in Tables TI.IO through T2.15. The evidence of damage to
sheet pile quay walls suggests that:
1) most damage to sheet pile quay walls is often associated with significant earth
pressures from a soft or liquefiable backfill soil (Tables T2.10 and T2.12),
and, hence, if liquefaction is an issue, implementing appropriate remediation
measures against liquefaction may be an effective approach to attaining
significantly better seismic performance (Tables T2.11 and T2.13);
2) if the anchor is embedded in competent soil, most failures of sheet pile quay
walls occur at the sheet pile wall due to excessive bending moment (Tables
T2.10 and T2.12), whereas, if the anchor is embedded in liquefiable soil, fail-
ures occur at the anchor, resulting in overall seaward movement and tilting of
the sheet pile wall and anchor system; and
3) structural damage to a sheet pile quay wall is governed by stress states rather
than displacements. It is therefore important to determine the preferred
sequence and degrees of ultimate states that may occur in the composite sheet
pile quay wall system.
T2.3 DAMAGE TO PILE-SUPPORTED WHARVES
As discussed in Section-l.S in the Main Text, a pile-supported wharf is composed
of a deck supported by a substructure consisting of piles and a rock dike/slope.
The unsupported pile length above the dike/slope surface is variable. When rock-
fill suitable for construction of the dike is uneconomical. a gravity or sheet pile
retaining structure is also constructed to replace a portion of the dike. The
seismic response of pile-supported wharves is influenced to by
interaCliQ!l.Q!!ring gr91!D.c! Typical failure modes
during earthquakes depend on the magnitude of the inertia force relative to the
ground displacement. In particular. the failure associated with the ground dis-
placement shown in Tables T2.16. T2.] 7, TI.20, and T2.22 suggests that specific
design consideration is required in the geotechnical aspect of seismic design.
Case Histories 95
Among various structural components in the pile-supported wharves. special
attention should be paid to the application of batter piles. Batter piles remain the
most efficient structural component for resisting lateral loads due to mooring.
berthing and crane operation. However, the batter pile-deck system results in a
much more rigid frame than one with vertical piles. As a result. large stress con-
centrations and shear failures of concrete batter piles have been observed during
earthquakes (Tables T2.16 and T2.18). This mode of pile failure demonstrates
that the structural design of batter piles used in regions of high seismicity must
account for displacement demand and appropriate ductility.
Structural damage to a pile-supported wharf is governed by stress/strain
. states rather than displacements. It is important to determine the preferred
sequence and degrees of ultimate states to occur in the composite pile-supported
wharf system. Most pile failures are associated with liquefaction of soil which
can result in buckling of the pile, loss of pile friction capacity, or development of
pile cracking and hinging. Hinging may be at the cap location or at an
interface between soil layers of differing lateral stiffness (Table T2.20). Damage
to piles in the soil may also occur about I to 3 pile diameters below grade in
non-liquefying soils due to the dynamic response of the wharf (see Technical
Commentary 5).
T2.4 DAMAGE TO CELLULAR QUAY WALLS
As discussed in Section 4.4 in the Main Text, a cellular quay wall is made of a
steel plate or sheet pile cell with sand or other fill. Resistance against inertia
forces and earth pressures is provided by the fill friction at the bottom surface of
the cell for a non-embedded cell, or by the resistance of the foundation subsoil at
the cell embedment. Typical failure modes during earthquakes depend on cell
embedment and geotechnical conditions, resulting in varying degrees of seaward
displacement, settlement and tilting. In addition to the gross deformation, warp-
ing of the cellular cross section can occur (Table T2.23). Such performance of
cellular quay walls should be considered in the seismic design criteria.
T2.5 DAMAGE TO CRANES
As discussed in Section 4.5 in the Main Text, a crane consists of an upper struc-
ture ~ o r handling cargoes, and a supporting structure for holding in place and
transporting the upper structure. The crane is generally made of a steel frame.
The supporting structure is either a rigid frame type or a hinged leg type,and
rests on rails through the wheels.
96 PlANe
A crane at rest is fixed to rails or to a quay wall with clamps or anchors. The
strengths of these clamps or anchors provides the upper limit resistance against
external forces acting on the crane. A crane in operation, however, is not supported
with clamps or anchors. The lateral resistance of the crane against external forces
during operation is only due to friction and resistance from the wheel flanges. ..
Typical failure modes during. earthquakes are derailment of wheels, detach-
ment or pull-out of the vehicle, rupture of clamps and anchors, buckling, and
overturning. Widening of a span between the legs due to the deformation of the
quay wall results in derailment or buckling of the legs (Table T2.26). Conversely,
narrowing of a leg span can also occur due to the rocking response of the crane
(Table T2.24). This is due to an alternating action of the horizontal component of
resisting forces from the quay wall during a rocking type response involving
uplifting of one side of the legs. The rocking response can also result in the
derailment, leading to tilting. If the tilting becomes excessively large, overturn-
ing of the crane may result (Tables T2.25 and T2.27). The overturning typically
occurs toward land because of the location of the center of gravity of the cranes
at rest, and differential settlement on the crane foundation associated with the
deformation of quay walls.
Crane rails are often directly supported either by a portion of a retaining wall,
or by the deck of a pile-supported wharf. When the width of the gravity wall is
small, or the quay wall is a sheet pile or cellular type, a separate foundation often
consisting of piles is provided to support the rails. In order to achieve desirable
seismic performance of quay walls with cranes, special consideration is required
for the rail foundation. One such consideration could include providing a dedi-
cated and cross-tied upper structure to support the rails.
T2.6 DAMAGE TO BREAKWATERS
As discussed in Section 4.6 in the Main Text. a breakwater is usually made of a
. J
rubble mound, a caisson, or a combination of both placed on the seabed. Stability
against a horizontal external load is maintained by shear resistance of the rubble.
friction at the bottom of the caisson, and with associated resistance to overturn-
ing and bearing capacity failure. Typical failure modes during earthquakes are
settlement associated with significant foundation deformation beneath the
caisson/rubble foundation (Table T2.28). It should be noted that even with
moderate earthquakes, failure can happen due to extremely soft clay deposits and
large seismic amplification within and beneath the rubble mound (Table T2.29).
Damage to breakwaters from a tsunami typically results in the caissons being
washed off or tilting into the rubble mound (Table T2.30). If the performance
objective of a breakwater includes preventive measures against a tsunami. appro-
priate design criteria need to be established with respect to both the earthquake
motion and tsunamis.
Case Histories 97
Table T2.1. Selected case histories of damage to port structures (1980-1999).
No. Port structure Earthquake Port
year (magnitude")
1 Gravity Block 1985 (A,t =7.8) San Antonio Port Chile
")
quay 1986 (.H\ =6.2) Kalarnata Port Greece
3 wall 1989 (M =6.0) Port of Algiers Alseria
c
4 Caisson 1993 UII!l =7.8) Kushiro Port Japan
5 1993 (M
l
=7.8) Kushiro Port Japan
6 1995 (M
l
=7.2) Kobe Port Japan.
7 Block 1999 (Mw=7.4) Derince Port Turkey
8 Caisson 1999 (M
s
=7.7) Taichung Port Taiwan
9 Sheet pile quay wall 1983 (Ml=7.7) Akita Port Japan
10 1983 (M
I
=7.7) Akita Port Japan
11 1993 (M
l
12 1993 (M
l
13 1993 (M
l
=7.8) Hakodate Port Japan
14 1993 (M
s
=8.1) Commercial Port, Guam USA
15 Pile- RC pile** 1989 (.W
s
= 7.1) Port of Oakland USA
16 supported 1990(M
s
=7.8) Port of San Fernando Philippines
17 wharf PC pile** 1994(M
s
= 6.8) Port of Los Angeles USA
18 1994 (M
s
=6.8) _Port of Los Angeles USA
19 Steel pile 1995 (M
l
=7.2) Kobe Port Japan
20 PC pile 1995 (M
s
= 7.2) Port of Eilat Israel
21 Dolphin Steel pile 1995 (lV!l =7.2) Kobe. Sumiyoshihama Japan
22 Cellular quay wall 1995 (M
l
=7.2) Kobe Port Japan
23 Crane 1983 (M
l
=7.7) Akita Port Japan
24 . 1985 (M
s
=7.8) San Antonio Port Chile
25 1995 (Ml =7.2) Kobe Port Japan
26 1999 (Mw =7.4) Derince Port Turkey
27 Break- Composite 1995 (Ml =7.2) Kobe Port Japan
28 water Rubble 1984(M =4.5) Patras Port Greece
29 Composite 1993*** (M, =7.8) Okushiri Port Japan
* For definition of magnitude. refer to Technical Commentary 3.
** RC: Reinforced concrete; PC: Prestressed concrete.
*** Damage due to Tsunami.
98 PlANe
Table T2.2. Case history No.1: San Antonio Port, Chile. (Ms =-7.8)
Structural Structure and location: Berths 1 & 2
and seismic Height: + 4.9 m
conditions Water depth: - 9.6 m
Construction (completed): 1918-1935
Earthquake: March 3, 1985 Chile earthquake
Magnitude: M, =7.8
Peak ground surface acceleration (PGA): 0.67 g
PGA evaluated based on strong motion earthquake records.
Cross section: Block quay wall
6.0
'9
'4
18
15
Scal. level I, T
1m) 1m) Soil type I
SPN
value
.4 7ft t, or. >n ,n 4n .. n
o - Sand and I"':
1 gravel fill :, ! :
2 2.S! : : : i :
3 1.'0 Coarse sane ;::.::
: : :
, If' I
5 Fine sand :,:, 1 '.: i, ;
6 '.20 1'9: :
Medium to ::: I
coarse sane ' ,
8 3.65 r:: :
9 '1 i
Clay 10",
10 I: :
l' :::
12 .7.50 witn silly 1m... 5: ,'o?
F'lne sane I
13 8.65 "" h stone ::: "PO
''l II.... I
. .
I I I
Fmeto : : :
medium sand iii : :
16 -12.00 : : I ';'0'
17 I I
I : I :
Fine sand ' .,
, : : ,pO;
(Unit: m)
Hydraulic fill
9.0
3.84 1.0
I..
0.16 1.0
o
N
-
+4.9
';7 000
-9.6
Concrete Backfill
blocks --+I..
r-.-..........,---t (5-100kg stones
with gravel)
+1.68
MS.l +0.9
_ ......
Boring log at the backfill
Features of damagelcomments
Collapse of block type quay wall over 60% of wharf length (452 m) due to strong earth-
quake motion and backfill liquefaction.
References: Tsuchida et al. (1986): Wyllie et al. (1986)
Structural
and seismic
conditions
Case Histories 99
Table T2.3. Case history No.2: Kalamata Port. Greece.
Height: + 2.1 m
Water depth: - 9.5 m
Earthquake: September 13, 1986 Kalarnata earthquake
Magnitude: M, = 6.2
Peak ground surface acceleration (PGA): 0.2 to 0.3 g
Peak ground surface velocity (PGV): 0.4 mls
PGA/PGV evaluated based on 1D & 2D equivalent linear analysis.
Seismic Horizontal displacement: 0.15 0.05 m
performance Tilt: 4 to 5 degrees
000
--------- .... -_... -....... ----
--_.. ----
3.0m
~
SG
(Unit: m)
SPT N-value
10 20 30 40 50
Fill
Index: SM=silty sand, SG=gravely sand
Features of damage/comments
Seaward displacement with tilt of block type quay wall constructed on firm foundation.
The backfill settled 20 cm behind the wall with no settlement at a distance of 30 to' 40
meters away. The quay wall remained serviceable.
Reference: Pitilakis and Moutsakis (1989)
100 PIANC
Table T2.4. Case history No.3: Port of Algiers, Algeria.
. Structural Structure and'location: Quay No. 34
Earthquake: October 29, 1989 Chenoua earthquake
Magnitude: M =6.0, Epicenter located 60 km to the west of Algiers
Seismic Horizontal displacement: 0.5 m
performance Vertical displacement: 0.3 m
-10.00
sz L.W.L. 0.00
""F"
o
"<:T
c:i
- - ~ - - - - - -
~ ~
Fill
Unit (m)
During this earthquake, 8.000 buildings were destroyed; there were ground ruptures,
landslides, and the coastal road was closed by falling rock. No additional detailed infor-
mation is available on the earthquake.
Reference: Manja (1999)
Case Histories 101
!able T2.5. Case history No.4: Kushiro Port. Japan.
. Seismic
performance
Structural
and seismic
conditions
-Structure and location: West No. I. West Port District
Height: + 3.0 m
Design seismic coefficient: k
h
= 0.20
Construction (completed): 1975
Earthquake: January 15. 1993 Kushiro-Oki earthquake
Magnitude: M
J
= 7.8 (1MA)
Peak ground surface velocity (PGV): 0.63 m1s
PGAlPGV evaluated based on strong motion earthquake records.
Horizontal displacement: 0.75 m
Vertical displacement: 0.2m
Tilt: 2%
Cross section: Caisson quay wall
SPT.N value
oto203040 SO
Silty sand
Sandy silt

S.ncly Itt
Fine sand
Fine sanCl
Fine sand
Fine Sand
Coarse sand
Coarse sand
Coarse sand
SOil type
Co..... Unci
mixed'" r.v"
0 sand
- - - Before earthquake
- After earthquake
15.00
Unit (m)
:!'-
r
I
\ \ .
H.W.l.+1.5 _ .....
V l.W.L.+O.O -- I-- I "
.0.70 I "
1- ..... T
I
I

I Caisson I .....
I
L 8 H ...
I 5.0X12.5xl0.5 IRubblestone ""
I I .".
..... I I.".""
..... .z...< I ,."....
..... -9.0 ,1- _ 1.... .........
......... -- - - -........ .".
' .....
Seaward displacement of caisson type quay wall constructed on firm foundation.
Liquefaction of backfill.
Reference: Iai et al. (1994)
102 PIANC
Table T2.6. Case history No.5: Kushiro Port, Japan.
Structural Structure and location: East Quay, Kita Wharf, East Port District
b
=0.15
Earthquake: JanuaryIS, 1993 Kushiro-Oki earthquake
Magnitude: M
J
= 7.8 (JMA)
performance Vertical displacement: 0.2 to 0.5 m
7.27 "\
\7+2.50 /"
Coarse
sand
Coarse
sand
Find
sand
Sandy
gravel
Gravelly
sand
Sandy
.20 ~ r a v e l
'0
Elevation(m)
Fill
Coarse
o sand
(Unit: m)
)7+0.80
: r-,
: t "..
~ : '''"
.: -,
. '.
.. "'<..
!
.
.
.
:.
!
, .99 !
'\7 H.W.l. +'.50 +1.50
l.W.l. :0.00
Seaward displacement of caisson type quay wall constructed on firm foundation.
Liquefaction of backfill. The displacement was larger presumably because of a smaller
width to hei.ght rgJio_thalLIDe case. history shown in the previous page (i.e. West No. 1
quay wall. West Port District: Kus-hffO Port, Japan).
Case Histories 103
Table T2.7. Case history No.6: Kobe Port. Japan.
Structural
and seismic
conditions
Seismic
perfonnance
Structure and location: Rokko Island. RC5
Height: + 4.0 m. Water depth: - 14.0 m
h
=0.15
Earthquake: January 17. 1995 Hyogoken-Nambu earthquake
Magnitude: M
1
= 7.2 (JMA)
PGAfPGV evaluated based on strong motion earthquake records.
Horizontal displacement: 4.2 to 5.2 m
Vertical displacement: 1.5 to 2.2 m
Tilt: 4 to 50
Unit (m)
Alluvial clay
layer
Backfill soil
'V-.o
24.00
Sand drain
v -33.00--35.00
30.50
'V-34.00--36.00
Backfilling sand
for repladng clay layer
7.0t
4oO
I 1300
20.0-17.0
7-14.50
Alluvial cl.ay layer
I

'-
..
~ .
Ie -
0
0
5 10 15 20 25 30 "d: residual honzontal
Thicknessof Backfill Sand displacement (rn]
for Replacing Clay Deposit (m) H: wall height 1m)
50 Features of damage/comments
Loose foundation soil below the wall
induced significant seaward dis-
placement and tilting, associated
with the deformation in the foundation
soil. Horizontal displacement of
caisson walls is closely correlated
with thickness of loose foundation
deposit as shown in the right figure.
Reference: Inagakiet al. (1996)
J04 PlANe
Table T2.8. Case history No.7: Derince Port, Turkey.
Structural Structure and location: No. 6 quay wall
Earthquake: August 17, 1999 Kocaeli earthquake
Magnitude: M... = 7.4 (USGS)
performance
Inside broken lines: Presumed by referring to typical quay walls in Turkey

+2.25m \ -----
------
_...... ---
6.5m
Rubble Mound
v+3.00m

+O.75m
O.OOm
The concrete blocks are put in
oblique to the horizontal in
Turkey pons as shown in the
figure right (a front view of a
block type quay wall in Turkey)
(Unit: em)
Reference: Suaano and lai (1999)
....
Case Histories 105
Table T2.9. Case history No.8: Taichung Port. Taiwan.
Structural
and seismic
conditions
Structure and location: No. 1 Quay Wall
Height: + 6.2 m
Desizn seismic coefficient: k
h
= 0.15
Earthquake: September 21. 1999 Ji-Ji earthquake
Magnitude: M, = 7.7
\
Native soils
i
Conveyor - - - - - - - - - -.,.
---- \
_------ f
l
"
\
- \
\
\ .
\
I
20.60
26.60
8.00
f.o,
s

SZ EL. O.OOm
EL.1300m
Displacement of caisson quay wall constructed on firm foundation. Liquefaction of back-
.; ' fill. Settlement at backfill was 1m maximum.
Reference: Sugano et aL (1996)
I.
: .- ' ..
r.
106 PlANC
Table T2.1O. Case history No.9: Akita Port, Japan.
Structural Structure and location: Ohama No.2 Wharf
h
= 0.10
Earthquake: May 1983 Nihonkai-Chubu earthquake
Magnitude: M
J
=7.7 (JMA)
. Seismic Horizontal displacement: 2.0 m (at the top of the wall)
performance Vertical displacement: Q.-3 to 1.3 m
Cross section: Sheet pile quay wall
Unit (m)
EIe

SPT NV.tu.
v.tton
.m) 102030.0 1O0
l:I ilL
.3 m
0
't:
-5 :":
-10

-15 1\........
:.::.::.:
:"::::':.: l-
.20 ... Id
fb: t-
::tl::::t
_--,+1.2
l:soM I
::!::
. 1'
::V')::
.. -f'
If Ill'

::e::

.... .
I'-'-
II .. ,
::e::
:;e::
0,1
:.::g: :
::--&: :
,. I'
I I 41)1 ,
.,-"
: (c.: :
CIo''
a.
Q.
Ci
tI>
ii5
2O.0m
Backfill sand
\tI steel tie rod
u'gh tensile streng
nl . . n 2 om
ttl 55mm. spaelll. .
+2.0 }_.- .._u --""""r---_
l.W.L :to.0"-:: ..-----
-+ -
Opening of a crack-
The sheet pile wall suffered significant damage due to liquefaction of backfill.
Reference: lai and Kameoka (1993)
Case Histories 107
Table T2.1 I. Case history No. 10: Akita Port, Japan.
Structural
and seismic
conditions
Structure and location: Ohama No. 1 Wharf
Height: + 2.0 m
Water depth: - 10.0m
h
=0.10
Earthquake: May 26, 1983 Nihonkai-Chubu earthquake
Magnitude: M
J
= 7.7 (JMA)
PGAJPGV evaluated based on strong motion earthquake records.
Seismic Horizontal displacement: 0 m (no damage)
performance Vertical displacement: 0 m (no damage)
; +0.00 high tensile strength steel tie rod.li: IV
. .
"\
20.00
r
0-
I
-
:J
x
$=
0
It)
I
0.

CIl
!!:..
s

'a
at
, 0.00

&:
.:.!.QJ..Q... U.

..
..
CIl
.5'22
-30
0442
4445
(Unit: m)
Although the cross section of this sheet pile quay wall is similar to that of Ohama No.2
Wharf (shown in the previous page), this quay wall suffered no damage. No signs of liq-
uefaction were evident at this quay wall. Occurrence or non-occurrence of backfill lique-
faction resulted in the significant difference of seismic performance of sheet pile quay
walls.
Reference: Iai and Kameoka (1993)
108 PlANe
Table T2.) 2. Case history No. 11: Kushiro Port, Japan.
Structural
and seismic
conditions
Structure and location: South Quay Wall. Fishery Port, East Port District
Height: + 2.7 m
Design seismic coefficient: k = 0.20
Earthquake: January 15, 1993 Kushiro-Oki earthquake
Magnitude: MJ = 7.8 (JMA)
Seismic Horizontal displacement: 0.4 m (at the top of the wall)
Soil type I SPTN value
o 10 20 30 ~ o se
Unit (m)
- - - Before earthquake
- After earthquake
Sheet pile Original ground
It'!
CD
-h
-7.5 ~
, L.W.L.O.O
+2.7
, HoW.L. .. , 05
Liquefaction of backfill resulted in opening of a crack in the sheet pile wall at an elevation
of-4mo
Case Histories 109
Table T2.13. Case history No. 12: Kushiro Port. Japan.
Seismic
performance
Structural
and seismic
conditions
Structure and location: South Quay, West No. 1 Wharf, West Port District
Height: + 3.0 m
h
=0.20
Earthquake: January 15, 1993 Kushiro-Oki earthquake
Magnitude: M
1
=7.8 (JMA)
Peak ground surface acceleration (POA): 0.47 g
Peak ground surface velocity (POV): 0.63 mls
POAJPOV evaluated based on strong motion earthquake records.
Horizontal displacement: 0 m (no damage)
Vertical displacement: 0 m (no damage)
Tilt: 0 (no damage)
SOil type
[
E..... bOn]

Coars. sand

.' mi.ed with
grave.
5
10
15
Unit (11:l)
I 34
30.0
1.5
20,0
2%
.

'i
in
-22.7
+3.0 n
+
H.WL +1.5 if r
.
Gravel mal an '\n
"
+2.0
L.W.L. ... O.O

-,
.' Tie-rod etc 1.98
C!
<II
...
"
:i:!:
., Gravel drain a..
Rubble -
-a.
: : m
....
'.
.-
&:-
..
"'TV", v"' v.,. -.
_______:Z.5..._____
;;
E :. Sand como.clion oile

&:
u
c:
-.:
4l:700mm -c
'P
'" ....:.1l:.L -12.0
;;;
..
II
...
'&
E
_ E
401.5 401.7

- aefor. compaction
--. Aher compaction
The quay walls with remedial measures against liquefaction suffered no damage.
Reference: lai et aI. (1994)
110 PIANC
Table T2.14. Case history No. 13: Hakodate Port, Japan.
Structural Structure and location: No.6 Quay Wall, Benten District
h
= 0.15
Earthquake: July 12, 1993 Hokkaido-Nansei-Oki earthquake
Magnitude: M
J
= 7.8 (JMA)
Peak ground surface acceleration (pGA): 0.12 g
Tilt: 15
1 n
....
... or;:
20 .. ' #1#
.0 t7 t'"
:. too
": ,:r.;. r-
30 ,. .',
., ::I(
..
'
'0 ::.' I
Depth(m) SPT N-value
o 0 204060

10 / )<
J
+2.3
+1.0
-6.0
Steel sheet pile
SSP IA
Semihigh tensile
steel tie rod
<I> =42mm etc 1.6m
&
+2,3 -, I
,..-r-----------+---=.;;,.---
H.W.L. +1.05
L.W.L. 0.00
-14.0
Unit (m) except for
etc: tie-rod spacing
(centre to centre)
, .0 ; f'
..'

:., -,
... "
. :..
70
':;:.H II
Despite the small PGA. in the order of 0.1 g, significant deformation occurred due to
liquefaction of backfill as well as foundation soil.
Reference: Inatomi et al. (1994)
Case Histories III
Table T2.15. Case history No. 14: Commercial Port. Guam. USA.
Structure and location: Berths F3 through F6
Height: + 2.7 m
Earthquake: August 8, 1993 Guam earthquake
Maznitude: M, =8.1
- .
Horizontal displacement: 0.6 m Seismic
performance
Structural
and seismic
conditions
(Unit: m)
Hyllraulic Fill
2440
Steel
Sheet
Pil ..
Temporary till remqved
after backfilling of
bulkhead and tie rods.
0.00 MLLW ,,--
_.. _.. _- --------.
"
"
"
"
"
"
"
"
-- "
"
"
"
. 1 ~ t M L L W ...:':- _
"
This quay wall was constructed using a three-step dredging sequence. The last step is
depicted in the figure.
References: Comartin (1995); Mejia and Yeung (1995); Vazifdar and Kaldveer (1995)
112 PIANC
Table T2.16. Case history No. 15: Port of Oakland, USA.
Structural Structure and location: 7th Street Terminal wharf
conditions Water depth: - 1].0 m
Earthquake; October 17-, 1989 Lorna Prieta earthquake
Magnitude: M, = 7.1 .
Seismic Horizontal displacement: 0.] m
performance
Cross section: Pile-supported wharf
(41ern square section reinforced concrete piles)
+6
..3
0
3
m
CD
<
III
6
s
:>
'3
9
,..
11
,..
12
-15
18
21
"
"
"
"
"
C(
Crane rail
Unit (rn)
"
"
"
"
"

"
:.:
..
..
v
..
:..
<t
Crane rail
Dense sands
Stiff silty clays and dense sands
V
....- . e,
Hydraulic sand fill ./4 Rock fill
/.&. --.-"*-. -
./. Hydraulic sand fill

Tensile failures at or near the tops of most of the rearmost line of batter piles (inclined
inboard). Evidence of liquefaction of the backfill. Hydraulic sand fill below the dike (i.e.
rock fill) is presumed to be the major geotechnical cause of damage.
References: Benuska (1990); Egan et al. (1992)
Table T2.17.
Structural
and seismic
conditions
Case Histories 113
Case history No. 16: Port of San Fernando. Philippines.
Structure and location: Government Pier No. 1
Earthquake: July 16. 1990 Luzon. Philippines. earthquake
Magnitude: M, =7.8
Seismic Horizontal displacement: Displacement along the longitudinal direction
performance of the pier toward the sea. Many gaps opened in the longitudinal direction
of the concrete deck. including a 0.7 m gap. Summation of all the gaps
amounted to about 1.5 m at the seaside end of the pier.
Cross section: Pile-supported pier (square reinforced concrete piles)
200 m long. 19 m wide pier that consists of a concrete deck supported
by vertical and battered concrete piles.
Om
o'
k
::
0'
k
::
.'
0'

"

..
....
..
..
..
.,
..
.,.
200m

Schematic figure. Arrows indicate presumed lateral deformation of the foundation ground.
Extensive opening of cracks on the pier deck. Cracks and chipping at pile caps. exposing
reinforcing steel bars. which were either bent or cut.
Damage was presumably caused by lateral deformation of the foundation ground as illus-
trated above.
Reference: Japan Society of Civil Engineers (1990)

1.9
4i
t;qts"
- --.
915
."
1]4 PIANC
Table T2.18. Case history No. 17: Los Angeles Port, USA.
Structural Structure and location: B127, APLTerminal
Construction (completed): 1960s
Eanhquake: January 17, 1994 Northridge earthquake
Magnitude: M, =6.8
PGA evaluated based on strong motion earthquake record.
Cross section: Pile-supported wharf (46 cm prestressed octagonal concrete piles)
20.28
1%
1.68
15.25
Ballast
Concrete slab
Pull-out or chipping of concrete off batter piles at pile cap.
References: Hall ( 1 9 9 5 ) ~ lai and Tsuchida (J 997)
(Unit: m)
Case Histories 115
Table T2.19. Case history No. 18: Los Angeles Port. USA.
Structure and location: BI 26(New). APL Terminal
Height: + 4.5 m
h
= 0.12. 0.04 in normal direction
Earthquake:January 17. 1994 Northridge earthquake
Magnitude: M, = 6.8
PGA evaluated based on strong motion earthquake record.
Horizontal displacement: 0.1 m
Structural
and seismic
conditions
Seismic
. performance
Cross section: Pile-supported wharf (61 cm prestressed octagonal concrete piles)
El 13 13
.,
Fender
U.S. pierhead
T
EL
.
O
.
O
Timber fender pile 61cm octagonal
prestressed
conc. pile I
25.62
Bulkhead line
& edge of conc.
(Unit: m)
Liquefaction of backfill.
Hairline crack at pile cap of most landward row of piles.
Horizontal displacement of 8 cm relative to B126(Old) at an expansion joint caused
damage to crane rails.
References: Hall (1995); Iai and Tsuchida (1997)
116 PIANC
Table T2.20. Case history No. 19: Kobe Port, Japan.
Structural Structure and location: Takahama Wharf
Design seismic coefficient: kh = 0.15
Earthquake: January 17, 1995 Hyogoken-Narnbu earthquake
Magnitude: MJ =7.2 (JMA)
Seismic Horizontal displacement: 1.5-1.7 m
performance Tilt: 3%
Cross section: Pile-supported wharf (D; = 700 mm, t
p
=10, 12, 14 mm, steel piles)
Horizontal
displacement
1.3-1.7m
+3.6
i HW.L. +1.6
L.W.L. +0.0
Pleistocene cia 1
Pleistocene gravel (2)
Pleistocene clay (2)
Cracking at
pilecap
- After earthquake
- - Before earthquake
Local bUckling
(Unit in meters)
of steel Eiles was found not only at the pile caps in the embedde<i12Q-
.JiQ.o. The damage was associated with significant deformation of alluvial sand layer
below the rubble mound/dike.
Reference: Iai (1998b)
Structural
and seismic
conditions
Case Histories 117
Table T2.2l. Case history No. 20: Port of Eilat. Israel.
Structure and location: Main Wharf
H
h ..,-
elg t: + _..) m
Earthquake: November 22. 1995 Earthquake
Magnitude: M, =7.2
The port was located more than 100 km away from the epicenter.
Cross section: Pile-supported wharf (46 em octagonal prestressed concrete pile)
Existing slope (Inspection 1976)
Existing slope (Inspection 1980)
Unit (m)
(LOader rail beam
11
II
11
5 to 15 nun displacements at the expansion joints. No damage to piles.
Reference: Buslov (1996)
1]8 PIANC
Table T2.22. Case history No.2]: Kobe, Japan.
Structural Structure and location: Sumiyoshihama district
and seismic Height: + 5.5 m, Water depth: - l2.5 m
conditions Design seismic coefficient: kh = 0.15
Construction (completed): 1982 .
Earthquake: January 17, 1995 Hyogoken-Narnbu earthquake
Magnitude: M
J
= 7.2 (JMA)
Peak ground surface velocity (PGV): 1.06 m/s
Seismic Horizontal displacement: 0.1 to 0.6 m landward
performance Tilt: 2 to 5 landward
Cross section: Pile-supported dolphin (steel piles) 160 m long and ]2.5 m wide
Moorage
face line
('
V +5.50
HW.L. +1.67
LW.L KP+O.
I
I
I
-12.00 -1140 I
__
" .
--1
25.0
117
Seawall
face line
("
.
- - - Beforeearthquake
- After earthquake
mudline elevation
surveyed earthquake
(Unit: m)
Steel piles buckled near the foundation. Lateral deformation was about l.Omat the buck-
ling point.
Retaining wall: horizontal displacement 1.5-1.7 m seaward. vertical displacement 0.6 to
l.Om. tilt: 3 to 4 seaward.
Damage to the pile-supported jetty was not only due 10 the seismic inertia force in the
deck but also due to lateral displacement of foundation soil toward the sea.
Reference: Nishizawa et al. (I 998)
Case Histories 119
Table T2.23. Case history No. 22: Kobe Port. Japan.
Structural
and seismic
conditions
Structure and location: Maya Wharf No. 1
Height: + 4.0 m, Water depth: - 12.0 m
h
= 0.15
Earthquake: January 17. 1995 Hyogoken-Nambu earthquake
Magnitude: M
J
= 7.2 (1MA)
Seismic Horizontal displacement: 1.3 to 2.9 m
Tilt: 11
0
(max)
Cross section: Steel plate cofferdam type
(Unit: m)
- - - . Before earthquake
- After earthQuake
Plan
Before earthguake
After earth uake
I Cell
'II II
"I '" Steel pile
'.1 II
Steel sheet pile

II J. II
II Sand fill II ",'"
W II ",
..-:-.--- 1lJ , '" 1\1 "'0
.. -14.50 /' v
;::.:t:r- - - - - - -
Replacement sandl: -------I
3.05 _ 1845 ...
Steel pile

HW.L. +1.70
i LW.L. 0.00
Note the warping of the cellular cross section as shown above.
Reference: Suganoet al. (1998)
Structural
and seismic
conditions
JLU PlANC
Table T2.24. Case history No. 23: Akita Port, Japan.
Structure and location: Nakajima No.2 Wharf (sheet pile quay wall)
Earthquake: May 26, 1983 Nihonkai-Chubu earthquake
Magnitude: M
J
= 7.7 (JMA)
The crane derailed and inclined 20 landward.
Cross section: Crane on sheet pile quay wall
About 20
-'.-,.
Vehicle detached from the leg
I...
12.5m
..,
Hinged leg
Landward rail Seaward rail
Due to the rocking response of the crane, the crane derailed. The derailed legs were
landed on liquefied backfill of a sheet pile quay wall, resulting in a 20 landward
inclination.
Reference: Tsuchida et al. (1985)
Structural
and seismic
conditions
Case Histories 121
Table T2.25. Case historv No. 24: San Antonio Port. Chile.
Structure and location: Berths 1 & 2
Earthquake: March 3. 1985 Chile earthquake
Magnitude: M, = 7.8
Cross section: Crane on gravity quay wall (rail span =5.25 rn)
Tilting and overturning of cranes due to rocking response and differential settlement at
the liquefied backfill.
Reference: Wyllie et aI. (1986)
122 PlANC
Structural Structure and location: Rokko Island, RC5 (Gravity quay wall)
and seismic Earthquake: January 17, 1995 Hyogoken-Narnbu earthquake
conditions Magnitude: M
J
= 7.2 (JMA)
Cross section: Container crane on gravity quay wall
116m
Land-side
-
I~
30m
Span
~ I
Sea-side
Buckling at upper portion of crane legs not only due to inertia force but also due to widen-
ing of the rail span associated with the movement of the caisson wall toward the sea.
References: Tanaka and Inatorni (1996): Scott (1997)
Structural
and seismic
conditions
Case Histories l13
Table T1.27. Case history No. 26: Derince Port. Turkey.
Structure and location: No.6 QuayWall (Gravity quay wall)
Earthquake: August 17. 1999 Kocaeli earthquake
Magnitude: M; = 7.4 (USGS)
Peak ground surface acceleration (PGAl: 0.2 to 0.25 g
Cross section: Crane on gravity quay wall (rail span = 5.5 rn)
Cranes derailed due to rocking response. Tilted toward the landside due to differential
settlement of backfill, resulted in collapse of one crane.
Reference:Sugano and lai (1999)
J24 PlANe
Structural
and seismic
conditions
Structure and location: Breakwater No.7
Height: + 5.0 m
Construction (completed): 1970-1972
Earthquake: January 17, 1995 Hyogoken-Nambu earthquake
Magnitude: MJ =7.2 (JMA)
Peak ground surface velocity (PGV): 1.06m/s
. performance Vertical displacement: 1.4 to 2.6 m
Cross section: Vertical composite type
Clay Clay
Unit (m)
- - . Before earthquake
.:-- After earthquake
~ . 6 4
'Or
CD
....5.00
.... 1
~ -= CWe OOO
JZ. HW.l. ...1.70
Settlement associated with significant foundation deformation beneath the caisson/rubble
foundation.
Case Histories 125
Table T2.19. Case history No. 28: Parras Port. Greece.
Structure and location: Southern extension
Height: + 2.0 m
Construction: 1984 (just completed)
Earthquake: Late February of 1984 Earthquakes
Magnitude: M =3.5 to 4.5
PGA evaluated based on 10 equivalent linear analysis.
Vertical displacement: 3.0 to 4.0 m Seismic
performance
Cross section: Rubble mound type
Structural
and seismic
conditions
Primary armour
Harbour
sz :0.00
.....
I..
8765
..I
/
(a) Typical cross section
Harbour
., .... _- - ....... ---
.; Natural seabe1 _ -- -- - -- - --- --
~ ~ ~ - - - - - - ' - ~ - - - - - - - - - -
I. 10m ..I
(Unit: m)
(b) Cross section before and after the failure
Abrupt settlements of the breakwater (3.0 to 4.0 m) took place after a series of moderate
earthquakes of magnitude 3.5 to 4.5. The damage was caused mainly by the large ampli-
fication (up to 2.5 times) of the accelerations within the mound.
The breakwater was constructed on a normally consolidated soft clay layer 30 to 38 m
thick with SPT N-values ranging from 0 to 15 and undrained shear strength ranging from
5 to 40 kN/m
2
, increasing with depth. It was underlain by a thick medium-stiff clay layer
with SPT N-value greater than 42 and undrained shear strength ranging from 120 to 250
kN/m
2
The breakwater was constructed in two stages; the level of the mount reached
11 m from its foundation at the first stage, and after a period of 6 months, the level
reached 18 m. Total settlement due to consolidation was about 3 m. .
Reference: Memos and Protonotarios (1992)
]26 PlANe
Table T2.30. Case history No. 29: Okushiri Port, Japan.
Structural
and seismic
conditions
Structure and location: North breakwater, Section C
Height: + 3.1 m
Construction: (completed): 1988-] 992
Earthquake: July 12, 1993 Hokkaido-Nansei-Oki earthquake
Magnitude: M
J
= 7.8 (JMA)
Cross section: Vertical composite type
8.0
7.8
lliocks
(Unlt:rn)
l e H
10X8 X6
5.0
+3.1
7.6
Concrete block
l e H
3x2.5 x 1
Rubble (30-300kgf)
Caissons were washed off or tilted into the rubble mound due to tsunami.
Reference: Inatomi et a1. (1994)
TC3: Earthquake Motion
T3.1
T
'"' '1
.:l._
T3.3
T3.4
T3.5
T3.6
Size of Earthquakes
T3.1.1 Intensity
T3.1.2 Magnitude
Strong Ground Motion Parameters
Seismic Source and Travel Path Effects
Local Site Effects
Seismic Hazard and Design Earthquake Motion
T3.5.1 Seismotectonic approach
T3.5.2 Direct approach based on seismic history
T3.5.3 Design accelerozrams
.... .....
Ground Motion Input for Seismic Analysis of Port Structures
T3.6.1 Gravity quay walls
T3.6.2 Sheet pile quay walls
T3.6.3 Pile-supported wharves
T3.6.4 Embankments and breakwaters
Earthquake Motion
Earthquakes are complex natural phenomena, with their origin in the release of
tectonic stress which has accumulated in the earth's crust. Their principal effects
on port structures are caused by oscillatory ground movement, which depends on
such factors as seismic source, travel path, and local site effects.
Each port. and to a certain extent each structure, requires a specific evaluation
of the design parameters of ground motion. This technical commentary outlines
the definitions of primary parameters, and makes recommendations pertaining to
the basic data to be collected and the analytical procedures to be followed in the
seismic design process.
T3.1 SIZE OF EARTHQUAKES
The basic parameters which characterise the size of earthquakes are intensity,
magnitude and energy release.
T3.1.1 Intensity
Intensity is a measure of the destructiveness of the earthquake, as evidenced by
human reaction and observed damage. It varies from one location to another,
depending on the size of the earthquake, the focal distance and the local site con-
ditions. Seismic damage and the corresponding intensity depend on characteris-
tics of seismic motion, (acceleration, duration and frequency content) as well as
the natural frequencies and vulnerability of the affected structures. Intensity is the
best single parameter to define the destructiveness of an earthquake at a given
site, but it cannot be used as input for dynamic analysis. In many cases, especially
for historic earthquakes, it is the only parameter available for characterising the
earthquake motion,
Several different seismic intensity scales have been adopted in different parts
of the world. The Modified Mercalli (MM) Scale (Wood and Neumann, 1931;
Richter, 1958) is widely used in North and South America The early intensity
130 PlANe
scales of Mercalli or Rossi-Forel have been updated to include the twelve degrees
(I to XII). The Medvedev-Sponheuer-Karnik (MSK) Scale (Medvedev and
Sponheuer, 1969), recently revised to the European Macroseismic Scale (EMS)
(Grunthal, 1998), is commonly used in Europe. These scales are roughly equiva-
lent. Japan maintains the Japan Meteorological Agency (JMA) Scale, that con-
sists of eight degrees (0 to VII). Relationships between the different seismic
intensity scales can be found in the publication by ISSMGE(TC4) (1999).
Intensity is used to estimate the locations of historic earthquakes, establish the
rate of occurrence of earthquakes as a function of their size, as well as to estimate
the strength and attenuation characteristics of seismic ground motions. Based on
intensities at different locations, a map of contours of equal intensity, called an
isoseismal map, is plotted. Figure T3.l shows examples of typical isoseismal
maps. Note that the contours could be affected by population centres, regions of
poor construction and areas of weak soils.
Maximum intensity in the epicentral region, 1
0
or Ie, is a measure of the size of
an earthquake. The use of intensity for seismic hazard analysis, can be found in
the publication by ISSMGE(TC4) (1999).
T3.1.2 Magnitude
Magnitude is a physical measure of the size of the earthquake, typically evalu-
ated based on the recorded data. There are several scales based on the amplitude
of seismograph records: the Richter local magnitude M
L
, the surface wave mag-
nitude M. the short-period body wave magnitude ms; the long-period body wave
magnitude mB and the Japan Meteorological Agency magnitude M
J
Moment
magnitude M; is calculated from the seismic moment. which is a direct measure
of the factors that produce the rupture along the fault.
The relationship between the various magnitude scales can be seen in Fig. T3.2.
M is presently preferred by seismologists. The use of M
L
for magnitudes up to
6 and M for magnitudes 6 to 8 does not significantly differ from the A1
w
values.
Earthquakes with magnitude less than 3 are considered as microtrernors, while
those measuring up to 5 are considered minor earthquakes with little associated
damage. Maximum recorded magnitude is about M
w
= 9.5 (e.g. Chilean earth-
quake of 1960).
T
'" ..,
.:>.- STRONG GROUND MOTION PARAMETERS
For the seismic analysis envisaged in this book, earthquakes are characterised by
the ground motions that they produce,- which is usually described by means of
one or several of the following parameters or functions:
Earthquake Motion 131
_ Peak Ground Horizontal Acceleration. PGA
H
or simply PGA, is the maxi-
mum absolute value reached by ground horizontal acceleration during the
earthquake. It is also called peak acceleration or maximum acceleration.
_ Peak Ground Vertical Acceleration. PGAv. is the same as above. but for verti-
cal movement.
30'
35'
O'
5" ~ 0
*
Faro
IX
\
Fyz ) ) VI
M"';''1fJ J
/ / 1
E
. ,VII / /
ssao 18 ,/'
(7) r Marrakeen VI /
J /8(7) /
Agadir" /' /'
7(6) / ~ /" /'
-> /"
. - - ~ __IV
10' ____
*:APproximate epicentre
VI : Observed MSK intensity
(a)
Fig. T3.1. Examples of isoseismal maps.
(a) 1755 Lisbon earthquake. MSK scale (after Elmrabet et al., 1991).
(b) 1964 Niigata earthquake. JMA scale (after Okamoto, 1973).
(c) 1989 Lorna Prieta earthquake. MM Scale (after Jones et al., 1995).
]32 PlANe
Fig. T3.1. Continued.
140'
(b)
142'
40'
38'
36'
*:Approximate epicentre
V : Observed JMA intensity
Accelerogram aCt) is the time history of acceleration. It can be horizontal or
vertical. It is usually an irregular function of time. Figure T3.3 shows some
examples of accelerograms from different types of earthquakes.
- Peak Ground Horizontal Velocity, PGVH, or simply PGV. is the maximum'
horizontal component of the ground velocity during the earthquake.
- Peak Ground Vertical Velocity, PGVv, is the same as above, but for vertical
component.
- Peak Ground Displacement, PGD, is the maximum displacement.
- Acceleration Response Spectrum SACT, D) represents the maxirnumaccelera-
tion (absolute value) of a linear single degree-of-freedom '(SDOF) oscillator,
with period T anddamping D% of critical, when the earthquake motion is
. applied to its base. The SDOFoscillator is the simplest model of a structure.
Thus the spectrum represents a good approximation of the response of the dif-
ferent structures when they are subjected to an earthquake. Similarly, there is
a Velocity Response Spectrum, Sv(T, D), and a Displacement Response
N
+
o 10 20 30 40 SOkm
I ! ! I I
Fig. T3,1. Continued.
(c)
121'30'
6 .
Brentwood
Manteca
Banta 6
6
Vernalis
6
VI
37'00'
6 Gonzales
*:Approximate epicentre
VI : Observed MM intensity
Spectrum So(T, D), although, in these cases, the velocity and the displacement
are not absolute but relative to the base, The response spectra usually refer
to horizontal movement, but can also be defined for vertical movement.
Figure T3.4 shows -exarnples of response spectra of different types of
earthquakes.
Fourier Spectrum represents the decomposition of an accelerogram as the sum
of a series of harmonic functions of variable amplitude, phase and frequency
(see Appendix of Technical Commentary 4),
Duration is a measure of the time between the beginning and the end of the
ground motion. Two types of duration are used in practice:
~ Bracketed Duration, associated with a givenacceleration 'a' (usually 5% or
10%-of 'g'), which is the time between the first and last occasions when
acceleration exceeds value 'a',
~ Significant Duration, associated with a given proportion of seismic energy
En (usually 90%), which is the time needed to build 'up between the
(100 - n)/2% and the (l00 + E
n)I2%of
the total seismic energy in the
accelerogram.

M
s
-ut:

::.:.L-:-:
--_.-
m
s
;/t,-.--

.'
.. ..
z..'
_- -I- --
- --
",:
.. m
b
_
.
..
..
v
,,.
. ,
.:
I
IY

I
V I
2
2 3 4 5 6 7 8 9 10
Moment magnitude M
w
Fig. T3.2. Relationship between moment magnitude Mw and various magnitude scales:
M
L
(local magnitude), M (surface wave magnitude), mb (shan-period body
wave magnitude), mB (long-period body wave magnitude), and M
J
(Japan
Meteorological Agency magnitude) (after Heaton et aI., 1982; as reponed by
Idriss, 1985).
- Equivalent number of uniform cycles, Ne{,j' is a parameter related to the dura-
tion of ground movement, used to approximate an irregular time history by N
eq
cycles with uniform amplitude. It is useful in calculating resistance to lique-
faction by some simplified methods. The parameter Ncq depends on the dura-
tion of the earthquake and on the relative amount of low and high peaks. It can
be computed for each time history. Some authors (Seed and Idriss,
Arango, 1996) have tabulated N
eq
as a function of magnitude.
T3.3 SEISMIC SOURCE AND TRAVEL PATH EFFECTS
The tectonic mechanism in the seismic source zone, the source-to-site distance,
and the attenuation characteristics of the motions along the travel path. influence
the resulting ground motion at the site of interest. Depending on the position of
the hypocentre and its relationship with the global tectonics processes, earth-
quakes can be classified as
>- interplate
> intraplate
Interplate earthquakes originate in the great stress accumulated in contacts
between tectonic plates (the principal fragments into which the earth crust is
divided). Most destructive earthquakes, with magnitude above 7 to 8. belong to
0.5.,.--------------------------.,------,
..L.
-villi"IJ '" .. I'
Ci 0.3
C
g 0.1
to
..
-0.1 +-:..I.<JtJ-J.:..::......;..--------------------------1
u
u
-0.3 +--fl------------------------------!
80
70 60 50 40
Time (s)
30 20 10
-0.5 +--I----.-----.----..,.------,.----.----.,..-------,.--__---l
o
0.5.,.--------------------------------,
0.3
e
o 0.1
';
..
'! -0.1
u
u
-0.3 +-;.....-----------------------------1
80 70 60 50 40
Time (s)
30 20 10
-0.5
o
(b)
0.5 ..,-------------------------------,
0.3
e
.2 . 0.1 +-------:-
;;
..
'! -0.1
u
u
-0.3
80 70 60 50 40
Time (5)
30 20 10
-0.5 +-----.----.,..------.----..,.---.......,...---...,..------,.------j
o
(e)
0.5 ...---------------------------------,
96 178
0.3
g _ 0.1
e
'! -0.1 +-I-t--------'It-&---it----'''--tjl-\f--f-..L------------+-+-I
u
-0.3 __....,....
180
90 80 70 60
Time(5)
so 40 30
-0.54---......,...---...,.----...,....,....---......---.,..----......,...----.----1
o
(d)
Fig. T3.3. Examples of accelerograms from different types of earthquakes and site con-
ditions.
(a) Parkfield, CA, USA, 1966-06-28. A record of almost a single shock.
(b) El. Centro, CA, USA, 1940-05-19. A record of moderate duration with a
wide band frequency content.
(c) Lima, Peru, 1966-10-17. A record with predominant high frequency
components.
(d) Mexico City. Mexico, 1985-09-19. A long duration record with pre-
dominant low frequency components.
/.)0 .t'lANC
7-r-----.,-------r-----,------,------,
5 4 3
Period T (s)
2
Lima. 1966
I
c
o 3
-nl
...
Q)
Q)
o

"t:l
Q)
III
E
2s+------t-----+-I---\r------+-------f-------!
-o
Q)
0..
III
Q)
4 +------/------t---+----T--;--------j--- --
o
0..
III
Q)
(/'J
Fig. T3.4. Examples of acceleration response spectra shapes.
this category. The earthquakes which originate in the subduction zones are a par-
ticular type of interplate earthquakes.
Intraplate earthquakes originate in faults within a plate and away from its
edges. The intraplate earthquakes typically do not reach such high magnitudes.
Magnitudes above 7.5 are unusual, although large historic earthquakes have
been documented in those regions (e.g. 1811-1812 New Madrid Earthquakes,
USA).
As shown in Fig. T3.5, the type of faults causing the earthquakes are usually .
classified as reverse, normal, or strike-slip. Tectonic movement in reverse and
normal faults occurs mainly in the direction of the dip, with a relatively high
vertical component. Faults of strike-slip type are typically associated with
horizontal movement. In general, reverse faults allow greater normal stress to be
.transmitted than normal faults and are capable of storing' and, consequently,
releasing in an earthquake a larger amount of energy. The type of fault also influ-
ences the -frequency content of the vibratory motion and the intensity of the
motions in the near field of the source (i.e. near-fault effects).
(al
(b)

(e)
Fig. T3.5. Types of seismic faults.
(a) Reverse.
(b) Normal.
(c) Strike-slip.
Another conventional way of classifying earthquake motions is related to
the distance between the hypocentre and the site where the earthquake is felt.
High frequencies attenuate faster than low ones. Near field ground motions are
traditionally distinguished from other types. The term "near field earthquake" is
useful in characterising the earthquake motion close to the seismic source, as rep-
resented by their higher peak acceleration and greater content in high frequen-
cies. However, a continuous variation of these parameters with epicentral dis-
tance is present!y accepted.
In practice, the effects of seismic source and travel path are taken into account
through magnitude and distance. The movements at the bedrock or at an out-
cropping rock has an amplitude that increases with magnitude and decreases with
distance. Predominant periods are influenced by the same factors. Generally, the
greater the magnitude or focal distance, the greater the predominant period.
Figure T3.6 compares the response spectra of several Mexican earthquakes. All
of the earthquake motions associated with these response spectra came from
essentially the same source and were recorded at about the. same distance. The
variations in amplitude and frequency content, particularly in the long-period
range, are apparent in Fig. T3.6. Figure T3.7 compares the median response spec-
tra for strike-slip events of magnitude 7 at several distances. The decrease of
spectral ordinates and the displacement of predominant periods towards long
period range are also apparent in Fig. T3.7.
138 PlANe
10
....... ""'"
...
...
1 0.1
0.1 L...-__~ " - " ; " ---::.......I "":"----.J
0.01
1000 ~ - - - - - : " - : - - - - - ~ - - - - - : - - : : - - " I
-
~
E
o
->
C/) 100
E
:::l
....
13
(l)
0-
m
(l)
m
c: 10
o
0-
m
(l)
....
.c
'>
o
CD
> 1
o
"0
:::l
(l)
m
0..
Period T (5)
Fig. T3.6. Effects of magnitude on response spectra (after Anderson, 1991).
T3.4 LOCAL SITE EFFECTS
There are important differences in ground motion depending on the local site
conditions. The main local factor that influences the movement of the surface
during an earthquake is the presence of soil (especially when soft) over the
bedrock. Effects of the topography (i.e. hill effects) and of the shape of the soil
deposits (i.e. basin effects) have also been noted.
The effects of a soft soil layer is usually the amplification of the vibration,
especially in some low frequencies. The main physical mechanism responsible
for the motion amplification at soft soil sites is related to the resonance of the
components of vibration with frequency near the natural frequency of the soil
layer. The fundamental resonant period for shear wave, Ts, propagating through
a soil deposit with a shear wave velocity of V
s
with a thickness of H, is given
by T, 4HIV
s
At the surface, the upgoing and downgoing waves of such a
frequency are in phase, resulting in a double amplitude wave. The resonance
amplification could be significantly modified by the damping and the non-linear
properties of the soil.
10 " T . " : " : ~ ~ ~ ~ : - : - - - - : - :---------.----------,
,----.....-._..-:--..~ _ . -;---;-- ---;--
....
"".
-
<:
Cf)
c
0
-c::
...
<ll
<ll
U
U
<{
c::
...
-to:
<ll
c,
0.1
Cf)
0.01..J.,--------:...------..........+---.........-...---------4
0.01
PeriodT(5)
* Bedrock deeper than 20m
Fig. T3.7. Effects of distance on response spectra (after Abrahamson and Silva, 1997).
Figure T3.8 shows the differences in observed surface accelerations depend-
ing on local site conditions. For low and medium PGA, acceleration in the soft
soil surface is larger than acceleration in the rock outcropping. However, for high
acceleration in the bedrock, the shear strength of soil limits the level of the accel-
eration, resulting in the decrease in acceleration amplification, and eventually in
the deamplification.
As shown in Fig. T3.9, the frequency content of the movement at the surface
of a soil deposit is different from that in a nearby outcrop rock or in the bedrock.
Typically, the predominant periods in rock is tenths of a second. Depending on
the thickness and stiffness of the soil layer, the predominant period at the surface
can be tenths of a second to a few seconds. Accordingly, the maximum seismic
vibratory effects for structures founded on rock can occur for natural periods of
tenths of a second (e.g. stiff buildings, gravity quay walls), while, at the surface
of a soil deposit, the most affected structures can be those with a natural period
of tenths of a second to a few seconds (e.g. high-rise buildings, offshore plat-
forms, pile-supported wharves).
.
...:
Earthquake magnifude=7
140 PIANC
0.6
0.5
-Ol
-
tf)
0.4
$
'ii5
'0
tf)
0.3
-
15
tf)
co
0.2
<t:
o
a..
0.1
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6
PGA at rock sites (9)
Fig. T3.8. Comparison of peak horizontal accelerations at soft soil sites with accelera-
tions at rock sites (after Idriss, 1991).
T - (4)(37) - 2
s- 75 - s
1.0 1.0 1.0 1.0
UNAM
Ts l
SCT
0.8
CAF CAO
0.8 0.8 0.8
0.6
5% Damping
0.6 0.6 0.6
SA(g)
0.4
SA (g) 0.4
0.4 0.4 l T
s
0.2
00 00
,
1 2 3 4 5 1 2 3 4 5
1 2 3 4 5 2 3 4 5
T (s) T (5) T (5) T (5)
0
I I I
:::37m
!:;:47m
Soft
:::57m clay
50m
,;/ ,;
//.' 1"1'/
/''/ ,'"
Vs::: 500 to 900 m/s
///. r r,
Cemented/compact
100m
sand and gravel
Fig. T3.9. Effects of soil conditions in recorded acceleration response spectra for
several sites in Mexico City during the 1985 earthquake (after Dobry, 1991).
T3.5 SEISMIC HAZARD AND DESIGN EARTHQUAKE NI0TION
Peak horizontal acceleration, peak horizontal velocity and response spectra
ordinates are commonly used parameters to characterise the seismic hazard at a
given site. There are two approaches to compute these parameters of earthquake
motions for the seismic analysis of port structures. They are a seismotectonic
approach and a direct approach based on seismic history.
T3.5.1 Seismotectonic approach
In the seismotectonic approach, there are two different, but complementary
methods:
- Deterministic method: The largest earthquake(s) which could affect a given
site is used. Commonly, the result of a deterministic analysis consists of the
Maximum Credible Earthquake (MCE), the maximum earthquake that
appears possible to occur under the known tectonic conditions, taking into
account seismic history of the region.
- Probabilistic method: Several values of the earthquake motion parameters are
used, usually acceleration or response spectra ordinates, associated with
annual exceedance probability.
The procedures for deterriIinistic and probabilistic analysis are shown
schematically in Figs. T3.10 and T3.11. The process includes the following steps:
Step 1. Identification of active faults and other seismic sources.
In both the deterministic and probabilistic methods, all sources capable
of producing significant ground motion at the site must be considered.
The locations and other parameters of active, and potentially active seis-
mic sources, should be identified.
In the probabilistic method, the temporal occurrence of earthquakes
should also be characterized. In addition to hazards associated with the
specific faults, broader seimotectonic provinces, i.e. regions with uni-
form tectonic and seismic conditions, are often defined: Recent earth-
quakes in several seismically active regions of the world demonstrate
that the current state of knowledge of both the spatial and temporal
occurrence of potentially damaging earthquakes can be incomplete.
This uncertainty in the characterization of the seismic hazard is
compounded in regions of low- to moderate-seismicity, and in areas
where the seismic sources ~ e not wen understood. In light of the
seismic hazard associated with unidentified sources, the inclusion of
areal, or 'random', sources is warranted in most regions of the world.
The distribution and rate of occurrence of earthquakes associated with
areal sources are specified based on the nature of the seismotectonic
province.
142 PlANe
Step 1
>-
~
ii
E
co
Y,
til
Q.
Y
2
c:
0
~ Yo
E
-e
c
:I
e
C>
Ro=O R
2
R,
Distance, R
Step 3
Step 2
Step 4
Fig. T3.1O. Main steps of a deterministic seismic hazard analysis.
,
2
o
Size of earthquake. m
Step 1 Step 2
Parameter value, y Distance, R
.Median-+, (1
---............ ,,/
<,{ /Medlan
..... .
..... "'>'... / Median ., (T
..... ....
" .......
....... .... ..
..... .
..... ,
--
For each source
Step 3 Step 4
Fig. T3.11. Main steps of a probabilistic seismic hazard analysis.
Earthquake Motion 1 ~ 3
The area studied should include seismic sources. both onshore and off-
shore. The methodology to be applied includes the interpretation of:
- Topographic and bathymetric maps;
- Seismicity maps;
- Geophysical surveys;
- Repeated high precision geodetic measurements;
- Aerial photographs;
- Geomorphological data;
- Stratigraphic correlations;
- Paleoseismicity (i.e. geologic guidance for pre-historic earthquakes).
Step 2. Characterisation of each seismic source activity
- Step 2a. In the Deterministic Method:
The maximum earthquake (Mj ) and the minimum distance (Rj ) from the
source to the site are defined. The term 'maximum earthquake' is not well
defined. It is preferable to refer to the Maximum Credible Earthquake,
MCE, the largest earthquake that appears capable of occurring in a
seismic source. As pointed out by Wyss (1979), Fig. T3.12, dimensions of
the fault segment, which could move at the same time, limit the size
(magnitude) of the associated earthquakes. If causative faults are not
known. the regional seismic history is the basic information which helps
the engineer to decide the maximum credible earthquake. for R;
00' ...
... 0 ...
...o ~ 0
... ...
...
.... ~ ..'"
.... t:. ...
.~ .
M
s
=log A + 4.15
7.0 8.0
Magnitude M
s
9.0
Fig. T3.12. Data and regression-for magnitude versus fault-rupture area (after Wyss,
1979).
144 PlANe
epicentral, hypocentral or fault plane distance may be used, depending on
the attenuation model considered in Step 3.
- Step 2b. In the Probabilistic Method:
The parameters of the earthquake occurrence statistics are defined, includ-
ing the probability distribution of potential rupture locations within each
source and the recurrence relationship. Commonly, it is assumed that all
points within the source have the same probability of originating an earth-
quake. The recurrence relationship specifies the average rate at which an
earthquake of a given size will be exceeded, and also, the maximum earth-
quake, the same as discussed in the deterministic method. ..
For modelling the occurrence of earthquakes of different magnitudes,
the conventional exponential model (Gutenberg and Richter, 1944),
or any of its variants are commonly used. They are based on the
Gutenberg-Richter relationship, that relates magnitude (or intensity)
M with the mean annual number of events, n, that exceeds magnitude (or
intensity) M:
log n = a - bM (T3.1)
The coefficients a and b must be obtained by regression of the data of each seis-
mic source. They could depend on the range of earthquake sizes used in the
regression. If the seismic catalogue is incomplete for small earthquakes, as it is
usual, only earthquakes with a size beyond a certain threshold level must be used.
Figure T3.13 shows the application of Gutenberg-Richter relationship to pri-
mary worldwide seismicity zones for earthquakes with magnitude m 6. This
Fig. T3.13. Gutenberg-Richter recurrence
law "applied to worldwide seis-
micity (after Esteva, 1-970).
10 8
Magnitude m
7 6
I
I
,
I
,
I
t I I I
- - ~ - - - - - ""'" - - - - - - ~ - - - - - -1- _
I I I
I I I
I , ,
I I I
I I I
I I I
I I I.
1
10
~
Ql
.0
E 0.1
:J
c:
co
:J
c:
~ 0.01
c
....
o
E 100
500 , - - ~ - - - . - - - , - - - - r - - - - ,
- -. --._..... .... __
model was first derived for a set of regional data. that included several seismic
sources as in seismotectonic provinces in regions of low seismicity, but it would
not be appropriate for the occurrence of large earthquakes in a fault. For those
cases, the elastic rebound theory predicts the occurrence of a large 'characteristic'
earthquake when the elastic stress in the fault exceeds the resistance of the rock.
Characteristic earthquakes repeatedly occur with a segment of a fault having the
maximum possible source dimensions.
The temporal model commonly used for the occurrence of earthquakes is a
Poisson model, in which each earthquake is a memoryless event, independent
of the previous and next earthquake at the same location. The elastic rebound
. theory suggests that the occurrence of characteristic earthquakes on a particular
fault segment should not be independent of past seismicity. A renewal process
better than a Poisson process would represent the occurrence of these large char-
acteristic earthquakes. However, the Poisson model seems adequate, physically
and observationally, to represent the smaller events, presumably due to local
readjustment of strain. secondary to main elastic rebound cycles. A hybrid
(Renewal-Poisson) model was proposed by Wu et al. (1995). Figure T3.14 illus-
trates this model. In this mode, small (to moderate) earthquakes follow a Poisson
process in time and the exponential part of a characteristic distribution in magni-
tudes. Characteristic earthquakes follow a general renewal process in time and
the non- exponential part of a characteristic distribution in magnitudes. Times
and magnitudes of all earthquakes are assumed mutually independent. The terms
mr and m: are the lower and upper bounds of the magnitudes of the small earth-
quakes; and m; are those of the characteristic earthquakes. The value of m:
is assumed less than or equal to
... Small earthquakes (P) tCharacteristic earthquakes (C)
-<
)0
Time
c p
f,Jm) f.,<t)
........ m t
Fig. T3.l4. Hybrid Poisson renewal model (after Wu et aI., 1995).
146 PlANe
Step 3. Determination of the attenuation relationship for the acceleration,
response spectra ordinates or other parameters of interest.
One attenuation relationship is typically used that should adequately repre-
sent the attenuation for the region. If a statistical analysis has been done, a _
median relationship would not be enough. A relationship should be
selected that corresponds to a conservative estimate of the design
parameters.
Step 3b. In the Probabilistic Method: ..,
Ground motion parameters (e.g., PGA, PGV, spectral acceleration) are
routinely estimated in practice based on the probabilistic evaluation of
mean values obtained from the attenuation relationships. In regions of
high seismicity, or in applications involving long exposure intervals that
approach the return period for the largest earthquakeis) expected in the
region, the standard deviation term established for the specific attenuation
relationship being employed will often be used. The application of the
standard deviation term in estimates of the ground motion parameters
accounts for the probability of experiencing greater than mean motions
during the period of interest. The mean attenuation function and the stan-
dard deviation should be computed through the statistical analysis of data
from earthquakes of the same' region or, at least, from earthquakes of
similar tectonic environment, recorded in stations with travel paths and
local ground conditions similar to those of the site of interest.
In recent years, it has become common practice to use specific atten-
uation relationships for each of the spectral ordinates of different periods.
A general expression for an attenuation relationship is:
log Yg =II (F
T
) + h (M) +13 (R) +~ (ST) + Ca (T3.2)
where:
)"g = ground motion p a r a m ~ t e r or response spectrum ordinate
F
T
- a set of discrete variables describing the fault type
M - magnitude
R - a measure of distance
ST - a set of discrete variables describing the site subsoil conditions
or a continuous variable depending on the average shear wave
velocity in the deposit
fi - functions, h. is often assumed linear in powers of M. f ~ depends
on R, log R and, sometimes, M
f<r - a random error term with zero mean and (J standard deviation
This procedure allows the inclusion offault type and distance effects and
supplies appropriate response spectra for rock. However, results for soil
sites are averages of values from different soil conditions and will not
- . .
represent any particular site. In many cases, it may be preferable to first
Fig. T3.15. Comparison of proposed
PGA attenuation with dis-
tance for several regions
of the world (after
ISSMGE (TC4), 1999).
' ~
1
M=7j'"
M=6
M=5
. I Ii. I ill I Iii i I II
I I
1.0=-
-
en
-::--
..
c
0
-co
"-
0.1=-
dl
dl
-
-
o
(.)
co
co
-
-C
0
0.01;:-
N
Italy
"-
~
-
-
0 ~
North America
::c
L..
L
Japan
L
I
0.001
10 100
Distance (km)
obtain the ground motion parameters In rock and then compute the
seismic response at the ground surface.
Figure T3.15 shows the comparison among the attenuation of PGAH
from Japan, Italy and North America. Selected attenuation relations for
spectral ordinates can be found in Abrahamson and Silva (1997), or in
Sabetta and Pugliese (1996).
Step 4. Definition of the seismic hazard
Ground motion, primarily described by PGA and spectral ordinates must
be defined.
Other parameters such as intensity or duration can also be obtained in a
similar way.
Calculate the values of the motion parameters at the site from each source
of the earthquakes and select the maximum.
- Step 4b. In the Probabilistic Method:
i) Calculate the annual number of occurrences of earthquakes from each
source which produce, at the site, a given value of the PGAH (or other
earthquake motion parameters).
ii) Calculate the total annual number, n, of exceedance.
iii) Calculate the reciprocal of the mean annual rate of exceedance for the
earthquake effect (PGAH):
T =.!-
R n (T3.3)
This is the return period of earthquakes exceeding that PGAH
iv) Calculate the probability PR(a
max
, T
L
) of the PGAH being exceeded in
the life TL of the structure.
148 PlANe
'.0.,.---------------,
"Oica
diZ
_--
Lisbon ! . . . G ~ _- - - -
---
-- .,
.;
.;
Fig. T3.16. Peak horizontal bedrock
O.O+-........-....--.--.......-......---.-.......-.................---l
o 100 500 '000 acceleration evaluated for
Return period (years) different regions.
c 0.8
.2
~
Q)
Q) 0.6
8
ttl
iti
0.4
N
~
~ 0.2
Q)
CL.
P
R
(Q
max
, T
L
) = 1- 0- n)T
L
::: TLn = i (::: for TL/T
R
-e; 1.0)
R
(T3.4)
The results of the analysis, sensitive to the details of the procedures used,
reflect the seismic hazard of each site. As presented in Fig. T3.16, the
seismic hazard is a variable in different parts of the world. e
Step 5. If the attenuation relationships do not match with the local site conditions
(e.g. if attenuation relationships for rock are used and there is a
surface soil deposit), convert the motion parameters to the specific site
conditions using empirical amplification ratios or numerical dynamic
soil response models.
For preliminary studies, or when data are scarce, the methodology previously dis-
cussed is simplified. The traditional method, which is still used. consists of
calculating the PGAH by means of a probabilistic or a deterministic study and
scaling a response spectrum to this PGA
H.
In this method, care must be taken in
the selection of the response spectrum shape, which depends on fault type. mag-
nitude, distance to rupture zone and local site conditions. Figure T3.l7 shows the
shapes of elastic response spectra proposed by Bea (1997).
T3.5.2 Direct approach based on seismic history
If a particular earthquake controlling the seismic hazard is defined directly from
the seismic history, and taking into account the tectonic characteristics of the
region around the site. the seismic hazard analysis is significantly simplified. An
example is the conventional practice in seismic design codes. where every time
that a set of the most recent strong earthquake motionsIlarger than the previously
(log scale)
10.------------------------,
~ = 2 0 T
adesign
SA
adeSign 1+-------{
~ = 2 . 5
adesign
'l'i''Vr 1
2.5
Period T (5)
'VS'VT
T
10
(log scale)
Soil condition
"'s
Tectonic condition
"'T
A-Rock Type A
(Vs > 750 mls) 1.0 Shallow crustal faulting zones 1.0
B Stiff to very stiff soils. gravels J'ypeB
(Vs =360 to 750 mls)
1.2 Deep subduction zones 0.8
C Mediumstiff to stiff clays TypeC
(Vs =180 to 360 mls)
1.4
Mixed shallow crustal & deep
subduction zones 0.9
D Soft to medium stiff days
TypeD
(H =1 to 20 m; Vs < 180 mls)
2.0
Intraplate zones
0.8
E Conditions other than above Site
Default value
1,0
specific
studies
required
Fig. T3.17. Design response spectra shape proposed by Bea (1997) with modification.
recorded ones) are obtained, the seismic code has been modified to incorporate
the response spectra of the last' recorded earthquake motions.
In this direct approach, the magnitude, the distance and the fault type of the
design earthquake are defined in the simplified manner described above.
Accelerograms recorded from earthquakes in these magnitude and distance
ranges, with the similar fault type, propagation path and recording site conditions
as the design earthquake, are then used to define response spectra at various per-
centile levels suitable for design.
T3.5.3 Design accelerograms
Depending on the type of the analysis, time histories might be defined. Either
recorded or generated accelerograms may be adequate.
. .'
]50 PlANe
Recorded time histories must be selected from the collections of accelero-
grams caused by earthquakes whose fault type, magnitude and distance are sim-
ilar to those of the design earthquake and can be scaled to the design PGAH.
Commonly it is necessary to use several accelerograms to cover the implicit con-
tingency in its selection. When the strong ground motion parameters of the
design earthquake are influenced by several seismic sources, it is not possible to
find one recorded accelerogram that incorporates all the characteristics of the
design motion, including all the spectral ordinates and duration. In such a case,
several records of each earthquake type should be used.
Typical sources of digitised accelerograms can be found at:
the National Geophysical Data Centre (http://www.ngdc.noaa.gov) In the
USA; and
- the K-NET (http://www.K-NET.bosai.go.jp) in Japan (Kinoshita, 1998).
In particular, Port and Harbour Research Institute, Japan, have operated the
strong motion recording network throughout port areas in Japan since 1964. The
recorded strong motion data have been compiled as annual publication (e.g. Sato
et aI., 1999), together with occasional compilation and update of local site con-
ditions at the strong motion recording stations (e.g. Ichii et aI., 1999). Opening
of a web site for easy access to these data are planned in near future.
F9r of structures, it is important that the frequency con-
tent of the recorded time histories matches the spectral shape of the design earth-
quake. If the maximum recorded acceleration is different from the design PGA
H,
the recorded accelerograrn can be scaled to this value. In some cases, a slight
modification of the time scale in the recorded accelerogram is also acceptable to
achieve a better agreement between the frequency contents of the recorded and
the design acceleration time histories. For liquefaction analysis. it is also impor-
tant that the duration of a recorded accelerogram matches the design duration.
Artificial time histories must match the PGA
H,
the duration and the response
spectrum of the design earthquake.
Formulation of design earthquake motions for near source regions, including
Level 2 earthquake motion, could require adequate modelling of fault rupture
process and travel path effects. Practice-oriented guidelines proposed by the
Japan Society of Civil Engineers are summarized by Ohmachi (1999).
The seismic hazard analysis discussed in Section T3.5.l can be generalized by
using eanhquake magnitude and source-site distance as primary earthquake
parameters (Kameda and Ishikawa. 1988). In this approach, the magnitude and
distance are evaluated as conditional mean values, given that the ground motion
intensity (e.g. PGA
H
) exceeds the level specified by the annual probability of
exceedance. Other ground motion parameters. including predominant period,
ground motion duration. response spectral ordinates. can be determined as a
function of these hazard-consistent and This approach also
forms the basis for establishing the link between probabilistic seismic hazard
analysis and the deterministic scenario earthquakes, and makes it possible to
associate the scenario earthquakes with the specific risk level (Ishikawa and
Kameda, 1991).
T3.6 GROUND MOTION INPUT FOR SEISMIC ANALYSIS OF PORT
STRUCTURES
As discussed in Chapter 5 in the Main Text, the seismic input to be used for
analysis and design depends on the type of port structure and on the type of
analysis being performed. For dynamic analysis of soil-structure interaction,
seismic motions at the bedrock is required. If the seismic data consist of free field
ground surface motions, it is necessary to compute the subsurface/bedrock
motion by numerical deconvolution techniques. For simplified analysis and
simplified dynamic analysis, the seismic input depends on the configuration of
the structure.
In some cases, the effects of elevation difference between the landward ground
surface and the seabed can be an important issue, that should be considered in the
development of the response spectra and other earthquake motion parameters.
There is not a criteria well established in design practice, and this issue needs
careful studies if higher reliability is required on the results of the simplified and
simplified dynamic analyses.
Simplified analysis of gravity quay walls assumes a uniform field of seismic
acceleration. Therefore, the design acceleration can be the average acceleration
over the heizht of the caisson and the backfill soil wedze areas. To estimate the
~ ~
average acceleration, one-dimensional site response analysis can be performed
using the geotechnical configuration at Band C locations shown in Fig. T3.18.
Simplified dynamic analyses include the Newmark-type rigid block sliding
and the simplified chart approaches based on the parametric studies (see
Technical Commentary 7). For the Newmark-type analysis, a time history of
acceleration averaged over the caisson and the backfill soil wedge can be used.
For simplicity, the time history at the elevation of the bottom of the caisson
scaled to the average of the peak acceleration over the sliding mass is often used.
For simplified chart approaches based on the parametric studies, the input
acceleration must be representative of the one used to develop the chan.
Simplified and simplified dynamic analyses of sheet pile quay walls assumes
uniform field of seismic acceleration on the active and passive soil wedges.
.' .. ~ .
]52 PlANe
Backfill
: l
I
I I
I
Caisson
I
I I
I I
I
--- ..
I
Rubble mound
I
Bedrock
-t---J------r---+-------,--:i-
I
--f Soil layers
I
A B c
Fig. T3.18. Earthquake motion input for seismic analysis of gravity quay walls.
r--"
l
I
>- Backfill
I
I ________ .J
'1
--.=::....,4----:-----I----...,.L-+----l..--I-II-----
I
I
I
Active soil wedge
I
I
I
--c----"----r----. --------J
i
( Soil layers
I I I I I
Fig. T3.19. Earthquake motion input for seismic analysis of sheet pile quay walls.
Therefore, the design acceleration can be the average over the active and passive
soil wedges. One-dimensional site response analysis can be performed using the
geotechnical configuration either in front of or behind the wall in locations Aand
B shown in Fig. T3.l9. There may be an option to adopt different accelerations
for the active and passive soil wedges.
For simplified and simplified dynamic analyses of pile-supported wharves, the
equivalent fixity pile model is often adopted as shown in Fig. T3.20. It is typical
Apron
H.W.
Soil layers
I , I
I I
I I I I I I
I I . I . I I I
~ ~ ~ ~ ~ ~ Bedrock
ABC
(a)
...,--------,,..- ----... --_.
l-....,..---..,..---..,.-J I "" ""
I, ,
I, ,
t' ,
I' ,
, ,
I , '
,,_.!._----'---.. "
, ,
4 ~ ~
7"'Vinual groUnd surface
13-----------
(b)
Fig. T3.20. Earthquake motion input for seismic analysis of pile-supported wharves.
(a) Pile-supported wharf.
(b) Modelling by a frame structure having equivalent fixity lengths.
- .... -----
.... _---
Soil layers
Bedrock
Fig. T3.21. Earthquake motion input for seismic analysis of embankments and break-
waters.
to obtain the seismic action at the base of the equivalent structure by an one-
dimensional site response analysis of the soil column in the middle portion of
dike/slope below the deck, in location B in Fig. T3.20a.
T3.6.4 Embankments and breakwaters
Simplified and simplified dynamic analyses of embankments and breakwaters,
often assumes a sliding failure block as shown in Fig. T3.21. In this case,
]54 PlANe
average acceleration over the sliding block may be adequate for sliding block
analysis. It can be obtained by a one-dimensional response analysis of a wedge
shear beam or averaging the results of the one-dimensional site response
analysis of soil columns A and B, shown in Fig. T3.21.
TC4: Geotechnical Characterisation
14.1 Mechanical Behaviour of Soil under Cyclic Loads
Ttl.l Soil behaviourin pre-failure conditions
T4.1.2 Soil behaviour at failure
14.2 Measurement of Soil Parameters
T4.2.1 In situ penetration tests
T4.2.2 In situ geophysical tests
T4.2.3 Laboratory tests
T4.2.4 Combined use of in situ and laboratory tests
T4.3 Evaluation of Pre-Failure Soil Properties through Empirical Correlations
T4.3.l Small strain stiffness
T4.3.2 Small strain damping
T4.3.3 Non-linear pre-failure behaviour
14.4 Assessment of Liquefaction Potential
T4.4.1 Liquefaction potential assessment using field data
T4.4.2 Liquefaction potential assessment combining field and labo-
ratory data
T4.5 Geotechnical Characterisation Strategy with Reference to the Analysis
Requirements
Appendix: Interpretation of Geophysical Tests
A.I Time domain analysis
A.2 Frequency domain analysis
A.3 Spectral analysis and propagation velocity
Geotechnical Characterisation
This technical commentary addresses characterisation of subsoil and earth-fill for
site response, liquefaction and soil-structure interaction analysis. The subsoil
refers to the natural soil deposit interacting with a structure foundation, and
extending from the ground surface to the bedrock as illustrated in Fig. T4.1. The
subsoil can significantly modify the seismic waves propagating from. the
bedrock, and affect the behaviour of a port structure under earthquake loading.
The earth-fill is the man-made deposit interacting with the structural surfaces of
a port structure. The earth-fill can be either a backfill (for quay walls as shown in
Fig. T4.l) or a foundation layer (for quay walls and breakwaters). In some cases,
e.g. for rubble mound breakwaters, the earth-fill can be the structural material.
Appropriate characterisation of subsoil and earth-fill is important for site
response and seismic analysis of port structures.
This technical commentary begins with reviewing the soil behaviour under
cyclic loads in Section T4.1, followed by the relevant field and laboratory tests in
Section T4.2. Empirical correlations are addressed with respect to deformation
properties in Section T4.3, and procedures for liquefaction potential assessment
are summarized in Section T4.4. The technical commentary concludes with the'
recommended strategy for geotechnical characterisation in Section T4.5, with an
Appendix. on the interpretation of geophysical tests. The contents of this techni-
cal commentary reflect the state-of-the-practice in the geotechnical characterisa-
tion around the world.
T4.1 MECHANICAL BEHAVIOUR OF SOIL UNDER CYCLIC LOADS
The effects of earthquake loads on a soil element for port structures can be
represented by a complex time history of shear stress 't(t). The pre-existing quasi-
static stress state due to gravity further affects the soil element behaviour.
Depending on the level of earthquake motions considered for design, the shear
strain level induced in the soil can vary. According to the strain level considered,
]58 PlANe
Fig. T4.1. Schematic view of
subsoil and backfill in a
typical quay wall.
soil behaviour may be characterised with increasing complexity, and accordingly
specified with different geotechnical parameters and variables. The soil behaviour
may be broadly classified into pre-failure and failure conditions, depending on the
strain level. The relevant soil behaviour is reviewed in the following subsections.
T4.1.1 Soil behaviour in pre-failure conditions
The level of shear strain y in a competent horizontally layered site is usually less
than 1% for low and moderate earthquake shaking. Even at such low to moder-
ate strain amplitudes, the soil exhibits non-linear behaviour, typically specified
by two parameters; the equivalent shear modulus G and equivalent damping ratio
D (Richart et al., 1970). These parameters define the characteristics of a shear
stress-strain loop of a soil element undergoing a cyclic simple shear as shown in
Fig. T4.2.
The equivalent parameters vary with increasing cyclic strain amplitude y, as
schematically shown in Fig. T4.3. With increasing strain amplitude, the non-lin-
ear, irreversible behaviour of a soil becomes more pronounced, resulting in the
decrease in the equivalent shear modulus G and the increase in the equivalent
damping ratio D (Fig. T4.3(a. In the following, the pre-failure soil behaviour
is described with reference to three strain levels: small strains (Fig. T4.3(b,
medium strains (Fig. T4.3(c, and large strains (Fig. T4.3(d)).
(1) Small strains
Figure T4.3{a) shows that the equivalent shear modulus G of a soil initially takes
a maximum value (Go or G
max
) , after that it decays with increasing shear strain
level. Although soil non-linearity starts from a very low strain level, it is common
practice to assume G(1J constaIl:t and equal to Go if the shear strain level is lower
Geotechnical Characterisation 159
y
---
---
---
1: t
"C
G =-PE..
Ypp
W
o
=areaD =energy damped in the loop
W
s
=area. =equivalent elasticstrain energy
Fig. T4.2. Typical shear stress-strain loop and definitions of the equivalent parameters G
andD.
than a threshold level designated by linear threshold YI (Vucetic, 1994). The
equivalent damping ratio D takes its minimum constant value Do in the same
strain range. Within this strain range, the r-y loops are approximately elliptical,
and soil response can be adequately represented by a linear model.
For clean granular soils, even at the highest loading rates, dissipative effects
are negligibly small (Do == 0) and the r-y relationship is practically linear (Fig.
T4.3(b. Therefore, an elastic model can be appropriate for the analyses..
Fine-grained soils can exhibit rate-dependent behaviour even in such a small
strain level; the equivalent parameters Go and Do increase with higher loading rate
(Shibuya et al., 1995; d'Onofrio et al., 1999a). This rate-dependency may be rep-
resented by a visco-elastic model.
(2) Medium strains
Beyond the linear threshold Yt. soil behaviour begins to show a marked non-
linearity and, as shown in Fig. T4.3(a), the equivalent shear modulus G gradually
decreases with increasing y. The shape of the r-j-Ioop changes, taking a more
.. " . .
pronounced spindle shape: the equivalent shear modulus G(y) decreases and the
energy absorbed per cycle (i.e. D(y) increases.
In the medium strain level, the soil behaviour is practically unaffected by the
previous loading history: regardless of the previous loading history, the same
]60 PlANe
Cyclic stress-strain behaviour: Pseudo-linear
~ ,
Medium
NOl)-linear
non-degradable
Large
Non-liollar
degradable
0.8 L.,....---__---.
15
5
*"
, 10 _
o
0.1
/'
/'
/
/
/
0.001 0.01
0[%]
~ , i
-----
0.2
.
o
C>
a0.4 ~ l - - - - __-..J
_0
~
::l
<J 0.0
(a)
[
. .. Bender elements IBEI. Field tests ICH. DH. SASWI
Resonant Column IRC) "
Testing techniques:
---------- Cyclic torsional shear ICTS) ----------
---------- .. to failure
Cyclic triaxial (CTXt and simple shear ICSS)
G(1)= .. G(N)
T
~
{.
.,
G
./
, -
(b)
v
(c)
\d)
T
Fig. T4.3. Variation of the equivalent shear modulus (G) and damping ratio (D) with the
cyclic shear strain amplitude.
(a) Typical G(1J, D(1J curves and related experimental techniques.
(b) Typical stress-strain loop at small-strains.
(c) Typical stress-strain loop at medium strains.
(d) Typical stress-strain loop at large strains.
stress-strain loop is obtained for a loading cycle at the same strain level
(Fig. T4.3(c)). Within this strain range, the cyclic soil behaviour can be termed
as 'stable': cyclic loads neither produce residual volumetric strains, e; in
drained conditions, nor appreciable pore-water .pressure build-up, L1u, in
undrained conditions. Therefore, the saturated soil behaviour can be repre-
sented by an equivalent one-phase medium, and total stress analysis may be still
applicable.
In practice. the stress-strain response can be analytically described by either
an equivalent-linear or a non-linear modeL An equivalent-linear model is typi-
cally specified by the parameters. G and D, referring to the current strain level y.
Alternatively. a non-linear elasto-plastic model, with adequate kinematic harden-
ing rules. may represent the stress-strain relationship throughout irregular
sequence of loading. If the kinematic hardening rule is based on Masings rule
(Hardin and Dmevich, 1972), appropriate modifications need to be introduced in
order to avoid undesirable effects, including ratcheting/drifting of hysteresis
loops (Pyke. 1979).
The equivalent linear model is typically adopted for site response analyses in
the frequency domain, using amplification (or transfer) functions, with iterations
on soil parameters to account for strain dependent properties (e.g. Schnabel et al.,
1972). The non-linear elasto-plastic approach is typically implemented in time-
domain analyses and performed by means of step-by-step time integration, with
progressive updating of tangential soil properties (e.g. Finn et al., 1977).
(3) Large strains
In the large strain levels, beyond the so-called volumetric threshold 1'v (whichis
typically greater than Yl by one to two orders of magnitude), the soil undergoes
irreversible micro-structural changes. The effects of repeated cyclic loading
become evident with cespect to the shear-volumetric coupling, inducing pore-.
water pressure build-up for undrained cyclic simple shear loads.
In such conditions, a progressive increase in the shear strain amplitude at con-
stant cyclic stress or, conversely, a reduction in stress at constant strain amplitude
can be observed (Fig. T4.3(d). As a consequence, the equivalent soil parameters
GandD progressively change with the number of loading cycles. As an analogy
to 'fatigue behaviour' for solid materials, these changes may be defined by cyclic
degradation index.es, 8
G
and 8
0
, defined as
...__..- .
8 _ G(N)
G - G(l)
8 = D(N)
o D(l)
(T4.1)
A corresponding degradation parameter (Idriss et al., 1978) may be defined for
the shear modulus G
log8
G
t
G
=
10gN
(T4.2)
At large strain levels, cyclic soil behaviour can no longer be described by the
same conventional Gel? and D(n relationships used for equivalent linear analy-
sis. Atotal stress analysis may be still appropriate, provided that the variation of
modulus and damping with the number of cycles is appropriately accounted for,
such as through the degradation parameters defined above. Alternatively, the con-
stitutive models based on an effective stress approach (e.g. elasto-plastic with
]62 PIANC
kinematic hardening) accounting for shear-volumetric coupling, cyclic degrada-
tion and stress-strain behaviour approximating failure may be required .(Sagaseta
et al., 1991). The analyses of a saturated soil need to be carried out by coupled
two-phase effective stress approaches (e.g. Ishihara and Towhata, 1 9 8 2 ~
Zienkiewicz and Bettess, 1982), which may incorporate the description of states
close to failure, as it will be more extensively discussed in Section T4.1.2.
The above conceptual framework associating strain levels with constitutive
models and types of analysis is shown in Table T4.1 (adapted after Ishihara,
1982).
T4.1.2 Soil behaviour at failure
As discussed in Section T4.1.1, large strain behaviour of soil is invariably asso-
ciated with marked cyclic degradation effects, resulting in significant irrecover-
able strains and/or dynamic pore-water pressures. Beyond the large strain level
near or at failure state, further excessive inelastic deformations and significant
loss of shear strength of the soil may occur, leading to significant damage to port
structures.
Permanent displacements and failure of saturated soils are a consequence of
the progressive pore pressure increase at large strains, beyond the volumetric
threshold ]'v. This build-up takes place in all granular soils, either in loose or
dense states, but the pattern of behaviour is quite different from each other: Loose
Table T4.1. Relationship between strain levels-soil models-methods of analysis.
Large
Soil properties Shear strain amplitude, g [%] (indicative values)
(parameters) 10-l 10-
3
10-2 10.
1
Small Medium
10
Failure
Stiffness
(Go, G)
Damping
(Do, D)
.......------------------
Pore pressure build-up
(.1u)
Irrecoverable strains
(cf.. ~ )
Cyclic degradation
(oG, 00) ~
Mechanical model
Method of analysis
Linear visco-elastic
Linear
Total stresses
Lin. eq. visco-elastic
Elasto-plastic
Linear equivalent
Total stresses
Non-linear
Incremental
Effective stresses
sand is typically contractive, and may abruptly lose its strength when. after some
cycles of shaking. the pore-water pressure build-up becomes equal to the initial
effective stress. i.e. u = (Fig. T4.4(a)). This phenomenon is called liquefac-
tion. Dense sand. on the contrary. is typically dilatant, and behaves more stable
and recovers its resistance even if it is momentarily lost. This behaviour is called
cyclic mobility. This behaviour is shown in Fig. T4.4(b), in which the abrupt
increase in shear strain typical of liquefaction is not recognised.
Under constant amplitude cyclic shear stress, gradual build-up in pore pressure
causes the effective stress path to approach near the origin of the failure envelope
-
Torsional shear
test
;.c/oo'=0.229
Or=47%
-
00'=98kPa
Ko=1.0
I\(\n nn

r
_ An
""V\J
Loose sand
time
OIL------- --.I
0.5 -

(a)
0.2
10
'-'" 5
o1-------_,.o.,.,-A.....J.-1I-+--l
5
10

"'c

s::..=
C/)(/)
(/)
(/)
0.1
en.
2
<,
...
I .... -0
Q) I-l
s:
C/) 0.1
time
..
n
..
t\ iI
n
.-
1\
"- f'I.
r
'"
\oj
\i
"
\I \.l
"
\, \,
\J
"
\j "
\.
"
Dense sand (\ (\ (\ r
!.o-..,..",,,,,"I"\r'\I"\I"\f'\(\{\
'-'''''''JVVVVV\JUUUVUVU
Torsional shear test Or=75% CJQ'=98 kPa
.
I

5
o
5
1.0
(b) 0
0.6
0 0.3
o
....... 0
-0
0.3
0.6
"'c 1._
a>e!
.t:_
C/)(/)
Fig. T4.4. Time histories of shear stress, strain, and pore water pressure ratio (after
Ishihara, 1985).
(a) Loose sand.
(b) Dense sand.
]64 PlANe
for loose sand (Fig. T4.5). Once this stress state is reached (either in exten-
sion/compression cycles or simple shear cycles), the shear strain amplitude sig-
nificantly increases and the slope of the peak-to-peak line of the hysteresis loops
becomes nearly horizontal, leading to a shear modulus nearly zero. The cyclic
mobility behaviour of dense sands (Fig. T4.5(b)) is similar in that the stress path
gradually approaches the failure envelope, but the changes in shear strain ampli-
tude and stress-strain loop are less distinctive: after many cycles of shaking, the
shear strain reaches a limiting value and the stress-strain loops take a hysteretic
shape of hard-spring type.
An alternative interpretation of this phenomenon may be given in terms of the
stress path position relative to the phase transformation lines as shown in Fig.
T4.6 (Ishihara, 1985). If the stress path is confined within the zone between the
phase transformation lines and the horizontal axis, the degradation of the soil is
very small (low to medium pore-water pressures and shear strain levels). Once
the stress path goes beyond the phase transformation lines, the pore pressures
increase rapidly, and large shear strains appear and continue to grow. When the
stress point reaches the failure envelope, a 'steady' (or "critical') state may be
reached, at which the soil continuously deforms at constant volume.
In engineering practice, onset of liquefaction is characterised in two ways,
depending on the mechanical variable used. If the pore-water pressure build-up, u,
is used for defming the liquefaction, the onset of liquefaction is characterised by
u
T. - -1
---
U ,
avO
(T4.3)
where r, is the pore-water pressure ratio which, for horizontally layered ground,
is typically defined with the venical effective stress ( ) ~ o . An alternative definition
assumes that liquefaction is triggered when the shear strain amplitude reaches a
limiting value; the value of 5%is typically adopted.
The difference in the definition of liquefaction is typically minor for loose
clean sand. However, for dense sand or soils other than clean sands, differences
may arise. Liquefaction resistance is typically summarized as the cyclic shear
stress ratio, 7:/ a ~ , o , plotted against the number of loading cycles to cause Iique- .
faction, Ns. as shown in Fig. T4.7. .
Analysis involving the liquefaction phenomenon is typically based on effec-
tive stress modelling. For horizontally layered ground, volume change and excess
pore-water pressure may be interrelated as suggested by Martin et al. (1975). In
undrained conditions (i.e. no volume change), the tendency of the soil skeleton
to contract is compensated by the elastic reboufia WIlli effective stress decrease
(Fig. 1'4.8). ThIS IS wntten as -
,-:::::: .
U = Kuc-vd (T4.4)
where E,-d is the volumetric densification of the soil (inelastic volumetric strainY
arid XU IS the rebound bulk modulus .
....
Stress-Strain Curve
(a)
i
i
Stress-Strain Curve
0.4
0.4
0.6
0.6 Stress Path
Torsional shear test
t,yl (J_'=O.111
Or=75'
rIM;=98kPa
Fuji river sand
(b)
--
--
- - '(kPa)
Lineof phase l ' ..
transformation
0"...:. (J.:: initial &current
confining stresses
Stress Path(extension positive)
0.6
0.4
0.4
0.6
Torsional shear test
r ...1
Or=47\
Ci_'::98kPa
Fuji river sand
o
be 0.2
<,
>-
....>< 0.2
4 6.5
r.
2
0.8,..--------,-------,
Stress-Strain Curve
0.6
0.4
o 0.2
bE
......
>-
....>< 0.2
0.4
0.6t1AZ1/fliWIVVlrVf

6.5
(d)
0.8 Computedfor
0.6 Fuji river sand
0.4 Or=75%
- 0 0.2
bE 0.0
<,
0.2
0.4
0.6

o 20
0.8
Computed for
0.6 Fup river sand
0.4
Or=47"10
0
0.2
bE
<, 0.0
>-
....><
0.2
0.4
0.6 Stress Path (J m'= ( (J ;. (J /Y2
0,8
0 20 40 60 80 100
a '(kP.)
..
0.8
Stress-Strain Curve
0.6
0
0.4
bE
<, 0.2
>-
....>< 0.0
0.2
0.4
8 6 4 2 0 2 4
(c)
r ell
..,
Fig. T4.5. Measured and computed cyclic behaviour of sand (see Fig. T4.4 for time
histories).
(a) Loose sand (measured) (after Ishihara. 1985).
(b) Dense sand (measured) (after Ishihara, 1985).
(c) Loose sand (computed) (after Iaiet al., 1992).
(d) Dense sand (computed) (after lai et al., 1992).
]66 PIANe
c
o
iii
Ul
o ~
J a.
~ E
II U
CT
-
Ul
Ul
OJ
"-
'iii
c
o
iii
c
~
~
Failure
line
Line of phase
transformation
Fig. T4.6. Phase transformation lines in the
stress plane (after Ishihara and
Towhata, 1980).
1000
Number of cycles, N
c
10 100
\
r\.
-,
1"-
1\
~
r-..
I"-
"
D
r=90%
I'-...
t--.
""'tt-t-
82
%
Ib--
r--t--
t::
I"-r-t--
I
"---r-Io-
68%
1--_
II ~ O / " II-
I
Ii
o
o
b
:: 0.3
.2
iii
a::
~ 0.2
~
Vi
"0
Cl.l
~ 0.1
8
Fig. T4.7 Typical relationship between cyclic stress ratio and number of cycles to
liquefaction (after De Alba et al.. 1976).
If drainage is permitted in a soil layer, the rate of pore pressure generation,
dU/dt, is balanced with the rate of pore pressure dissipation, due to consolidation
of the layer. The governing equation is:
dUd =c d"1
Ud
+ au (T4.5)
dt v dz"1 at
in which c" is the coefficient of consolidation and Ud stands for the pore-water
pressure at an elevation: in the drained layer. The effects of pore-water pressure
0
-
.-
<:
e c:
0
-,
Q
-
0
>
CURVE
, NOUCED
PORE WATER
PRESSURE
6u-
a
l
o
PRESSURE
Fig. T4.8. Mechanism of pore pressure generation (after Seed and Idriss, 1982).
redistribution within the soil deposit play an important role in the development
of pore-water pressures. For highly permeable deposits, pore-water pressure dis-
sipation can occur at a faster rate than pore-water pressure generation, thereby
preventing the sand from liquefying.
For more complicated situations (e.g. involving a thin sand layer sandwiched
between impervious materials, irregular stratigraphy or complex. soil-structure
systems), more sophisticated analyses of liquefaction, based on inelastic two-
phase mediumapproaches, rna):' be needed. In such cases, the constitutive model
should incorporate all the fundamental features of the liquefaction phenomenon,
including strain-softening of the soil skeleton, hysteresis of stress-strain loops,
anisotropy and coupling between the solid and fluid phases. This will lead to
non-linear effective stress analyses based on elasto-plastic models generalized
for two- or three-dimensional analysis.' "
A suitable elasto-plastic model in effective stresses should be simple, numer-
ically robust, yet sophisticated enough to reproduce the essential features of
soil behaviour described in the beginning of this subsection.' The
168 PlANe
essential requirements for such a constitutive model should ideally include the
ability to: .
follow the stress path close to the shear failure line during cyclic loading of a
dense saturated sand, such as shown in the upper plot in Fig. T4.5(b);
reproduce the hysteresis loop of a hardening spring type, such as shown in the
lower plot in Fig. T4.5(b);
reproduce the progressive increase in the shear strain amplitude, such as
shown in the lower plot in Fig. T4.5(a, b);
analyse the cyclic behaviour of an anisotropically consolidated soil.
An example which meets these requirements is the constitutive model based on
a multiple shear mechanism (Towhata and Ishihara, 1985) among others. This
model was coded into a computer code (FLIP, see Iai et al., 1992) and found to
be capable of simulating the behaviour of sand discussed earlier and as shown in
Fig. T4.5(c, d). While these models rely on a fundamental physical understand-
ing of liquefaction phenomena, they still require careful calibration and field val-
idation studies in order to routinely apply them in practice (Blazquez, 1995).
Nevertheless, research is very active in this field and a growing number of com-
puter programs, e.g. DIANA (Zienkiewicz et al., 1999), DYNAFLOW (Prevost,
1981), GEFDYN (Aubry et al., 1985), TARA (Finn et al., 1986), among others,
have been made available for numerical analysis.
T4.2 MEASUREMENT OF SOIL PARAMETERS
The experimental determination of the cyclic soil properties defined in Section
T4.1, requires the use of conventional and/or ad-hoc testing procedures, which
can be grouped into the following main categories:
in-situ penetration tests;
in-situ geophysical tests;
laboratory tests;
Table T4.2 shows each test type and its applicability for a specific range of strains
and frequencies designated, and whether or not the test allows for direct or indi-
rect determination of soil properties.
T4.2.1 In-situ penetration tests
The static (CPT) and dynamic (SPT) penetration tests are used for determining
parameters for site response analysis, soil-structure interaction analysis and
liquefaction assessment of both subsoil and earth-fills. In particular:
strength parameters: the shear strength of soils is typically estimated
from CPT/SPT data based on semi-empirical background knowledge and
Table1'4.2. Field and lnborutory investigation techniques and related parameters.
Test class Test type Consolidation Shear strain Frequency Stiffness Damping Strength
stress state rl%] flHz] C F
rPenetration { SPT
Lithostatic - - Go - <P'
CPT Lithostatic
- - Cjc -) V
s
-) Go - <p' Cu
Field
{ Down-Hole
Lithostatic <10-
1
10-100 V
s

Geophysical Cross-Hole Lithostatic <10-
3
10-100 V
s
-cc, possible
SASW Lithostatic <10-
3
10-100 V
R
-) V
s
Go
( Tliaxial
Axisymmetric >l<r
1
0.01-1 q.e; G Wp/Ws -) f) qler: : N.
Cyclic Simple shear Axisymmetric >10-
2
(U)I-I
Wr/W
s
-) D ito; : N.
Torsional shear Axisymmetric

0.01-1 r:r-) Go, G Wt/W
s
-) f)
Laboratory ( or true triaxial
{ReSonant column
Axisymmetric

>10 ./; -) Go, G H.p., R.I'. -) f)
Dynamic or true triaxial
Bender elements Axisymmetric
< 10'\
> 100 V
s
-) Go
Legend: Vs =shear wave velocity; VR =Rayleigh wave velocity. j, =resonant frequency; l I.p. =half-power method; R.I'. =resonance
factor method; N = SPT blowcount; q; = CPT tip resistance; <P' =friction angle in effective stresses; C
u
= undrained shear strength;
C = coarse-grained, F =fine-grained soils; =deviator/radial stress rntio; tio; =shear/vertical stress ratio; N. =number of
cycles.
o
n-
o
,..,
o
()
::r
::3 _.
()
a
()
::I"
..,
()
,...
o
::J.
Vl
,..,
o'
::I
- 0-.
\0
]70 PlANe
experience. The relevant parameters are traditionally specified in terms of
. either effective (coarse grained soils) or total stresses (fine-grained soils);
shear-wave velocity profiles: regional and worldwide experience has led to the
proposals of a large number of correlations between penetration resistance
(qc for CPT, N for SPT) and shear-wave velocity (V
s
) or initial shear modulus
(Go). Examples of selected correlations based on CPT tip resistance and SPT
blow count are given in Table T4.3;
liquefaction resistance: as discussed later in Section T4.4, liquefaction charts
based on field correlation of penetration test data are widely used in design
practice, as often specified in codes and guidelines (e.g. ISSMGE(TC4),
1999). However, the penetration test procedures are different depending on the
regional practice. Table T4.4 shows the variability of the SPT procedures and
energy efficiency in various regions around the world. As an attempt to reduce
the variability, Seed et al. (1985) made recommendations for North-American
practice as shown in Table T4.5. If the energy efficiency of a particular proce-
dure is different from that adopted in the proposed empirical correlation, the
measured penetration test data should be corrected accordingly (see Section
T4.4).
Table T4.3. Examples of regional correlations between penetration resistance and shear
modulus or shear wave velocity.
Source Expression Coefficients Soil
Rix and Stokoe (1991) Go =k(qct( O":-o)b
Mayne and Rix (1993) Go =k(qct(e)b
Ohta and Goto (1978) Vs =
Holocene clays
Holocene sands
Pleistocene clays
Pleistocene sands
Holocene
Pleistocene
Clay
Fine sand
Medium sand
Coarse sand
Gravellv sand
Gravel
Holocene soils
Clay
Quartz sand k = 1634
a = 0.25, b = 0.375
k=406
a = 0.695, b =- 1.13
a = 102, b = 0.29
a = 81. b = 0.33
a = 114, b = 0.29
a = 97, b = 0.32
a= 1.0
a= 1.3
f3 = 1.00
f3 = 1.09
f3 = 1.07
f3= 1.14
f3= 1.15
f3 = 1.45
K = f(soil)
,h
Vs=aA
Baldi et al. (1988)
lmai (1977)
Key: Go, q., 0":-0, in kPa; N = SPT blow count at 60% maximum energy delivered:
:: =depth, in m; Vs =shear wave velocity in mls.
Table T.+--+. Variability of regional SPT procedures (Seed et al., 1 9 8 5 ~ Skernpton, 1986;
after ISSMGE (TC4). 1999).
Country Hammer type Hammer release Estimated rod
energy ER
m
( %)
Correction
factor for
60CJc rod
energy
Donut
Donut"
Tombi 78 1.30
Rope and pulley with 67 1.12
special throw release 1.2 (average)
Safety" Rope and pulley 60 1.00
Donut Rope and pulley 45 0.75
Argentina Donut" Rope and pulley 45 0.75
Donut" Free-fall" 60 1.00
Donut Rope and pulley 50 0.83
Pileon Trip 60 1.00
Old standard Rope and pulley 60 1.00
China
U.K.
U.S.A.
"Japanese SPT results have additional corrections for borehole diameter and frequency
effects.
'1>revalent method in each country today.
'Pilcon type harrimers develop an energy ratio of about 60%.
Table T4.5. Recommended SPT procedure for use in liquefaction assessment (Seed
et al., 1 9 8 5 ~ as reported ISSMGE (TC4), 1999).
Borehole
Drill bit
Sampler
Drill rods
Energy delivered
to sampler
Blowcount rate
Penetration
resistance count
4 to 5 in. diameter rotary borehole with bentonite drilling mud for
borehole stability
Upward deflection of drilling mud (tricone of baffled drag bit)
G.D. =2.00 in.: 1.D. =1.38 in. - Constant (i.e., no room for liners in
barrel)
A or AW for depth less than 50 ft: N or NW for greater depths
2520 in.-lbs.(600/0 of theoretical maximum)
30-40 blows per minute
Measures over range of 6-18 in. of penetration into the ground
T4.2.2 In-situ geophysical tests
In-situ geophysical tests permit direct measurement of shear-wave velocity V
s
,
from which the initial shear stiffness Go can be derived through the well-known
relationship
.,
Go =pVs (T4.6)
where p denotes the bulk soil density.
....:.....
]72 PlANe
As shown in Fig. T4.3(a) and Table T4.2, the strain levels attained by the geo-
physical testing are never higher-than 0.001% (i.e. within the linear range), and
the frequencies used are usually higher than those typical of earthquakes.
The field geophysical tests can be classified into surface and borehole tests,
with obvious implications in tenus of cost of test execution proportional to
drilling and related operations.
(1) Surface tests
The well-known seismic reflection and refraction surveys give only a rough
delineation of major units of soil layering, for instance allowing for the Iocalisa-
. tion of a bedrock. A description of the testing procedure and its interpretation is
reported in most textbooks (e.g. Kramer, 1996) and will not be presented here.
Worth remembering though is that these geophysical tests are most commonly
adopted for undersea investigations.
Higher resolution and degree of reliability among the surface geophysical tests
has been achieved by. the Spectral Analysis of Surface Waves (SASW) method
(Stokoe et aI., 1994). This test is based on the analysis of the propagation of
Rayleigh waves generated at an appropriate distance by applying a vertical pulse
on the surface of a subsoil (Fig. T4.9). The receivers are located symmetrically
with reference to the vertical profile investigated. The depth of subsoil surveyed
by this method is in proportion to the wave length AR = VR!feV
R
,f= velocity and
frequency of surface waves). By decomposing transient surface waves into
Fourier components and computing their phase velocity, V
ph
(see Appendix), it is
possible to describe an experimental dispersion curve (Vph:A) representative of
the subsoil shear stiffness profile. A numerical inversion procedure then needs to
be used, by assuming a layered wave velocity profile that most closely fits to the
experimental dispersion curve.
SASW (Spectral Analysis of Surface Waves) I
Section
x X/2
CL
X/2
:.. :1 :.:.::: ...::
... 1 ::::::: ".".'.".".'. :::::::::.:. 4 .
.. . :,. ::. ".".".".".".".".".",".".".".".".".". :::::.. .'.: ... 1 .
. :. :::::::.. :::::::::::::::::::::: . ' . .j : :::::: : ...: ".".".
Index
Source. Receiver 0
Fig. T4.9. Experimental setup for
SASWtests.
Even though the SASW data reduction procedure may be cumbersome. this
technique is operationally inexpensive. since it does not require drilling of a bore-
hole. In addition, its execution has been successfully tested underwater (e.g. Luke
and Stokoe. 1998).
(2) Borehole tests
The borehole tests can be classified according to the position of the impact
source. In the Down-Hole (DH) technique, a horizontal impact source is placed
on the surface and one or more receiver geophones are installed in the borehole.
The receivers are then lowered as shown in Fig. T4.10. Shear wave pulses with
predominant horizontal components (SH waves) are generated from the surface
and recorded at every geophone location. In the conventional test interpretation
practice. the shear-wave velocity V
s
results from the ratio between the inclined
ray path length and the first arrival time at the receiver (see Appendix).
The Cross-Hole (CH) test is carried out by drilling two (or more than two)
boreholes. in which a vertical impact source and one (or more) receiver geo-
phones are installed (Fig. T4.11). Shear-wave velocity Vs is given by the ratio
between the source-receiver distance and the first arrival time at the receiver (see
Appendix). The damping ratio (Do) can be evaluated by using two receiving bore-
holes (Mok et aI., 1988).
IDown-Hole I
Plan view
Sect/on
I Horizontal
, . tramduc813
I
I ... ~ ) " "
/ ' ~ j ~ ~ ~ 1 _. "'.: ..
DHsource/ t .....Borehole
Index
Source. Receiver a
Fig. T4.1O. Experimental setup for
Down-Hole tests.
174 PIANC
Section
ICross-Hole I
Index
Source. Receiver 0
Fig. T4.11. Experimental setup for
Cross-Hole tests.
An updated version of the above techniques, the so-called Seismic Cone (SC)
utilises a cone penetrometer tip as a receiver, to avoid drilling of the correspon-
ding borehole (Robertson etal., 1985). The seismic cone enables measurementof
both the shear-wave velocity and the tip resistance at the same depth, thus per-
mitting calibration of direct correlations between qc and V
s
. This technique has.
been successfully utilised in sea-bottom investigations, since it overcomes some
of the disadvantages (e.g. site accessibility and deployment) of other geophysical
tests such as cross-hole and down-hole.
The data recorded in borehole tests are analysed according to well-established
procedures, including manual picking of first waveform arrivals and numerical
Fourier analysis of the digitised records. More details are given in the Appendix,
which is located at the end of this technical commentary.
For most offshore geophysical tests carried out in the sea floor. the following
issues should be considered:
1) the boreholes have to be cased with either steel or plastic, which limits the
type of sondes that have to be used;
2) the high water content of sea floor deposits can create problems in the inter-
pretation of the logs, eventually 'masking' some of the responses to geophys-
ical logging;
3) practice has shown that the sondes have to be run very slowly through the sea
floor material to ensure that an adequate response is recorded.
T4.2.3 Laboratory tests
Laboratory tests can characterise soil behaviour over a wide strain range which.
according to the selected technique, spans from yas low as 0.0001% up to the
non-linear range and towards failure (see Fig. T4.3(a) and Table T.+.2).
- ~ ."
The laboratory testing techniques can be classified into:
cyclic tests. in which inertia forces in the soil element are negligible. and the
mechanical response is .characterised by direct measurement of stress and
strain values:
dynamic tests, in which the response needs to be interpreted by means of a
dynamic analysis of equilibrium of the soil element.
(l) Cyclic tests
a) Cyclic triaxial test (CTX)
The cyclic triaxial test (CTX) involves the application of cyclic compression-
extension loads on a cylindrical specimen (Fig. T4.12(a)). controlling stress or
strain amplitudes. There are some studies that suggest that strain-controlled tests
may be preferable to the frequently used stress-controlled tests because they may
be less sensitive to experimental factors (Dobry et al., 1979, 1980; Talaganov,
1986). The most up-to-date versions of mechanical and hydraulic loading
systems allow the application of arbitrary stress and strain paths to cylindrical
specimens. Since the principal stress directions alternatively switch between
horizontal and vertical, this testing pattern cannot fully reproduce the ideal
simple shear loading conditions acting on a soil element during an earthquake.
In the CTX, the axial .stiffness and damping parameters can be directly
obtained by analysing the deviator stress-axial strain loops (see Fig. T4.12(b) and
Table T4.2). CTX tests are traditionally oriented to analyse cyclic failure behav-
iour, described by the relationship between the deviatoriclradial effective stress
ratio (q!G:) and the number of cycles at liquefaction (Nc). However, experimen-
tal limitations (such as equipment compliance and resolution in the axial strain
Extension t
~ q
Compression
~ q l l l l
(a)
Fig. T4.l2. Cyclic triaxial test.
(a) Loads on soil specimen.
(b) Interpretation.
Ca,pp
(b)
176 PIANC
measurement) may limit the reliability of CTX tests for the characterisation of
pre-failure behaviour at small and medium strain levels.
Recent research developments in advanced triaxial equipments (e.g. Tatsuoka
et aI., 1997) showed that shear strains as low as 0.001% can be resolved in static
(e.g. Goto et al., 1991) andcyclic (e.g, Olivares, 1996) triaxial tests by using local
displacement transducers (LDT). However, these techniques have not yet been
commonly adopted in design practice.
b) Cyclic simple shear test (CSS)
The cyclic simple shear test (CSS) is performed by applying a cyclic tangential
load on the boundaries of a short-height specimen (Fig. T4.l3(a, the shape of
which can be either prismatic (Cambridge type) or cylindrical (NGI and SGI
types, shown in Fig. T4.l3(a. The non-compliant cell walls permit K
o
consoli-
dation state before the application of the cyclic shearing, which is usually strain-
controlled.
. The stiffness and damping parameters can be directly obtained by interpreting
the shear stress-strain loops (Fig. T4.13(b and defined in Fig. T4.2. Since the
state of shear stress and strain within the specimen can be far from uniform, the
reliability of measurements can be much reduced at small and medium strains.
As in the CTX tests, this laboratory technique can be used to investigate lique-
faction potential, described by the relationship between the shear/vertical effec-
tive stress ratio ( T / ( J ~ , ) and the number of cycles at liquefaction (Nc).
c) Cyclic torsional shear test (CTS)
The cyclic torsional shear test (CTS) is performed by applying a cyclic torque to
the upper base of a cylindrical slender specimen (Fig. T4.14(a. The possibility
7
\
I
I
I
I
\
t
I
a
I
B
.
l
~ 7(t)
0.
~ ~ - - - - - - - - ~ 0.
r . . r
la\
Fig. T4.l3. Cyclic simple shear test.
(b) Interpretation.
r
G=2E.
'Ypp
(b)
of using hollow specimens. and independently controlling inner and outer con-
fining pressures (Fig. T4.14(a. permits the application of triaxial consolidation
prior to the. application of shear loads. However. testing on natural soils is prac-
tically feasible only for solid specimens (Fig. T4.14(b)). The test is normally
stress-controlled. with torsional loads applied by means of several kinds of
devices. more frequently electro-magnetic (e.g. Isenhower. 1979; dOnofrio
et al., 1999b) or electro-mechanical (e.g. Bolton and Wilson, 1989; Alarcon
et al., 1986).
As in the CSS test, the stiffness and damping parameters can be directly
obtained by interpreting the shear stress-strain loops according to their defini-
tions (Fig. T4.14(c. Due to the high degree of uniformity and to the elevated
resolution in the measurement of shear stress and strains in the specimen. the
reliability of pre-failure equivalent parameters evaluated (G and D) is expected to
be high at small and medium strains. Normally, with electro-magnetic devices, it
is not possible to bring a specimen to failure state; in such a case, the CTS test is
not suitable for assessing liquefaction potential.
(2) Dynamic tests
a) Resonant column test (RC)
The resonant column test (RC) is currently performed with essentially the
same electro-magnetic driving devices adopted for CTS tests (Fig. T4.15(a)).
a, a,
a, a,
(a)
(b)
T
Ws
't
D= W
o
T
llll
G =-IT..
Ypp
41tW
s
'tpp
(e)
Fig. T4.14. Cyclic torsional shear test.
(a) Loads on hollow cylinder specimen.
(b) Loads on solid cylinder specimen.
(c) Interpretation.
178 PlANe
A cylindrical specimen is loaded with a constant amplitude torque with variable
frequency until the resonance of the assembly of the soil specimen and the driv-
ing system is achieved (Fig. T4.15(b)).
The soil parameters are obtained from the interpretation of the frequency
response curve (Dmevich et aI., 1978): the shear-wave velocity V
s
(hence, the ..
shear modulus G) is obtained from the measurement of the resonant frequency fr,
and the damping ratio Dfrom the half-power frequencies f1 and f2. The damping
ratio can also be measured from the acquisition of the free vibration decay curve
(Fig. T4.15(c. However, with the RC test, as with the CTS test, it is not possi-
ble to bring a specimen to failure. Consequently, the test is only suitable for
investigating deformation properties.
b) Bender elements (BE)
The bender elements (BE) are active and passive transducers, made with piezo-
ceramics or piezocrystals (Dyvik and Madshus, 1985), which can be installed at
the top and at the bottom of a specimen mounted in a triaxial cell (Fig. T4.l6(a.
Recently, this technique has been successfully introduced in large-scale triaxial
equipment that was sized to test coarse-grained soils such as gravels and rockfills
(Modoni et aI., 2000).
When the active (transmitter) transducer is excited, it generates a single pulse
shear-wave that propagates along the specimen length and is recorded at the other
end by the passive (receiver) transducer. The direct interpretation of the test is
G=pV
S
2
o
D:-
2n
Mit)
1=1,
0
\ II
11 1\ 11
Molt)=O
t
V~ ~ V V ~
.,
G=pV
S
-
D:/
2
-
f L
2f
T
I
-=FtanF
1
0
V
s
:: 2r:frL
F
Molt) --increasing 1 -
e.... 1./2
L
Oa
10
tal ih) Ie)
Fig. T4.15. Resonant column test.
(b) Interpretation of resonance test.
(c) Interpretation of free vibration decay test.
r - : ~ . ------I
o 1000 III l..---.:' _
, ,
;--,ts
R
L
Q;
>
'Qj
u
..
a:
S
n
I
t
L
.'-;':':::-::::->."::;'
..
l
.._ ~ " . ' : " :.: .....
,.
}:',':'\It",.":
u
R
I I
(a) (b)
Fig. 14.16. Bender element test.
(b) Interpretation.
therefore similar to that of the geophysical field tests. The visual analysis of time
records of vibration amplitudes permits the measurement of shear-wave velocity
Vs. from which the small-strain shear modulus Go is -calculated through
Eqn. (T4.6). Wave propagation analysis in the frequency domain (such as that
adopted for in-situ tests) is also recommended for a more accurate investigation
(Viggiani and Atkinson, 1995).
T4.2.4 Combined use of in-situ and laboratory tests
It is important to remember in the geotechnical design practice, that field and lab-
oratory investigation techniques have their own advantages and disadvantages.
In-situ geophysical tests can measure soil properties in their natural state. This
is achieved by interpreting vibration records, based on the theory of propagation
of seismic waves within a layered half-space. However, these methods do not
allow the measurements of non-linear soil properties since the vibration energy
level of geophysical tests is typically not strong enough to allow the development
of large strains.
Laboratory tests allow the measurements of. soil behaviour in a strain range,
much wider than the in-situ tests. The laboratorycharacterisation of an earth-fill
material before its placement may be achieved through a reliable reproduction of
the in-situ conditions by means of appropriate specimen preparation, compaction
and re-consolidation procedures. However, for natural subsoil, the laboratory
results are often affected by soil disturbance during sampling. Other disadvan-
tages include limits in reproducing in-situ stress states, and scaling effects due
to the limited size of soil specimen tested for seismic excitation. The sample
]80 PlANe
disturbance can significantly affect the test results, especially for underwater sed-
iments. Care should be taken for choosing the most appropriate sampling tech-
nique among those available in local practice; regional recommendations are
available on the suitable sampling methods depending on the soil type and den-
sity (e.g. CEN, 1997; Port and Harbour Research Institute, 1997)
In view of these advantages and disadvantages, a combined use of in-situ and
laboratory test results may be recommended as follows.
(l ) Pre-failure conditions
The shear modulus and damping ratio at small-strain, Go and Do, are sensitive to
the sampling procedure. In the medium to large strain ranges, the normalised
equivalent shear modulus, GIGo, is relatively insensitive to sampling disturbance.
Therefore, the design strain dependent soil stiffness may be best evaluated
through
(T4.7)
In Eqn. (T4.7), (GO)field is preferably determined from the shear-wave velocity
measured in-situ, while the strain-dependent normalised shear modulus GIGo
should be obtained by the laboratory tests. '
The D( nrelationship can be expressed as
D(y)=(D
O
) +D(y)
(T4.8)
(T4.9)
and it is typically determined from laboratory tests only. If Dois measured in-situ
(e.g. from CH tests performed with at least two receiver boreholes). then Eqn.
(T4.8) should read:
D(y) = (Do )field +[D(y)] lab
where [D(y)] represents the laboratory measured damping ratio scaled to the rel- .
evant small-strain value Do.
(2) Failure conditions
With reference to failure conditions. the following laboratory tests have been
developed to characterise soil properties:
direct liquefaction tests (stress or strain-controlled). with pore-pressure
recording on undrained saturated samples:
indirect liquefactiontests (constant volume tests). with densification recording
on dry soil samples.
By cyclic laboratory tests, a considerable saving in testing effort may be
achieved. since the pre-failure properties of the soil may be obtained along with
.the cyclic strength characteristics. However. it is most often the case that these
laboratory test results are strongly affected by many testing factors. frequently
leading to low reliability in the test results. Figure T4.17 shows examples of the
effects on the cyclic strength curves of experimental factors such as compaction
procedures (Fig. T4.17(a and re-consolidation techniques (Fig. T4.17(b).
For the reasons stated above, in-situ estimators of liquefaction susceptibility
(by means of static and dynamic penetration resistance, or V
s
values) are cur-
rently used (see Section T4.4). Each one of these field measurements has advan-
tages and disadvantages, and their suitability is also soil-dependent (Table T4.6).
0.2
1

l!: t-
iii
.2
o 0.1
0.5 I
Low frequency Dr=50%
vibrations on dry
samples (Jo'=8 psi
'00 0.4 (55.2 kPa)
C\l vibrations on dry
""lJ samples
0"Q
+l 0.3
o
...
3 10 30 100 300
Number of cycles to initial liquefaction and 2.5% axial strain
oL.
1

(a)

Undisturbed samples
Samples tested after
re-applyingpreconsolldation
history
00 0
0.6
l:l 0.3
l!:
iii
0.2
0.1
4 10 20 40 100 200 400
(b) Number of cycles required tocause initial fiquefaetion
Fig. T4.17. Effects on the liquefaction resistance.
(a) Effects of sample preparation techniques (after Mulilis et aL, 1975).
(b) Effects of re-consolidation techniques (after Mori et aI., 1978).
]82 PlANe
Table T4.6. Comparative performance of various field tests for assessment of liquefac-
tion potential (Youd and Idriss, 1997).
Feature Test Type
SPT CPT V
s
Number of test measurements Abundant Abundant Limited
at liquefaction sites
Type of stress-strain behaviour Partially drained, Drained, large Small strain
influencing test large strain strain
Quality control and repeatability Poor to good Very good Good
. Detection of variability of soil Good Very good Fair
deposits
Soil types in which test is Non-gravel Non-gravel All
recommended
Test provides sample of soil Yes No No
Test measures index or Index Index Engineering
engineering property property
To summarise, site and laboratory tests are complementary. By combining the
test results of both investigations, an adequate description of soil behaviour for
seismic design may be achieved. The most appropriate experimental strategy may
be chosen in accordance with the analysis requirements; this point will be dis-
cussed in Section T4.5.
T4.3 EVALUATION OF PRE-FAILURE SOIL PROPERTIES THROUGH
EMPIRICAL CORRELATIONS
Conventional practice in geotechnical characterisation suggests that, whenever a
soil property is directly measured or indirectly evaluated, it must be reviewed in
a light of pre-existing test data and experience. In compiling the knowledge gath-
ered on natural soils, pre-failure soil properties are often correlated with grain
size, micro- and macro-structure, and geologic age. In earthquake geotechnical
design practice, empirical correlations with index properties are commonly
referred to, and often suggested by relevant codes or guidelines (e.g. EPRL 1993:
PHRI. 1997). Some of these useful empirical correlations are summarised and
discussed below.
T4.3.1 Small-strain stiffness
The small-strain shear modulus Go can be expressed as a function of the state
variables (mean effective stress pi, void ratio e, and overconsolidation ratio OCR)
as (Hardin. 1978)
Go =S(LJ' !(e)OCR
m
e. Pa
(T4.10)
where pa =atmospheric pressure, fee) is a decreasing function of e, and S, n, m
are dimensionless parameters. The function f(e) varies depending on the soil
types. The most common relationship forfee) is that given by Hardin and Black
(968)
f
(2.973 _e)2
( e ) = ~ - - -
l+e
(T4.11)
Taking this function as a reference for many natural soils, Mancuso et al. (1997)
obtained the variation of Sand n with I p shown in Fig. T4.18. For non-plastic
granular materials, the parameter S varies from 200 to 400; for fine-grained soils,
it gradually decreases with increasing plasticity index I p (Fig. T4.18(a. The
parameter 11 increases with I p from an initial range 0.4-0.6 (gravels and sands) to
about unity for high plasticity clays (Fig. T4.18(b)). The plot showing the
dependency of the third parameter, m, on I p (Fig. T4.18(c)) was originally repor-
ted by Hardin and Black (1969).
For normally consolidated soils (OCR = 1), e and p' are correlated with each
other (i.e. the normal compression line); therefore, the dependency of Goon the
state variables can be simplified as a power function of the mean effective stress
p' as follows (Rampello et al., 1994)
Go =s(lt-Jn
Pa Pa
(T4.12)
The small-strain stiffness data on marine clays were collected from existing
literature by d'Onofrio and Silvestri (2001), compiled as shown in Fig. T4.19.
The stiffness coefficient S decreases with le (Fig. T4.20(a)) in the similar manner
as shown in Fig. T4.18(a), but with different meaning because of the different
formulation of Eqn. (T4.12) with respect to Eqn. (T4.10). The trend' for n in
Fig. T4.19(b) is almost the same as in Fig. T4.18(b).
The small-strain stiffness Go can be also expressed as a function of soil
density p and shear-wave velocity V
s,
through Eqn. (T4.6). The case histories
indicate that, at moderate depths, the typical values of p and V
s
for the majority
of soils are those reported in Table T4.7. These values can be useful for
obtaining a rough estimate of Go.
T4.3.2 Small-strain damping
The laboratory tests show that, irrespective of the soil type, the damping ratio at
small-strains, Do, decreases with the mean effective stress, p', Based on the
I
1
,: , I'
," .
I
I
184 PlANe
reviewof literature data, the influence of soil type on Do is reported by Mancuso
et al. (l997) as follows:
for granular soils (sands, gravels, rockfills), the range of variation of Do with
stress state and history is very narrow and approaches zero;
50 40 20 30
plasticity index, Ip
'0
(a)
600
C/)
500
gravels and sands
~ silu and clays

c
0
ll>
400
'0
;;::
....
ll>
300 0
o
III
III
200
ll>
.E
....
.;:
III
1.2 r---'"-------------------,
50 40
gravels a ~ d sandsl
silts and clays i
o i
'-------'! l
! !
20 30
plasticity index. Ip
10
,
c
x
~ 0.8
of
CIl
CIl
.E 0.6
....
-;:;
CIl
(b)
0.7 r----------------------,
0.6
E
0.5
~
c
~ 0.4
o
~ 0.3
ll>
5 0.2
o
o. ,
(c)
20 40 60
plasticity index, Ip
80 100
Fig. T4.18. Variation of stiffness parameters with plasticity index Ip for natural soils
(after Mancuso et al., 1997).
(a) Parameter S.
(b) Parameter 11.
(c) Parameter m.
50 40
!
i
... :
I
!
20 30
plasticity index, I p
10
!'-..------".c----;-------------
1500
1250
(/)
'E
C1l
1000
'0
c
Qi
0
750
c
III
VI
C1l
c:
500
~
iii
250
0
0
(a)
,
I
i
,
; I
!
_.
,
1- -
.-
--
1
......
I
, I
! ...
I t
j
I
...
...
I
I -...
,
I
i
A
...
...
...
!
... :
...
I
...
,
I
! ,
I
I
I
i
t i
I
f
: I
,
!
0.8
c:
"Ii
CD
0.6
"0
.::
CIl
CIl
C1l
0.4
c:
;;:
iii
0.2
0
(b)
o 10 20 30
40 50
Fig. T4.19. Soil stiffness parameter variation with plasticity index I p for normally con-
solidated marine clays (after d'Onofrio and Silvestri, 2001).
(a) Parameter S.
(b) Parameter n.
(T4.13)
for fine-grained natural soils, the average values of Do and the rate of decrease
with stress level become higher going from stiff to soft clays; also, at the same
stress state and history, Do increases with plasticity index;
the values of Doof medium/fine-grained compacted materials are often high-
er than those typical of natural clays, due to the relatively young age of the
man-made soil deposits.
For the same normally consolidated marine soils analysed to obtain the correla-
tion in Fig. T4.19, d'Onofrio and Silvestri (2001) used the simple power function
relationship
Do=Q(:J
186 PlANe
5r------r----r---,.-----,-----,
50 40 20 30
10
(a)
-

-

"0
E
Cl.l
o
C,) 2 ..........'i__---':::_""'::::::..---1
Cl
c:
'5.
E

o
60 50 10 20 30 40
plasticity index. I p
I
I
I
!
,
I
-
I
A
A I
AA
A
I
I
-
A
I I
i
A I
I
.-
i
, j
(b)
o
o
0.5
0.1
0.4
x
0.3
Cl
c:
c.. 0.2
E
co
"C
Fig. T4.20. Soil damping parameters variation with plasticity index I p for normally con-
solidated marine clays (after d'Onofrio and Silvestri, 2001).
(a) Parameter Q.
(b) Parameter j .
to relate the decrease of the small-strain damping to the increase of mean effec-
tive stress. The correlation with the plasticity index lp of the damping parameters
Q (representing the value of Do. expressed in %. at a p' =pa) and j, shown in Fig.
T4.20 was obtained. The plots confirm that both the values of Do and its rate of
decrease with p' increase with soil plasticity.
T4.3.3 Non-linear pre-failure behaviour
.Several correlations have been proposed in the literature for specifying the strain-
dependentproperties G(J1 andD(J1 using the plasticity index I.
Table T4.7. Typical ranges of soil density and shear wave velocity.
Soils Typical ranges
---:--------------
p
Organic soils. peats
Very soft clays
Loose volcanic silts/sands
Soft clays. loose sands
Dense/cemented volcanic silts/sands
Stiff clays. dense sands
Cemented sands and gravels
Intact rocks
l.0 to 1.3
1.4 to 1.7
1.0 to 1.7
1.6 to 1.9
1.2 to 1.9
1.8 to 2.1
1.9 to 2.2
2.0 to 3.0
<100
100 to 200
200 to 400
400 to 800
>800
Based on the test data found in literatures, their own laboratory test
data, Vucetic and Dobry (1991) report the average G(y) and D("/) curves shown
in Fig. T4.21. Both plots show that the degree of non-linearity (hence, the
energy dissipation) increases with decreasing plasticity. The granular soils with
[p = 0, shownon-linearity over a linear threshold strain y, of about 0.001%. Note
that the curves relevant to the damping ratio are not scaled to the small-strain
value, Do, and that, according to the observations in Section T4.3.2, such small-
strain value should increase with I p and decrease with p',
The strain dependent curves for Japanese marine clays have been proposed by
Zen et al. (1987) as shown in Figs. T4.22 and T4.23. The effects of variations in
the state variables (e, pi, OCR) on the shape of nonnalisedcurves were found to
be negligible for fine-grained high 'plasticity soils (Fig. T4.22(a. On the other
hand, the increase in stress level can significantly shift to the right the G(y)
relationship for a medium to low plasticity soil (Fig. T4.22(b, c)). In contrast to
the soils compiled by Vucetic and Dobry (1991) (Fig. T4.2l), the Japanese
marine clays do not appear to exhibit dependency on plasticity for [p > 30. Again,
the degree of non-linearity is seen to increase with decreasing plasticity. The
influence of state and consolidation variables on the increase of damping ratio
with shear strain was not clearly established, and the average D(i? curves given
in Fig. T4.23 were suggested.
The linear threshold strain ri, beyond which soil non-linearity is observed, can
be also correlated with the plasticity index [P. Figure T4.24(a) shows that,
depending on particle grading and microstructure, Yl may vary from 0.0001 % to
more than0.01%. The values of YI for granular soils increase with decreasing par-
ticle size; those for volcanic sands and silts are higher with respect to those typ-
ical of other non-plastic soils, due to the stronger inter-particle bonds provided
by cementation and/or particle interlocking. The degree of non-linearity of com-
pacted fine-grained soils may be moderately affected by plasticity (Mancuso
et al., 1995), whereas that of natural clays can be strongly affected by
]88 PlANe
0.001 0.01 0.1
CYCLIC SHEAR STRAIN.'" (%)
0.8
c.s
-C) a.'/'
0.2

(a)
IOCR =1-151
Ip =200
3

10
10
15
30
50
100
0.001 0.01 0.1
CYCLIC SHEAR STRAIN .., (%)
IOCR=1-S!
aL..-__--"__
a.OOO1
(b)
25
5
20
e
o
615
.-
<l:
e::
10
0-
:E
<l:
Cl
Fig. T4.21. Strain dependent stiffness and damping (aflerVucetic and Dobry. ]99]).
(a) Normalised equivalent shear modulus G(y).
(b) Equivalent damping D(.".
macro-structure. For instance, the linear threshold level of a fissured clay can be
much lower than that of homogeneous reconstituted samples of the same soil (see
Fig. T4.24(a)).
The volumetric threshold level }\., beyond which soil dilatancy is observed.
increases with I in a similar way as that of yl, and is about 1 to 2 orders higher
than YI as shown in Fig. T4.24(b).
The average dependency of the degradation index tG (defined in Section
T4.1.1) on the strain level is shown in Fig. T4.25 for normally consolidated soils
with different plasticity (Vucetic, 1994). A significant degradation occurs when
the strain exceeds the volumetric threshold j; (increasing with lv): thereafter. IG
increases with the strain level. At a given strain amplitude, the rate of degrada-
tion of shear modulus decreases with soil plasticity.
0.8
0.6
o
~
0.4
0.2
I
p>30
(No influence of (J m'l
......
---
0.8
0.6
o
8
0.4
0.2
(b)
0.8
0.6
o
o
~ 0.4
0.2
(c)
10.5
10-4
r
r
10-3
10-
3
I
p=9.4
to 30
10-
2
Ip=N.P. to 9.4
10-
2
10-
1
10-
1
Fig. T4.22. Normalised equivalent shear modulus stiffness G(y) of Japanese soils (after
Zen et al., 1987).
(a) Soils with plasticity index Ip ~ 30.
(b) Soils with plasticity index 9.4 ~ t, ~ 30.191.
(c) Soils with plasticity index Ip ~ 9.4 including non-plastic soils.
190 PIANC
o
0.2
0.1
-----
---
----
---
--
10-' 10-
2 10-
3
r
10-4
............................a.--_......:.......J
10-6
Fig. 14.23. Equivalent damping ratio D(YJ of Japanese soils (after Zen et al., 1987).
0.1
0001
.verage values compacted subgrades
(Vueetie &. Dobry, 1991) (Kim,.' 991)
plio-pleist. nat. clays
plio-pleis\,rec. clays
100 80
AVERAGE
20 40 60
__..._litt_--L---......J
0.0001 0.001 0.01 0.1 1.0
(h) CYCLIC THRESHOLD SHEAR STRAINS, n and yo. (%)
0.000
,
0
(:1)
60
-0.
50
x
ill
Cl 40
Z
30
0
t- 20
(f)
<:
-J
c, 10
Fig. 14.24. Influence of grain size and microstructure on the threshold strains.
(a) Linear threshold shear strain YI (after Mancuso et al., 1995).
(b) Volumetric threshold .shear strain r (after Vucetic, 1994).
0.10 r-,------,--r----------.
Fig. T4.25. Influence of plas-
ticity index on
strain-dependent
parameter tc (after
Vucetic, 1992).
1
OCR=1
0.5
Cyclic shear strain amplitude. Yc(%)

0.0
o
..:
Q.)
-
Q.)
E
(tj
0.
c 0.05
o
g>
.0
T4.4 ASSESSMENT OF LIQUEFACTION POTENTIAL
Studies on liquefaction potential assessment have been primarily based on labo-
ratory tests. These have lead to a development of conventional test procedures
that approx.imately represent equivalent field stress conditions. For example,
irregular time histories of earthquake motions have been often approximated by
equivalent uniform cyclic loads in the laboratory (Seed et al., 1975; Annaki and
Lee, 1977). Hence, soil liquefaction is characterised by cyclic resistance curves
(see Fig. T4.7) and pore-water pressure time histories (see Fig. T4.4).
Despite the further developments in laboratory liquefaction test procedures
(Seed, 1979; Ishihara, 1985), the laboratory-based approach has been known to
be sensitive to many factors and uncertainties. The liquefaction resistance can be
quite sensitive to physical properties of soils, including fines content and degree
of cementation. They are also highly sensitive to small variations of the degree of
saturation in the soil specimen (Fig. T4.26). Thus, the laboratory tests are often
not performed in routine design practice; liquefaction potential is typically
evaluated based on field investigation results (Section T4.4.1). If required, the
liquefaction potential assessment can be improved by combining additional
information obtained from laboratory test (Section T4.4.2).
T4.4.1 Liquefaction potential assessment using field data
Assessment of liquefaction potential is typically accomplished based on
simplifiedprocedures correlating undrained cyclic resistance at in-situ conditions
with field performance (i.e. case histories of liquefaction observed during
192 PlANe
earthquakes). These procedures are available in seismic codes and technical rec-
ommendations (e.g. CEN, 1994; Youd and Idriss, 1997; ISSMGE(TC4), 1999).
In these procedures, two variables need to be evaluated: the seismic load
(cyclic stress ratio, CSR) and the capacity of the soil to withstand liquefaction
(cyclic resistance ratio, CRR). soil will be then determined liquefiable if .
CSR> CRR.
(l) Evaluation of seismic load
There are several ways to define the cyclic stress ratio CSR. The most common
expression of CSR is given by the following equation (Seed and Idriss, 1971):
CSR = =0.65 amax(surfacc:) a 'd
(jvO g avO
(T4.14)
where amax(surface) is the peak horizontal acceleration at the ground surface, g is the
acceleration of gravity, avO and are respectively, the total and effective verti-
cal overburden stresses. The coefficient rd is a stress reduction factor at the depth
of interest, which accounts for the defonnability of the soil profile. The range of
variation of rd with depth, as proposed by Seed and Idriss (1971), is shown in
Fig. T4.27. Analytical expressions to approximate the mean rd(Z) profile are
suggested by the consensus made at the workshop on evaluation of liquefaction
resistance of soils at the National Center for Earthquake Engineering Research,
hereafter called 'NCEER workshop' (Youd and Idriss, 1997). As indicated in
the figure, the validity of these profiles becomes uncertain at depths greater
than 15m. When possible and applicable, it is suggested to assess the value
assumed for I'd through the site response analysis.
c
.:.
0.2
\..

1 3 10 30 100
Number of cycles to cause liquefaction
Fig. T4.26. Influence of the degree
of saturation on the
liquefaction resistance
(after Martin et al.,
1978).
., .
27
Fig. T ~ . 2 7 . Variation of the stress
reduction factor rd
with depth within soil
profile (after Youd and
Idriss, 1997; adapted
from Seed and Idriss,
1971).
Avcraue values
- \
\
Mean values of rd \
Range fordifferent
soil profiles
<>, : .. : ....
. '.
30 '. I .' .,
24 . regIOn. . .
-- E 12
--
..c 1:>-
Q.
~
_0 18
21
6
9
(2) Correction and normalisation of field data
The liquefaction resistance CRR is typically evaluated using empirical charts to
be discussed in the subsection (3). In these charts, the in-situ test data (such as
SPT N-values, cone penetration resistance, or the shear-wave velocity) should be
normalised to account for variability in the reference depth and corrected to a
reference standard testing procedure, as shown below.
As mentioned in Section T4.2.1 and shown in Table T4.4, the energy efficien-
cy of SPT procedures are different in the various regions in the world. The same
table suggests broadly that energy efficiencies, ER
m
are 67 to 78% for Japanese
correlations and about 60% or even less for most other countries. If the energy
efficiency of a particular procedure, ER
m
is different from that adopted in the
proposed empirical correlation, the measured blow count N should be scaled as
follows
N =ER
mN
ER ER m
(T4.15)
(T4.16)
where ER and N
ER
denote the energy efficiency and blow count used for estab-
lishing the correlation between the cyclic strength and the SPT value'. For exam-
ple, the measured SPT blow count, N; in North American practice is typically
normalised to 60%energy efficiency as follows (Seed et al., 1985)
(N
l
)60 = C
N
. ER
m
Nm
60
]94 PIANC
where (N)6() is the reference blow count (at an overburden effective stress of 1
atmosphere (= Ikgf/crrr' = 98 kPa) with 60% energy ratio, CN a stress correction
factor depending also on relative density (Fig. T4.28(a.
A more general expression has been proposed by the NCEER workshop (Youd
and Idriss, 1997), namely
(N)60 =CNCECBCRCsNm (T4.l7)
where the correction factor CN accounts for overburden stress, CE for hammer
energy ratio ER
m
C
B
for borehole diameter, CR for rod length and Cs for samplers
with or without liners. The suggested ranges of values for these factors are listed
in Table T4.8. The expression of the stress correction factor C
N
in Table T4.8
refers to the overburden effective stress normalised with respect to the atmos-
pheric pressure P (Liao and Whitman, 1986). An alternative expression for the
correction factor C
N
may be given by
C
_ 1.7
N-
( ) ~ +0.7
(T4.18)
(T4.l9)
where c r ~ o is expressed in kgf/cm
2
(Meyerhof, 1957).
A similar normalisation procedure should be applied to the measured tip
resistance q., to obtain the reference value qcJ through the expression
C qc
qcl = q-
Pa
(b) c, or C
q
00 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
'"
-
~
--
100 ~
:...
'" :::; /
u: /
v- I'
<; Dr =60 to 80 %
u
,
...
C
Q
, /
c,
200
,
~
I
~
I
...
:::; I
::
I
~ 300 I
::
I
I
U I
;;.-
I
-
400
I f'I' . ER" . ~
u
(, I )Oll ~ c , 00' \n,
:.-
u;
;.,
;..
~
~
~ 400
=200
::.
i::
,.
;..
~ 300
-
,:,-
(a)
Fig. T4.28. Stress correction factors for SPT and CPT.
(a) C" for SPT blowcount (after Seed et al., 1985).
(b) C
II
for CPT tip resistance (after Seed et al., 1983).
Table T4.8. Correction factors for SPT blow count (Robertson and Wride. 1997).
Factor Equipment variable Correction Value
Overburden stress C
N (P3/ cr:ot
5
s 2
Energy ratio Donut hammer CE
0.5 to 1.0
Safety hammer 0.7 to 1.2
Automatic-trip 0.8 to 1.3
donut-type hammer
Borehole diameter 65 mm to 115 mm CB
1.0
150mm LOS
200mm 1.15
Rod length 3 m to 4 m C
R
0.75
4 m to 6 m 0.85
6 m to 10 m 0.95
10 m to 30 m 1.0
>30 m > 1.0
Sampling method Standard sampler Cs
1.0
Sampler without liners 1.1 to 1.3
The curve indicated by Seed et at (1983) for C
q
is shown in Fig. T4.28(b);
NCEER workshop (Youd and Idriss, 1997) suggested to use the generalized
expreSSIOn
c =
q a'.
va
(T4.20)
with n dependent on grain characteristics of the soil, and ranging between 0.5
(clean sands) and 1.0 (clayey soils).
The measured shear-wave velocity V
s
can be normalised to a reference effec-
tive overburden stress through the expression
(T4.21)
where n can be 3 (Tokimatsu et al., 1991) or 4 (Finn, 1991; Robertson et al.,

(3) Use of liquefaction charts
A liquefaction chart is usually expressed as a curve in a plane defined with the
cyclic stress ratio on the ordinate and the selected field estimator of liquefaction
resistance on the abscissa. The curve limits the boundary between the experi-
mental points representing observed liquefaction cases (above) and non-liquefied
cases (below). Thereforecit empirically represents the locus ofthe eRR, i.e. the
196 PlANe
minimum CSR required to produce liquefaction of a soil layer characterised by
the reference field property.
Figures T4.29 through T4.31 show the criteria obtained as the consensus at
the NCEER workshop (Youd and Idriss, 1997) for evaluation of liquefaction
resistance, respectively, based on normalised and corrected values of SPT blow
count (N
1
) 60, CPT tip resistance qa, and shear-wave velocity V
S 1
The curves in
these figures are given for earthquakes with moment magnitude M
w
= 7.5, on
gently sloping ground (slopes lower than about 6%) with low overburden pres-
sures (depths less than about 15m). The values of CRR for moment magnitudes
other than 7.5 need to be corrected by a scaling factor, CM, decreasing with the
design earthquake magnitude. Table T4.9 shows the values originally suggested
by Seed and Idriss (1982), together with the upper and lower bounds suggested

25'
Percent Fines =35 15 s 5
I
I
0.51------4-----fl'----t"--i-----+------l
CRR curves for 5.15. and
35 percent fines. respectively
I
I
I
I
, AIO
I
20
Adjustment
Recommended
ByWorlcshop"
.50+
FINES CONTENT 5%
Modified ChineseCodeProposaJ (claycontent== 5%) @
0.1
Marginal No
Liquefaction Liquefaction Liquefaction
Pan American data e
Japanese data 0 e
Chinese data. A
0 O.4I------+-----f-:::+--++----+------l
t{
<,
>
N"
.2
-
Cl:l
0:::
(/)
(/)
E
....
CI:l

U
10 20 30 40
Corrected BlowCount, (NJ)6O
50
. Youd and ldriss (1997)
Fig. Evaluation ,ofCRR from SPT N-value (Youd and Idriss, 1997, modified after
Seed et al., 1985).
Fig. T4.30. Evaluation of eRR from
CPT data (Robertson
and Wride, 1997).
300
1:--/0 Li4uelaction I
6. 6.
B
o 0
Oqj
o
6.
t:O
l'i<:W !'cn"ormancc l.i'l.:"o Liq_
!>Iark &: Uk,., 11'1'1; i 0
SuzukI et ,d. (\ QQS \ .. 6.
\
' 0.25 < D501mm) < 2.01
. Fe (%) <5 I
y..",10%;<:; 10\'::::3%
I I
, J
I
I J
I I
",I", J
I
so 100 150 200 250
Corrected CPT tip resistance. qel"

Workshop
\:1 -=- -.5 .

0.1
0.5
O.n
"'. . .
""Llquclactlon
,
0""
... ,.,
t "'",.J
It. 0
.. ,,,' 0 0
te oW 4J '" I?'o
... ., 10. t:I:) co
t>A
r7
:-:-::---:--'=-_ --:-:-_ ---,--,
-
-
YoU<! and ldriss (l9'>i)
as a consensus made in the NCEER workshop (Youd and Idriss, 1997) on the
basis of more updated contributions by several researchers.
In the charts in Figs. T4.29 and T4.31, additional curves relative to soils other
than clean sands are suggested, referring to variable fines contents, Fc, defined as
the percentage finer than 0.075 mm. Several correction procedures were sug-
gested to take into account in a general way, the increase of eRR with increased
fines content percentage (Seed et al., 1985). For the application of the SPT-based
method on soils with non-plastic fines (lp < 5), Robertson and Wride (1997) sug-
gested to convert the calculated (N
1
)60 to an equivalent value for clean sand,
(N
1
)6Ocs
(T4.22)
K, = 1 +0.025(F
c
- 5) (T4.23)
where Fe is the fines content. To date, a further correction factor (larger than Ks)
for soil with plastic fines has not been adopted by the geotechnical engineering
community due to the insufficient availability of data.
Similarly, for the application of the CPT-based method to sands with fines,
Robertson and Wride (1997) suggested to convert the normalised tip resistance,
qclN, to an equivalent value for clean sand, (qC\N)cs
'. (qeIN)cs =KeqelN (T4.24)
where the correction factor Kc- is expressed by a function of the. so-called 'soil
behaviour type index', L; as described in Fig. T4.32(a). The index Ie can be
estimated from the apparent fines content from the plot of Fig. T4.32(b), or
calculated by means of the combination of measurements of both tip and sleeve
resistance, with the procedure fully described by Robertson and Wride (1997).
198 PLANe
0.8r-----,----r-----r-----,.---r------,
300
No
Liquefaction

I I I
I I I
I I I
Liquefaction
-
--
100 200
Stress-Corrected Shear Wave Velocity, m/s
.........----.....---'---_..Jo--_--'-___..J
o
0.6
0
t{
<,
>
w'"
6
....
ca
r;x:
0.4
Vl
(I)
lI,)
'""
...
tI;)
0
<3
>.
U
0.2
Boundaries Definedby:
CRR =a(Vslll00)2 +b[l/(VSIc-VSl)-lNslcl M
w
= 7.5
a =0.03,b =0.9
V
S1e
=220 mls for sands and gravels withFC 5%
V
s
Ie =210 mls for sands and gravels withFe It: 20% .
VS
1e
=200 mls for sands and silts withFe 35%
Chan Based on:
Average values ofVs
1
& a
max
Uncemented, Holocene-age soils
Average fines content(FC)
Fig. T4.31. Evaluation of eRR from shear wave velocity (Andrus and Stokoe. 1997).
Table T4.9. Magnitude scaling factors required to modify CRR for M :t 7.5.
Seed and Idriss (1982) Youd and Idriss (1997)
Magnitude, M
w
5.5
6.0
6.5
7.0
7.5
8.0
8.5
CM
1.43
1.32
1.19
1.08
1.00
0.94
0.89
eM
2.20 to 2.80
1.76 to 2.10
1.44 to 1.60
1.19 to 1.25
1.00
0.84
0.72
More sophisticated correction factors.to account for layer thickness. high over-
consolidation states, static shearstress for sloping ground, and age of deposit,
have begun to appear in various publications, but they still belong to the research
domain. The topic was. extensively discussed during the NCEER workshop
4 3.5
lc=26
1 1.5 2 2.5 3
Soil Behaviour Type Index, Ie
0.5
Gravelly Sands Sands Sud

MiXlW'CS
,
,
-I 4 J 1 tI
K=.0.4031.; + 5.5811c - 21.63I
c
+ 33.75Ie- 17.88
<.
vi
Sik Clays

l<qCINks= Kc'lcINI
/
.
(a)
u
5
4
8-
?::
...
::s 2
o

.c
4,)
c::l
-1
'S
en
Zone 2: organic soils - peats
fiI'CS (I p,? -
..
.. .
Zone 3: silty clay to clay
...
.. ..
..
.. -
I4tiDrI
Zone 4: clayey silt to silty clay
..
en4
c4
GcncrGl carre
Rcco
lltJll
- ________ Zone S: silty sand to sandy silt.... -
- "r 5) .. _ -.
./ ..r<o.......
, .. .. .. - - .. .... Zone 6: clean sand to silty sand
, -.
....
..
.
N.B. The correlationis approximate
sinceIe depends on other factors such
Zone 7: gravelly sand to dense sand
as plasticity. mineralogy. sensitivity
andstress history.
o
o
(b)
10 20 30 40

50
Fig. T4.32. Correction factors for CPT resistance (Robertson and Wride, 1997).
(a). Correction factor K;
(b) Soil behaviour index lc.
(Youd and Idriss, 1997); nevertheless, most of these latter corrections appear of
minor importance for conditions specific to port structures.
T4.4.2 Liquefaction potential assessment combining field and laboratory data
A methodology combining field and laboratory data has been developed and
. . .
..ex.tensively used in standard design practice in Japan (Port and Harbour Research
". Institute, 1997; Yamazaki et al., 1998; Ministry of Transport, Japan, 1999). In this
200 PlANe
procedure, liquefaction potential assessment takes place in two steps. The first
step uses the grain size distribution of soil and standard penetration test blow
counts (SPT N-value) for a soil layer. If the results of this first step are close to
the borderline between liquefaction and non-liquefaction conditions, then the
first step can be supplemented with the second step, cyclic triaxial tests.
(1) Grain size distribution and SPT N- value (I st step)
Initially, a soil is classified according to the grain size distribution zones shown
in Fig. T4.33. The soil is considered non-liquefiable if the grain size distribution
falls outside of the liquefaction possibility zone in Fig. T4.33. For the soil with
grain size distribution curve falling inside the liquefaction possibility zone, the
liquefaction potential is evaluated based on the following procedure.
a) Equivalent N-value
An equivalent N-value, N
65
, is calculated by correcting for the effective overbur-
den pressure of 65 kPa using the following expression
N = N
m
- 0.019 (1: - 65)
65 0.0041 r: -65) +1.0
(T4.25)
0.1 1.0
Grain size (mm)
10
Gravel
Fig. T4.33. Grain-size distribution
of soil having the pos-
sibility of liquefaction
(Ministry of Transport,
Japan, 1999).
10
Gravel
0.1 1.0
Grain size (mm)
Soil with high coefficient of unifonnity Uc>3.5
: Clay
0.005
where Nm =measured SPT N-value of a soil layer (obtained with typical energy
efficiency in Japan). o; =effective overburden pressure of a soil layer (kPa). cal-
culated with respect to the ground surface elevation at the time of the standard
penetration test.
b) Equivalent acceleration
An equivalent acceleration. Qeq, is estimated by the following equation
a
=07.r=.0
eq , 0
Cf
v
(T4.26)
where r
max
=maximum shear stress in the layer obtained from site response
analysis. c r ~ =effective overburden pressure calculated, this time, with respect to
the ground surface elevation at the time of earthquake.
c) Liquefaction potential assessment for clean sands
For clean sands (i.e. soils with fines content Fe less than 5%), liquefaction poten-
tial is determined by using the chart in Fig. T4.34. The zone to which a soil layer
belongs is determined from the equivalent N value, N
65
, and the equivalent accel-
eration, Qeq. Liquefaction potential is assessed as follows:
- Zone I has a very high possibility of liquefaction. Decide that liquefaction will
occur. Cyclic triaxial tests are not required.
- Zone II has a high possibility of liquefaction. Decide either to determine
that liquefaction will occur, or to conduct further evaluation based on cyclic
triaxial tests.
30
25
to
co
Z
<D
20
::s
lij
~
Z
15
-
c
<D
cti
. ~
10
::s
0-
W
5
IV
/
I
ill
V,
./
/ .:
y
I
/;
/
./
P
100 200 300 400 500
Equivalent accelerationaeq(gal)
600
Fig. T4.34. Liquefaction potential
assessment based on
equivalent acceleration
and equivalent N-values
(Ministry of Transport.
Japan, 1999).
202 PlANe
Zone III has a low possibility of liquefaction. Decide either to determine that
that liquefaction will not occur, or to conduct further evaluation based on
cyclic triaxial tests. When it is necessary to allow a significant safety margin
for a structure, decide either to determine that liquefaction will occur, or to ,-
conduct further evaluation based on cyclic triaxial tests.
- Zone IV has a very low possibility of liquefaction. Decide that liquefaction
will not occur. Cyclic triaxial tests are not required.
d) Correction for silty or plastic soils
For a soil with fines content higher than 5%, the SPT N value is corrected accord-
ing to the following three cases:
Case 1: The plasticity index Ip < 10 or the fines content 5 < Fe < 15%.
An equivalent N-value obtained from Eqn. (T4.25) is corrected through the
following equation:
N;s = N
6SlCFC
(T4.27)
where: N;s = a corrected equivalent N-value, C
FC
= a correction factor based on
the fines content (Fig. T4.35). Using the corrected equivalent N-value, the
liquefaction potential is evaluated in the same manner as c) by referring to
Fig. T4.34.
Case 2: The plasticity index 10 < I p < 20, and the fines content r, > 15%.
Two corrected equivalent N-values are calculated as follows:
.1N = 8 + 0.45(1p - 10)
(T4.28)
(T4.29)
(T4.30)
Fines. content Fe (below O.075mm) (%)
Fig. T4.35. Reduction factor
for critical SPT
N-\'alue based on
the fines content
(Ministry of
Transport. Japan.
1999).
20 15 10 5
o
o
Ci3

'C 1.0 t----_..
o b:
at)
-(I)
:J
0_
-co
o >
co I 0.5
-z
c::

=n..
gCl)
"0
Q)
cr:
The two corrected equivalent N-values are plotted in Fig. T4.34 against the
equivalent acceleration, and the zone to which a soil layer belongs is deter-
mined as follows:
,. When N ~ ; is in Zone 1. liquefaction potential is evaluated by Zone I.
,. When N ~ ; is in Zone II, liquefaction potential is evaluated by Zone II.
,. When N ~ ; is in Zone III or IV, and when N;5 is in Zone 1. II, or III
liquefaction potential is evaluated by Zone III.
,. When N ~ ; is in Zone III or IV, and when N;s is in Zone IV, liquefaction
potential is evaluated by Zone IV.
Case 3: The plasticity index lp >20, and the fines content Fe > 15%.
Acorrected equivalent N-value is calculated using Eqns. (T4.29) and (T4.30),
and then liquefaction potential is evaluated using Fig. T4.34.
(2) Cyclic triaxial test (2
nd
step)
When the liquefaction potential cannot be determined from the grain size distri-
bution and SPT N-value (i.e. for zones II and II!), it is evaluated based on
undrained cyclic triaxial tests using undisturbed soil samples. Based on the
relationship between the cyclic stress ratio and number of cycles to cause lique-
faction in the laboratory tests, the cyclic resistance ('l",1( ) ~ ) N l = 20 for use in the
liquefaction potential assessment is obtained by reading off the cyclic stress ratio
at 20 cycles of loading (Fig. 1'4.36). Using this cyclic resistance, the in-situ
liquefaction resistance R
max
is obtained using the following equation:
(T4.31)
0.5
0.4
b(J
<,
~ 0.3
0
;
e
(/)
0.2
(/)
~
- en
0.1
,
,
,
,
R
max
(impact type:'
Rmax (vibration type: C
k=o./7)
....
.... -
---
------------
Correction for Ko (=0.5) stress state
1000
Nt=20
1 10 20 100
Number of cyclic loads
O'-- ......J- L.---I. --'- ~
0.1
Fig. T;4.36. Correction of laboratory liquefaction resistance for in-situ condition
(Ministry of Transport, Japan, 1999).
In this equation, the following corrections are applied
- Stress correction: The stress correction for in-situ (Ko) conditions from the triax-
ial test condition (isotropic consolidation with mean confining pressure of
- Input motion correction: The correction for irregularity of earthquake motion
(e.g. impact type/vibration type) from the harmonic motion in the cyclic tri-
axial test. The correction factors are given as
Impact type input motion Ck = 0.55
Vibration type input motion Ck = 0.7
The seismic stress ratio Lmax = is calculated through site response
analysis. The liquefaction potential (safety factor) F
L
is given as
F. =R
max
(T4.32)
L L
max
If F
L
< 1.0, the soil layer is expected to liquefy.
Laboratory data on pore-water pressure increase can be also obtained and used
for design of remediation methods against liquefaction (see Technical
Commentary 6).
T4.5 Geotechnical Characterisation Strategy with Reference to the Analysis
Requirements
In seismic analysis of port structures, appropriate geotechnical characterisation
should be made with respect to the following:
- soil profile and ground water table;
- initial effective stress state;
- physical and mechanical properties of soils.
In particular. a reasonable strategy should be established for adequate geotechni-
cal characterisation with respect to the mechanical properties of soils required for
seismic response analysis. According to the hierarchy of analysis procedures
addressed in Section 5.1 in the Main Text, the following stratezv mav be SUg-
'-.. eI
gested (see Table T4.10):
(l) Simplified analysis
In simplified analysis, local site effects are evaluated based on the thickness of
the deposits and the average stiffness to a specified depth (generally few tens of
meters), or over the entire deposit above the bedrock. This information is then
used to establish the site classification, leading to the use of specified site ampli-
fication factors or site dependent response spectra. .
The subsoil characterisation required for the simplified analysis may be
achieved based on the existing regional information on site category or by field
correlation using standard penetration tests (SPT) or cone penetration tests (CPT)
data. Liquefaction potential of sandy soils may also be evaluated based on
SPT/CPT through empirical liquefaction criteria. At this level, index properties
and conventional static tests may effectively support the geotechnical character-
isation andior the site classification.
(2) Simplified dynamic analysis
In simplified dynamic analysis. local site effects are typically evaluated using
equivalent linear modelling in total stresses. The subsoil is idealised into hori-
zontal layers of infinite lateral extent and one-dimensional (l D) wave propaga-
tion analysis is used. The pre-failure. strain-dependent parameters (G and D) can
be adequately characterised by means of geophysical tests (Section T4.2.2),
cyclic/dynamic laboratory tests (Section T4.2.3), and combining the field and
laboratory measurements through the methods suggested in Section T4.2.4. In
the absence of direct measurements by suitable field and laboratory tests, it is
common practice to use empirical correlations with penetration test data and
index properties (Section T4.3). These correlations are typically shown in design
guidelines (e.g. EPRL 1993; Port and Harbour Research Institute, 1997). The
applicability of these correlations should be confirmed with regional data before
using them in practice. This is especially important when special or regional soils
are encountered in design, including highly organic soils, cemented soils, soils
with fissures, weathered soils, or soils of volcanic origin.
In simplified dynamic analysis.Tiquefaction potential can be evaluated based
on a comparison of computed shear stresses during the design earthquake and the
results of cyclic laboratory tests, and/or based on SPT/CPT data. The liquefac-
tion potential evaluation based on the shear-wave velocity Vs may also be advan-
tageous. The advantages of using Vs may be distinctive for gravelly soils because
their penetration resistance may be misleading unless measured by the Becker
Penetration Test (BPT) (Youd and Idriss, 1997). However, caution must be exert-
ed with the straightforward application of shear-wave velocity (valid for small-
strains) to liquefaction analysis (which is a large strain mechanism). For instance,
the liquefaction potential of weakly cemented soils may be misinterpreted by the
method using V
s
, since these soils typically show unusually higher Vs relative to
their likely liquefaction resistance. It is preferable to apply two or more field-
based procedures to achieve consistent and reliable evaluation of liquefaction
resistance.
(3) Dynamic analysis
Dynamic analyses of soil-structure interaction (SS!) attempt to account for the
combined, or coupled, response of the structure and the foundation soils. Unlike
the simplified procedures wherein the response of the structure is evaluated using
the free-field soil ~ e s p o n s e as input, SSI analyses incorporate the behaviour
of both soil and structure in a 'single model. Numerical finite element (FEM)
or finite differencetf'Dlvl) methods are commonly used for advanced SSI analy-
ses. In this type of analysis, local site effects and liquefaction potential are
often not evaluated independently but are evaluated as a part of the soil-structure
Tnhlc T4.10. Geotechnical invcstigntion methods with reference 10 analysis requirements.
IV
0
0-
Type or Aspects Analysis method Input Soil Geotechnical investigation methods
parameters
"tl
analysis
SCISllllC
Gathering Field Laboratory
- :>
action
existing
PT GT Index Static Cyclic/Dynamic
Z
o
data
Prop. DT LT
Simplified SRA Silc category til Soil profile
.1 II
bedrock and
classification
LJQ SPT/CPT lique- at Nlqc
...

faction charts surface
.; -,
SSI Quay wulls/ Pseudo-slat icl Output or c, 1>, pn, D
...

Breakwaters Empirical SRA
.
Piled wharf Response Output of p : y curves
...

spectrum SRA
Simplified SRA ID total stress a(t) at Vs(z) profile
-
--
dynamic analysis bedrock G:y,D:y
..

curves
LJQ Field based SPT/CPTIVc;
at NlqJVs
.11111
approach Iiqucfaction surface
charts
Lab. bused IJ) tolal stress rCf) al Cyclic
II II
approach analysis surface strength
curves
SSI ()lIa)' walls/ Newmark/ ( )111 pill r, q" u. /)
...

Breakwaters Parametric or SRA
Piled wharf Pushover/ Output I' : Y curves
.1111111111111
Response ofSRA
spectrum
Key: SRA=Seismic Response Analysis; LIQ=Liquefaction; SSI =Soil Structure Interaction. PT =Penetration Tests. GT =Geophysical
,Tests, DT =Pre-failurelDeformation Tests, LT =Failure/Liquefaction Tests.
_iii. used as a standard
... use depends on design conditions
SRA Not included ID total stress
LIQ Not included 10 total stress
in SSI model analysis
10 effective
stress analysis
Included in FEMIFDM
SSI model
a
t'1l
o
(;
o
:r
::l

n
E:.
-
'1. " -, . '\-". ,.i... 'fl.' _

. "r , Plll
'i . ',' .}.,
'/'.
'

, Ll'.' f
1."' .. ,
. - -:
1
" .1. <

"+J
fIlftftl .. _
mIIIi m111
lfim
Vs(z) profile
c.r.o.r
curves
Cyclic
undrained
stress-strain
properties
bedrock
a(t) at
a(t) at
bottom
of analysis
domain
FEM/FOM
in SSI model analysis
10 effective
stress analysis
Included in FEM/FOM
SSI model
SSI
Dynamic
, .
n
....
(1)
::1.
Vl
....
o'
::l
IJ
o
-.l
208 PlANe
interaction analysis of port structures. If the bottom boundary of the analysis
domain for soil-structure interaction analysis differs from the bedrock (i.e. when
the bedrock is too deep for soil-structure interaction analysis), the local site
effects below the bottom boundary of the soil-structure interaction analysis
domain may be evaluated based on ID non-linear (effective stress) or ID equiv-
alent linear (total stress) analysis.
In this category of analyses, soil is idealized either by equivalent linear or by
an effective stress model, depending on the expected strain level in the soil
deposit during the design earthquake. Depending on the analysis requirements,
laboratory cyclic tests may often be required for evaluating undrained cyclic
properties in terms of effective stress-strain behaviour. Combined with
.field investigations, including SPT/CPTIV
s
, laboratory cyclic tests should achieve
the most adequate geotechnical characterisation for dynamic analysis of port
structures.
APPENDIX: INTERPRETATION OF GEOPHYSICAL TESTS
The interpretation of borehole geophysical tests can be performed by the anal y-
sis of the recorded signals in the time domain (time domain analysis) or in the
frequency domain (frequency domain analysis). .
A.l Time domain analysis
Figure A1 shows a typical recording of the source signal (an impact) and receiv-
er signals (generated waveforms) from a CH test in a soft clay deposit (Simonelli
and Mancuso, 1999). Visual analysis allows the determination of the travel times
of the waves between the source and the receivers, by picking the trigger time (T)
at the source, and the first arrivals of the waveforms at the receivers (P for com-
pression waves and S1 for shear-waves). Moreover, the travel times between the
receivers, located along a straight array, can be determined by comparing the first
arrivals S1, or even by picking the times of characteristic points of the waveforms
as peaks (see open circles in the figure) or zero-crossing points.
In the conventional procedure. the shear-wave velocity V
s
is given by the ratio
between the travel path of the waveform L; , and the travel time T;
V
s
=L
w
(A.])
T
w
Direct wave velocities (between the source and the receivers) and interval
wave velocities (between the two receivers) can be determined; their values
can be different, due to the 'spreading' of the waves during propagation (Hoar.
1982).
200
>
Ql
"0
::l
C.
E
-c
Trigger
200
150 t (rns) 200
First receiver
100
100

o
1550 r-n--r--r--r--r--r--r---r---r---r---r--.--.--.---.--.-""-..,.--r--1
-
>
Ql
"0
Ol-:l--+--""""----------@
c..
E
-c
Fig. A1. Example of time domain analysis of a CH test with two receivers.
The waveform velocity between two receivers located along a straight line can
be determined by the Cross Correlation function, provided that the signals are
acquired in digital form, The Cross Correlation [CCt).('r)] of two signals x(t) and
y(t) is defined as
+-
CC.t:'.(r)= Jx(t)y(t+r)dt
(A.2)
where r is the variable time delay between signals x(t) and yet).
Hence, a Cross Correlation value is given by the sum of the products of the
two signal amplitudes, after having shifted the second record, yet), with respect
to the first, x(t), by a time interval 1:. The Cross Correlation function yields dif-
ferent values by varying the time delay 1:, and reaches its maximum CCm:u when
the degree of correlation between the signals x(t) and y(t+r) is maximum (i.e.
when 'the waveforms are practically superimposed). The time delay 1:
max
corre-
sponding to CC
max
is the best estimate of the travel time of the whole waveform
between the two receivers,
The Cross Correlation shown in Fig. A2 is obtained using two receiver signals
(5 waves) of Fig. AI. The Cross Correlation values have been normalised with
., .. _".' re
210 PlANe
50 100 150 't'(ms)
200
Fig. A2. Cross-correlation function of the receiver signals in Fig. AI.
respect to the maximum CC
rnax
, hence the function varies between -1 and 1. The
travel time is approximately 62 ms.
A.2 Frequency domain analysis
By means of the Fourier analysis, a generic signal x(t) can be decomposed into a
sum of a series of sinusoidal functions, each one characterised by its amplitude
(A), phase (4)) and frequency (f) (see Fig. A3). The sine functions are the fre-
quency components of the signal x(t), and their complex representation in the fre-
quency domain is defined as the Linear Spectrum (Lx) of the signal xU).
The transformation of a signal in the correspondent Linear Spectrum can be
performed by means of the algorithm Direct Fourier Transform (DIT); inverse-
ly, it is possible to convert a Linear Spectrum into the correspondent time record
by means of the algorithm Inverse Fourier Transform (Iff). The frequency con-
tent of a signal is effectively represented by the amplitude versus frequency plot
of the Linear Spectrum; in Fig. A4, such diagrams are plotted for the receiver
(RI and R2) signals in Fig. AI.
Each sine function of a Linear Spectrum can be effectively represented by a
vector (with length proportional to the amplitude, and direction equal to the phase
of the sine function), and therefore, by a complex number. The rules and tools of
complex algebra are applicable to signal analysis in the frequency domain. In the
following paragraphs, the most useful functions for frequency domain analysis
are reviewed.
The Cross Power Spectrum CPS
xy
of two signals x(t) and yet) is the product of the
correspondent Linear Spectra L, and 4
(A.3)
Each frequency component (f) of the CPS has an amplitude [AcpsCfl] equal to
the product of the amplitudes of the same frequency components of the linear
spectra. and phase [4>cps(f)] equal to their phase difference. The Acps(j) allows
A
+
+
A
A
At
O==C>
- ~
=
t
-
I
t
+
A
+
Fig. A3. Transformation of an amplitude-time signal to an amplitude.frequency signal
by means of the OFf.
>
E
.....
IV
'"0
:;,
.-
Q.
E
<
50
z:6(m)
R1
100 150 200
--
>
E.
Q.
E
<
50
z:6(m)
R2
100 150 200
frequency (Hz) Frequency (Hz)
Fig. A4. Linear Spectra of two signals propagating between two receivers in a Cross-
Hole (CH) test.
one to determine the main frequency components of the propagating waveform,
while the (/Jcps(f) allows one to determine the delay of phase and then the travel
time of each component.
Generally, the CPS function is repeatedly applied to several couples of signals,
relative to the same propagation phenomenon. In fact, repeating the signals and
2]2 PlANe
averaging them drastically reduces the noise which affects signals during
propagation.
The CPS obtained from the signals similar to those of Fig. Al is shown in
Fig. A5. In the upper part, the Acps(j) shows that the main frequency components
of the propagating waveform lie in the range from 25 to 75 Hz. In the lower part,
the values of the tPcps(j) are conventionally plotted between actually the
real diagram should be obtained by translating the points A, B, C etc., respec-
tively, in A', B', C' etc.; the tPcps(f) values start from zero and gradually decrease
as frequency increases.
The Coherence function [COxy(j)] between x(t) and yet) signals is an algo-
. rithm which allows one to determine the level of noise affecting the frequency
components of the propagating waveform.
The COxy function, as the CPS, is applied to repeated couple of signals, rela-
tive to the same propagation phenomenon. The coherence and the corresponding
signal to noise ratio [S/R]xy are determined as:
--*
COr. (1) = (f) . CPS (f)
. APSyy(f)' APSi.r(f)
(A.4)
where APS
u
and APS
yy
are the Auto Power Spectrum of each signal, given by the
square of the linear spectrum of the signal (e.g. APS'CX= L.r(f) L.:(f. The bar over
the symbols CPS and APS indicate the averaged value of repeated signals.
The values range from 0 to 1, as shown in Fig. A6, for a series of two
signals from CH tests. The diagram clearly shows that the effect of noise is very
low in the range from 25 to 75 Hz.
Q> 0
Ul
III
s:
a.
-180
0 50
f (Hz)
0
I
i
!
-
to
"tJ
Q>
"tJ
-l
:J
-40
-
c
0>
III
-80
!
0 25 50 75 100
180
Fig. AS. Cross Power Spectrum of two signals from a CH test.
Ql
U
C
e 0.5
1
o
o
50
Frequency {Hzl
Fig. A6. Coherence of two signals from a CH test.
75 100

. -:j "-'
, .
915
Ql
U
e
0.5 Ql
..
00
180
Ql
III
ClI
.I:.
0.
100
>
o I
o
100
100
F" A7 Phase velocity diazram from a CH test.
19. .
214 PlANe
A.3 Spectral analysis and propagation velocity
The Cross Correlation function (previously defined in the time domain analysis
paragraph), and hence the travel time and the correspondent waveform velocity,
can be immediately determined utilising the frequency domain analysis. The
Cross Correlation of two signal x(t) and y(t) is given by the Inverse Fourier
Transform of the Cross Power Spectrum of the two signals
CC.l).(r) =Jx(t)y(t+ T)dt =IFT[cPSyx(!)]
(A.5)
.The Cross Power Spectrum allows for the determination of the velocity of each
of the frequency components of the propagating waveform, by means of its phase
[(/)cps(f)]. Using the delay of phase t1cfJ(f) for a given frequency component (f),
wave propagation between two points at known distance is represented by the
corresponding number of wavelengths N: given by
NJ!) = - ~ ~ )
(A.6)
If rl2 is the distance between the two points, the wavelength A(J) and the veloci-
ty Vph(f), defined as phase velocity, of the component (f) are given by
(A.?)
The phase velocity diagram Vph(f), calculated for a series of signals x(t) and yet)
from a CH test is shown in Fig. A7. together with the relative Acps(f), COxy(f)
and (JJcps(j). It is clear that, for frequencies greater than 25 Hz. the V
ph
(! ) dia-
gram is practically horizontal, slightly varying around the value of about 78 m!s,
which can be assumed as the group velocity of the whole waveform.
TCS: Structural Design Aspects of Pile-Deck Systems
T5.1
T5.2
T5.3
T5.4
T5.5
T5.6
Design Performance and Earthquake Levels
T5.1. 1 Performance goal
T5.1.2 Design earthquake levels
T5.1.3 Limit states
T5.1"+ Seismic load combinations
Modelling Aspects
T5.2.1 Soil-structure interaction
T5.2.2 Movement joints
Methods of Analysis for Seismic Response
T5.3.1 Method A: Equivalent single mode analysis
T5.3.2 Method B: Multi-mode spectral analysis
T5.3.3 Method C: Pushover analysis
T5.3..+ Method D: Inelastic time-history analysis
T5.3.5 Method E: Gross foundation deformation analysis
Structural Criteria for Pier and Wharf Piles
T5.4.l Deformation capacity of pile plastic hinges
T5.4.2 Implication of limit states
T5.4.3 Moment-curvature characteristics of piles
T5.4.4 Primary structural parameters and response of piles
T5.4.5 Plastic hinges
T5.4.6 Shear strength of piles
T5.4.7 Design strength for deck members
T5.4.8 Batter and timber piles
PilelDeck Connection Details
T5.5.l Steel-shell piles
T5.5.2 Prestressed piles
T5.5.3 Practical connection considerations
T5.5.4 Capacity of existing substandard connection details
Existinz Construction
....
T5.6.1 Structural criteria for existing construction
T5.6.2 Strengthening of an existing structure
T5.6.3 Deterioration of waterfront structures
TECHNICAL CONIMENTARY 5
Structural Design Aspects of Pile-Deck Systems
This technical commentary will further illustrate the general design methodo-
logies developed in the Main Text and present some numerical guidelines for the
performance of pile-deck structures. Piles are a common and significant element
of waterfront construction. This technical commentary will provide guidance in
pile design.
The following guidance is directly based on Technical Report TR 2103-SHR
'Seismic Criteria For California Marine Oil Terminals' by J. Ferri tto,
S. Dickenson, N. Priestley, S. Werner and C. Taylor, July 1999. This report
was sponsored by the California State Lands Commission, Marine Facilities
Division. The structural criteria defined herein are based largely on the work
of: Nigel Priestley. These criteria are developed from a compilation of current
practice in North America, with state-of-the-art technology for evaluating
seismic damage potential. Although the guidance addresses structural design
of concrete, steel and wooden piles, the focus of attention is directed to
concrete piles, reflecting the practice in North America. Design criteria adopted
in other regions, where steel piles have been typically used, notably in Japan,
might be different from this guidance as briefly discussed in Technical
Commentary 7.
T5.1 DESIGN PERFORMANCE AND EARTHQUAKE LEVELS
T5.1.1 Performance goal
The general performance goals were discussed in Chapter 3 in the Main Text.
This technical commentary focuses on pile-deck structures, specifically piers and
wharves. The reader is referred to Tables 3.1 to 3.3 in the Main Text. In this tech-
nical commentary, piers and wharves are as Grade A structures and
are expected to perform as follows:
( 1) To resist earthquakes of moderate size that can be expected to occur one
or more times during the life of the-structure (Level 1) without structural
damage of significance (Degree I: Serviceable).
........ ,
218 PlANe
(2) To resist major earthquakes that are considered as infrequent events (Level 2),
and to maintain environmental. protection and life safety, precluding total
collapse but allowing a measure of controlled inelastic behaviour that will
require repair (Degree II: Repairable).
T5.1.2 Design earthquake levels
This section utilizes the following earthquake levels presented in Chapter 3 in the
Main Text as defined events.
. Level 1 - (Ll) Ground motions with a 50% probability of exceedance in
50 years exposure. This event has a return time of 75 years and is considered a
moderate event, likely to occur one or more times during the life of the facility.
Such an event is considered a strength event.
Level 2 - (L2) Ground motions with a 10% probability of exceedance in
50 years exposure. This ~ v e n t has a return time of 475 years and is considered a
major event. Such an event is considered a strength and ductility event.
T5.1.3 Limit states
Serviceability Limit State (Degree I: Serviceable) All structures and their
foundations should be capable of resisting the Level I earthquake without sus-
taining damage requiring post-earthquake remedial action.
Damage Control Limit State (Degree II: Repairable) Structures and their
foundations should be capable of resisting a Level 2 earthquake. without collapse
and with repairable damage, while maintaining life safety. Repairable damage to
structures and/or foundations, and limited permanent deformation are expected
under this level of earthquake.
T5.IA Seismic load combinations
Wharves and -piers may be checked for the following seismic load combinations,
applicable to both Level 1 and Level 2 earthquakes:
(l + kJ(D ffi rL) ffi E (T5.1)
(1 - kJD ffi E (T5.2)
where D = Dead Load
L'= Design Live Load
r = Live Load reduction factor (dependson expected L present in actual
case; typically 0.2 but could be higher)
Structural Design Aspects of Pile-Deck Systems 219
E =Levell or Level 2 earthquake load. as appropriate
k, =Vertical seismic coefficient (typically one half of the effective peak
horizontal ground acceleration in g)
EB = Load combination symbol for vector summation
Note: seismic mass for E should include an allowance for rL, but need not
include an allowance for the mass of flexible crane structures.
Effects of simultaneous seismic excitation in orthogonal horizontal directions.
....
indicated by x and y in Fig. T5.1. may be considered in design and assessment of
wharves and piers. For this purpose, it will be sufficient to consider two charac-
teristic cases: .
1000/c E/B 30% e,
30% E
f
$ 100% s,
(T5.3)
(T5.4)
where Ex and E; are the earthquake loads (E) in the principal directions x and y,
respectively.
Where inelastic time history analyses in accordance with the requirements of
Method D in Section T5.3 are carried out, .the above loading combination may
be replaced by analyses under the simultaneous action of x and y direction
components of ground motion. Such motions should recognize the direction-
dependency of fault-normal and fault-parallel motions with respect to the
structure principal axes, where appropriate.
T5.2 MODELLING ASPECTS
This section further illustrates the modelling aspects of pile-deck structures sum-
marized in Technical Commentary 7.
I
Landward F.dL'C

,.

x
......

(. (
Seaward Edge
(a)
1--------
I
I
I
I
I
I
I
I
(b)
1--------------...
I I
I .IJ
1\ ~
__-r...z.;yOYx L ~ I -i. 6n
-1\-6xy
(c)
Fig. T5.1. Plan view of wharf segment under x and y seismic excitations.
(a) Plan view.
(b) Displaced shape, x excitation.
(c) Displaced shape, y excitation.
...' ....
220 PlANe
T5.2.1 Soil-structure interaction
Figure T5.2(a) represents a typical transverse section of a wharf supported on a
soil foundation comprised of different materials, including alluvial sand and
gravel, perhaps with clay lenses, r i p ~ r a p , and other foundation improvement
materials. The most precise modelling of this situation would involve inelastic
finite element modelling of the foundation material to a sufficient depth below,
and to a sufficient distance on each side, such that strains in the foundation
material at the boundaries would not be influenced by the response of the
wharf structure. The foundation would be connected to the piles by inelastic
Winkler springs at sufficiently close spacing so that adequate representation
of the pile deformation relative to the foundation material, and precise definition
of the in-ground plastic hinging, would be provided. Close to the ground surface,
where soil spring stiffness has the greatest influence on structural response,
the springs should have different stiffnesses and strengths in the seaward,
landward, and longitudinal (parallel to shore) directions, as a consequence of the
dike slope.
The pile elements .would be represented by inelastic properties based on
moment-curvature analyses. Since the key properties of piles (namely strength
and stiffness) depend on the a ~ i a l load level which varies during seismic
response, sophisticated interaction modelling would be necessary.
It is apparent that structural modelling of the accuracy and detail suggested
above, though possible, is at the upper limit of engineering practice. and may
be incompatible with the considerable uncertainty associated with the seismic
input. In many cases, it is thus appropriate to adopt simplified modelling
techniques. Two possibilities are illustrated in Fig. T5.2. In Fig. T5.2(a), the
complexity may be reduced by assuming that the piles are fixed at their bases
F4
AS
F5
4 3 2
....
....
f2A2[ <; ~ .........
) ...
A4........ -
F3
FI
AI
1
- -- .....
Wharf Deck
Boundary
2 4
Finite element
soil continuum
(a) (bl
Fig. T5.2. Modelling soil-structure interaction for a wharf.
(a) Soil-structure continuum model.
(b) Equivalent depth-to-fixed model.
with the seismic input applied simultaneously and coherently to each pile base.
The soil springs also c0n.nect the piles to the rigid boundary. Thus the assump-
tion is made that the deformations within the soil are small compared with those
of the wharf or pier.
In Fig. T5.1(b), the complexity is further reduced by replacing the soil
spring systems by shortening the piles to 'equivalent fixity piles, where the
soil is not explicitly modelled, and the piles are considered to be fixed at a depth
which results in the correct overall stiffness and displacement for the wharf or
pier. Where different soil stiffnesses are appropriate for o p p o ~ i t e directions of
response, average values will be used, or two analyses carried' out, based on the
two different stiffnesses, respectively. It should be noted that, though this
. modelling can correctly predict stiffness, displacements and elastic periods, it
will over-predict maximum pile in-ground moments. The model will also require
adjustment for inelastic analyses, since in-ground hinges will form higher in the
pile than the depth of equivalent fixity for displacements.
Geotechnical guidance will be needed to determine the appropriate depth to
fixity (e.g. i, in Fig. T5.2(b)), but an approximate value may be determined from
the dimensioniess charts of Fig. T5.3. A typical value of IJ =5D
p
where D; is the
pile diameter, may be used as a starting point for design.
T5.2.2 Movement joints
Long wharf structures, and less commonly piers, may be divided into segments
by movement joints to facilitate thermal, creep and shrinkage movements.
.. Typically, the joints allow free longitudinal opening, but restrains transverse
displacement by the incorporation of shear keys. Modelling these poses
problems, particularly when elastic analysis methods are used. Under relative
longitudinal displacement, the segments can open freely, but under closing
displacements, high axial stiffness between the segments develops after the
initial gap is closed. Relative transverse rotations are initially unrestrained,
but, after rotations, are sufficient to close one side of the joint. Further rotation
implies relative axial displacement, resulting in axial compression between
the segments. The behaviour can be modelled with a central shear spring
allowing longitudinal displacement, but, restraining transverse displacements
as per the two axial springs, one at each end of the joint, as suggested in
Fig. T5.4. These two springs have zero stiffness under relative opening
displacement, but have a very high compression stiffness after the initial gap .de
is closed.
Clearly, it is not feasible to represent this nonlinear response in an elastic
modal analysis. Even in an inelastic analysis, accurate modelling is difficult
.because of the need to consider different possible initial gaps, and difficulty in
modelling energy dissipation by plastic impact as the joint closes.
222 PlANe
Q.
c
4.5
"C
....
Q)
4.0
III
is
Q)
3.5
C

3.0
-
s:
Q.
c3 2.5
H=20
p
140 120 40 60 80 100
1000 x Kh'SubDIl6/ O"El
eff
20
2.0
o
(a)
(b)
140
I Ii ,.'
120
Ii', . , ., I ' ,I., I
60 80 100
6
1000 xKh.subOp / 0 El
eff
it ,( j. i.
40
s: 3.0J
Q.
Q ,
o -I
2.51
2.0 iii ii ii' i' i iiii

o 20
....
(.)
o
E 4.5
ro
6
Q 4.0
a:::
>-
x 3.5
u:
o
-
e ::]".,,'" i
-0-
5.0
Fig. T5.3. Equivalent depth to fixity for CIDH piles.
(a) Free-head pile.
(b) Fixed-head pile.
T5.3 METHODS OF ANALYSIS FOR SEISMIC RESPONSE
This section further illustrates the methods of analysis for pile-deck structures
discussed in Chapter 5 in the Main Text and in Technical Commentary 7.
According to the three analysis types discussed in the Main Text, Methods A
throuzh E discussed below mav be classified as follows:
- ..
F
-j f- to g(gap)
1-----,
(al
High como.
Stiffness
-..-+-++-+Compression
a
"" : Shear key: Relative longitudinal, but no
"lIof relative shear displacement permitted
Highcomp.
Stiffness
---........++-... Compression
to
(b)
Fig. T5.4. Modelling of movement joints.
(a) Joint details.
(b). Joint modelling.
Simplified Analysis: Method A
Simplified Dynamic Analysis: Method C combined with Methods A and/or B
Dynamic Analysis: Method D
Method E is a gross deformation soil analysis and can be used with all the meth-
ods. It may take the form of a simple Newmark type analysis as in Simplified
Dynamic Analysis or a complex nonlinear effective stress analysis as in Dynamic
Analysis.
The use of Method A for Simplified Analysis is justified for preliminary analysis
or Grade C structures, provided that the parameters used in Method A have been
derived based on a comprehensive set of Method C type parameter studies for
typical pile-deck systems proposed for design (see Section T7.l.3 in Technical
Commentary 7). A pushover analysis can be used in Method A to establish struc-
ture stiffness and resistance.
In comparison with many other structures such as multistory buildings or
multispan bridges, wharves and piers are frequently structurally rather simple. As
such, analyses to determine seismic response can often utilize comparatively
simple analysis procedures, as discussed in Sections T5.3.1 through T5.3.4,
without significant loss of accuracy. Complexity tends to come more from the
high significance of soil-structure interaction, from significant torsional response
- resulting from the typical increase in effective pile lengths from landward to
seaward sides (or ends) of the wharf (or pier), and from interaction between adja-
cent wharf or pier segments separated by movement joints with shear keys, rather
than from structural configuration.
224 PlANe
It has been well documented that the peak response computed for pile-deck
structures will v.ary depending on the analytical method employed (e.g., Methods
A, B, or D in the following text). It should be noted, however, that the modeling
of member stiffness and soil-structure interaction effects has been found to yield
a greater influence on the computed peak response of the structure than the vari-
ations in computed response due to the application of various analytical tech-
niques. Indeed, there is no point in carrying out a sophisticated time-history
analysis unless detailed and accurate simulation of member stiffness and strength
has preceded the analysis.
T5.3.1 Method A: Equivalent single mode analysis
Long wharves on regular ground tend to behave as simple one-degree-of-freedom
structures under transverse response. The main complexity arises from torsional
displacements under longitudinal response, and from interaction across shear
keys between adjacent segments as discussed above. An individual segment will
respond with a significant torsional component when excited by motion parallel
to the shore or along the wharf. Dynamic analyses by Priestley (1999) indicate
that the torsional response is reduced when segments are connected by shear
keys, and with multiple-segment wharves, the torsional response of the inner
segments is negligible. It is thus conservative to estimate the displacement
response based on the behaviour of a single wharf segment between movement
joints except when calculating shear key forces.
Extensive inelastic time-history analyses of single and multi-segment wharves
have indicated that an approximate upper bound on displacement response of the
piles could be established by multiplying the displacement response calculated
under pure transverse (perpendicular to the shore) excitation by a factor, taking
into account the orthogonal load combinations of Eqns. (T5.3) and (T5.4),
including torsional components of response. For the critical shorter landward line
of piles
(T5.5)
where
~ = displacement under pure translational response
e; = eccentricity between centre of mass and centre of rigidity
LL= longitudinal length of the wharf segment
Note that the comer piles of the line of piles at the sea edge of the wharf will
be subjected to slightly higher load/displacement than given by Eqn. (T5.5), but
these piles will not be critical because of their greatly increased flexibility and
hence increased displacement capacity compared with the landward piles.
The shear force across shear keys connecting adjacent wharf segments cannor
be d i ~ e c t l y estimated from a Method A analysis. However. time-history analysis
of multiple wharf segments indicates that the highest shear forces will occur
across movement joints in two-segment wharves, and that these forces decrease.
relative to segment weight, in inverse proportion to wharf segment aspect ratio
LJLr, where L
t
and 4 denote the longitudinal and transverse plan dimensions of
the wharf segment pile group. The following expression may be used as an
approximate upper bound to the shear key force V'k:
v, =1.5(ejL
t
) VolT (TS.6)
where V.JT is the total segment lateral force found from a pushover analysis at the
level of displacement t1T calculated for pure translational response at the appro-
priate limit state.
Method A is of adequate accuracy for design of many simple structures when
supplemented by a Method C pushover analysis, and is particularly useful for the
preliminary stages of design, even of relatively important and complex struc-
tures. Final design verification may be made by one of the more sophisticated
analysis methods.
The approximate lateral stiffness of a wharf structure analyzed using Method
A should be determined from a pushover analysis, as discussed in relation to
Method C.
TS.3.2 Method B: Multi-mode spectral analysis
Spectral modal analysis is the most common method used for estimating
maximum displacement levels, particularly when deck flexibility is significant.
However, as noted above, when multiple segments of long wharves are connected
by rnovementjoints with shear keys, it is difficult to model the interactions occur-
ring at the movement joints adequately because of their non-linear nature.
It is thus doubtful if it is worth modelling the joint, particularly if conditions
are relatively uniform along the wharf. It should also be recognized that it is
unlikely that there will be coherency of input motion for different segments of a
long wharf. Longitudinal motion of the wharf is likely to be reduced as a conse-
quence of impact across joints and restriction of resonant build-up.
It is thus reasonable for regular wharf conditions to consider an analysis of
single wharf segments between movement joints as 'stand-alone' elements. It
will also be reasonable in most cases to lump several piles together along a given
line parallel to the shore and to provide an analytical 'super pile' with the
composite properties of the tributary piles in order to reduce the number of
structural elements, which can he excessive, in a multi-mode analysis. Since the
modal analysis will be used in conjunction with a pushover analysis, it will be
worth considering a further reduction in the number of structural elements by
226 PlANe
representing the composite stiffness of a transverse line of piles, found from the
pushover analysis, by a single pile, and using a damping level appropriate to the
expected displacement as discussed in more detail in the following section.
In most stand-alone analyses of single wharf or pier segments, there will be
only three highly significant modes: two translational and one torsional. These
will generally be closely coupled. As a consequence, it will not generally be dif-
ficult to satisfy the 95% participating mass requirement, which is larger than
commonly specified in design codes. Note that, to correctly model the torsional
response, it is essential to model the torsional inertia of the deck mass. This can
either be done by distributing the deck mass to a sufficiently large number of uni-
formly distributed mass locations (with a minimum of 4 located at the radius
of gyration of the deck area), or by use of a single mass point with rotational
inertia directly specified. Deck mass should include a contribution for the mass
of the piles. Typically, adding 33% of the pile mass, from deck level to the point
of equivalent foundation fixity, is appropriate.
Because of the typical close coupling of the key modes of vibration. the modal
responses should be combined using the Complete Quadratic Combination
(CQC) rule (Wilson et al., 1981).
T5.3.3 Method C: Pushover analysis
Typically, as a result of large variations in effective depth to fixity of different
piles in a given wharf or pier, the onset of inelastic response will occur at greatly
differing displacements for different piles. This is illustrated in Fig. T5.5,
representing the response of a 6 m segment of a typical wharf loaded in the
transverse (perpendicular to shoreline) direction. There are a number of impor-
tant consequences to this sequential hinging:
It is difficult to adequately represent response by an elasto-plastic approxi-
mation, which is the basis of the common force-reduction factor approach to
design.
. The appropriate elastic stiffness to be used in analysis is not obvious.
Different piles will have greatly different levels of ductility demand. with the
shortest piles being the most critical for design or assessment.
The centre of stiffness will move from a position close to the landward line of
piles to a position closer tothe centre of mass as inelasticity starts to develop
first in the landward piles.
As a consequence, it is not directly feasible to carry out an elastic analysis
based on the typically assumed 5% equivalent viscous damping. and then deter-
mine member forces by reducing elastic force levels by a force-reduction factor.
Among other failings, this will grossly underestimate the design forces for the
longer piles. It is thus strongly recommended that a key aspect of the design or
analysis process be a series of inelastic pushover analyses in both transverse and
y
F
-
.:. 3 4
H2 H3 H4 H5
,
,
:.2
,
-- A3
,
x, ,
x,
A4
,
-
xl
AS
X,
'"
(:1)
x
H5 A4
'I
"II
"
iI
"!I
I It
!I
t i1
" II
\I
1I
II
\I
Displacement 6.
4
5
3
2
,
II
\I
HS
II
II
II
iI
iI
\I
\I
"
..
" II
t
"
"
Ii
l'
6.cc
Pile Displacement
" .
t I I'
: ~ ' A ~ 1 ~ : _'-+----:"7,"""----
(d)
A
H
(C)
Fig. T5.5. Pushover analysis of a wharf segment.
(a) Transverse section for analysis.
(b) Force-displacement response and hinge sequence for transverse
excitation.
(c) Pile model for longitudinal analysis.
(d) Individual pile longitudinal force-displacement response.
longitudinal directions, where two dimensional (2D) sections of the wharf or pier
are subjected to incremental increases in displacement allowing the inelastic
force-displacement response (e.g. Fig. T5.5(b)), the sequence of hinge formation,
and the magnitude of inelastic rotation e
p
developing in each hinge to be deter-
mined. The results of such analyses can be used to:
determine an appropriate stiffness for a Method A or B analysis,
determine appropriate damping levels for elastic analyses, and
determine peak plastic rotations of critical hinges at the maximum displace-
rnents predicted by the Method A or B analyses.
These aspects are discussed in more detail in the following sections (1)
through (4).
...._-
. '(I) Elastic stiffness from pushover analysis
It is recommended that elastic analyses used for Method A or B analysis be based
on the substitute structure analysis approach (Gulkan and Sozen, 1974), where
the elastic characteristics are based on the effective stiffness to maximum
228 PlANe
displacement response anticipated for the given limit state, and a corresponding
level of equivalent viscous damping based on the hysteretic characteristics of the
force displacement response (e.g. Fig. T5.5(b)). Thus, for the transverse
response, the stiffness for a 6 m tributary length of wharf would be k
s
or k
oc
for
the serviceability and damage control limit states, respectively, based on the
expected limit states displacements Lls and .1
oc
Since these displacements will
not be known prior to a Method A or B analysis, some iteration will be required
to determine the appropriate stiffnesses.
For a Method A analysis, the transverse period can then be directly calculated
from the stiffnesses and tributary mass in the usual manner. For a Method B
analysis, using a reduced number of piles as suggested above, longitudinal
pushover analyses will be required on characteristic sections of a wharf or pile.
As suggested in Fig. T5.5(c), the characteristic element for a wharf may be taken
as a pile plus deck section extending midwayto adjacent piles on either side. A
set of pushover analyses for each of the characteristic piles, A through E, can then
be carried out and plotted as shown in Fig. T5.5(d). Again, based on the expected
limit states displacements, the total stiffness of the tributary length of wharf con-
sidered can be calculated as
(T5.7)
where n, is the number of piles in row i in the tributary width considered, and
V
j
is the pile shear force at the limit state deflection ..1.,.
The centre of rigidity measured relative to the arbitrary datum shown in
Fig. T5.5(a) is given by
(T5.8)
Thus, for a Method B analysis, the complete longitudinal and transverse stiffness
of a given length of wharf may be represented by a single 'super pile' located at
the longitudinal centre of the length modelled by the pile, and located trans-
versely at the position defined by Eqn. (T5.8), with stiffness values as defined
above. Note that different stiffnesses and centres of rigidity will normally apply
for the serviceabilirv and damage control limit states.
'" ...
(2) Damping
The substitute structure approach models the inelastic characteristics of struc-
tures by elastic stiffness and damping levels appropriate for the maximum
response displacements rather than using the initial elastic parameters. The
advantage to using this approach is that it provides a more realistic representation
of peak response, and eliminates the need to invoke force-reduction factors to
bring member force levels down from unrealistically high elastic values to
realistic levels. The damping level is found from the shape of the complete
(T5.9)
Monotonic(envelope) CUNe
6 . ~ s
Displacement
Monotonic (envelope)CUNe
Fig. T5.6. Equivalent viscous damping for substitute-structure analysis.
hysteresis loop for a single cycle of displacement at maximum response, as
illustrated in Fig. T5.6. In this, the skeleton force-displacement curve for both
directions of response is calculated by a pushover analysis, as illustrated in
Fig. T5.5(b). The unloading curve is based on a modified Takeda approach
(Takeda et al., 1970), where the unloading stiffness k
u
is related to the structure
ductility j.1,j' defined in Fig. T5.6, and the initial stiffness k, by
k k
-112
u = I J 1 ~
The residual displacement Ll
r
is thus given by
(T5.10)
The remainder of the stabilized hysteresis loop is constructed, as shown in
Fig. T5.6. The equivalent viscous damping, as a percentage of critical damping,
is then given by
(T5.11)
where A
hs
is the area of the stabilized loop, shown by hatching in Fig. TS.6.
For initial analyses of wharves or piers supported on reinforced or prestressed
concrete piles, damping values of 10% and 20% at the serviceability and damage
control limit states will generally be appropriate.
(3) Capacity design checks
When the pushover analyses are based on the conservative estimates of material
properties for the wharf or pier elements, the member forces resulting from the
230 PlANe
analyses represent lower bound estimates 'of the forces developed at the appro-
priate limit state. Although this may be appropriate for determining required
strength of plastic hinge regions, and for ensuring that calculated limit state dis-
placements are not exceeded, the results will not be appropriate for determining
the required strength of members. It is clear that if material strengths exceed the
specified strength, then the flexural strength of plastic hinges may significantly
exceed the dependable strength, and as a consequence, the shear forces in piles
and deck members may be as much as 20-30% higher than predicted by the
pushover analysis.
In order to ensure that a conservative upper limit on the possible strength of
members or actions, which are to be protected against inelastic response, is
achieved, a second set of pushover analyses should be carried out where the
material strengths adopted represent probable upper bounds. This approach, based
on capacity design principles, is discussed in detail in Priestley et al. (1996).
(4) Iteration considerations
When checking a completed design, iteration will be needed to ensure compa-
tibility between the Method A or B elastic analysis used to determine response
displacements, and the stiffness and damping values determined from the
pushover analysis. Typically, the necessary adjustments to the substitute structure
characteristics take very few cycles to converge with adequate accuracy.
In the design process, the limit state displacements will generally be known
before the design strength is established, provided the type and diameter of pile
is known. In this case, the iteration required will be centred on determining the
correct number and location of piles in order to limit the response displacements
to the structural limit state displacements.
T5.3.4 Method D: Inelastic time-history analysis
Inelastic time-history analysis is potentially the most accurate method for
estimating the full seismic response of a wharf or pier. It has the capability of
determining the maximum displacements and the inelastic rotations in plastic
hinges from a single analysis. However, to do this, it is necessary to model each
pile individually and to" simulate the pile/soil interaction by a series of Winkler
springs as discussed in Section T5.3.1. This can result in unacceptable matrix
sizes for analysis of wharves or piers with very large numbers of piles. An alter-
native is to combine the analysis with inelastic pushover analyses and represent
groups of piles by equivalent 'super piles' for modal analysis as discussed in
Sections T5.3.2 and T5.3.3. Although this somewhat reduces the attraction of the
time-history analysis. it makes the analysis more tractable, and still enables
several advantages of the method, not available with Method A or B. to be
retained.
A disadvantage of the method is that considerable variations in response can
be obtained between two different spectrum-compatible acceleration records. As
a consequence. it is essential to run an adequately large number of simulations
usinz different acceleration records, and to average the results. A minimum of
~ ~
five spectrum-compatible records is recommended.
The following points need to be considered before undertaking a time-history
analysis:
1. A full simulation of the wharf or pier will require use of a computer program
capable of modelling three dimensional (3D) response. There are compara-
tively few 3D inelastic time-history programs available at present (March
2000), and experience with them outside of research applications is rather
limited. Frequently, approximations must be made relating to hysteresis rules
and strength interactions in orthogonal directions which make the added
sophistication of time-history analysis of reduced utility.
2. If the deck has sufficient rigidity to justify its approximation as a rigid element
both in-plane and out-of-plane, a two dimensional (2D) plan simulation may
provide adequate accuracy. However, special multi-direction spring elements
with realistic hysteresis rules are required for such a simulation.
3. Time-history. analysis enables different stiffnesses to be used for different
directions of response. Thus, provided the necessary hysteresis rules are avail-
able, it will generally be possible to model the higher stiffness for movement
into the shore than for movement away from the shore, and to have a separate
stiffness for longitudinal response.
4. It is comparatively straightforward to model the interactions across shear
keys, using inelastic time-history analysis.
5. When modelling reinforced or prestressed concrete members, degrading stiff-
ness models such as the Modified Takeda rule should be adopted. There is
little point in carrying out time history analyses if simplified rules such as
elasto-plastic or even bilinear stiffness rules are adopted.
6. Care needs to be exercised into how elastic damping is handled. It is common
to specify 5% elastic initial stiffness related damping. This can greatly over-
estimate the damping at high ductilities. In fact, since the hysteretic rules
available in the literature have generally been calibrated to experimental
results, the justification for adding elastic damping, which should be apparent
in the experimental hysteresis loops, is of doubtful validity. There 0 is thus a
case for ignoring elastic damping if levels of inelastic response are high.
However, at low levels of displacement response, the simplifications inherent
in the hysteresis rules generally mean that the damping is underestimated for
o 'elastic' or near-elastic response.
7. Results from a time-history analysis should always be compared with results
from a simplified approach (e.g. a Method. A or B analysis) to ensure that rea-
sonable results are being obtained. o.
232 PlANC
T5.3.5 Method E: Gross foundation deformation analysis
As discussed above, the assumption will normally be made that the foundation
material is competent when carrying out any of the analyses described above.
However, examination of the performance of wharves and piers in past
earthquakes reveals that liquefaction and foundation sliding or slumping are
common as discussed in Technical Commentary 2. Analysis techniques to
estimate the sensitivity of the foundation to such failures, and the extent of
deformation to be expected, are dealt with in Technical Commentary? Where
geotechnical analyses indicate that moderate permanent deformations are to be
expected, the structure should be analyzed under the deformed soil profile to
estimate the influence on the structure. This will generally require at least a 2D
analysis incorporating discrete modelling of piles and soil springs, with
foundation deformations applied at the boundary ends of the springs as
appropriate.
Structural deformations due to inertial effects alone occur relatively early in
the time history as the peak ground motions are experienced. Conversely,
structural behavior associated with ground failures, soil settlement and gross
foundation deformations are likely to take place much later in the time history, or
even well after the strong ground motions have ended. Indeed, liquefaction
failure may well occur some minutes after the ground motion has ceased.
Consequently, it is not necessary to combine the results from a Method E
analysis with the results of a Method A to D analysis. It will be sufficient to
confirm that response of each type is individually satisfactory.
T5.4 STRUCTURAL CRITERIA FOR PIER AND WHARF PILES
T5.4.l Deformation capacity of pile plastic hinges
The ability of wharves and piers to respond inelastically to seismic excitation
depends on the displacement capacity of pile plastic hinges. This displacement
capacity will depend on types of piles used in the structure, their length. cross-
section dimensions, axial load, and material propeities. A brief discussion of the
deformation capacities of different pile types is included below. This is based on
an examination of their moment-curvature characteristics.
Until recently, piles were designed neglecting the confinement effects of the
reinforcement in the concrete. Work by Joen and Park (1990) shows the signifi-
cant increase in moment capacity by considering the effect of spiral confinement
on the concrete. This work uses the 'Mander et al. (1988) concrete model for
confined and unconfined concrete. Typically, the ultimate compressive concrete
strain of unconfined concrete is about 0.003 for use in computing flexural
strength as reported by Priestley et al. (1992). For confined' concrete, the
following ultimate compressive strain <', may be used (Priestley et al., 1992):
(T5.12)
where p,
hh
effective volume ratio of confining steel
yield stress of confining steel
strain at peak stress of confining reinforcement. 0.15 for grade
40 and 0.12 for grade 60 (see Table T5.1 for designation of
reinforcing bars)
confined strength of concrete approximated by 1.5!c'
f/ ultimate concrete compressive stress at 28 days
T5.4.2 Implication of limit states
Two levels of design earthquake are considered for this study.
Level 1 Earthquake; Serviceability Limit State Level 1 has a high prob-
ability of occurrence during the life of the pier or wharf. Under this level of
excitation, the structure should satisfy' the serviceability criterion of continued
functionality immediately after the Level 1 earthquake. Any repairs required
should be essentially cosmetic. Structural damage requiring repair is not penn-
itted. Note that this does not imply a requirement for elastic response which
would limit concrete strains to about 0.001 and steel strains to yield strain.
Concrete structures may be considered serviceable, without any significant
decrement to structural integrity, provided concrete strains reached during maxi-
mum seismic response to the Level 1 earthquake do not cause incipient spalling
of concrete cover and if residual crack widths after the earthquake are small
enough to negate the need for special corrosion protection measures such as
crack grouting. Note that this latter requirement implies that significantly larger
crack widths might momentarily exist during seismic response since these have
no effect on corrosion potential. .
Table T5.1. Tensile and yield stresses of concrete reinforcing bars in North America
(1.0 psi =6.9 kPa).
Designation Minimum yield (psi) Minimum tensile (psi)
Grade 40*
Grade 50**
Grade 60
Grade 70
Grade 75
40.000
50,000
60.000
70.000
75.000
70.000
80.000
90.000
80,000
100.000
*Previously designated as Intermediate.
**Previously designated as Hard.
234 PlANe
Thus, the performance of potential plastic hinges should be checked under
Levell earthquake to ensure that maximum material strains do not exceed the
limits defined for structural performance limit states.
Note that this check will normally only be required at the pile/deck interface,
since curvatures will be higher there than in any potential in-ground hinges. Also,
the cover concrete of the potential in-ground plastic hinges are confined by the
lateral pressure developed in the soil, and therefore the concrete strain at which
spalling initiates is higher for the in-ground plastic hinges (Budek et aI., 1997).
This will increase the corresponding serviceability curvatures for the in-ground
plastic hinges, particularly for prestressed piles, where steel strains are unlikely
to be critical.
Level 2 Earthquake; Damage-control Limit State Level 2 earthquake repre-
sents an earthquake with a low probability of occurring during the design life of
the structure. Repairable damage to structure and/or foundation, and limited per-
manentdeformation are acceptable under this level of earthquake
Since the design philosophy is to restrict inelastic action to carefully defined
and detailed plastic hinges in piles, strain limits may again be used to define
acceptable response. Distinction needs to be made between the pile/deck hinge
and the in-ground hinge locations, since access to the latter will frequently be
impractical after an earthquake. As a consequence, the in-ground hinge should
satisfy serviceability criteria even under Level 2 earthquake. Thus, the steel strain
should not exceed 0.01. However, as noted above, passive confinement by the soil
increases the spalling strain and a concrete strain of 0.008 may be adopted. The
pile/deck connection hinge will have allowable concrete strains, dependent on the
amount of transverse reinforcement provided. As a consequence of these actions,
the maximum strains calculated under a Level 2 earthquake must not exceed the
strain capacities defined in Section T5.4.3.
T5.4.3 Moment-curvature characteristics of piles
The key tool for investigating the deformation characteristics of piles is moment-
curvature analysis. The analysis adopted must be capable of modelling the full
stress strain curves of reinforcement and concrete realistically. For the concrete.
this entails discrimination between the behaviour of unconfined cover concrete,
if present, and that of the confined core, which will have enhanced strength. and
particularly enhanced deformation capacity. Elastic, yield plateau, and strain-
hardening sections of the steel stress-strain curve should be modelled separately;
a simple elasto-plastic or bi-linear representation will be inadequate for mild steel
reinforcement, but may be adequate for prestressing steel. The computer code
used should provide output of moment and curvature at regular intervals of a
sequential analysis to failure, and should identify the peak values of extreme fibre
concrete strain, and maximum reinforcement and/or prestressing strain at each
increment of the analysis, so that curvatures corresponding to the serviceability
and damage-control limit state can be identified.
Curvatures at the serviceability limit state are based on strain limits suffici-
ently low so that spalling of cover concrete will not occur under the Level 1
earthquake. and any residual cracks will be tine enough so that remedial
grouting will not be needed. These are taken to be
'-' .....
Concrete extreme fibre compression strain:
Reinforcing steel tension strain:
...
Prestressing strand incremental strain:
...
Structural steel (pile and concrete-filled pipe):
Hollow steel pipe pile:
0.004
0.010
0.005
0.008
0.008
Serviceability
Limit Strains
Levell
The specified incremental strain for prestressing strand is less than for reinforc-
ing steel to ensure that no significant loss of effective prestress force occurs at the
serviceability limit state. For steel shell piles, either hollow or concrete filled, it
is recommended that the maximum tension strain in the shell be also limited to
0.01, to ensure that residual displacements are negligible. Note that, if the steel
shell pile is connected to the deck by a dowel reinforcing bar detail, the connec-
tion is essentially a reinforced concrete connection confined by the steel shell. As
such, the limit strains for reinforced concrete apply.
A sample moment-curvature response appropriate for a reinforcing steel dowel
connection between a prestressed pile and a reinforced concrete deck is shown in
Fig. For structural analysis, it will often be adequate to represent the
response by a bi-linear approximation as shown in the figure. The initial elastic
portion passes through the first yield point (defined by mild steel yield strain, or
1000
E
z
'E
E

y s .o.s
I
------------
- - - first yield
200
- - -cracking
Fig. Pile-cap moment curvature.
0.16 0.08 0.12
CulVature (11m)
0.04
o-+-.................. ........""T"""".......................,.........--.-..........,.,
o
236 PlANe
an extreme fibre concrete compression strain of 0.002, whichever Occurs first),
and is extrapolated to the nominal flexural strength. This defines the nominal
yield curvature, (A.. The second slope of the bi-linear approximation joins the
yield curvature and the point on the curve corresponding to the damage-control
limit-state strains. These are taken to be
Damage
Control Limit
Strains Level 2 0.04
0.015
0.025
Concrete extreme fibre compression strain:
Pile/deck hinge: Value given by Eqn. (T5.12), but <0.025
In-ground hinge: Value given by Eqn. (T5.12), but <0.008
Pile/deck reinforcing steel tension strain: 0.05
In-ground reinforcing steel tension strain: 0.01
Structural steel (pile and concrete filled pipe): 0.035
Prestressing strand:
Pile/deck hinge:
In-ground hinge:
Hollow -steel pipe pile:
Equation (T5.12) defines a safe lower bound estimate of the ultimate compres-
sion strain of concrete confined by hoops or spirals (Priestley et aI., 1996). Actual
compression failure, initiated by fracture of spiral or hoop confinement, will typ-
ically not occur until strains are on average 50% larger than the value given by
Eqn. (T5.12), providing an adequate margin for uncertainty of input.
Figure T5.8 shows examples of theoretical moment-curvature curves for
different types of piles, each with an outside diameter of 61 em, and each
subjected to different levels of axial load. Comparison of the curves for the two
reinforced concrete doweled connections of Figs. T5.8(a) and T5.8(b) show the
significant influence of the concrete cover on the moment-curvature characteris-
tics. Flexural strength is significantly higher with the reduced concrete cover of
Fig. T5.8(b). Sections with high cover (10 em) show significant reductions in
moment capacity when spalling initiates, for all levels of axial load, while piles
with 6.4 em cover exhibit much more satisfactory response. Similar behaviour is
apparent for circular or octagonal prestressed piles, as shown for two different
values for concrete cover in Figs. T5.8(c) and T5.8(d). Moment-curvature curves
for hollow and concrete-filled steel shell piles are shown in Figs. T5.8(e) and
T5.8(f). For the hollow steel shell pile, the influence of axial load is rather lim-
ited with axial compression tending to reduce the flexural strength. The influence
of internal concrete is to increase the moment capacity significantly, particularly
when axial load is high. Another advantage of the concrete infill is that it reduces
the sensitivity of the steel shell to local buckling.
Rectangular section reinforced or prestressed concrete piles confined by
square spirals and a rectangular distribution of longitudinal reinforcing bars
or prestressing strand should not be used for new construction, since the con-
finement provided to the core concrete by the rectangular spirals is of very low
1000
800
E
z
.><:
600
Qj
E
0
2
"'..
, \
~
'" N=3100kN
..... '" -.,"'''' ..... ...... _ _ N-44\"fO kN
I' ' ..... .-.- .. - . ~ r ~ : 1300 kN
.. ........ -
~ ..,. --.-....... ~ _ ..-- N=450 kN
~ ~ _-_ ... - ..- ..-.-P_...
~ ~ N=-450 kN
.
(a)
o
o 0.04 0.08 0.12
Curvature (11m)
0.16
0.20
__----- N=-450 kN
E
z
..><:
1200
~ __N=3100 kN
f
-_... -N=2200 kN
---
.......... ..- ... --
I ..... ...... .. N=1300 kN
1
~ .......--_....._..
~ ..--.......-............... .. __ N=4S0 kN
I.' _... _r---
f" ---
r ......... ~ - . - _ ... -"
.. ..
~ r:
~ :
l ~
~
(b)
0,
a
I
0.04 0.08 0.12
Curvature (11m)
0.16
0.20
Fig. T5.8. Moment-curvature curves for 610 mm diameter piles.
(a) Cover = lO2 nun (4 in.) RC dowel connection 8#10 bars.
(b) Cover = 64 nun (2.5 in.) RC dowel connection 8#10 bars.
(c) Prestressed pile, FIA =6.4 MPa, cover = 76 mID.
(d) Prestressed pile, FIA =6.4 MPa, cover =25 rnm.
(e) Steel shell pile, t
p
= 12 mm.
(t) Concrete filled steel shell pile, t
p
= 12 mm.
238 PlANe
N=O
"
-- _----N=O.2f
e'Ag

, ,
0.04 0.06 0.08 0.10 0.12 0.14 0.16
Curvature (11m)
(c)
1200 ........----------___:J
1100
1000
900
- 800
E
700
-
C 600
G>
E 500
o
400
300
200
100

0.00 0.02
0.08 0.12 0.16
Curvature (11m)
0.04
N=0.4fe'Ag
_------ N=0.3f:'A;
N=02f
e'A;

__---------N=O
ld)
1500
1400
1300
1200
1100
_1000
900
.:> 800
700
g 600
500
400
300
200
100
__.,..-
0.00
efficiency. For assessment of existing structures, the concrete core maximum
strain corresponding to the damage-control limit state should not be taken larger
than 0.007 at the pile/deck hinge. Higher maximum concrete strains. as given by
Eqn. (T5.l2), are appropriate if the core of the rectangular pile is confined by a
circular spiral and the longitudinal reinforcement or prestressing is also circularly
disposed. However, such sections typically have low ratios of concrete core area
to grOSS section area, resultinz in flexural strength of the confined core being less
- ......... - .
than that of the unconfined gross section. .
Hollow prestressed piles are sometimes used for marine structures. These,
however. have a tendency to implode when longitudinal compression strains at
1200
N=O
---
f
-
E N=1500 kN
z
...
800
c
E
c
..
400
0.12 0.10
, .,
0.04 0.06 0.08
Curvature (11m)
0.02

0.00
1500
,..
e
Z
... 1000
c
,..
c::
0
50
(f)
0.00 0.04 0.08 0.12
Curvature (11m)
0.16 0.20
the inside surface exceed 0.005. Consequently, the damage control limit state
should have an additional requirement to the strain limits defined earlier that
inside surface compression strains must not exceed 0.005. Note that to check for
this condition, the moment-curvature analysis must be able to model spalling of
the outside cover concrete when strains exceed about 0.004 or 0.005. The outside
spalling can cause a sudden shift of the neutral axis towards the centre of the
section resulting in strains on the inside surface reaching the critical level soon
after initiation of outside surface spalling. More information on piles is available
in Priestley and Seible (1997).
(T5.l3)
(T5.l4)
240 PlANe
T5.4.4 Primary structural parameters and response of piles
(l) Elastic stiffness
The effective elastic stiffness may be calculated from the slope of the 'elastic'.
portion of the bi-linearapproximatiori to the moment-curvature curve (e.g.
Fig. T5.7), as
El
eff
=MN1y
where leff effective moment of inertia
MN nominal flexural strength
The value of the yield curvature, y' for a reinforced concrete pile or pile/deck
connection is rather insensitive to axial load level or longitudinal reinforcement
ratio. Results of analyses of a large number of cases indicate that Eqn. (T5.13)
may be approximately expressed as a fraction of the gross section stiffness as
Eleff/EIgross =0.3+ Ntt];AgrosJ
where N is the axial load level, and Agross is the uncracked section area.
For prestressed piles, the effective stiffness is higher than for reinforced
concrete piles, and values in the range 0.6< EleiE1gross < 0.75 are appropriate. for
prestressed piles with reinforced dowel connections to the deck, the effective
stiffness should be an average of that for a reinforced and a prestressed connec-
tion, or a short length, approximately 2D
p
long of reduced stiffness appropriate
for reinforced pile, should be located at the top of the prestressed pile.
(2) Plastic rotation
The plastic rotation capacity of a plastic hinge at a given limit state depends on
the yield curvature, Y' the limit-state curvature. s for Levell or LS for Level 2,
and the plastic hinge length L
p
and is given by
Gp=pLp= s or LSJ - )L
p
(T5.l5)
(3) Plastic hinge length
The plastic hinge length for piles depends on whether the hinge is located at the
pile/deck interface or is an in-ground hinge. Because of the reduced moment
gradient in the vicinity of the in-ground hinge. the plastic hinge length is signi-
.ficantly longer there. For pile/deck hinge locations with reinforced concrete
details. the plastic hinge length can be approximated by
SI units 4 = 0.08L + 0.022f.d
h
>O.044hdh (MPa, mm)
US units L; = 0.08L + 0.15fd
h
> 0.30h.do (ksi, in.) .
(T5.l6a)
(T5.16b)
where I, is the yield strength of the dowel reinforcement. of diameter do. and Lis
the distance from the pile/deck intersection to the point of contraflexure in the pile.
For prestressed piles where the solid pile is embedded in the deck (an unusual
detail in the USA). the plastic hinge length at the pile/deck interface can be taken
as
(T5.17)
For in-ground hinges, the plastic hinge length depends on the relative stiffness of
the pile and the foundation material. The curves of Fig. T5.9 relate the plastic
hinge length of the in-ground hinge to the pile diameter, D
p
, a reference diameter
D* =1.82 m, and the soil lateral subgrade coefficient, (N/m
J
) .
For structural steel sections and for hollow or concrete-filled steel pipe piles,
the plastic hinge length depends on the. section shape and the slope of the
moment diagram in the vicinity of the plastic hinge, and should be calibrated by
integration of the section moment-curvature curve. For plastic hinges forming in
steel piles at the deck/pile interface, and where the hinge forms in the steel sec-
tion rather than ina special connection detail (such as a reinforced concrete
dowel connection), allowance should be made for strain penetration into the pile
cap. In the absence of experimental data, the increase in plastic hinge length due
to strain penetration may be taken as 0.25 D
p
, where D; is the pile diameter (with
a circular cross section) or pile depth (with a non ciruc1ar cross section) in the
direction of the applied shear force.
(4) In-ground hinge location
The location of the in-ground plastic hinge for a pile may be found directly from
an analyses where the pile is modelled as a series of inelastic beam elements, and
2,00
...
q)
1.80
ell
'6
.s 1.60
1.40
1.20
z
1.00
ell
a:
H=20
p
0.80
70 60 20 30 40 50
\000 x / OB
aff
10
o.60
o
(H=height of contraflexure point above ground, Dp=pile diameter)
Fig. T5.9. In-ground plastic hinge length.
242 PlANe
the soil is modelled by inelastic Winkler springs. When the pile/soil interaction is
modelled by equivalent-depth-to-fixity piles, the location of the in-ground hinge
is significantly higher than the depth to effective fixity, as illustrated in Fig. T5.10
by the difference between points F, at the effective fixity location, and B, the
location of maximum moment. Note that when significant- inelastic rotation is
expected at the in-ground hinge, the location of B tends to migrate upwards to a
point somewhat higher than predicted by a purely elastic analysis. It is thus
important that its location, which is typically about 1-2 pile diameters below
ground surface (such as dike surface beneath wharf), should be determined by
inelastic analysis.
(5) Pile force-displacement response
The information provided in the subsections (1) through (4) enables an inelastic
force displacement response to be developed individually for each pile. This may
be directly carried out on a full 2D section through the wharf, involving many
piles, as part of a pushover analysis, or it may be on a pile-by-pile basis, with the
pushover analysis assembled from the combined response of the individual piles.
With respect to the equivalent-depth-to-fixity model of Fig. T5.l0, the pile is
initially represented by an elastic member, length L, with stiffness El
ctf
given by
Eqn. (T5.13) or (T5.l4), as appropriate, and the deck stiffness represented by a
.
spring k
d
, as shown in the same figure. Often, it will be sufficiently accurate to
assume the deck to be flexurally rigid, particularly with longer piles.
The deflection and force corresponding to development of nominal strength
M
N
at the pile/deck hinge can then be calculated. Note that, for elastic defor-
mation calculations, the interface between the deck and pile should not be
I
l===::::;:;:=====*=*A
R.
F
(a)
B
(b)
1.1..
Q)
::
o
1.1..
3
Displacement 1!.
(c)
Fig. T5.1O. Force-displacement response of an isolated pile.
(a) Pile equivalent-depth to fixity model. .
(b) Displacement profiles.
(c) Force-displacement response.
(T5.18)
Structural Design Aspects of Pile-Deck Systems 2.+3
considered rigid. The effective top of the pile should be located at a distance
O.022!rdh into the deck to account for strain penetration. This is particularly
important for short piles. This additional length applies only to displacements.
Maximum moment should still be considered to develop at the soffit of the deck.
The elastic calculations above result in a pile displacement profile marked I in
Fig. T5.10(b), and the corresponding point on the force displacement curve of
Fig. T5.10(c). For the next step in the pile pushover analysis, an additional spring
k
pl
must be added at A, the deck/pile interface (i.e. the deck soffit), to represent
the inelastic stiffness of the top plastic hinge. This stiffness can be determined
from Fig. T5.7 as
k =(Mu-MN )
pI (tfu -tP
y
)4
Essentially, this spring is in series with the deck spring. Additional force can be
applied to the modified structure until the incremental moment at B is sufficient
to develop the nominal moment capacity at the in-ground hinge. The deflection
profiles and force-displacement points marked 2 in Figs. T5.10(b) and (c) refer
to the status at the end of this increment. Finally, the modified structure, with
plastic hinges at the deck/pile interface and B, is subjected to additional
displacement until the limit state curvature tPLS is developed at the critical hinge,
which will normally be the deck/pile hinge. Note that the inelastic spring
stiffnesses at the two hinge locations will normally be different, due to
differences in the structural details between the in-ground and hinge locations,
and due to the different plastic hinge lengths.
The procedure outlined above is sufficiently simple to be carried out by
spreadsheet operations, and has the advantage that the post yield moment-
rotation stiffness of the hinges can be accurately modelled. This function is not
available in all pushover codes.
T5.4.5 Plastic hinges
(1) Material properties for plastic hinges
For both design of new structures and the assessment of existing structures, it is
recommended that the moment-curvature characteristics of piles be determined
based on probable lower bound estimates of constituent material strengths. This
is because strength is less important.to successful seismic resistance than is
displacement capacity. For the same reason, there is little value in incorporating
flexural strength reduction factors in the estimation of the strength of plastic
hinges. Flexural strength reduction factors were developed for gravity load
design where it is essential to maintain a margin of strength over loads to avoid
catastrophic failure. Inelastic seismic response, however, requires that the
flexural strength be developed in the design level earthquake. Incorporation of a
. .
(T5.19)
244 PlANe
strength reduction factor will not change this and at best may slightly reduce the
ductility demand. It has been shown (Priestley, 1997) that the incorporation of
strength reduction factors in the design of plastic hinges may reduce the ductility
capacity slightly, thereby negating any benefits that may have been derived from
additional strength.
On the other hand, the use of nominal material strengths and strength reduc-
tion factors in design or assessment will place a demand for corresponding
increases in the required strength of capacity protected actions, such as shear
strength. This is because the maximum feasible flexural strength of plastic
hinges, which dictates the required dependable shear strength, will be increased
. when design is based on conservatively low estimates of material strength. This
will have an adverse economic impact on the design of new structures and may
result in an unwarranted negative assessment of existing structures.
As a consequence, the following recommendations are made for the determi-
nation of moment-curvature characteristics of piles (Priestley et aI., 1996):
Concrete compression strength s: =1.3f:
Reinforcement yield strength I; = 1.1fy
Prestress strand ultimate strength hue = 1.0fpu
where j/ is the specified 28 day compression strength for the concrete. and1;. and
I; are the nominal yield and ultimate strength of the mild steel reinforcement and
prestressing strand, respectively. For assessment of existing structures, higher
concrete compression strength will often be appropriate due to natural aging, but
should be confirmed by in-situ testing.
In both new design and assessment of existing structures. the flexural strength
reduction factor for pile plastic hinges should be taken as unity.
(2) Confinement of pile plastic hinges
Research by Budek et a1. (1997) has shown that lateral soil pressure at the
in-ground plastic hinge. location helps to confine both core and cover concrete.
This research found that. as a consequence of this confinement, for both rein-
forced and prestressed circular piles, the plastic rotation capacity of the in-ground
hinge was essentially independent of the volumetric ratio of confinement pro-
vided. As a consequence of this, and also as a result of the low material strains
permitted in the in-ground hinge at the damage-control limit state, much lower
confinement ratios are possible than those frequently provided for piles. It is
recommended that. unless higher confinement ratios are required for pile-
driving, the confinement ratio for the in-ground portion of the pile need not exceed
4 ~ p
p = . =0.008
, (D - "c )s
p - u
where A,p =area of the spiral or hoop bar
Dr = pile diameter
c, =concrete cover to centre of hoop or spiral bar
s =spacing of spiral or hoop along the pile axis.
In the vicinity of the potential plastic hinge at the top of the pile. the amount
of spiral or hoop reinforcement can be adjusted to ensure that the ultimate
compression strain given by Eqn. (T5.12), is adequate to provide the required
displacements at the damage-control limit state. Thus, the designer has some
ability to' optimize the design of pile confinement, dependent on the displace-
ment requirement predicted under the lateral response analysis. The calculated
value should be supplied for at least 2D
p
from the critical section. Because of
uncertainties associated with the final position of the tip of a driven pile prior to
driving, a longer region of pile should have the increased confinement
determined from the above approach. In many cases, the full pile length 'will
conservatively be confined with the volumetric ratio required at the deck/pile
hinge location.
As an alternative to the approach outlined above, a prescriptive require-
ment, modified from bridge design, and defined by Eqn. (T5.20) may be
adopted
f [ 1.25(N + F.) J
I U p. +O.13(PI-O.OI)
I; he
(T5.20)
where N; =axial compression load on pile, including seismic load
F, =axial prestress force in pile
PI =longitudinal reinforcement ratio, including prestressing steel.
The volumetric ratio of transverse reinforcement given by Eqn. (T5.20) may
be used as a starting point, but the adequacy of the amount provided must always
be checked by comparing displacement demand with capacity.
The pitch of spiral reinforcement provided for confinement should not exceed
6d
b
nor D15, where db is the diameter of the dowel reinforcement. For in-ground
hinges in prestressed piles, the pitch should not exceed 3.5d
p
where d, is the
nominal diameter of the prestressing strand.
(3) Addition of mild steel reinforcement to prestressed piles
The use of mild steel dowels to provide moment-resisting connections between
piles and decks is common. It is also common to provide additional mild steel
reinforcement throughout the length of the pile in the belief that this will enhance
the performance of the in-ground hinge. Tests by Falconer and Park (1983) have
shown this additional reinforcement to be unnecessary. Provided adequate
confinement is provided at a pitch not greater than 3.5 times the prestress strand
nominal diameter, .dependable ductile response can be assured. Figure T5.11
compares lateral load-displacement hysteresis loops of prestressed piles with and
246 PlANe
without additional longitudinal mild steel reinforcement, and subjected to high
axial load levels. Both piles were able to sustain displacement ductility levels of
10 without failure. The pile with additional mild steel reinforcement was, as
expected, stronger than the pile without additional reinforcement, and the loops
indicated somewhat enhanced energy dissipation. However, the in-ground hinges
will normally only be subjected to moderate levels of ductility demand, for which
the added damping provided by the mild steel reinforcement will be of only
minor benefit.
It is thus recommended that additional longitudinal mild steel reinforcement be
provided in piles only when there is a need to increase the flexural strength.
1kN =225/b
1mm=0.0394in
11=6 11=8 11=10
I I I

effect
IUNlT1\
500
IJ=2
tal

effect
'theoretical
11 =6 JJ =8 IJ=10
I I I
IUNlT21
Lateral
Load
(kN)
timtJ.:.
1kN =225/b
1mm =0.0394 in
I I I
11 =10 IJ=8 II =6
(0)
F.i: 1. Load-displacement hysteresis loops for prestressed piles (after Falconer and
"-Park. ]983).
(a) Fully confined pile with additional longitudinal mild steel reinforcement.
(b) Fully confined pile without additional longitudinal mild steel re-
inforcement.
--- - ~ - .... -._....._-.,..
(4) Capacity protection of elastic actions and members
The essence of modern seismic design, is the precise determination of where and
how inelastic actions may occur (i.e, by inelastic flexural rotations in specified
plastic hinge locations). and the protection of other locations (e.g. the deck) and
other actions (e.g, shear) to ensure these remain elastic. This is termed 'capacity
design' (Priestley et al., 1996) and is done by ensuring that the dependable
strengths of the protected locations and actions exceeds the maximum feasible
demand based on high estimates of the flexural strength of plastic hinges. Since
development of flexural plastic hinge strength is certain at the design seismic
input, the consequence of material strengths significantly exceeding design val-
. ues will be that corresponding increases will develop in the forces of capacity
protected members.
The most consistent method for determination of the required strength of
capacity protected actions and members is to carry out a second series of
pushover analyses, or dynamic time-history analyses, where the moment curva-
ture characteristics of the pile plastic hinges are based on realistic upper bound
estimate of material strengths. The following values are recommended:
Concrete compression strength
Reinforcement yield strength
Prestress strand ultimate strength
f c ~ =1.7f:
fyro =1.3fy
fpum = l.lfpu
The design required strength for the capacity protected members and actions
should then be determined from the pushover analyses at displacements
corresponding to the damage control limit state. Since these force levels will be
higher than those corresponding to the serviceability limit state, there is no need
to check capacity protection at the serviceability limit state.
A simpler, conservative approach to the use of a second 'upper bound'
pushover analyses is to multiply the force levels determined for capacity pro-
tected actions from the initial design analysis by a constant factor, representing
the maximum feasible ratio of required strength based on upper bound and lower
bound material strengths in plastic hinges. This ratio should be taken as 1.4
(Priestley et al., 1996).
T5.4.6 Shear strength of piles
(1) Shear strength of concrete piles
The requirement for capacity protection is that the dependable strength exceeds
the maximum feasible demand. Hence, shear strength should be based on nomi-
nal material strengths, and shear strength reduction factors should be employed.
Most existing code equations for shear strength Of compression members,
including the Americal Concrete Institute's (ACI's) 318 equations, which are
248 PlANe
widely used in the USA, tend to be unreasonably conservative, but do not
adequately represent the influence of reduction of the strength of concrete shear
resisting mechanisms. An alternative approach (Kowalski et aI., 1998), which has
been widely calibrated against experimental data and shown to provide good
agreement over a wide range' of parameter variations, is recommended for
assessing the strength of piles. This approach is based on a three parameter
model, with separate contributions to shear strength from concrete (V
e
) ,
transverse reinforcement (VI)' and axial load (V
p
)
V
N
= V, + VI + V
p
(T5.21)
A shear strength reduction factor of 0.85 should be applied to Eqn. (T5.21) to
determine the dependable shear strength.
(T5.22)
where k, = factor dependent on the curvature ductility within the plastic hinge
region, given by Fig. T5.12 .
f; = concrete compression strength in MPa
A
e
= O.8A1!TO<s is the effective shear area.
The reduction in k, with increasing curvature ductility J 1 ~ = /" occurs as a
result of reduced aggregate interlock effectiveness as wide cracks develop in the
plastic hinge region. For regions further than 2D
p
from the plastic hinge location,
the flexural cracks will be small. and the strength can be based on flo = 1.0.
Different values of k, are provided in Fig. T5.12 for new design and for assess-
ment. It is appropriate to be more conservative for new design than for assess-
ment, since the economic consequences of extra conservatism are insignificant
when new designs are considered. but can be substantial when assessment of
existing structures are concerned. Different values are also given, depending on
whether the pile is likely to be subjected to inelastic action in two orthogonal
directions (biaxial ductility) or just in one direction (uniaxial ductility).
Concrete Mechanism Strength: The strength of the concrete shear resisting
mechanisms, which include the effects of compression shear transfer, aggregate
interlock, and dowel action, is given by
V
e
= keF: . A
e
Transverse Reinforcement (truss) Mechanism: The strength of the truss mech-
anism invol ving transverse spirals or hoops is given by:
Circular spirals or hoops
IT
\\ = 1 ~ p f Y h (Dp -C
ex
-co)(cot a
t
) / S (T5.23)
where Asp = spiral or hoop cross section area
f'h = yield strength of transverse spiral or hoop reinforcement
. .
Dr = pile diameter or gross depth (in the case of a rectangular pile with
Spiral confinement)
C
O
\ = depth from extreme compression fibre to neutral axis at tlexural
strength (see Fig. T5.13)
c, =concrete cover to centre of hoop or spiral (see Fig. T5.13)
= angle of critical crack to the pile axis (see Fig. T5.13) taken as 30
for assessment. and 35 for new desizn.
.
s = spacing of hoops or spiral along the pile axis.
Rectangular hoops or spirals
Vi =AJh(D
p
- C
ex
- co) (cot a)/s (TS.24)
.where = total area of transverse reinforcement, parallel to direction of applied
shear cut by an inclined shear crack.
Axial Load Mechanism: The presence of axial compression enhances the shear
strength by development of an internal compression strut in the pile between the
compression zones of plastic hinges, whose horizontal component V
p
opposes the
applied shear force. Axial prestress also acts in similar fashion to enhance.shear
strength. Thus. with reference to Fig. T5. 14, the shear strength provided by axial
compression is
V
p
=<Pp(N
u
+ F
p)
tan a"" ., (T5.25)
where N, =external axial compression on pile including load due to earthquake
action
F, =prestress compression force in pile
.5
0.6
0.5
1.2
1.0
3.0
20 18 16 8 10 12 14
Curvature Ductility Factor
6 4 2
I
I
i k,.S[ I us uniJ
I
Curvacun: Biaxial Dw:tiliev Uniaxial DuaiiiN
Ductility i
I-
I
FaclOr
Assessment Design Asscument I Design
\\ \
0.29 / 3.5 I 0.25/ 3.0
- 3
\ 0.29/3.5 0.25/3.0
I
l 0.29/3.5 I 0.25/3.0
5 0.\/1.2 0.083/ \.0 I
\
t\\
l\
7 ! 0.\/1.2 I 0.083/\.0
,
13 0.05/0.6 0.042/0.5 I I
-, 15
: 0.05/0.6 I O.042!O.S
,
I I
"\
"\\
I I
;
'niaxial
[vc=k.:.Jf'"c A.l
"
I
'
----- Design
Assessment
1
I- " -
t:---
'"--
t:---
. -
-----
t:::---
1-_-
r::--:: --.
I-
-----
:. .._--
1-----
,
-----
o
o
0.05

O.
0.083
0.25
"'iii 0.2
Q.
::E
0.15
0.35
0.3
Fig. T5.12. Concrete shear mechanism.
250 PlANe
au=angle between line joining the centres of flexural compression in the
deck/pile and in-ground hinges, and the pile axis (see Fig. T5.14)
ct>p =1.0 for assessment, and 0.85 for new design.
Fig. T5.13. Transverse shear mecha-
nism.
I
c
ex
I
'j
,/
NA
../
\.
I
,
r\
I
\
\
I
. ~
'\
51 ~ Y '
\.!
<,
;:..:<::: N.A.
<,
I DECK
c
~ _ - - - l DecklPile Hinge
..
Nu+F
p
Compression force
Vp=d> (Nu+Fp>t.an CLax
In ground hinge ..
Fig. T5.14. Axial force shear rnecha-
msm.
(2) Shear strength of concrete-tilled steel shell piles
The flexural and shear strength of concrete-tilled steel shell piles can be deter-
mined based on normal reinforced concrete theory. assuming full composite
action between the shell and the infill concrete. Although some slip between the
shell and concrete may occur. it does not appear to significantly int1uence flex-
ural strength or displacement capacity.
Shear strength can be determined using the equations above, and considering
the steel shell as additional transverse hoop reinforcement. with area equal to the
shell thickness. and spacing along the pile axis of s =1:0. The contribution of the
shell to the shear strenzth is thus
'-
V,h'lI = - Ccx - cJ cot at (TS.26)
where ' is the thickness of the steel shell.
T5.4.7 Design strength for deck members
The required strength for deck members should be determined by adding the
actions resulting from the 'upper bound' pushover response, at the displacement
levels corresponding to the Level 2 earthquake, to those resulting from gravity
action. Particularly in assessment of existing structures, some redistribution of
.
design actions is appropriate, provided overall equilibrium between internal
actions and external forces is maintained.
Dependable strength for flexure and shear actions can be determined in accor-
dance with ACI 318 principles.
T5.4.8 Batter and timber piles
(1) Assessment of wharves and piers with batter piles
Batter piles primarily respond to earthquakes by developing large axial compres-
sion or tension forces. Bending moments are generally of secondary importance.
The strength in compression may be dictated by material compression failure, by
buckling, or more commonly by failure of the deck/pile connection, or by exces-
sive local shear in deck members adjacent to the batter pile. Strength in tension
may be dictated by connection strength or by pile pull out. In assessing the
seismic performance of wharves and piers' with batter piles, the following
additional items should be considered
Pile pullout is ductile and has the potential to dissipate a considerable amount
of energy. Displacement capacity is essentially unlimited.
In compression, displacement capacity should consider the effect of reduction
in pile modulus of elasticity at high axial load levels, and the increase in effec-
tive length for friction piles, resulting from local slip between pile and soil.
252 PlANe
Typical calculations indicate that displacement ductility of two or more are
possible. .
Where the prime concern is the prevention of total collapse, it should be
recognized that failure of the batter piles does not necessarily constitute fail-
ure of the wharf or pier, nor the initiation of collapse. It is possible that, after
failure of the batter piles, the wharf or pier may be capable of sustaining higher
levels of seismic attack. Although the strength will be diminished as a conse-
quence of the batter pile failure, the displacement capacity of the wharf or pier
will generally be greatly increased before the secondary failure stage, involv-
ing the vertical piles, develops. Consequently this system, involving only the
vertical piles, should be checked independently of the batter pile system.
(2) Assessment of wharves and piers with timber piles
There is little reported evidence of damage to timber piles in earthquakes, despite
their wide spread usage for wharf and pier supports for the past 150 years on the
West Coast of the USA. This is due to their large displacement capacity, and the
typically low mass of the supported wharf, if also constructed of timber, though
there are cases of timber piles supporting concrete decks.
Extensive testing of timber piles, both new and used, in both dry and saturated
conditions, has been carried out by British Columbia Hydro (B.C. Hydro, 1992).
No significant difference between results of new and old, wet and dry. piles was
found. The tests were carried out on nominal 12-in. (30 em) diameter peeled
Douglas fir piles, though actual dimensions were as large as 14 in. (35 em) at the
butt end, and as small as 9 in. (22.5 em) at the tip end. Three types of tests were
carried out
1. A simple bending test using a non-central lateral load on a pile simply sup-
ported over a length of 27.2 feet (8.28 m).
2. Cantilever bending tests on piles embedded in concrete pile caps with a free
length of about 4 feet (1.2 m).
3. In-situ testing of piles embedded in a finn silty-sand foundation with a free
length of about 16.4 feet (5 m).
In all cases, the piles were subjected to cyclic loading. Only in the first series
of tests were the displacements equal in the opposite loading directions. All piles
were subjected to axial loads of about 20 kips (89 kN) throughout the testing. The
results indicated that the piles typically exhibited ductile behaviour. Failure
generally initiated by compression wrinkling at the critical section. followed by
a period of essentially plastic deformation at approximately constant lateral load,
and terminated in a tension fracture. The first series of tests gave the lowest
results in terms of displacement capacity. This may have been due to the steel
loading collar. which applied lateral force to the pile at the location of maximum
moment. causing local distress because of the sharp edges of the collar. In fact,
failure was not achieved in the other two test series before actuator travel capac-
ity was reached, despite displacement of 1.0 m (40 in.) for the in-situ pile tests.
Back-analyses from the more conservative. (and more extensive) first series
(simply supported beams) indicated that the following data might conservatively
be used for checking the capacity of existing Douglas fir piles:
Modulus of Elasticity: 10 GPa (1.5 x 10
6
psi)
Modulus of Rupture: 35 GPa (5000 psi)
Serviceability limit strain: 0.004
Ultimate limit strain: 0.008
Damping at ultimate limit: 2%
The fracture strain has been back calculated from the maximum displacements
assuming a linear curvature distribution along the pile. This is because there is
insufficient data to define a true ultimate strain, and an effective plastic hinge
length. Thus displacement capacity should be based on the same assumptions.
Note that this will result in the ultimate displacement capacity of a 12-in. (30 em)
pile with an effective length of 25 feet being about 40 in. (100 cm).
Since the effectiveness of lateral bracing in timber wharves will generally be
at best suspect, it is recommended that as a first approximation it conservatively
be ignored when assessing seismic performance. This assumes that the bracing
connections, or members, will fail or soften to the extent that they will be in-
effective after the initial stages of lateral response. Similarly, it may be sufficient
to assume that the connections between batter piles and the wharf fail, and to
check the 'worst case' condition without the batter piles. Generally, calculations
based on these conservative assumptions will still produce displacement
demands that are significantly less than displacement capacity.
T5.5 PILEJDECK CONNECTION DETAILS
Connection details between the pile and the deck depend on the type of pile used
to support the wharf or pier. Although connection details for buildings and
bridges are often detailed to act as pinned connections, piles for wharves and
piers are almost always designed for moment resistance at the pile/deck connec-
tion. Consequently, only moment-resisting connections will be considered in this
technical commentary.
T5.5.1 Steel-shell piles
Steel shell piles will normally be connected to the deck via reinforcing bars and
a concrete plug, even when the concrete infill is not continuous down the height
of the pile. If the concrete plug is only placed in the vicinity of the connection,
care is needed to ensure that shear transfer exists between the concrete and the
steel shell, Although this may often be adequately provided by natural roughness
254 PlANe
of the inside surface of the steel shell, some more positive methods of transfer
should be considered. One possibility is the use of weld-metal laid on the inside
surface of the steel shell in a continuous spiral in the connection region prior to
placing the concrete plug (Park et al., 1983). Dependable flexural strength and
extremely large ductility capacity could be achieved for concrete filled steel shell
piles when the steel shell was embedded 5 em inside the deck concrete, and the
connection was made by dowel reinforcement properly anchored in the deck.
An example of the force-displacement hysteresis response is shown in
Fig. T5.l5. It will be observed that the flexural strength considerably exceeds the
nominal strength, denoted HM;I' when P-!! effects are included. This is partly a
result of the enhanced concrete strength resulting from very effective concrete
confinement by the steel shell, and partly due to the steel shell acting as
compression reinforcement by bearing against the deck concrete. The remarkable
ductility capacity of the pile apparent in Fig. T5.15 has been duplicated in other
tests with discontinuous casings at the interface. This pile, subsequent to two
cycles at a displacement ductility of J1,j = 18, was subjected to 81 dynamic cycles
at u, =20, and then a static half cycle to p,j= 40 without significant strength
degradation.
T5.'5.2 Prestressed piles
In the USA, prestressed piles are normally connected to the deck with mild-steel
dowel reinforcement, as discussed in the next section. However, other details are
possible, and have been successfully tested under simulated seismic conditions.
Pam et al. (1988) tested a number of connection details appropriate for
prestressed solid piles. The details were based on 40 em octagonal piles with
. 0 ,30 ,20 "0 "'f' 90 -80 -70
/ 'I
I '!heo I
""-t-;. __ P-lleffect I
---.-- '"
- - - - - - - - - - - - 1::...:1---+:; ......
Post -
I loading I lit lit
-40 -18 -/6 -10 -8 -8 -tI 2:p
2 4 6 B 10 16 IB=p
I I I I I I I Theorelical

P-ll - -
effect
H"CI=160 kN
!::'y =3.2 mm .
Fig. T5.15. Force-displacement response of concrete-filled steel shell pile connected to
deck by dowels.
10-12.5 mm tendons. but the results can be safely extrapolated to the 60 cm piles
more commonly used in wharf structures. Two units were tested with the solid
end of the pile embedded 80 ern into the deck. with a 10 mm diameter spiral at
15 cm pitch placed around the embedded length of the pile. Two further piles
were tested with the strand exposed. enclosed in a 12 mm spiral at 47 mm pitch.
and embedded straight in the deck for a distance of 60 ern. In each of the pairs of
piles. one pile contained only strand, while the other included 10-20 mm diam-
eter mild steel dowels. thus increasing the flexural strength of the piles. All pile
units were subjected to an axial load ratio of =0.2. in addition to the
axial compression resulting from prestress.
Results for the force-displacement response of these four units are shown in
Fig. T5.16. In each case, the piles were capable of developing the theoretical
moment capacity of the connection, and to maintain it to high ductility levels.
The apparent strength degradation in Fig. T5.16 is a result of P-L\effects from
the moderately high axial load.
The first sign of crushing of the concrete in the plastic hinge region was noted at
displacement ductility factors of about 2.0 for the embedded pile and about 2.5 for
the embedded strand. The difference was due to the higher total compression force
(a) (b)
Z 3.1
u
a
2.1 I 4.2 6.3 8.6
'a I I I l
120 :..-. C:
'iii ._ ::,ot,
.! AS
l
t
60 Theoretical
M
u

-200
IUNIT PC2!
A8

I I I
-8.6 -0.3
-7.3
100 200
Lateral displacement
(mm)
Z 2.8
ua1.8!3.7 5.5 7.4 9.2
." I I , I I
Theoretical M
u
- C,.J
60 . ....0 r
3 SA
lUNIT PC1 I
200

I I I I II .120
-9.2 .7.4 -5.5 -3.7 1.8=u
2.8
6.3 8.6
I I
200
Lateral displacement
(mmj
C First visible crushing of
cover concrete
Onset of of cover
o First observed buckling of
tendon
I -2.2="
-3.6
(d)
IUNIT PC41
A8
0' _

fractured -9.0
8.0 10.7 Theoretical
I I M
u
--
8
-300
(c)
\UNITPC3\
Fig. T5.16. Lateralforce-displacement response for four pile-pile cap connection details
(D
p
= 400 mm) (after Pam et al., 1988).
(a) Solid pile embedded 800 rnm, no dowels.
(b) Solid pile with standard dowels embedded 800 mm.
(c) Exposed strand embedded 600 mm, no dowels.
(d) Exposed strand and dowels embedded 600 rom.
256 PlANe
at the critical section (due to the pile prestress force) when the full pile section was
embedded in the pile cap. Since the piles were very slender, with an aspect ratio of
7, these values might be considered as lower bound estimates for a serviceability
criterion for the piles. However, this would be compensated by increased yield dis-
placements if the flexibility of the deck had been included in the tests.
In all cases, the strand was fully developed, despite the rather short develop-
ment length of 48d
b
in the exposed strand tests, where db denotes the bar diame-
ter of the prestressing strand. These results indicate that strand is sufficient to
develop moment capacity at the deck/pile interface, and that mild steel dowels are
not necessary.
. The test units by Pam et al. (l990b) had rather light joint reinforcement sur-
rounding the pile reinforcement in the joint region. However, it must be realized
that the deck in these tests was rigidly connected to a reaction frame, which
reduced stress levels in the joint region. Practical details should thus not be based
onthe joint rebar in Pam's tests.
T5.5.3 Practical connection considerations
The most common connection detail consists of dowel reinforcing bars, typically
of size #8, #9, or#10, inserted in ducts in the top of the pile after the pile has been
driven to the required level, and the top cut to the correct final elevation.
Typically the dowel bars in the past have been bent outwards, with the top of the
bend below the level of the top mat of the deck reinforcement, as suggested in
Fig. T5.17(a). In the past, the top of the bend has often been much lower than
shown in this figure.
If hooks are provided on dowels, they must have the tails bent inwards, rather
than outwards. The reason for this is that if significant tension force is transferred
up to the hook, which is bent outwards, it adds tension stress within the joint
region which is already subjected to high tension stress as a result of joint shear
forces. There is then a tendency for the diagonal joint shear crack to propagate
and bend horizontally outside the hook, particularly if the hooks are below the
top layer of deck reinforcement, as illustrated in Fig. T5.17(a). The problem is
compoundedif the top of the dowel hooks is lower in the deck than shown in
Fig. T5.17(a). Proper force transfer between the pile and deck becomes increas-
ingly doubtful, and spalling of the soffit concrete may OCCUI. as has been
observed in response to moderate earthquakes in Southern California.
If the hooks are high, and are bent inwards as shown in Fig. T5.17(b). the bend
results in anchorage forces directed back towards the compression comer of the
deck/pile connection, resulting in a stable force-transfer mechanism.
Although satisfactory force transfer is expected with the detail of
Fig. T5.l7(b), it can be difficult to construct. This is because tolerances in the
final position in plan of a driven pile may be as high as +/-15 em. In such cases.
... 01._ ........,_.... .. -- .
T
'--'" .
M
(b)

"'"""
<;
/
z
I
I
:
-::
i
<::::::>----
-T
(al
'--"" +
M
i

I
"
<}" .,.
?
;:.
",tri
v
,
-
ott

""
Fig. T5.l7. Anchorage with hooked dowels.
(a) Hooks low, bent out.
(b) Hooks high, bent in.
interference between the bent dowels and the deck reinforcement can cause
excessive placing difficulties, which may result in one or more dowel bars being
omitted. Design details need to be developed that are simple and insensitive to
pile position. This means that bent dowels should not generally be used.
Two alternative details, developed for the Port of Los Angeles (Priestley, 1998)
are shown for typical 60 em prestressed piles in Fig. T5.l8. The first uses straight
bar development up as high as possible into the deck, but with bars not terminat-
ing more than 10 em below the top surface. This is combined with spiral rein-
forcement to control the potential joint shear cracking. The second uses reduced
embedment length of the dowels, with partial anchorage provided by enlarged
end upstands (end bulbs), with bars lapped with headed vertical bond bars. The
lap and joint are again confined by a spiral.
Straight-bar Embedment Detail: The straight bar embedment detail of
Fig. T5.18(a) is directly analogous to details used for moment-resisting
column/cap-beam connections for seismic response of bridges (Priestley et al.,
1996). The required embedment length, le' is thus given by
SI units I, =O.3d, JJ: (MPa, mrn) (TS.27a)
US units I, =O.3d, JJ: (psi, in.) (TS.27b)
where the dowel bar diameter db is in mm or inches, and the material strengths
are in MPa or psi.
For typical concrete strengths of about fe' =30 MPa and dowel strength of
he = 450 MPa, Eqn. (T5.27) indicates that it is feasible to anchor #8 to #10
258 PlANe
extra tum top & bottom
r10cm maximum ./1for anchorage.
n n 7nln
I
-.--

I
"'"
'7
#[email protected]
==6Ocm
~
1
"-
iii ~
>
l,---" 60em pile
-
......
' - - ~ .........
L...--
1
90cm beam
!
(a)
dowels with bulb ends
!
60cmslab or
90cm beam
#9 headed bond bars
extra tum top & bottom
with bulb ends '\ I
/l for anchorage
\i.. ~
T
II
61
'"
~
#[email protected]
i.
I
e
J-
~
-
....--60cm pile
#10. or #11
i-
els
......
.......
L...-
#9.
dow
<h)
(Note: Take dowels up to as high as possible: 7.5cm from C.G. top or bottom mat okay)
#9headedbond bars; 75cm longfor 90cmbeam; 45cm long for 60cmbeam
Fig. T5.18. Anchorage details for dowels.
(a) Anchorage detail for dowels in 90 cm beam.
(b) Anchorage detail for dowels with upstands.
dowels by straight embedment in a 90 cmdeep deck beam. allowing 10 cm. from
the top of the dowel to the deck surface. Anchorage of a # 11 bar in a 90 ern
deck would require a somewhat higher deck compression strength. Note that the
use of the specified 28 day concrete compression strength in Eqn. (T5.27) is
very conservative. given expected conservatism in concrete butch design. and
expected strength gain before occurrence of the design earthquake, and it
would be more realistic, particularly for assessment of existing structures to use
a more characteristic strength, say 1.2f' This would still be considerably less
than the probable strength at the age of the concrete when subjected to seismic
loading. .
Priestley (1998) requires that. if additional external joint reinforcement is not
provided. the anchored bars must be enclosed in spiral or hoop confinement with
the effective volume ratio of the confining steel, given by
=0.46A,c [/,:]
P. ta, f.
(T5.28)
where !r: =l.4.fv is the dowel overstrength bar capacity
I. = 0.0015E
d
l, =actual embedment length provided
A
sc
=total area of dowel bars in the connection
Ed =dowel modulus of elasticity
D' =diameter of the connection core, measured to the centreline of the
spiral confinement.
The detail shown in Fig.. T5.18(a) is adequate for embedment of 8 #11 dowels
in a 90 em deck beam.
Headed Rebar Detail: This detail, shown in Fig. T5.18(b), is designed to allow
the dowels to terminate below the top layer of beam reinforcement. Anchorage is
improved by using an upstand on the end of the dowel. In this design, 50% of the
dowel anchorage force is transferred by the upstand, and 50% by bond, using
special headed reinforcement bond bars. This allows the minimum embedment
length to be reduced to 50% of that given by Eqn. (T5.27). The top of the dowel
should be as high as practicable, but in this case, may be below the top mat of
reinforcement. With this detail, all bars sized up to #11 can safely be anchored in
a 90 em deck beam, and all sizes up to # I0 can be anchored in a 60 ern deep
deck beam.
The total required area of the headed bond bars is AI= 0.65A<e' Spiral confine-
ment around the connection must be provided in accordance with Eqn, (T5.28).
Details similar to those shown in Fig. T5.18(a) have been tested for bridge
designs (Sritharan and Priestley, 1998) and. found. to perform well, with dis-
placement ductilities typically exceeding Jl4 == 6. The detail of Fig. T5.18(b) was
260 PlANe
Y.":
,
: ~ : ~ ; ' recently tested for the Port of Los Angeles, and found to have excellent response
(see Fig. T5.19), with failure finally occurring outside the connection at high
displacement ductility. Only minor cracking developed in the joint region.
Steel H-Section Piles: Connection of steel H-Piles to pile caps or decks is nor-
mally by partial embedment in the deck. Full moment-resisting connections are
rarely attempted with this detail. If moment-resistance is required of the connec-
tion, sufficient transverse reinforcement must be placed around the pile head to
enable the pile moment to be transferred by bearing in opposite directions at the
upper and lower regions of the pile embedment.
I
I
I
I
T5.5.4 Capacity of existing substandard connection details
Many existing piers and wharves will have connection details similar to that
shown in Fig. T5.17(a), where the dowels bend outwards from the pile centreline,
and the top of the 90 bend, h
d
above the soffit, is well below the deck surface.
Typically.the dowels will not be enclosed in a spiral within the joint region, and
there is thus aprobability of joint shear failure.
There are no known test data related to the detail of Fig. T5.17. but poor
performance of such details in the recent moderate Northridge earthquake, and
similarity to bridge T-joints, which have been extensively tested, leads to a need
for conservative assessment.
-4 -3 -2 u 1- ----Theoretical predictionI
10 -8 -6
-500
Drift(%)
600
-6 -4 -2 0 2 4 6
I
500
234 6 8 10 u
400
300
-
Z
200
.::t:
-
Q)
100
~
0
u,
0
"0
.9:? -100
a.
a.
-600 I I
-0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1
Displacement (m)
Fig. T5.19. Force-displacement response of pile connected to deck with headed rebar
(detail of Fig. T5.l8).
!
;L-...
(T5.29) .
A relevant procedure is available in Priestley et al. (1996), and is adapted in
the following for wharves and piers with reinforced concrete decks.
1. Determine the nominal shear stress in the joint region corresponding to the
pile plastic moment capacity.
tv/"
v =-=--
j J2h,jD:
where tvf is the overstrength moment of the plastic hinge, and h
d
is the depth
of the pile cap. The overstrength moment is the maximum possible moment
that could occur in the pile. It is an upper bound assessment of capacity to
ensure adequate shear and is usually based on multiplying the ultimate
moment by some factor (usually 1.4 times the nominal momentcapacity).
2. Determine the nominal principal tension stress in the joint region
(T5.30)
where
(T5.31)
is the average compression stress at the joint centre, caused by the pile axial
compression force N. Note, if the pile is subjected to axial tension under the
seismic load case considered, then the value of N, and thusI., will be negative.
3. If PI calculated from Eqn. (T5.30) exceeds 0.42 JJ: MPa, joint failure
will occur at a lower moment than the column plastic moment capacity
M'. In this case, the maximum moment that can be developed at the pile/deck
interface will be limited by the joint principal tension stress capacity, which
will continue to degrade as the joint rotation increases, in accordance with
Fig. T5.20. The moment capacity of the connection at a given joint rotation
can be found from the following steps:
a) From Fig. T5.20, determine the principal tension ratio P, I fJ:
corresponding to the given joint rotational shear strain, referring to the
T-joint curve.
b) Determine the corresponding joint shear stress force from
vJJPt(PI - fJ (T5.32)
wheref. is determined from Eqn. (T5.31).
c) The moment at the pile/deck interface can be approximated by
Mt; = S MO (T5.33)
.... -.
...: ."
262 PlANe
,
-.. ~ ....
1
I
I
I
I
I
I
'1
I
I .
I
I
I
I
I
o I
Q 3.5
10
~
5.0----------------------------------------- ().42
0.00015 0.0007 0.01
Joint Rotational ShearStrain(radians)
0.04
Fig. T5.20. Degradation of effective principal tension strength with joint shear strain
(after Priestley et al., 1996).
This will result in a reduced strength and effective stiffness for the pile in a
pushover analysis. The maximum displacement capacity of the pile should be
based on a drift anale of 0.04.
...
T5.6 EXISTING CONSTRUCTION
I
I
i
I
I
The criteria presented above were developed from.a compilation of current prac-
tice by many agencies combined with state-of-the-art technology for. estimation
of seismic damage potential. It is not a revolutionary step forward but rather an
evolution of design. The developed integrated criteria specify:
1. The required structure performance under expected loads.
.., Specification of the expected loads.
3. Specification of strain limits to ensure structural response limits to achieve
performance requirements.
The overall effect on the design. selection of pile sizes and cost of the pier is
not expected to be great; however. the assurance in meeting performance goals is
thought to be substantially enhanced.
In the application of these criteria to existing construction, it is thought that
the objective of a uniform set of performance goals should be maintained
across the waterfront. Where adequate capacity is lacking in the existing
system. it is thought better to strive for the performance goal. develop
several candidate upgrade alternatives, and then perform an economic/risk
analysis to determine what is the most cost effective solution, considering
the potential for a damaging earthquake and the existing lateral force system.
This approach varies from alternative procedures outlined in current
codes, wherein retrofit and restrengthening measures are initially designed
using criteria for new construction, then the computed response (e.g., loads.
moments. drift) is reduced by the application of semi-empirical reduction
factors for existing structures. The adherence to performance standards for
design is recommended over the application of simplified code-based reduction
factors, as any single reduction coefficient is not optimal over a range of
structures.
T5.6.1 Structural criteria for existing construction
The discussion of existing structures is of major importance, since there are many
existing facilities in use and relatively few new facilities being constructed. Many
of the existing structures were built during periods when seismic standards were
not well defined. In general, existing-structure performance criteria may be
related to new-structure performance criteria in terms of the degree of hazard, the
amount of strength required, the extent of ductility demand allowed, or the level
of design ground motion. The structure once built does not 'know' that it is
expected to perform to a 'new' or an 'existing' structure criteria; it responds
according the principles of structural dynamics.
The approach taken herein is to recognize that the goals for both new and
existing facilities should be the same. The structural parameters that are used to
control the response should be the same. What is of significance is that existing
structures have been in place for a period of time and have a shorter remaining
life than newfacilities. Thus, existing facilities face a shorter exposure to natural
hazards. This major factor suggests that the design ground motions be allowed to
differ based on the differing remaining life spans of the structures. A prudent
course must be charted to select reasonable goals for existing structures to the
adverse impact on the economics of port operations while ensuring public man-
dates for preservation of the environment.
The approach taken in this criterion is to utilize a factor, a.; to relate the
existing-structure exposure time to the new-construction exposure time taken as
50 years. The value of ely is equal to or less than 1.0. The value of ely is used to
define the exposure time for use in the Levelland Level 2. earthquake return
times, as shown in the following sections. Determination of a
y
establishes the
seismic loading. The performance goals and response limits for existing con-
struction remain the same as for new construction; only the loading is reduced.
This applies to all elements including the structure, the foundation and all
associated lifelines. ;
Seismic reviews of existing: waterfront construction directed by requirements
of regulatory agencies can utilize the above criteria for new construction as the
..
j s, ~ ....
" -.' ~
264 PlANe
target goal requirement for performance. In general, the existing structure must
satisfy the new structure performance limit states under earthquake peak ground
acceleration levels corresponding to reduced exposure times as follows:
Existing-Structure Exposure TIme = ely(New-Construction Exposure TIme)
where New-Construction Exposure TIme is 50years.
. The requirement for evaluation of the seismic resistance and possible upgrade
is triggered when the load on the structure changes, such as when the operation
of the structure is changed, or when the structure requires major repairs, or mod-
ifications to meet operational needs, or when recertification is required.
Determining ~ . When it is shown to be impossible or uneconomical to achieve
new construction levels of performance, (ely= 1), an economic/risk analysis using
procedures shown by Ferritto.et al. (1999) can be performed to determine the
most appropriate facility exposure time and level of seismic design upgrade.
Various alternative upgrade levels are considered ranging from the existing con-
dition to the maximum achievable. Each alternative should be examined to deter-
mine the cost of the upgrade, the cost of the expected earthquake damage over
the life of the structure, the impact of the damage on life and the environment,
and the structure's operational requirements. The choices of upgrade levels are
made by the design team based on a strategy consistent with requirements of life
safety, operational needs, cost effectiveness, and protection of the environment.
In no case should CXy be less than 0.6 of the 50-year exposure. time for new con-
struction (3.0 years).
As a minimum, analysis should be conducted and data developed for the cases
of ~ equal to 1.0 and 0.6. Additional cases are encouraged:
Levell - An earthquake with a 50% probability of exceedance in c, x (50 years)
exposure.
Level 2 - An earthquake with a 10%probability of exceedance in a, x (50 years)
exposure.
Table T5.2 shows the shortest allowable return times.
Peer review: Review of the results of the analysis by an independent peer review
panel of experts in structural engineering, geoscience, earthquake engineering.
seismic risk analysis, economics, and environmental engineering/science may
be required. In addition, the findings from the analysis can be reviewed by the
public and other stakeholders.
Allowance for Deterioration
In evaluating existing construction, it is most important to:
Evaluate the actual physical conditions of all structural members to determine
the actual sizes and condition of existing members.
Provide an allowance for corrosion and deterioration.
Table T5.2. Shortest allowable return periods for existing structures.
Probability of Exposure
Return Nominal return
non-exceedance time
time time
(%)
(years) (years) (years)
50 30 ..\.3 60
10 30 285 300
Evaluate the properties of the construction materials, considering age effects in
computing yield strengths. Average actual material properties should be used
in the evaluation.
Evaluate the existing structure details and connections. since this is often the
weakest link and source of failure.
Determine the design methodology used by the original designers at the time
the structure was designed and constructed.
....
Evaluate displacement demands and capacities. Previous code requirements
did not emphasize the need for ductility and the failure to include shear and
containment reinforcing is most common in existing construction. This has led
to numerous structure failures especially when batter piles have been used.
T5.6.2 Strengthening of an existing structure
Various methods of strengthening existing structures are possible. They include:
Plating. Where the top of flange is not accessible for adding cover plates,
reinforcement can be added to the web plate. The beam should be relieved of load
before the reinforcement is added. When cover plates are added, the flange to
web connection and the web plate stresses at the toe of the flange should be
investigated.
Composite Action. Beam section properties can be materially increased by
causing the concrete slab to act as a composite with the beam. The slab serves as
a top cover plate.
Prestressing. Jacks can be used effectively to reduce stresses in existing flanges.
Cover plates should be welded before removing jacks.
Shear Reinforcement. Vertical stirrups serve as hangers that support the beam
from the uncracked portion of concrete near the column.
Flexural Reinforcement. Longitudinal reinforcement can be added effectively,
if positive means for preventing separation and for transferring horizontal shear
are used. Composite materials can be glued to the underside of pier decks to
increase section capacity. Composite rods can be inserted and epoxied in groves
cut in the topside of pier decks to add reinforcement.
Pile/Column Reinforcement. Pile/Column .sections may be strengthened by
adding concrete with longitudinal and lateral-reinforcement, or by adding unre-
inforced concrete restrained by hoop bars. Wrapping by composite materials has
{ i.. 'I,;"
" .,.... ~
266 PlANe
been used very effectively for bridge columns. Mandrels have been applied to
piles and filled with concrete.
Other design considerations for existing structures include:
Compatibility. The design details should encompass any inherent incompatibil-
ity of old and new materials. Provision should be"made to resist separation forces.
Newconcrete should have a different modulus of elasticity, coefficient of thermal
expansion, and shrinkage than old concrete. Consider differing expansion effects
due to differing absorption of moisture. Provide resistance against 'curling' due
to thermal gradients. Compatibility of connectors must be considered. For exam-
ple, rivets or bolts are not compatible with welds. Friction bolts are not compat-
ible with rivets. Creep is an important factor.
Dead Load Versus Live Load Stresses. Unless the load on a structure is relieved
(for example, by removal or by jacking), the existing framing will continue to
. carry:
a) the full dead load of the construction,
b) any part of the live load which is in place when the new framing is
connected, and
c) a proportionate share of the live load subsequently added.
The new framing will carry only a part of the live load. As a result, under the final
loading condition, the stresses in the new and existing material of the same or
similar members will be different, often radically so. For example, assuming a
1:1ratio of dead to live load and of new to existing material in the cross section
of a given member and "disregarding plastic deformation, the stress in the
existing material would be three times the stress in the new. As a result, the new
material cannot be stressed up to allowable values without simultaneously
overstressing the existing sections. It is necessary either to provide an excess of
new material, or to relieve the load on the structure before strengthening. This
may not apply. if plastic deformation of the structure (and its associated,
increased deflection) can be permitted.
T5.6.3 Deterioration of waterfront structures
(1) Causes of deterioration
The more common causes of deterioration associated with steel. concrete, and
timber waterfront structures are as follows.
Steel Structures. For steel structures deterioration is caused by:
a) corrosion,
b) abrasion,
c) impact.
Concrete Structures. In concrete structures deterioration iscaused by:
a) corrosion of reinforcement, ..
b) chemical reactions,
c) weathering.
d) swelling of concrete, and
e) impact.
Timber Structures. In timber structures deterioration is caused by:
a) corrosion and abrasion of hardware,
b) borer attack.
c) decay. and
d) impact.
(2) Preventive measures in design and construction
Steel Structures All parts that will be subject to corrosion, should be accessible
for inspection and repair. If not, then they should be encased with concrete
or some other long-life, high-resistance type of coating. Shapes should be
selected that have a minimum of exposed surface. Detailing should be designed
so that accumulations of dirt and debris will be avoided. Avoid narrow crevices
that cannot be painted or sealed. Draw faying surfaces into tight contact by use
of closely spaced stitch rivets, bolts, or welds. Prime faying surfaces before
assembly. In general, the ability of framing to shed water is the single most
important factor in inhibiting corrosion and deterioration of coatings. If the
potential for ponding is unavoidable, provide drain holes. Drain holes should be
a minimum of 47 in. (l0 ern) in diameter to prevent clogging. The use of sacrifi-
cial metal should be avoided in favour of using protective coatings.
The thickness of metal and section properties should be determined by
considerinz the loss of the section as established in MIL-HDBK-lOO213, Steel
...
Structures (U.S. Navy, 1987), unless corrosion protection is provided. Typical
average corrosion rates for bare carbon steel are shown in Table T5.3.
The minimum thickness should not be less than 0.40 in. (1 em). When the
required minimum thickness is excessive, corrosion protection using approved
products or cathodic protection should be used. When coatings are used, care
must be exercised in driving the piles to preclude damage to the coatings.
Additionally, consideration must be given to abrasion of the piles by contact with
a fender. Tips of all steel H piles, having a thickness of metal less than 0.5 in.
(1.3 ern) and driven to end bearing on sound rock by an impact hammer, should
be reinforced. .
Concrete Structures Good quality is. the important factor in obtaining a dense
concrete. This, in tum, is the most important factor in preventing penetration of
moisture, which is the primary cause of deterioration of concrete. Do not use
poorly graded aggregate, or a water-cement ratio greater than 6 gal (22.71 l)/sack
of cement (94 pounds = 418 kN), reduce to 5 gal (18.92 1)/sack of cement for
thinner sections such as slabs and wherever clear cover over reinforcement is 2
in. (5 em) or less. Watertight concrete can be obtained by using air entrainment
(maximum 6% by volume) and a water-cement ratio not greater than 5 gal/sack
of cement.. .
\
i
1
\
I
!
I
I
L O ~ PlANe
Table T5.3. Typical corrosion" rates for bare carbon steel (1.0 ft =0.3048 m).
Zone (elevation from pile tip) Average corrosion rate (mmlyeat)
Imbedded Zone (0 to 15 ft) 2
Erosion Zone (15 to 21 ft) ]0
Immersed Zone (2] to 29 ft) 8
Atmospheric Zone ( ~ 9 to 35 ft) ]2
Types of cement (ACI Committee 116, 2000) affect the deterioration of con-
crete. Type III (high early strength) cement is excessively susceptible to sulphate
attack, and should not be used. In general, avoid the use of Type I cement in a
saltwater environment. Type IX (sulphate-resistant) cement should be used. The
use of Type V (high resistant) cement is seldom required.
Provision should be made for an adequate number of expansion joints. Use
types of expansion joints such as double bents with movement taken up by bend-
ing of the piles or electrometric pads with some form of joint sealer.
In tropical climates and in areas subject to salt spray, consider the use of gal-
vanized or plastic coated reinforcing bars. If plastic coated bars are used, atten-
tion should be given to bond stresses. Excessively rich mixes, over 6 bags per yd
3
(0.764 m'), should be avoided, as excess cement tends to enhance the potential
chemical reaction with seawater. For most aggregates, alkali-aggregate reaction
can be prevented by specifying maximum alkali content of the cement (percent
N3z0' plus 0.658 times percent K
2
0 ) not to exceed 0.6%. In a surf zone, the con-
crete cover and streamline sections should be increased to prevent abrasion.
Calcium chloride (as an accelerator) should not be used in prestressed concrete
and concrete exposed to seawater. The use of calcium nitrite or other chemicals
as a deterrent to corrosion of embedded reinforcing steel, is not an adequate
substitute for good quality concrete and adequate cover. This is not to say that
these additives do not have merit; however. use of coated reinforcing bars may
be required. Timber jackets for concrete piles and stone facing for concrete
seawalls work extremely well to prevent deterioration due to corrosion of rein-
forcement, weathering, and chemical attack. They tend to isolate the concrete
from chemical constituents in the environment. insulate against freezing, and
keep the reinforcing bars free of oxygen.
Reinforced concrete has been used as one of the major construction
materials at the waterfront. Since concrete is much weaker in tension,
cracking would be expected to occur when the tensile stresses in the
concrete were exceeded. typically at a numerical level equal to about 10%
of the maximum compressive stress. Cracking is a normal occurrence in
concrete members under flexural load. When the concrete cracks. the section
moment of inertia is reduced: generally the cracked moment of inertia is
about 30 .to 60% of the gross moment of inertia. depending on the axialload
level and reinforcement content.
. ,
In a marine environment, it is desirous to control the cracking to prevent
corrosion of the reinforcing steel. Confining steel is used to increase concrete
strength. ductility and shear strength. An initial prestress force is used in piles
as a mechanism for improving concrete performance by keeping the cracks
closed. It has been noted that crack widths of 0.18 to 0.22 mm are sufficientlv
- -
small to preclude deterioration of the reinforcement so that an allowable crack
width may be approximated at about 0.25 mm. It is not possible to directly equate
the crack width to an allowable tensile strain since crack spacing is not known:
however. corrosion has not been a problem when reinforcing stress has been
restricted to a tension of 117 kPa or less under service loads. At concrete
compressive strains below 0.0021, the compression concrete does not evidence
damage and crack widths under cyclic load should be acceptable. Occasional
larger loads may be sustained without deterioration, as long as a permanent
offset does not occur and the prestress forces can close the cracks. Reinforcement
deterioration is most pronounced in the presence of oxygen, such as in a pier pile.
where the pile is freestanding out of water or in the splash zone. At deep-water
depths or in soil, the oxygen content is reduced such that pile reinforcement dete-
rioration is less. Large loads causing loss of the concrete cover result in loss of
pile capacity and facilitate deterioration; such conditions can be repaired if
accessible by jackets around the pile. Loss of concrete cover may begin at
displacement ductility of about 1.5 to 2.0.
Timber Structures. Timber structures should conform to the following criteria:
a) Design detail should minimize cutting, especially that which must be done
after treatment.
b) Design detail should provide for ventilation around timbers. Avoid multiple
layers of timbers, as decay is enhanced by moist conditions at facing surfaces.
Curb logs should be set up on blocks. Walers (i.e. edge pieces of timber such
as timbers used as a fender parallel to the pier) should be blocked out from
face of pier. Thin spacers between chocks and wales (i.e. essentially the same
as walers), and gaps between deck and tread planks should be provided.
TC6: Remediation of Liquefiable Soils
T6.1 Overview
T6.1.1 Soil improvement and structural solutions
T6.1.2 Standard design procedure for liquefaction remediation
T6.2 Outline of Remedial Measures against Liquefaction
T6.3 Compaction Method: Design and Installation
T6.3.1 Determination of the SPYN-value to be achieved by compaction
T6.3.2 Designing compaction
T6.4 Drainage Method: Design and Installation
T6.5 Premix Method: Design and Installation
T6.5.1 Principle and application of the method
T6.5.2 Design method
T6.6 Preload Method: Design and Use
T6.6.1 Liquefaction characteristics of overconsolidated soil
T6.6.2 Design and installation procedure
T6.7 Design of Liquefaction Remediation
T6.8 Influence on Existing Structures during Soil Improvement
T6.8.1 Influence on bending stress in a sheet pile wall by compaction
T6.8.2 Excess pore water pressure and vibration resulting from compaction
T6.8.3 Influence of earth pressure increased by compaction
Remediation of Liquefiable Soils
Remediation of liquefiable soils can be an effective technique to improve the per-
formance of port structures subject to earthquake loading. This commentary out-
lines current techniques of seismic remediation, based on Japanese practice, and
proposes design approaches for areas of soil improvement and techniques to
assess the influence of remediated zones on existing structures. The approach dis-
cussed here draws upon the methods presented in the Handbook on Liquefaction
Remediation of Reclaimed Land by the Port and Harbour Research Institute,
Japan (1997).
T6.1 OVERVIEW
T6.I.l Soil improvement and structural solutions
Remediation solutions against liquefaction can be divided into two broad cate-
gories:
I) soil improvement to reduce the probability of liquefaction;
2) structural design to minimize damage in the event of liquefaction.
Both approaches can be further subdivided. Soil improvement techniques are
aimed at:
a) improving the performance of the soil skeleton under earthquake shaking;
b) increasing the rate of dissipation of excess pore water pressure; or
c) a combination of a) and b).
Similarly, structural solutions can be divided into i) those that reinforce the
structure in terms of strength and stiffness, and ii) those that effectively reduce the
. Consequences. of liquefaction in neighbouring or underlying ground through
the use of flexible joints or other structural modifications. Again, in practice a
combination of these two measures is often adopted.
The basic strategies for liquefaction remediation are summarized in Fig. T6.1
a ~ d these follow the outline laid down above. The following sections describe the
f different approaches currently in use. Before discussing these in detail, however,
....
.~ .
274 PlANe
T6.1.2 Standard design procedure for liquefaction remediation
it is important to summarize the standard design process or procedure for lique-
faction remediation.
front areas underground wan
Examples:
To lower the under-
9round water level
(reduce shear stress
andincrease effective
vertical stressof soil
below the water table)
I
Liquefaction
I
remediation
I
Soil I
I
Structural I
improvement design
I I
I I J
To improvethe soil
To achieverapid
To maintain stability
To relieve external
so that the soil grain
dissipation of forcebysoftening
skeletonwiD not col-
excessporewater
by reinforcing
or modifying
lapse under earth-
pressure
structure
structure
quakeloading
..
Examples: Examples: Examples:
Drain Reinforcement of Adjustment of bulk
Replacement pilefoundation unit weight of buried
(with gravel) (increased number structures
andthickness of Anchorage of bur-
I I
piles, install bracing ied structures
To increasethe To reduce members) Change ofs/ope
liquefaction earthquake Reinforcement of angleto flatterangle
strengthof the induced shear soil deformation for embankments
soil stress ratio withsheetpile and anddikesin water-
Examples:
Compaction
Cementation
Preload
Replacement (refilling
withmaterial whichwill
not undergo liquefaction)
Fig. T6.1. Basic strategy for liquefaction remediation.
i
J
1
\
I
I
!
!
I
1
\
I
\
I
i
l
\
A flowchart illustrating a standard design procedure for liquefaction remediation
is shown in Fig. T6.2. Once the strategy has been determined, then it is common
practice to select a method first, and then to compare the advantages and disad-
vantages of different solutions for the particular project. Typically, a number of
different solutions will need to be assessed before a final decision can be reached.
Sometimes, solutions will be combined to achieve the optimum design. The pro-
cedure illustrated in Fig. T6.2 is intended to be used as a guide only: it may often
be more efficient to screen out remediation methods that are clearly unsuitable at
an early stage based on a rough assessment of the likely area to be treated and any
project specific constraints.
A combination of two or more remediation methods is often very effective. A
typical example might be the combination of a low noisellow vibration method,
such as the use of drains, combined with compaction around the improved area,
to constrain or confine the overall site. At present, however, formal procedures for
. .
Remediation of Liquefiable Soils 275
Comparison and selection of
remediation measures (reliability,
constructability, cost, etc.)
In the case of liquefaction
remediation of both subsoils
and structure .
Selection of a structural
remediation method
Structural design for each
method
Design of each remediation
method (determination of
degree of compaction, drain
installation spacing. etc.)
Design of the soil improvement
area against liquefaction
Selection of a soil improvement
method
I
I
I
I
I
I
I
I
I
I
I
t
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
~ - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - ~ - - - - - - -
Fig. T6.2. Standard procedure for liquefaction remediation.
selecting the most efficient combination of methods for a specific site have not
been developed and the engineer must use judgement and experience to reach the
optimum solution.
T6.2 OUTLINE OF REMEDIAL MEASURES AGAINST LIQUEFACTION
(Japanese Society of Soil Mechanics and Foundation Engineering, 1988; 1991)
Remediation methods can be grouped into different categories, depending on the
objective of the treatment. In this section, eight different categories are proposed,
and these are discussed below, Table T6.1 provides a summary of a wide range of
current techniques based on these categories.
(l) Compaction
Methods based on compaction are used to reduce the susceptibility of the
soil to liquefaction by densifying sandy soil using vibration and impact. The
- - - - - - - - - - - ~ - - - - ..---_._---- ...._-----_. __.._-------_.
._---------.....;._...... _ - ~ - - - - - - - - - - - - ~ - - - - - - - - - - - _ . - - _ . - -------_... _......_-------------'-._-
, T"-
' ~ ... , ~ .
Table T6,/. Remedial measures agClillsl Iiquclaction.
N
-....I
0"
Vibro-rod method About
GL-20 111
Principle
Compaction
Methods
Dcsignation
Sand compact ion
piIc method
Vi bronotation
method
Depth
About
GL-35 m
About
GL-20 m
Summary
Insert steel pipe casing
undcrground
Install sand compaction pile
by forcing out sand during
extraction of casing, and
cornpact the sand pile
and the natural soil in a
horizontal direction
simultaneously
Compaction of soil is
conducted using vibratory
penetration of .1 rod and
filling with additional
sand from the ground
surface
The surrounding soil is
compacted with water
sprayed from nozzle tip
end and with horizontal
vibration from vibrator
with a built-in eccentric
Ioat!
Additional sands arc tilled
into the voids in the subsoil
profile
Influence on the
surrounding areas
This method produces high
levels of noise and vibration
The extent of noise and
vibration differsaccording to
the type of construction
equipment used
This method produces noise
and vibration that arc slightly
less than those produced
by the sand compaction
pile method
This method produces
little noise and vibration
compared to other methods
based on compaction
Remarks
Compaction efficiency is
high for soil with fines.
content of less than
25 to 30%. SPT N-value
increases to about
25 to 30
high with respect to soil
having fines content of
less than 15 to 20%.
SPT N-vaJue increases
to about 15 to 20
high with respect to soil
having fines content
of less than 15 to 20%.
SPT N-vulue increases
to about 15 to 20
::s!
~
o
Attachment of
drainage device
for steel pipes or
sheet piles
Deep mix method About
GL-30 m
Pore water
pressure
dissipation
Cementution
and
solidification
Dynamic
compaction
method
Gravel drain
method
About
GL-IO m
About
G L ~ 2 0 m
The soil is compacted with
an impact load by dropping
a weight of 10 to 30 tf from
a height of about 25 to 40 m
above the ground
The gravel pile is installed
by inserting gravel into a
casing placed at prescribed
position and then extracting
the casing. Excess pore
water pressure is dissipated
through the gravel pile
during earthquakes
Piles with a drainage device
are inserted into the subsoil
as liquefaction remedial
measures
Siubilizing material such
as cement is mixed and
solidified in the soil. There
are two approaches:
one is total improvement
which solidities all the soil,
the other is partial
improvement which makes
a solidified wall in the
subsoil profile
This method produces high
levels of vibration and large
impact
Influence on the surrounding
areas is minimal
Noise and vibration depend
on the installation method
used
Influence on the surrounding
areas is minimal
Compaction is difficult
when the fines content is
high
This method is oftcn used
when compaction is
difficult. If tines content of
the subsoil is high and
permeability is low,
application of this
method is difficult
The pile prevents uplift
and seulement of
upper-structures.
Liquetaction remediation
is possible even when
there is a possibility of
liquefaction in the soil
under the existing
structure, such as for an
embankment, by installing
an improvement wall
at the periphery
10
o
3
o
0-
~ .
...
o'
;:)
c
......
l'
.E'
c:
o
::n
-
-o
C/)
9,
-
CIl
I..,)
......)
......)
--.-------------------------------_.
. - ~ - - - - - - - - - - - - - - - _ . _ - - - - - - - - - - ~ - - - - - - - - - " . _ - - - - - - - - - - . _ - - - - - -
, ("-
~ ...~
t
.
c
..
...
~
~
(
Table T6. J. Continued.
Principle Methods Depth Summary Influence on the Remarks
Designation surrounding areas
Premix - Landfill is deposited by Water quality control is This method can reduce
method adding and mixing a necessary since-sand the construction period
stabilizing material such mixed with a stabilizing because liquefaction
as cement in advance in the material is dropped into remediation can be
soil. Liquefaction water accomplished
remediation following the simultaneously with
landfill becomes unnecessary landfllling, and treatment
can be perfonned.in a
large volume
Replacement Replacement About The liquefiable soil is Handling of the excavated The construction is
method GL-5 m replaced with a non- soil is necessary reliable since the
liquefiable material liquefiable soil is replaced
with non-liquefiable
material
Lowering of Decp well W.ller Thc perimeter of thc Caution should be exercised This method can be
ground water method level liquefiable soil area is with regard to consolidation applied to.the subsoiIs
level lowering surrounded with a cut and settlement of the soil helow existing structures.
of ahollt off retaining wall and Influence on the The long-term operuting
J5-20 III constructed with sheet piles, surrounding area from the cost should be considered
and the ground water level lowering of the ground
is lowered using deep wells water level
Shear Underground - Liquefaction remediation Influence on the surrounding This method can also
strain wall can be achieved by areas is minimal prevent the propagation of
restraint method surrounding the liquefiable excess pore water pressure .
soil using underground from the surrounding
walls made of grouting or liquefied soil
sheet piles, and restraining
the shear deformation of
subsoil during earthquakes
Preload Soil overlay Depending Preload is applied by Intluence on the surrounding The stress propagation in
on overlaying soils in areas is minimal the soil should be
embank- order to overconsolidate considered since there is
ment width the subsoil load dispersion
Deep well Water The ground'water level of Caution should be exercised This method can be
method level liquefiable soil is lowered with regard to consolidation applied to the subsoils
lowering temporarily and thereafter and settlement of the soil below existing structures
of about restored and influence on the
15-20 m surrounding area from the
lowering of the ground water
~
~
Icvel 3
~
Structural Pile The strength of the pile Influence on the
e:
- -
~
...
strengthening foundation increases resistance even surrounding areas is minimal
.......
0
::;)
if liquefaction occurs 0
.....
Sheet pile Shear deformation after There is noise and vibration This method can also
l'
-
.......
,.D
liquefaction is restrained during sheet pile installation prevent propagation of
c
~
by surrounding the soil excess pore water
::n
p:l.
with sheet piles pressure from the
sr
-
(lI
surrounding liquefied soil C/l .
o.
Top-shaped Used for spread foundation Influence on the surrounding
....... .
-
This method is applied 10 -
VI
foundurion hy laying top-shaped nreas is minimal sm4l1l huildings such us
Iv
concrete blocks on the soil residential housing
--.J
----
1.0
!
\
t
I
I
I
. I
280 PlANe
effectiveness of compaction depends on the soil gradation; when the
fines content of the soil is large, compaction is difficult. Compaction is
used extensively, and there are many case histories. Examples of compaction
methods include sand compaction piles (Mizuno et al., 1987), vibrating rod
(Sato et a1., 1988), vibroflotation (Shimizu and Takano, 1979) and dynamic
compaction (Tanaka and Sasaki, 1989). These methods usually cause vibration
and noise during the compaction process and 'result in increased lateral earth
pressures; hence the influence of such methods on the surrounding, ground
and adjacent structures needs to be carefully considered. Recent developments,
of the vibroflotation method are aimed at reducing these environmental
impacts.
(2) Pore water pressure dissipation
A second approach is to control the development of excess pore water pressure in
sandy soils by increasing the rate of dissipation through the installation of
permeable stone or gravel columns or drains. The gravel drain and plastic drain
methods (Iai and Koizumi, 1986) are typical examples. The installation of drains
usually causes less vibration and noise than the compaction method and hence
has the advantage that the influence on the surrounding area is reduced. This
approach requires complex calculations and the measurement of many parameters
to establish the design pile installation spacing. In some cases, for example in
soils with low permeability and a high fines content, the drainage method may be
impractical.
The installation of drains does not generally densify the natural ground and the
method relies on drainage of excess pore water pressures to reduce the likelihood
of liquefaction. In the event that the excess pore water pressures generated are
greater than anticipated and approach 100% of the initial vertical effective stress,
then Yoshimi and Tokimatsu (1991) have warned that large deformations may
suddenly be induced. and suggested that a site improved in this way may lack
'ductility'. Drainage solutions are therefore often used when the application of
other methods (such as compaction) is considered too difficult..
Variations of the drainage method have also been developed. Noda et al.
(1988). for example, proposed a water drainage device attached to a pile that com-
bines the benefit of increased pore water pressure dissipation in the ground with
the pile itself.
(3) Cementation and solidification
Liquefaction remedial measures based on solidification include the deep mix
method (Suzuki et aI., 1989) and premix method (Zen, 1990). The deep mix
method improves the ground by mixing a stabilizing material into sandy soil to
cement the soil skeleton. The method is versatile and can be used in different
ways. such as directly below existing structures. or to create a continuous under-
ground 'wall' around a structure or site, see (6) and (8) below.
The premix method works in a similar manner to deep mix methods by mixing
stabilizing materials with natural soil. Premix is particularly relevant to the
improvement of soil where substantial landfill is taking place. such as filling to
create reclaimed land. The stabilizing material is added to the landfill soil or sand
in advance and then the treated soil is placed underwater using a tremie chute or
bottom dump barge. The use of pre-mixing can reduce construction periods by
removing the need to treat the ground after fill placement.
Aspects that require particular attention under this approach are quality control
.of the treated subsoil profile, mix controlof the stabilizing material, and control
of water quality in the surrounding area.
(4) Replacement (Japanese Society of Soil Mechanics and Foundation
Engineering. 1982)
The replacement method replaces the original-liquefiable soil with non-liquefiable
material. This may comprise gravel, or soil mixed with cement. Consideration
must be given to the sourcing of suitable replacement material and to the disposal
of the excavated soil (spoil). The stability of the excavation itself requires careful
assessment.
(5) Lowering of ground water level (Omori, 1988)
Lowering the ground water level within a soil profile provides two benefits.
Firstly, the effective confining stress is increased at depths below the original
ground water table, and secondly, the upper soil layers can be effectively
de-watered.
The ground water level can be reduced by active or passive means. In most
circumstances, an active system (such as pumping from wells) is necessary to
maintain a lowered ground water level at or below the design requirement. Active
systems require to be operational at all times to provide security. In some
circumstances, passive drainage can be used. Both approaches require mainte-
nance and will therefore impose operating costs; these need to be assessed.
(6) Shear strain restraint (Yamanouchi et al., 1990)
A continuous underground wall can be installed around a liquefiable site with the
intent of reducing soil shear strain levels to acceptable levels under earthquake
loading. The method has advantages where access to the liquefiable soils or foun-
dation is difficult, such as under an existing building or structure.. The principal
disadvantage of this approach is the complexity of the analysis and design process
and justification of the proposed solution.
(7) Preload (Matsuoka, 1985)
Preload methods overconsolidate the soil, thus increasing the liquefaction resist-
ance. This is typically achieved using a surcharge on the ground surface, but can
also be achieved by temporarily lowering the ground water level (e.g. beneath
l
.... ~ ~ . ~
l :."
~ . '.......
I
I
1
1
I
282 PlANe
existing structures). The method is applicable to soil with a large fines content.
Detailed evaluation of the underlying soil strata is necessary, and it may not be
practical to achieve large overconsolidation ratios where the existing vertical load
is large. Settlement of existing structures needs to be evaluated. Case histories of
this approach. are ~ i m i t e d and design procedures are not yet fully developed.
(8) Structural remedial measures
... .
An alternative to remediating or constraining the liquefiable soil deposits is to
strengthen the foundation system. Piles or sheet piles can be used to found the
structure on denser deposits at depth, reducing the consequences for the structure
of liquefaction in the soil beneath (Nasu, 1984). Installation of piles can also
densify looser soils through increased lateral confinement, giving a further
benefit. The design of the strengthened foundation system needs to take into
account the changing dynamic response of the structure as liquefaction takes
place and the consequences of lateral or downslope ground movements and settle-
ments. The effects of liquefaction on services need to be assessed.
The methods outlined above at (l )-(8) and summarized in Table 6.1 provide a
general overview of remediation solutions. In the following sections, four of these
approaches are discussed in detail: compaction, drainage, premix and
preload.
T6.3. COMPACTION METHOD: DESIGN AND INSTALLATION
Compaction of ground or fill as a remedial measure against liquefaction can be
achieved by a range of techniques, such as vibrating a steel rod pushed into the
ground or by forcing sands or gravel into the liquefiable subsoils using a steel
casinz.
....
T6.3.l Determination of the SPT N-value to be achieved by compaction
The SPT N-value of the subsoils to be achieved following compaction can be
determined using the flow chart shown in Fig. T6.3. This illustrates how the effect
of compaction affects the earthquake response characteristics of the- subsoil
profile.
The minimum area of soil improvement necessary for a specific structure. may
often be limited to the area under the structure itself (Section T6.7). In such cases,
because of the proximity of the unimproved soil to the foundation, there may be
some interaction between the liquefied and non-liquefied areas, leading to further
degradation of the foundation. In these cases, it is particularly important that the
degree of compaction of the improved soil is adequate and that liquefaction of the
A rough estimate of the equivalent N-value to be
achieved by compaction is detennined for each soil
layer by using Fig. T4.33 in TC4 from the equiva-
lent acceleration obtained by response analysis of
the subsoil before compaction
The SPT N-value after compaction is obtained for
each soil layer by using Fig. T4.34 in TC4 from the
equivalent SPT Nvalue and the effective vertical
pressure
Response analysis of the subsoil after compaction
is accomplished using the SPT N-value obtained
for computing equivalent acceleration .
End of calculation for SPT N-value after compaction
The equivalent SPT N-value to be achieved
by compaction is determined from the newly
obtained equivalent acceleration for each soil
layer by using Fig. T4.33 in TC4
No
Fig. T6.3. Determination procedure of SPT N-value to be achieved by compaction.
foundation is fully prevented. The equivalent N-value to be achieved in order to
satisfy this condition is 16 or higher, which corresponds to a relative density of
80% or higher at an effective vertical pressure at a depth less than about 20 m.
Such a target value for the SPT blowcount is near the upper limit of what may be
achieved by the compaction method.
In other situations, an impermeable underground wall may be constructed
around the perimeter of a structure, greatly reducing the possibility of liquefaction
in the improved soil in the interior. In this case, where liquefaction of the outly-
ing areas will have only minimal impact on the improved soil underneath, then the
equivalent N-value to be achieved by compaction may be determined following
Section T4.4.2 (Technical Commentary 4). In particular, the target blowcount can
be determined from the boundary of areas IT and III in Fig. T4.34. For important
structures requiring a considerable rnargin of safety, the equivalent N - v ~ l u e tope
achieved by compaction is determined from the boundary of zones III and IV in
Fig. T4.34.
There may be cases where the SPT N-value assessed following the guidelines
above is unlikely to be achievable using the compaction method. One approach
is to make a more detailed assessment of the liquefaction resistance of the soil by
I
\"., : .... ,
t ~ "
,... ......
\
I
\
I
1
1
I
\
!
I
\
j
I
I
\
,
\
284 PlANe
cyclic triaxial testing, Section T4.4.2 (Technical Commentary 4), with a view to
developing a revised SPT N-value. Alternatively, the use of other remediation
methods such as preload, or cementation and solidification methods may be
considered.
T6.3.2 Designing compaction
The degree of compaction attainable by compaction methods is affected by many
factors, including those listed below (Mitchell, 1981).
i) Characteristics of the soil (grain size distribution and fines content, i.e. the
percentage by weight of grains smaller than 0.075 mm)
ii) Degree of saturation and level of the ground water table
iii) Relative density of the soil before compaction'
iv) Initial stresses in the soil layer before compaction (effective vertical stress)
v) Skeleton structure and cementation of soil before compaction
vi) Distance from a source of vibration during compaction
vii) Properties of sand fill
viii) Characteristics of compaction .methods (types of installation equipment,
vibration capacity of the equipment, installation method, expertise of the
operating engineers)
Because of the range of parameters affecting the degree of compaction that can
be achieved, the design of a compaction method should be based on case histories
and recent field testing wherever possible. Field tests should be performed at a test
site where the characteristics of the soil are demonstrably similar, using compa-
rable compaction plant and techniques. Field tests of the proposed compaction
method considerably enhance the reliability of the design.
T6.4 DRAINAGE METHOD: DESIGN AND INSTALLATION
The objective of the drainage method is to' enhance the overall drainage capacity.
of the subsoil profile by installing drains made of very permeable materials below
the ground surface. An individual drain is often similar to a pile in shape. formed
using gravel and crushed stone and installed in a distributed layout similar to a
pile configuration (Seed and Booker, 1977). Other types of drains. however, have
been proposed such as drains shaped like a wall, drains surrounding a structure
(Sakai and Taniguchi, 1982) and drainage devices that are attached to the back of
steel piles or sheet piles (Noda et al., 1990). Rubble or gravel, used as backfill
behind quay walls may also be considered as providing drainage if a permeable
sheet is used to prevent the migration of sand. In recent years, pipes made of
synthetic resin (i.e. plastic) with many small holes or perforations have also been
used.
{
Pil e drain .
Wall drain
Drain ..
Drainage zone surrounding structures
Attachment of drain function to structures
(a)
{
Gravel drain
Drain
Plastic drain
(b)
{
Method relying primarily on the drainage effect of the drain
Drain
Method which simultaneously achieves an increase
'ln soil density with installation of the drain
(e)
Fig. T6.4. Types of drainage methods used as liquefaction remedial measures.
(a) Types of drainage methods according to shape.
(b) Types of drainage methods according to materials used.
(c) Types of drainage methods according to installation methods.
The remainder of this section focuses on the design and installation of a pile
drain using gravel and crushed stone. ..
There are a number of variations of the method of forming pile shaped drains.
One variant is the gravel compaction method (Kishida, 1983) which combines
drainage and compaction. In recent years, another method has been developed
which increases soil density by pushing gravel into the soil with a steel rod. Care
must be taken with these methods to avoid liquefaction of the surrounding soils
and clogging of the drain due to strong vibrations during installation. Various
drainage methods are summarized in Fig. T6.4.
Vibroflotation also uses gravel and crushed stone as fill material to form stone
columns. This is similar in concept to the gravel column used in the drainage
method. With vibroflotation, however, there is a high possibility of sand migration
into the stone columns during installation. Vibroflotation, therefore, is not used as
a method to improve drainage but only to compact the surrounding ground.
In the drainage method, slight settlement accompanying water drainage and a
slight increase in excess pore water pressures are inevitable. Compared to the
compaction method, however, the drainage method is a relatively recent develop-
ment; there are few documented case histories of their performance in earth-
quakes and design approaches are still being debated.
Yoshimi and Tokimatsu (1991) have also questioned the deformation charac-
teristics of loose sandy soil improved by the drainage method. They suggest that
286 PlANe
Site
Soil surveyl I I Choosing drain 1
response
soil test
material
analysis }----- --I
t
IDrain material I
I
Iliquefaction I
II Permeability I I
chaeteristics
i
I
IPermeability of I
,
Compressibility I
I
the drain .
!
I
!
(3)
I
i
I
Time to attain
I'
f\-
Radius of
-1
liquefaction
the drain
I
without drainage
,
I
Depth
I
+
r- (length) of I
the drain I
ITime factor
I
IWell resistance I
I
Allowable excess
i
I
porewater pressure
I I
l
I
ratio
I
I
IDrain installation I I
spacing I
the 'ductility' of the ground "may be limited, implying that as excess pore water
pressures increase under stronger ground shaking, there may be a sudden sharp
increase in the level of deformation in the soil. A slight change in the level of
shaking caused by earthquakes may lead. to a significantly increased level of
deformation. . . "
For new constructions where there are minimal restrictions on the choice of
method, or for the remediation of existing structures for which a large safety
factor is considered appropriate, other approaches such as replacement with
gravel, or stone backfill are considered preferable.
Pile drains are designed using the procedure shown in Fig. T6.5. As shown in
this figure, the sequence of activities to be followed is (1) decide on the drainage
material, (2) conduct soil surveys and in-situ tests to establish the natural subsoil
profile, and finally (3) establish the installation spacing..
The recommended design approach is based on establishing a balance between
the rate of excess pore water pressure increase due to earthquake shaking and the
rate of dissipation of pore water pressure due to drainage. In soils of high perme-
ability, drainage of excess pore water pressure is coupled with the generation of
excess pore water pressure. The rate of increase of excess pore water pressure is
increased by high amplitudes of earthquake motion, sudden and intense shaking,
i
j
I
I
I
i
i
I,
i
I
I
:,
I
j
1
I
1
I
!
I
I
r
i
I
1
I
l
(2)
Fig. T6.5. Design procedure for drainage method.
(1)
.Remediation of Liquefiable Soils 287
and in loose saturated soils. The rate of dissipation of pore water pressure
increases if the coefficient of volume compressibility of the natural soil is small
and the coefficient of permeability is high. The object of the drainage method is
to increase the natural rate of dissipation by installing closely spaced drains with
drainage material of higher permeability than the surrounding ground.
Conditions under which the use of drains is likely to be less effective may be
summarised as follows:
a) low natural permeability of the subsoil profile;
b) high coefficient of volume compressibility of the natural soil;
c) low liquefaction strength (cyclic resistance) of the natural soil;
. d) high amplitude earthquake motion;
e) sharp or impact-type earthquake motions;
f) deep liquefiable soil layers.
T6.5 PREMIX METHOD: DESIGN AND INSTALLATION
The premix method uses a small amount of stabilizing material and a chemical
additive, mixed with landfill, backfill, or fill inside cellular structures to create a
. new fill material that is less susceptible to the generation of excess pore water
pressure and hence more resistant to liquefaction. After processing, the new fill is
transported and placed at the designated location to create a stabilized reclaimed
land area (Zen, 1990). .
Unlike other liquefaction remediation measures, which are only relevant to
existing subsoil profiles, the premix method provides a remediation option for
new landfill, where the proposed filling method is anticipated to result in material
which would otherwise be susceptible to liquefaction.
T6.5.1 Principle and application of the method
The premix method improves the soil matrix by partially cementing the points of
contact of the individual soil particles. This effectively reduces the capacity of the
soil to generate excess pore water pressure by increasing its stiffness and simul-
taneously provides some cohesion or cementation of the landfill, such that it can
sustain higher levels of excess pore water pressure without exhibiting large
deformations.
The method is applicable to construction sites, where it is being proposed to use
. .
imported sandy fill material. Examples of applications are shown in Fig. T6.6,
which includes hypothetical construction cases. Careful consideration must be
given to the specific conditions at each site or project since the installation method
and characteristics of the treated sand often differs widely. It is recommended that
','
"
,,00 YIANC
Original seabed/ground surface
(a)
~ I
~ ~
Utility duct
/
Original seabed/ground'
(b)
Original seabed/ground
(c)
Fig. T6.6. Application of premix method. '
(a) Sheet pile.
(b) Cell structure.
(c) Utility duct, reclaimed land.
(d) Caisson.
Original seabed/ground surface
(d)
the premix method should be used only after performing laboratory and/or field
tests to confirm the suitability of the approach..
There are a number of advantages to the premix method:
1 remedial treatment after the landfill is completed become unnecessary, and
therefore time on site can be shortened and site costs may be reduced,
2 the strength of the treated soil can be reasonably predicted,
3 dredged sand can be exploited as fill material.
4 large, deep-water installations can be completed readily (using a conveyor,
chute or bottom dump barge),
5 commonly available plant can be used throughout the process,
6 noise and vibration during fill placement are greatly reduced compared to the
conventional vibro-rod and sand compaction methods, '
7 the effects on adjacent structures caused by the placement and compaction are
reduced compared to other methods,
8 lateral earth pressures on adjacent structures in the long term are also likely
to be reduced.
There are, however, four disadvantages:
1 there may be substantial variation in the strength of the treated subsoil pro-
files, depending on the type of sand and placement procedure,
") the stabilizing material or chemical additives can be costly,
3 the environmental impact on water quality may need to be considered..
4 the method is limited to sites where the subsoil is to be excavated or replaced
or to newly reclaimed land.
T6.5.2 Design method
Figure T6.7 outlines the design procedure. The premix method differs from other
remedial treatments against liquefaction in that the treatment of the fill Occurs
prior to placement, rather than afterwards. Quality control procedures can be used
to test the fill prior to placement. The method can then reduce the length of time
required on site for reclamation projects, resulting in ground with the required
strength being achieved soon after the completion of filling. Determining the
required strength of the treated fill is the first consideration in the design process,
Presurvey I testing of untreated
sand and treated sand
Determination of load and
external force .
Determination of angle of internal
friction the treated sand
Estimate of coheslon (C;. i=n)
of the treated sand
Estimate of the improved area
(s,. i=n) in the treated subsoil
Evaluation of liquefaction
potential and stability
slope stability
bearing capacity
earth pressure
settlement
earthquake resistance
liquefaction potential
No
Yes
Evaluation of the cost (19 by
tentative determination of the
cohesion of the treated sand and
improvement area (K;:c;. Sj' i=n)
i=n+1 No
Yes .
Determination of cohesion of the
treated sand and improved area
Fig. T6.7. Design procedure of premix
method.
"' "" : .... ,
'- -.... ~ , '
I
I
i
i
I
I
I
I
\
\
I
I
i
I
\
I
\
i
!
!
I
I
I
i
290 PlANe
and this will depend on the type of structure and facilities. ~ o be constructed.
Laboratory and/or field tests may be necessary to confirm design assumptions
concerning the strength of the fill that can be achieved using the proposed
additives. Lateral earth pressures on adjacent structures, bearing capacity and
liquefaction susceptibility of the. treated fill must be determined. The required
quantity of cement or chemical additive can then be calculated and the installation
cost estimated. However, soil strength and the area and volume of fill to be treated
will vary depending on the proportion of additive used. It is recommended, that
the effects on cost and construction programme of using a range of different
percentages of additive are considered before a final design is reached.
T6.6 PI.lliLOAD METHOD: DESIGN AND USE
The preload method for liquefaction remediation is based on laboratory findings
that overconsolidation of sand increases liquefaction resistance (Yamazaki et al.,
1992; Ishihara and Takatsu, 1979.;. Matsuoka, 1985). In the field, overconsolida-
tion can be achieved either by placing a surcharge or by lowering the ground water
level by pumping.
The preload method is recommended for liquefaction remediation in the
following four situations:
1 when the subsoil profile consists of sandy soil with a high fines content and
it is difficult to apply compaction and/or drainage methods,
2 when the subsoil profile consists of combined layers of sand and clay and it
is necessary to improve the clay layer,
3 when it is necessary to lower the ground water level to facilitate excavation,
and
4 when it is required to improve the ground under an existing structure.
As with the premix method, it is recommended that in-situ field tests be used
to verify the preload method in advance of construction.
\ T6.6.1 Liquefaction characteristics of overconsolidated soil
(I) Increase in liquefaction resistance by overconsolidation (Yamazaki et al..
1992)
The liquefaction resistance of sandy soil increases with overconsolidation.
Figure T6.8 shows cyclic triaxial test results for Niigata Eastern Harbour sand.
The overconsolidation ratio (OCR) is defined by the ratio of ( j ~ over o; where ( J ~
is the maximum effective confining pressure that the sample has been subjected
to previously and (Je' is the current effective confining pressure as shown in
Eqn. (T6.1).
(T6.1)
,
OCR = O'p
,
O ' ~
In laboratory element testing, the cyclic stress ratio (CSR or liquefaction
strength). r./O'; is defined as the stress ratio to cause liquefaction (+/- 5% strain)
in 20 load cvcles. Figure T6.8 shows how the CSR varies as a function of number
. ....
of load cycles. The overconsolidation ratio also has an int1uence on the CSR, as
shown in Fig. T6.9. Ishihara and Takatsu (1979) proposed that the ratio of the
cyclic stress ratio to reach liquefaction of overconsolidated sands compared to
normally consolidated sands was in proportion to the square root of the overcon-
solidation ratio
(T6.2)
Fig. T6.8. Cyclic shear stress ratio
and number of load
cycles.
0.5
Niigata eastern harbour sand
0.4
(medium density)
u
b
<,
t:;.
..:-
0
0.3
:;:;
tU
....
OCR=2
CJ)
CJ)
0.2
(56%)
~
U5
0.1
0
1
10 100
Numberof cycles N.Q
0.5...----------------,
0
1
2 3
. Overconsolidation ratioOCR
Fig. T6.9. Overconsolidation
ratio and liquefac-
tion strength.
o
O:Dr=80%
6.:Dr=60%
O:Dr=50%
( rJ / ac')oc=jOCR( r, / ac')NC
Niigata eastern harbour sand ac'= 98kPa
g 0.4
tUo
(j)N
'(j) \I
~ z ' " 0.3
.--.
c' o
.Q b
-,
~ ...... 0.2
Q)-
::J
0"
::J 0.1
292 PlANe
(rl a;)NC: Liquefaction strength for normally consolidated sands
(r/a;)oC= Liquefaction strength for overconsolidated sands
This function is represented by the solid line in Fig. T6.9. For an OCR = 2, for
example, the liquefaction strength is increased by 40%; if the OCR = 4, then the
liquefaction strength is doubled.
(2) Consolidation or stress history
The influence of consolidation history (cycles of pre-consolidation) in compari-
son to the overconsolidation effect noted above appears to be minimal. The 0
mark in Fig. T6.8 indicates the result of an experiment when the consolidation
history (a; =100- 200 MPa, OCR =2) was repeated 5 times. It is apparent by
comparing this result with the data point for a single pre-consolidation cycle that
the repetition of cycles has had lIttle effect.
(3) Influence of fines content
The effect of consolidation history on the mechanical behaviour of soil is
generally. greater in clay than sand. If sandy soil contains a high fines content,
Fe' of clay and silt, then the overconsolidation effect may. be expected to be
higher. Figure T6.10 shows results from the work of Ishihara et al. (1976) on
the relationship between the fines content and the overconsolidation ratio
expressed in terms of the liquefaction strength (here defined as the cyclic
shear stress ratio to cause liquefaction in 30 cycles). It is apparent in.this figure
that the effect of overconsolidation on liquefaction strength increases with'
increasing fines content.
Fig. T6.1O. Fines content and liquefaction strength.
0.6
o
0.1
o 10 20 30 40 50 60 70 80 90 100
Finescontent Fe (%)
SAMPLE
OCR-
-
1.0 1.5 2.0
Takasagosand 0
G
Koto sandA 0
g
.. Koto sandB 0
~
'OCR : overconsolidation ratio
I-
:::--- :;::::::::::=-.-
:=: 0-. <>0 Q
0
c-
-
N,=30
e =0.7-0.9
. .
0.5
Q)
o
c:
go
. ~ C")
(/) II 04
Q) ""
~ z
c:-=---
g b
U
0.3
0,
ctS ""
Q) I-> 02
::)..... .
tT
:.:J
;
l'
: J .
1
\
I
L- ~ __
T 6 ~ 6 . 2 Design and installation procedure
A flowchart illustrating design and installation for the pre-load method is shown
in Fig. T6.11. The procedure is summarised as follows:
( I) Estimate the overconsolidation ratio (OCR) of the natural ground
(a) Undisturbed samples are obtained from the field before improvement. Cyclic
triaxial tests are performed on normally consolidated specimens to obtain the
liquefaction strength of the insitu soils.
(b) Using the cyclic triaxial test results, the liquefaction safety factor F
L
is
obtained following the procedure .described in Section T4.4.2 (Technical
Commentary 4).
Presurvey
Obtain undisturbed sample
from the field
Perfonn cydic triaxial test
Calculate the FI. value based on
the cydic triaxial test data
Establish relationship between
OCR and liquefaction resistance
Select preloading method at the
site and setting of preload
Installation, control during installation. measurement of ground water level,
pore water pressure. earth pressure, etc.
Postsurvey (verification of improved result)
Obtain undisturbed sample of improved soil
Calculate the FI. value based on cyclic triaxial test
Fig. T6.11. Design/installation flow chart.
,(
'- ~ ... -
"
I' "
1
\
L
294 PlANe
(c) The natural deposits are susceptible to liquefaction under earthquake loading
for FL < 1. For such deposits, the liquefaction strength must be improved by
overconsolidation by a factor of at least l/F
L
to achieve a factor of safety
greater than 1 against liquefaction. .
(d) The relationship between the overconsolidation ratio and liquefaction
strength is then obtained from further cyclic triaxial tests on overconsolidated
specimens of the field samples.
(e) The overconsolidation ratio for the subsoil profile with respect to the speci-
fied liquefaction strength can then be deduced.
(2)' Select method of preloading
Two methods are generally adopted to achieve overconsolidation of a subsoil pro-
file; firstly, the application of a surcharge by embankment load, and secondly, the
temporary lowering of the ground water level using deep wells (Japanese Society
of Soil Mechanics and Foundation Engineering, 1988).
For the surcharge approach, the vertical effective stress increment within the
soil is calculated from the prescribed overconsolidation ratio and then the height
and area of the embankment are determined. Stability of the embankment and
foundation against slip failure should be checked.
Lowering of the ground water level is appropriate for larger vertical effective
stress increments. The permeability of the soil must be determined and a seepage
flow analysis carried out in advance.
The choice of preload method may depend on other construction aspects of the
project. For example, if sand drains are to be installed to assist drainage from
in situ clay layers, then the surcharge method may be more appropriate. If the
construction works involve dewatering to facilitate excavations, then the dewater-
ing system may also enable overconsolidation of the liquefaction susceptible sand
layers to be achieved in parallel.
T6.7 DESIGN OF LIQUEFACTION REMEDIATION
The design requirements for a specific soil improvement area will be based
on an assessment of the mechanism of damage likely to be induced in the
natural soils by liquefaction. In particular. it is critical to determine whether
the damage will be caused primarily by reduction in the shear resistance of the
liquefied soil, or by additional external forces due to excessively large dis-
placements. Large displacements of the liquefied subsoil may affect the found-
ations of existing and new structures, or cause serious damage to lifeline facilities
(such ~ s gas, water and telecommunication pipelines and conduits), and special
consideration will be necessary in respect of these cases. There are many other
cases. however, in which remediation measures against loss of shear resistance
will provide benefits against the risk of large displacements; ground compaction
being one example.
There is a relatively long experience of the use ofremediation measures against
loss of shear resistance compared to experience of measures to resist excessivelv
. ~
large displacements, and techniques to achieve this are still being studied and
developed. For this reason, this section focuses on the design of a soil improve-
ment area to resist loss of shear resistance.
Generally. even if the soil is predicted to undergo liquefaction over a wide area,
it may be possible to limit the area requiring soil improvement to the area that
controls the stability of the structure. For example, the zone within the subsoil that
contributes predominantly to the stability of spread foundation structures is the
part directly below and immediately around the structure; ground far away from .
the structure does not contribute to the same extent. A key issue is therefore to
establish how far the soil improvement needs to extend laterally from the struc-
ture. This can be determined by following the procedure summarised below (Iai
et aI., 1987; lai et al., 1991).
Secondly. it is necessary to determine the extent of the zone of soil improve-
ment in the vertical direction. Standard practice is to improve the soil down to the
deepest part of the liquefiable soil layer. If soil improvement is performed only
within the top section of a soil layer that then experiences liquefaction below this
elevation, the upward seepageflow of pore water generated may still result in loss
of strength and stiffness and even liquefaction in the upper, improved sections.
The design procedure for a soil improvement area presented below is based
around use of the compaction method. The procedure may, however, be adapted
for other methods with appropriate modification by considering the appropriate
characteristics of the improved soil (such as permeability and cyclic strength!
deformation). (In the following discussion, 'liquefaction' is defined as that state
in which the excess pore water pressure ratio of loose and medium sand reaches
1.0, i.e, when the pore water pressure u equals the vertical effective stress a;.)
(1) Propagation of excess pore water pressure into the improved zone
Outside the zone of improved soil, excess pore water pressures will exceed those
inside, resulting in a hydraulic gradient driving fluid into the zone of improved
soil. This is a complex issue, as the deformation characteristics of dense saturated
sand are highly non-linear. As a simplified design procedure, this phenomenon
may be addressed as follows:
For excess pore water pressure ratios uta; < 0.5, the effect of the excess pore
water pressure increase may be ignored: because laboratory test data indicate a
very small strain generation below this level. For uta; > 0.5, however, it is
necessary to take into account the effect of excess pore water pressure increase.
Shaking table tests and seepage flow analyses suggests that the pore water
pressure ratio is uta; > 0.5 within an area defined by the square ABeD in Fig.
T6.12. In this area, the soil shear resistance must be reduced for the purposes of
j
I
1
296 PlANe
the design. The tests also indicate that an area defined by the triangle ACD exhib-
ited unstable characteristics. This area should therefore be assumed in the design
to be liquefiable and treated accordingly.
The exception to this recommendation is when a drain or impermeable sheet or
zone has been installed at the perimeter of the improved area in order to shut out
the inflow of pore water into the improved area. Under these conditions, the area
corresponding to the square ABCD need not be included' as part of the improve-
ment plan.
I
I
I
1
i
I
,I
I
!
I
,
(2) Pressure applied by the liquefied sand layer. .
At the boundary between liquefied andnon-liquefied ground, there is a dynamic
force as indicated in Fig. T6.13, and a static pressure corresponding to an earth
pressure coefficient K = 1.0 which acts on the improved ground due to the lique-
faction of the surrounding soils. These forces may greatly exceed the forces
acting in the opposite direction from the non-liquefied ground. For a retaining
structure backfilled with soil and subject to active static and dynamic lateral earth
pressures at EF as shown in Fig. T6.14, the area of soil improvement must be large
enough that there is no influence. of liquefaction in the active failure zone. To
accommodate the net outward force, it is essential to check that sufficient shear
resistance can be mobilised along the passive failure surface GC.
(3) Loss of shear strength in liquefied sand layer
In the simplified design procedure, unimproved soil ..of loose or medium relative
density should be considered to have negligible shear strength after liquefaction
(i.e. the soil is treated as a heavy fluid). Since the shear strength of the improved
ground in triangle ACD also cannot be relied upon (see (1) above). then the
improvement area should be wide enough to obtain sufficient bearing capacity
from the shear resistance alone the solid lines EFG and HI in Fia. T6.15.
~ -
In practice, lateral pressure from surrounding liquefied sand layers may con-
tribute to the stability of certain structures. Figure T6.15 shows how the dynamic
earth pressure may be subtracted from the enhanced static lateral eanh pressure
Liquefaction
C Nonliquefiable Fig. T6.12. Area of softening due to
impermeable layer seepage flow.
I Unimproved soil _. .
o
B
A
Improved soil
//
.:::: ".. \ " : : ' : \ ~ : ? : ? )
.. ,
:
j
!
1
i
i !
!
; 1
j
: 1
Improved soil
Remediation ofLiquefiable Soils 297-
Unimproved soil
.0. : :.:".
0" . e."0":: "0.:
0 0
H
r bH (static pressure)
y
.-rl--...... 7 a G'i
H
8" 9 ~ a l . . , y n
(Dynamic pressure)
o
Non-liquefiable impermeable layer
a
max
=Maximum acceleration
g =Acceleration of gravity (980 gal)
r sat= Unit weight of saturated sand
r b =Buoyant unit weight of saturated sand
Fig. T6.13. Pressures applied at the boundary of improved soil.
Caisson
I'
E
F
Soil improvement area
B C
Pressure from liquefied layer
Fig. T6.14. Schematic diagram for investigation of stability with respect to pressures
applied from the liquefied sand layer.
(based on an earth pressure coefficient K = 1.0) in certain stability calculations.
The section o n ~ which the pressure from the liquefied sand layer is applied, can be
assumed to be along the lines GG' or II'. Although this figure shows only five
examples of foundation types, the same principles can be applied to other design
arrangements.
(4) Change in dynamic response of the improved soil
The dynamic response of the improved (compacted) soil area will be altered by
the influence of liquefaction in the adjacent unimproved soil and by the migration
298 PlANe
fa)
E
Dynamicpressure
(b)
C1-,..--+--1
Sialic pressure
Dynamicpressure
Ie)
,
of excess pore water pressures. This can result in a change in the inertia (body)
force applicable to the compacted zone. In order to assess the influence of this
change on the stability of the structure and the soil improvement zone. a separate
detailed evaluation may be necessary.
Schematic diagrams for investi-
gation of stability for determin-
ing the soil improvement area.
(a) Shallow foundation.
(b) Foundation with shallow
embedment.
(c) Foundation with deep
embedment.
(d) Pile foundation.
(e) Underground structure
(when passive resistance
is necessary in order to
maintain stability of the
structure).
Fig. T6.15.
A r D
It
IImprovement area I
~ . ~ /" /, ", ," I / / B
\ ~ )
ld)
I
I
!
, I
!
The nature of the evaluation depends on the extent of the area over which the
excess pore water pressure ratio exceeds 0.5 and the general subsoil profile..
If this area is small compared to the entire improved area, and. if the
surface profile of the entire improved area is relatively flat, then the change in the
dynamic response of the improved zone is likely to be small (see for example
Fig. T6.14). However, in extreme cases, where high excess pore water pressures
are predicted close to a structure or with sloping ground or excavations, a two-
dimensional earthquake response analysis may be required.
(5) Design of soil improvement area for port and harbour structures
Key parameters for any assessment of the area of soil improvement include the
angle of internal friction for sands and the anticipated intensity of earthquake
shaking. A stability analysis method will also need to be specified. Standard
design procedures adopt a simplified, pseudo-static approach to the prediction of
earthquake loads and response. Despite this, experience suggests that with appro-
priate care and selection of parameters, these methods generally provide satisfac-
tory design solutions.
For example, the seismic coefficient and stability analysis methods used for
port and harbour structures in Japan are described in the Technical Standard
for Port and Harbour Facilities and Commentaries (Ministry of Transport,
Japan, 1989). In this procedure, earth pressures caused by earthquakes loading
are computed by the Mononobe-Okabe theory (Mononobe, 1924; Okabe,
1924) and stability analysis is performed by Bishop's method (Bishop,
1955) using the specified angle of internal friction for sands and design seismic
coefficients.
The soil improvement area required for a typical gravity quay wall based on
this approach, is shown in Figure T6.14. For anchored sheet pile walls, the zone
of improvement in the vicinity of the anchor wall must be carefully considered.
For flexible anchor walls, soil improvement must extend to sufficient depth below
the tie rod level to ensure that the adequate capacity is maintained to resist the tie
rod force.
T6.8 INFLUENCE ON EXISTING STRUCTURES DURING SOIL
IMPROVEMENT
One further aspect of a remediation design is to consider in detail the influence of.
. the installation method on existing structures. Particular experience has been
gained in cases where the compaction method has been used to improve the
backfill sand behind flexible (sheet pile) quay walls. Several case histories are
discussed below..
300 PlANe
T6.8.1 Influence on bending stress in a sheet pile wall by compaction
Figure T6.16 illustrates the increase in bending stress in a steel pipe sheet pile
wall as a function of the proximity to the wall of compaction using the vibro-rod
method (Toyota et aI., 1988). In this example, the main wall comprised a steel
pipe sheet pile quay wall. (to a depth of - 10m), as shown in Fig. T6.17. Soil
improvement was carried out by compacting the ground using the vibro-rod
method over the area and depths indicated by the broken line in the figure (with a
maximum depth of -20 m). The soil profile is shown in Fig. T6.18 (Toyota et al.,
1988; Sato et aI., 1988). . ..
The bending stress in the wall shows an increase per rod installation, with the
distance from the wall being the distance measured from the driven position of the
vibro-rod. Clearly, the bending moments in the main wall will also be affected by
rod installation to either side along the quay, and these results are therefore rep-
resentative of the plane strain condition. The bending stress in the sheet pile wall
increases gradually with the number of rod installations.
The results reported by Toyota et a1. (1988), do not address the effect of com-
paction at distances closer than 5 m to the main wall. The Ministry of Transport,
Japan, First District Port and Harbour Construction Bureau (1984) has published
data for the increase in bending stress caused by vibro-rod and gravel drain instal-
lation at a distance of just 2 m from the main wall. These data show an increase
of 30 MPa in the bending stress in the steel sheet piles per rod installation using
the vibro-rod method and an increase of 25 MPa per gravel pile caused by the
Fig. T6.16. Increase in bending stress in a steel pipe pile wall during installation of the
rod compaction method,
Remediation ofLiquefiable Soils 301
20.0.
,
I I I I
I
,
I I I I

. .
I I I I

, ,
I I I I

. ,
-9.0
I I I I
-10.5

I I I
I
Steel pile
I I I
I
I I

I I
I
,
I
I I
9 -25.1
25.0
,:. ,..
Unit (rn)
Fig. T6.17. Location of instruments for monitoring the effect of compaction installation
for a quay wall.
installation of gravel drains. When compaction was carried out by the vibro-rod
method behind the wall following the installation of gravel drains, no significant
additional increase in bending stress was found. This study, therefore, suggests
that the combined use of gravel drains and compaction may be an effective solu-
tion for soil improvement close to existing structures.
Such case histories provide valuable guidance for remediation design.
However, the actual stresses and strains generated in an existing structure due to
the compaction effect will vary depending on a wide range of factors and it is,
therefore, recommended that a test installation is conducted in all cases to quan-
tify any potentially damaging effects from the proposed improvement method.
T6.8.2 Excess pore water pressure and vibration resulting from compaction
Figure T6.19 shows the excess pore water pressures and ground accelerations
measured during soil improvement compaction works at a number of ports and
harbours in Japan. Though the accelerations shown in the figure are quite large,
the frequency of vibration generated by compaction is relatively high, at around
10Hz; and direct comparison of these accelerations with levels of earthquake
shaking used in earthquake resistant design of' structures is inappropriate.
Nevertheless, the figure .shows a rapid or steep increase in the ground acceleration
at distances closer than 5 to 10m from the installation. Clearly, the effects of such
vibrations need to be considered carefully in the design of any installation close
to existing structures. .
I "
302 PlANe
Fraction
STP N-value of fines
content (%)
10 2030 405( 0 1020 3C
Fig. T6.18. Soil conditions at monitoring site.
Similarly, Fig. T6.20 shows the excess pore water pressure that may be gener-
ated due to compaction. These observed levels of excess pore water pressure
also suggest that appropriate care must be taken to assess the influence of the
installation process on any structures within a distance of around 5 to 10m. The
installation of drains may be used to control the effects of subsequent vibro-rod
compaction, but in this case, care must be taken to ensure that the drains do
not become clogged by fines driven by the high hydraulic gradients local to the
installation.
T6.8.3 Influence of earth pressure increased by compaction
Compaction of soil will enhance its resistance with respect to liquefaction,
because both the relative density of the soil and the horizontal effective stress (and
therefore the confining pressure) are increased. However, the increase in horizon-
tal stress or earth pressure may also have an effect on existing structures nearby.
such as a retaining wall or quay wall, potentially affecting the capacity of the
Of Liquefiable Soils 303
600 Remediation measure method : Vibro-rod compaction method
Average SPT
. N-value
10.8
10.8
2.75
6
9.0m
3.6m
t t.om
-3.5m
-9.1m
-6.5m
Depth of Improvement
monitoring depth
-4.1m 9.0m
Port A
Port B
Port C
PortA
Code Port
5 10 15
Distance from installation X (m)
).llj \
.;.:\'
.,
' ..
, . ', .. \ \
." ': \ 1\
'
, .
",- ... -.
, ,. ..
r:-:.: :'.:' .
-.
If
m
nn
_
:" ;-...
..
...: :...::':..:..:.:.. -. : .....-:.::- : . :-"':""'*
- - --.:... - --.:-.& ....
0"--------'-- 1...- --'
o
500
400
300
200
100
Fig. T6.19, Attenuation of acceleration.
.......... -.
- Vibro rod compaction methOd port A GL -4.1m
- - - Vibro rod compaction method port A GL -9.1m
- - Vibro rod compaction methOd port B GL -3.5m
---- Gravel compaction methOd port B GL -3.5m
5 10 15
Distance from installation X (m)
\
! I \
,. \
\ \ \
,: \
\ \ \
: \
\ \ \
, \
U \
,: \
\\ \.
.. "
\'
'\
'----
---.._---
o'--------L- ...;;;;:::.:..c:=-- -.L.
o
30
20
10
Fig. T6.20. Attenuation of excess pore water pressure.
structure under earthquake loading-, It may not be clear whether this effect is
positive or negative and therefore, in the remediation design, the consequences
of any' increase in lateral earth. pressure on adjacent structures must be carefully
assessed,
"I .
.....'.
304 PlANe
Kameoka and lai (1993) report an analytical study carried out to quantify this
effect for a selected configuration of a quay wall. Forced displacements corre-
sponding .to the deformation of an initially loose soil (relative density 35%)
caused by compaction were generated within a zone behind a wall as shown in
Fig. T6.2.1. This forced displacement resulted in increased earth pressures on the
sheet pile wall as shown in Fig. T6.22(the open circles in the figure indicate the
initial earth pressure prior to compaction). An earth pressure ratio was defined to
indicate the ratio of the earth pressure induced by compaction over the earth
pressure without compaction. The effect of the increase in relative density of the
soil on the earth pressure was ignored. --
Figure T6.23 shows the results in terms of bending moment and displacement
from several analyses of a quay wall with various initial earth pressures. As the
initial horizontal earth pressure is increased (by simulating the process of com-
paction of the fill behind), the initial (pre-earthquake) bending moment in the wall
also increases, as expected. The earth pressure after the earthquake, however,
decreased. A similar tendency was noted in the displacements of the sheet pile;
here it was found that the outward displacements caused by the earthquake
-10m
-23m
o
E -5
-
.J::
0..
co
0-10
Anchor
Fig. T6.21. Cross section of quay
walls for the analyses.
Earth pressure
ratio R*
T 0 1.0
2.3
o 3.3
4.1
Do 4.8
-20
-60 -40 -20 0 20 40 60 80.,
Earth pressure (kPa)
Fig. T6.22. Earth pressure distribution after induced displacement.
Remediauonof Liquefiable Soils 305
6
4000 r-------r======:::::;,
Z 3500
3000
2500
2000
1500
.s 1000
"'0
500
co
o
Q) 1.6
.c
r:; 1.4

CQ)
Q):: 1.0
EO.
Q) Q) 0.8
uQ)
0.6
g.cn 04
(5 0:2
0
0
Fig. T6.23. Relationship .between the
initial soil pressure and
6 earthquake-induced bending
moment and displacement.
shaking were reduced compared to the outward displacement without com-
paction. This effect is similar to the concept of pre-stressing of structures, and was
also noted in centrifuge model tests of cantilever retaining walls retaining dry
sand, where densification on successive earthquake shaking events increased the
in-situ lateral earth pressure (Steedman, 1984). In some cases, the increase in
bending stress and displacement caused by compaction may significantly reduce
the structural capacity of the wall (or tie backs) or cause unacceptable local dam-
age to services. However, provided this is not the case, then there may be a bene-
ficial effect on the overall resistance of sheet pile walls to earthquakes from this
form of soil improvement.
TC7: Analysis Methods
T7.1 Simplified Analysis
T7.1.1 Loads on gravity/sheet pile/cellular quay walls
T7.1.2 Displacement of walls
T7.1.4 Breakwaters, embankments and slopes
T7.2 Simplified Dynamic Analysis
T7.2.4 Breakwaters; embankments and slopes
T7.3 Dynamic Analysis
T7.3.1 Analysis procedures
T7.3.2 Examples
T7.3.3 Remarks on dynamic analysis of slopes and walls
T7.4 Input Parameters for Analysis
. TECHNICAL COMMENTARY 7
Analysis Methods
..
This technical commentary addresses analysis methods for port structures. These
methods are classified into three types according to the level ofsophistication as
discussed in Chapter 5 in the Main Text. Simplified analysis is appropriate for
evaluating approximate threshold limit for displacements and/or elastic response
limit and an order-of-magnitude estimate for permanent displacements due to
seismic loading. Simplified dynamic analysis is possible to evaluate extent of dis-
placement/stress/ductility/strain based on assumed failure modes. Dynamic
analysis is possible to evaluate both failure modes and the extent of the displace-
ment/stress/ductility/strain.
T7.1 SIMPLIFIED ANALYSIS
T7.1.1 Loads on gravity/sheet pile/cellular quay walls
The conventional practice for evaluating seismic stability of retaining walls,'
including gravity, sheet pile, and cellular walls, is based on pseudo-static
approaches. In this procedure, aseismic coefficient, expressed in terms of the
acceleration of gravity, is used to compute an equivalent pseudo-static inertia
force for use in analysis and design. The actual dynamic behaviour of retaining
walls is much more complex than treated in the pseudo-static approach. However,
this approach has been the basis for the design of many retaining structures in
North America, Japan and other seismically active regions around the world
(Whitman and Christian, 1999; Tsuchida, 1990).
(l) Stability assessment
In the pseudo-static approach, the stability of gravity type walls is evaluated with
respect to sliding, overturning, and bearing capacity. Sliding is often the critical
condition for a wall having a large width to height ratio. Overturning or loss of
bearing capacity becomes the critical condition for a wall having a small width to
height ratio. In the limit state under strong shaking, the instability with respect to
r'
.(
'. ',<
" -,,",
\
I
I
\
\
I
\
\
\
1
I
i
\
\
I
I
\
\
I
i
\
\
\
\
- - .. AIA..l" '\-r
overturning and/or bearing capacity is much more serious than thar for' slidi:
because tilting of the wall, if excessive, will lead to collapse. Thus, it is comrm
practice to assign a higher safety factor for overturning and bearing capacity th.
for sliding (Ministry of Transport, Japan, 1989; Ebeling and Morrison, 1992)..
addition to these limit states, estimates of tilting are warranted for evaluating pos
earthquake stability of walls (Whitman and Liao, 1984; Prakash et al., 1995). Lo,
of bearing capacity has been conventionally evaluated based on the circular sf
analysis. The design parameters, including the internal friction angle of found:
tion soil, for use with this simplified analysis can be found in the relevant desig
codes (e.g. Ministry of Transport, Japan, 1999).
Stability of anchored sheet pile walls is evaluated with respect to gross stabilit
and stresses induced in structural components. In particular, gross stability is eva]
uated for a sheet pile wall to determine the embedment length of the sheet pile
into competent foundation soils. Stability is also considered for an anchor of th.
sheet pile wall to determine the embedment length and the distance from the wall
Stresses are evaluated for the wall, anchor, and tie-rod. In the ultimate state, the
rupture of tie-rods results in catastrophic. failure of the wall and, therefore, thi:
mode of failure must be avoided. Thus, it is common practice to assign a large
safety factor for tie-rods (Ministry of Transport, Japan, 1989; Ebeling anc
Morrison, 1992). A less established issue is whether the wall or the anchor shoulc
be the first to yield. Excessive displacement of the anchor is undesirable (Gazetas
et al., 1990). Balanced movement of the anchor, however, reduces the tension in
the tie-rods and the bending moment in the wall. Thus, there are differences in the
design practices of anchored sheet pile quay walls (Iai and Finn, 1993). These dif-
ferences can result in a variation of up to a factor of three in the forces used for
designing tie-rods and anchors.
(1) Active earth pressures
In the pseudo-static approach, the earth pressures are usually estimated using the
Mononobe-Okabe equation (Mononobe, 1914; Okabe, 1924). This equation is
derived by modifying Coulomb's classical earth pressure theory {Coulomb, 1776;
Kramer, 1996) to account for inertia forces. In the uniform field of horizontal and
(downward) vertical accelerations. khg and k.g, the body force vector. originally
pointing downward due to gravity. is rotated by the seismic inertia angle, lJI.
defined by (see Fig. T7.1)
. ...
lfI = tan-I [ kh. ] (T7.1 )
1- k\.
The Mononobe-Okabe equation is. thus. obtained by rotating the geometry of
Coulomb's classical solution through the seismic inertia angle, lfI. and scaling the
magnitude of the body force to fit the resultant of the gravity and the inertia forces
(T7.3)
Analysis Methods 311
__ wet
X:X:X:I
(a)
.0
r sat
J
r sat
o
(b)
Fig. T7.1. Active and passive earth pressures.
(a) Gravity quay wall.
(b) Sheet pile quay wall.
(Mononobe, 1924; Whitman and Christian, 1990). For a vertical wall
having a friction angle, 8, between the backfill and the wall, and retaining a
horizontal backfill with an angle of internal friction, , the dynamic active earth
pressure coefficient, K
ae
, is given by
K = cos
2
( -1jf)
at [ sin(th +8) sineth -111) ]2 (T7.2)
cos 1jf cos( 1jI +8) 1+ 'f' 'f' '1'
cos(8 +1jI)
For completely dry soil, the dynamic active earth thrust, which acts at an angle, 8,
from the normal to the back of the gravity type wall of height, H, (see
Fig. T7.l(a is given by
Pac =; trd(l-kJH
2
where J'd is the unit weight of the dry backfill. In case of a uniformly distributed
surcharge qsup J'd should be substituted with (XI +(qsu/ H. The surcharge qsur
used for seismic design is typically half of the surcharge used for static stability
design. Other horizontal loads, including a fraction of the horizontal bollard pull,
can also be considered in the design.
(T7A)
1
!
\
\
"
3]2 PlANe
(3) Passive earth pressures
A similar expression is given for the dynamic passive earth thrust, Ppe. The
dynamic passive earth pressure coefficient, Kpe, for the soil in front of the wan
with a level mudline is given by Eqn. (TI.2) with a minus sign in front of the
square root symbol. In this case, H in Eqn. (Tij) is the depth of the embedment
of the wall, D
emb
(see Fig. T7.1(b). In both active and passive cases, the point of
application of the resultant force is typically chosen at a level of OAOH to DASH
(Seed and Whitman, 1970).
Another approach that gives similar results to the Mononobe-Okabe method is
that of Seed and Whitman (1970). This approach calculates separately the
dynamic component of the active and passive earth pressure coefficients as
3 17
J1K
ae
= 4k
h
, J1Kpe = - gk
h
This dynamic component acts at a point O.SH to O.6H from the base of the wall.
A complete set of equations may be found in the design codes and manuals
(e.g. Ministry of Transport, Japan, 1989; Ebeling and Morrison, 1992). In
referring to these codes and manuals, a careful look at the sign convention
of the friction angle, 8, is needed. In much of the Japanese literature, the
friction angle, 8, is defined as positive whenever the soil drags the wall down.
The same is true for American literature for the active earth pressure, but for
the passive earth pressure, the friction angle, 8, is defined as positive when the
soil drags the wall up, as shown in Fig. T7.1(b). Since this paper is written in
English. the sign convention of American literature is followed here.
(T7.5)
\
I
\
\
. i
\
1
\
t
(4) Fully and partially submerged fills
The Mononobe-Okabe equation was derived for dry yielding backfill retained
by a rigid wall. When the backfill is saturated with water, it has been common
practice 10 adopt the assumption that pore water moves with the soil grains..
Considering a fully saturated Coulomb wedge. the horizontal inertia force is
proportional to the saturated total unit weight, ~ a l ' and the vertical gravity force is
proportional to the buoyant unit weight (Jb = 'Kal - Yw)' Thus. the modified
horizontal seismic coefficient is given by (Amano et al., 1956)
i; = Ysat k
h h
Yh
Using the modified horizontal seismic coefficient in Eqns. (T7.l) through (T7.3)
with a unit weight Jb in Eqn. (T7.3), will give P
ae
and Ppe for a fully submerged
backfill. The vertical seismic coefficient k; will not be modified for the submerged
condition.
Analysis Methods '313
For a partially submerged soil wedge as shown in Fig. T7.l (a). the equivalent
unit weight for the soil wedge may be obtained by computing the weighted
average of the unit weights based on the volume of soil in the failure wedge above
and below the ground water table. This results in the equivalent unit weight as
(e.g. Ebeling and Morrison, 1992)
(T7.6)
The modified seismic coefficients will be obtained by modifying Eqn. (T7.5) as
follows:
where
'J( = tqsur H+ tYweI H s ~ r +Ywet HsubHsur + tYsat H::b 'st
t qsur H + t Ywet Hs: r + Ywet H
sub
H
sur
+ t Yb H
S
: b
(T7.7)
qsur: a uniformly distributed surcharge
Yb : buoyant unit weight
The active and passive earth thrusts, Pae and Ppe' for the partially submerged
backfill are obtained through Eqn. (T7.3) by using a unit weight %and k ~ .
An alternative procedure for taking into account a partially submerged soil
situation is to idealize the effect of the overlaying soil layer as a surcharge to
the soil layer of interest (e.g. Ministry of Transport, Japan, 1999). Since the
contribution for surcharge from the layer above the water table to the saturated
layer below is equal to YwetHsup I'd in Eqn. (T7.3) should be substituted by
(Jb+(qsur+YwetHsur)IHsub)' This leads to an alternative equation to define the
modified seismic coefficient as
(T7.7')
These results can be further generalized for multi-layered partially submerged
soils.
(5) Coarse-grained fills
For extremely coarse-grained fill (e.g. gravel and rock fill), the soil skeleton and
pore water act independently. The permeability k is often used as a primary
.parameter to dictate the pore water movement relative to the soil skeleton
(Matsuzawa et al., 1985). Values ranging from 10-
2
cmls to 10 cmls have been
quoted to define the transition zone between the two states where water and soil
move independently, and where the two move as a single body. However, no clear
consensus in practice exists for the limiting value of permeability for these
(17. 8)
j ,.
1.'.
I
I
I
I
i
I
j
i I
j
j
I
\
~ j
\
"'Jq. t'lANC
'conditions. Within the. transition zone, a percentage of the pore water remains
attached to the soil, and the rest moves independently. The pore water movement,
however, is also affected by other factors, including the compressibility co-
efficient of the soil skeleton, the rate of cyclic loading, and boundary conditions
for pore water flow around the porous fill. These issues need to be thoroughly.
investigated before making any recommendations in design practice.
It is difficult to develop a completely logical set of assumptions for extremely
coarse-grained fill. One procedure is to assume that the total thrust is made up of:
1) a thrust from the soil skeleton, computed using Eqns. (T7.1) through (T7.7)
with idin place of Ysat in Eqn. (T7.7) or Eqn. (T7./); and
2) the effect of the pore water by assuming Westergaard dynamic water loading
(Westergaard, 1933; to be discussed later, see Eqn.' (T7.8)).
This procedure, however, is not totally consistent, since the effect of the increased
pore pressures due to the dynamic water pressure is ignored in the computation of
the thrust from the soil skeleton (Ebeling and Morrison, 1992).
(6) Effects of excess pore water pressure
Another significant, yet poorly defined issue involves the procedures recom-
mended for evaluating the effects' of excess pore water pressures generated in sat-
urated backfill on the dynamic earth pressures exerted on the wall. One approach
is to use the reduced internal friction angle which produces the same shear resist-
ance of soil under a reduced confining pressure (Ebeling and Morrison, 1992).
This approach may work if reasonable engineering judgments are used in deter-
mining the level of excess pore water pressure. Many laboratory tests indicate,
however, that the pore water pressure generally will not remain constant once the
soil begins to deform. Dilatancy of sand at large strain levels complicates the
behaviour' of saturated sand under cyclic loading. Effects of the initial deviator
stress, anisotropic stress conditions. and the rotation of the principal stress axis
also complicates the cyclic behaviour of the sand. All of these issues may be best
addressed in the context of seismic analysis of the soil-structure interaction phe-
nomena based on cyclic behaviour of sand using effective stress analysis (lai and
Ichii, 1997).
(7) Hydrodynamic pressure
During seismic shaking, the free water in front of the structure exerts a cyclic
dynamic loading on the wall; the critical mode occurs during the phase when suc-
tion pressure is applied on the wall. The resultant load can be approximated by
(Westergaard, 1933):
7 .,
e: = 1 2 ~ Y w H :
_ ...J ....... ~ , J ...... _ ........... - ......... ..a__
where
Xv: unit weight of seawater
Hi: water depth
The point of application of this force lies O.4H
w
above mudline.
The sea water level in front of the wall can be that most conservative in
design. The most conservative water level can be a low water level. a high
water level. or the level between them, depending on the design conditions of the
wall.
(8) Equivalent seismic coefficient
For simplicity, it is often assumed in design practice that k; = 0 (e.g. Ministry of
Transport. Japan. 1989). The factor of safety against earthquake for induced
deformation is defined by
F = ~
S k
e
(T7.9)
where k; threshold seismic coefficient from pseudo-static analysis
k
e
: equivalent seismic coefficient based on design level ground motions
The threshold seismic coefficient k, is the limit defined for a structure, beyond
which a structure-soil system will no longer maintain the pseudo-static equilib-
rium. For example, the seismic stability of a gravity quay wall is evaluated for
sliding, overturning and bearing capacity failure. By increasing the magnitude of
the seismic coefficient k
h
for use in the pseudo-static approach, the first failure
mode to yielda safety factor of unity is identified as the mostcritical. The seis-
mic coefficient representing this threshold for the most critical failure mode
defines the threshold seismic coefficient k;.
.The equivalent seismic coefficient k
e
is a pseudo-static acceleration, specified
as a fraction of gravity, for use in the pseudo-static analysis to represent the effects
of an actual earthquake motion. The equivalent seismic coefficient is not always
equal to the design level PGNg due to the transient nature of the earthquake
motions. The ratio of k, to PGNg used in practice has been based on empirical
data and engineering judgement. For example, the threshold seismic coefficients
obtained by back-analyses of a damaged and non-damaged structure at sites of
non-liquefiable soils, provide a lower and upper bound estimate for an equivalent
seismic coefficient. The threshold seismic coefficient of a damaged structure pro-
vides a lower bound estimate because the effective seismic coefficient of the
earthquake should have exceeded the threshold seismic coefficient for damage to
occur. Conversely, the threshold seismic coefficient of a non-damaged structure
provides an upper bound estimate because the effects of the earthquake motion
represented by the effective seismic coefficient should have remained below the
instability threshold.
316 PlANe
0.5
"
~
"
"
,
" ~
~
0.4
1 Muroran
:2 KU5hinl
3Akk5b1
4 Uno
S Kiriuppu
6 Nemunl
7Hanauki
I<Dmal!ujima
I
0.3
1 D 1
k =- (-!!.!!l.)1 D
max
~ 0.2g
e3 g
0.2
1,;,.........
0.1
D
max
< 0.2g
O ~ ....:- --.l. .l..- ...:... --'
o
Peak Ground Acceleration (3
ma
/ g)
Fig. T7.2. Equivalent seismic coefficient k, for retaining walls at non-liquefied waterfront
sites (after Noda et aI., 1975).
Figure T7.2 shows a summary of the lower and upper bound estimates based on
the case histories of 129 gravity type quay walls during 12 earthquakes (Noda
et al., 1975). The roots of the arrows in the figure, rather than the points, show the
exact values. The arrows pointing up indicate lower bound estimates; those point-
ing down indicate upper bound estimates. The equation for an upper bound enve-
lope was given by Noda et a1. (1975) as
a
k - max
---
e
s
k,= ~ ( a ; r
a max < 0.2g
a max ~ 0.2g
(T7.10)
(T7.1 n
A similar study for anchored steel sheet pile quay walls was performed using data.
from 110 case histories. It was concluded that the above relation is also applica-
ble for the anchored sheet pile quay walls (Kitajima and Uwabe, 1979).
As shown in Fig. T7.2, the relation in Eqn. (T7.l0) is an envelope. An average
relationship between the effective seismic coefficient and the peak ground accel-
eration may be obtained as
k, = O.6( a; )
Anatysis Methods 3-17
T7.1.2 Displacement of walls
(1) Non-liquefiable sites
On the basis ofearthquake case histories at non-liquefaction sites, a set of empir-
ical equations were derived through regression analysis to define horizontal
displacement, settlement, and displacement ratio (i.e. horizontal displacement
normalized by wall height). The variable used in the empirical equations is the
safety factor F's defined by Eqns. (T7.9) and (T7.10), which defines the threshold
level for the first failure mode among sliding, overturning and bearing capacity
failure. The empirical equations are defined as a linear function of (IIF
s
) (see
Fig. T7.3 and Tables T7.1 and T7.2) (Uwabe, 1983) as
Damage Degree'
0
1
)(
2
4-
3
0
0 4
+(7
f---
---
--
---
.,A--
i-- 4- 4-
4-)( Cl.
)( )(
-(7
I
,..x
o b- 0 I
-.3---
1
- 600
E
o
-
C
a>
E 400
CD
u
a.
en
is 200
c:
o
N
'C
o
J: 2
1/Fs
.. refer to Table T7.3 (a) and (b)
(a)
3 4
1/Fs
.... refer to Table T7.3 (a) and (c)
{b)
Damage Degree-
o 1
)( 2 n
4- 3
+a.:"
c 4
"
"
,,"/
...
...-: cr,,""
" ..
.>:
"
...
,,"
4-
-"
.>.
...
...
"
" )( "
"
;/./
...
<;l
...
"
8 ...
1/

)(
I ca
1-'''.
I
I I
250
-
E

-c:
CD
E
150
J!1
c,
en
is 100
g
c:
o
SO
o
J:
o
o 1 2 3 4 5
Fig. T7.3. Statistical summary
of case history data
for non-liquefaction
sites (after Uwabe,
1983).
(a) Gravity quay
wall .
(b) Sheetpile quay.
wall.
3]8 PIANC
Table T7.1. Empirical equations for a gravity quay wall (Uwabe, 1983).
Deformation Empirical equations** Correlation
coefficient
Standard
deviation
Max. horizontal displacement (em) .d =-74.2 + 98.2(l1F;) 0.34 130
Settlement of gravity wall (em) s =-16.5 + 32.9(lIFJ 0.50. 30
Normalized displacement* (%) r =-7.0 + 10.9(1IF;) 0.38 13
* Maximum horizontal displacement normalized with a wall height H.
** F;,: Factor of safety against earthquake =kjk
e
Table T7.2. Empirical equations for a sheet pile quay wall (Uwabe, 1983).
Deformation Empirical equations** Correlation Standard
coefficient deviation
Max. horizontal displacement (em) d =-1.6 + 34.9(l/F;,) 0.68
Settlement of sheet pile wall (em) s = -5.3 + 14.7(lIFJ 0.40
Normalized displacement* (%) r = -1.5 + 5.8(11F;,) 0.65
* Maximum horizontal displacements normalized with a wall height H.
** F;,: Factor of safety against earthquake = k/k
e
d = a + b / ~
where
27
20
5
(T7.12)
(T7.13)
d: horizontal displacement, settlement, or displacement ratio
a, b: empirical constants
F
s
: factor of safety during an earthquake
Field performance data from numerous sites in Japan compiled by Kitajima and
Uwabe (1979) have been used to derive another empirical method for
estimating the approximate degree of damage to anchored sheet pile walls during
earthquakes (Gazetas et al., 1990): Two non-dimensional factors, the effective
anchor index (AI) and .the embedment participation index (EPI) have been
related to the degree ofdamage recorded for the sheet pile quay walls. Referring
to Fig. T7.4, EAI is used to quantify the amount of available anchor capacity with
AI =Dane
H
The EPI provides the contribution of the wall embedment as
EPI=Ppe(l+ Dr 1
r: Dr+H
(T7.14)
where Dr is the depth of the effective point of rotation of the sheet pile wall below
the mudline. The effective point of rotation is approximated, using the effective
seismic coefficient, k;, and the angle of internal friction of the backfill soil. . as
D, = [0.5(1 + - 0.02(1)- 20)]H
Fig. T7.4. Relevant parameters
for defining effective
anchor index. EAI.
and embedment par-
ticipation index. EPI
(after Gazetas et al.,
1990).
(T7.l5)
The effective point of rotation, D
f
, can be conservatively estimated as the embed-
ment length of the sheet pile below the mudline, D
cmb
'
For uniform soils, EPI can be approximated as
(T7.16)
The seismic active and passive earth pressure coefficients can be readily estimated
from Eqn. (T7.2). The angle, aae' from a horizontal plane to the active
failure plane is approximated as
=45 + epl2 - (T7.17)
The relationships between the EAI and EPI factors were plotted for 75 Japanese
case histories in Fig. T7.5, with the descriptions of the degrees of damage given
in Table T7.3. ..
(2) Liquefiable sites
When sites contain saturated loose to medium sandy soils, it is necessary to
consider the consequences of liquefaction. The simplified geometry of the loose
saturated sand relative to the cross section of the wall shown in Figs. T7.6 and
T7.7 was used for classifying the case histories for gravity and sheet pile quay
walls. Table T7.4 shows the catalogue of the case histories with supplemental
information shown in Fig. T7.8 (lai, 1998a). A summary of the normalized dis-
placements of walls at liquefaction sites are shown in Tables T7.5 and T7.6 for
gravity and anchored sheet pile walls. The intensity of the earthquake motions can
be categorized in two levels. The first level (E1)'..is defined as the earthquake
motion level being equivalent to the design seismic coefficient. The second level
(E2), is defined for those about 1.5 to 2.0 times larger. It should be noted that these
earthquake levels are defined for sites of non-liquefiable soils. If a site
consists of liquefiable soils, a nearby site of non-liquefiable soil is used for
defining the earthquake levels. For evaluating the levels of earthquake motions
1" 320 PlANe
1.
1
,
I
Degree of damage
o 0
1
"'
!
I
I
I
\
2
1.5
-
~
)(
-8 1
.5:
...
o
.c:
o
c:
III
CD .
.~
13 0.5
CD
::
UJ
o
..().S
n
o
o
o 0
.0-
o
o
o
o
o
o
o
o 0
o
o
,
,
,
,
\
\
o
o
-
o
1
2
3
4
o
0\
\
\
\
l
I
o
o 0.5
Embedment participation index (EPI)
* refer to Table T7.3 (a) and (d)
1
\
I
;
!
Fig. T7.5. Developed seismic design chan (Gazetas et al.. 1990).
relative to the seismic coefficient used for design, the average relationship shown
in Eqn. (T7.11) is used. The normalized displacement dlH (%) is defined as the
ratio of a horizontal displacement of a quay wall over a wall height from
mudline.
From the numbers shown in Tables T7.5 and T7.6, it is possible to obtain a
rough estimate of displacements based on the wall height. Case history data of
anchored sheet pile walls at liquefied sites are limited, and a statistical summary
for E2 motion has not yet been available.
In the simplified analysis of pile-deck systems of pile-supported wharves. seismic
inertia force is applied to the deck. where the mass of the pile-deck system is
Table T7.3. Qualitative and quantitative description of the reported degrees of damage
(after Uwabe. 1 9 8 3 ~ Gazetas et al., 1990).
(a) Qualitative description of the reported degrees of damage (Uwabe. 1983: Gazetas
et aI.. 1990)
Degree of Description of damage
damage
o no damage
1 negligible damage to the wall itself; noticeable damage to related structures
(i.e. concrete apron)
2 noticeable damage to the wall itself
3 general shape of anchored sheet pile preserved. but significantly damaged
4 complete destruction, no recognizable shape of wall remaining
(b) Quantitative description of the reported degrees of damage for gravity quay wall (after
Uwabe, 1983) .
Degree of
damage
o
1
2
3
4
Maximum residual
displacement at
top of wall (ern)
o
<25
25 to 70
70 to 200
>200
Average permanent
displacement at top of
wall (em)
o
<25
25 to 40
40 to 200
>200
(c) Quantitative description of the reported degrees of damage for sheet pile quay wall
(after Uwabe, 1983)
Degree of damage
o
1
2
3
4
Maximum residual
displacement at
top of sheet pile (em)
o
<30
30 to 100
100 to 200
>200
Permanent
sheet pile (em)
o
<10
10 to 60
60 to 120
>120
(d) Quantitative description of the reported degrees of damage for sheet pile quay wall
(after Gazetas et al., 1990)
Degree of damage
o
1
2
3
4
Permanent displacement at top of sheet pile (em)
<2
10
30
60
120
I ,'. 322 PlANe
Non-Liquefiable soil
(a)
(b)
Fig. T7.6. Generalized soil conditions for gravity quay wall at liquefiable site.
(a) Loose sand in backfill only.
(b) Loose sand in both backfill and foundation.
(T7.18)
where
1: flexural rigidity (kNm
2
)
;: depth from the ground surface (rn)
concentrated. Stresses in the pile-deck system computed in the analysis are com-
pared with the limit stresses at initiation of yield and at plastic hinge formation in
the piles (see Fig. T7.9). The seismic inertia force used in the simplified analysis
is typically a fraction of gravity. often specified in codes and standards.
Comparison of computed stresses in the pile-deck system with limit stresses may
result in an approximate estimate of the threshold level for pile-supported
wharves.
The embedded portion of the piles are typically idealized as a beam on a
Winkler foundation described by
d ~ . 1
EI-=-P=-pD
d ; ~ p
i
I
1
I
I
.1: lateral displacement of a pile at the depth (m)
P: subgrade reaction per unit length of a pile at the depthc (kN/m)
p: (= P/D
p
) subgrade reaction per unit area of a pile at the depth (kN/m
2
)
Dp:pile diameter (or equivalent pile width) (m)
In North American practice, the relationships between p and .1 outlined in
American Petroleum Institute (API) (1993), called p-y curves, have been widely
1\
-
Non-liquefiable
soil
Non-liquefiable Soil
(a)

Non-liquefiable
soil
Non-liquefiable soil
(b)
(c)
Fig. T7.7. Generalized soil condi-
tions for sheet pile quay
wall at liquefiable site.
(a) Loose sand behind
the wall only.
(b) Loose sand backfill.
(c) Loose sand in both
backfill and found-
ation.

_
40
50
C6
--
c ,0
o 0"
N - 30
.;:: E
o Q)
.c E
~ ~ 20
:.= co
coC5..
E . ~ 10
0
0
z
i
!

I
!
:
I
I I
~
,-
1
!.

I
I I
5 10 15 20 25 30
Thickness of Liquefiable Sand
beneath Wall (m)
Fig. T7.8.. Horizontal displacement nor-
malized with respect to wall
height. correlated with thickness
of loose soil deposit below the
wall (after Iai. 1998a).
Table T7.5. Normalized displacement of gravity walls at liquefied sites.
Nonnalized displacement d/H (%)
0-5 5-10 10-20 20-40
During Design Level Earthquake Motion (E1)
Non-liquefaction
Loose sand at backfill. only
Loose sand at both backfill and foundation
During 1.5 or 2.0 as high as E1 Motion (E2)
Non-liquefaction
Loose sand at backfill only
n
*fi.
Table T7.6. Normalized displacement of anchored sheet pile walls at liquefied sites.
During Design Level Earthquake Motion (E1)
Nonnalized displacement d/H (%)
Non-liquefaction
Loose sand behind the wall only
Loose sand at backfill including anchor
0-5 5-15 15-25 25-50
*!J!
A M
.....
WS"
adopted for use at ports. The API equations evolved from Reese et al. (I 974). The
p-y curves for soft clay are given as shown in Fig. T7.10(a) and (b) and the rele-
vant parameters, Pu' 4:, and ~ r ' are evaluated from standard laboratory tests or
empirical correlations with soil parameters obtained from in-situ tests. The
required geotechnical information consists of buoyant unit weight, Jb (for a
326 PlANe
SIMI Pile O.,700rnm
(a)
(c)
Virlual groundsurfaoe
/
curve. or coefficient 01 subgradereaction
(b)
grgund SUr180e

Fig. T7.9. Modelling of pile-deck system of a pile-supported wharf.
(a) Pile-supported wharf.
(b) Modelling by a frame structure supported by Winkler foundation.
(c) Modelling by a frame structure having equivalent lengths.
completely submerged soil deposit), undrained shear strength, c
u
' from vane shear
or triaxial tests, and the strain at 50% peak shear resistance. e; The ultimate resist-
ance, Pu' increases from 3c
u
to 9c
u
as increases from 0 to as shown in
Fig. T7.10(c). The parameters zi, and are given by
(T7.19)
(T7.20)
where J is a dimensionless empirical constant with values ranging from 0.25 to
0.5 determined from field testing.
l J ..o - -
P
Pu
1.0
0.5

I
I
I
I
'(p) :
- ='0.5 --
Pu : :1c
I
I
I
I
I
I
o 1.0 5.0 8.0
p
Pu
1.0
(a)
10.0
(b)
, I I

I
I 1
I I() ()3
: :..E... =0.5 A
I I Pu
I I
I
01.0 3.0 5.0
0.5
0.72
o
o
(c)
9c Pu
Fig. T7.1 O. p-y curves for soft
clay (API, 1993).
(a) Static loading.
(b) Cyclic loading.
(c) Variation of Pu
with depth.
The API p-y curve for sand is defined by
p = Ap, tanh ( )
(T7.21)
where
A=O.9
A= (3.0 -0.8
for cyclic loading
for static loading
328 PlANe
The required soil parameters include the buoyant unit weight of soil, 'Yb' and the
angle of internal friction, t/J. The angle of internal friction is commonly
estimated using empirical relationships based on in-situ penetration resistances
(SPT or CPT). The relevant parameters, including the ultimate resistance, Pu' ini-
tial modulus of s u ~ g r a d e reaction k
sub
, are then determined from relationships with
t/J (API, 1993).
In Japanese practice, (equivalent) linear or non-linear relationships outlined in
Ministry of Transport, Japan (1989) have been widely adopted. The linear rela-
tionship, widely known as Chang's method in Japan, is defined by (Chang, 1937)
P = Est1 = ktrsub
D
p
t1 (T7.22)
where Es(=kh.suJ)p): equivalent subgrade elastic modulus (kN/m
2
) is assumed con-
stant.
The non-linear relationships evolved from Kubo (1962) are defined as
k
A 0.5
P= ctype,u
(for S-type subsoil)
(for C-type subsoil)
(T7.23)
(T7.24)
where S-type subsoil refers to that with linearly increasing SPT N-values with
depth and C-type subsoil refers to that with constant SPT N-values with depth.
The coefficients k
h
-
sub
, k
s
.
type
and kc-type are correlated with SPT N-values (Ministry
of Transport, Japan, 1989).
In both the North American and Japanese practices, these pseudo-static
methods are applicable for cases involving piles embedded in (a) horizontal soil
layers with negligibly small deformation relative to the displacement of the piles
due to the inertia load applied at the deck, and (b) soils that do not lose
appreciable strength during cyclic loading.
It is quite common to have pile-supported structures that are founded on slop-
ing embankments. In order to take into account the effect of sloping embank-
ments, a virtual (or a hypothetical) embankment surface, which is set below the
actual embankment surface. mav be assumed for evaluating the lateral resistance
~ -
of piles embedded in an embankment as shown in Fig. T7.9(b) (Ministry of
Transport, Japan, 1989).
The stresses in a pile-deck system can be computed either as a frame structure
supported by linear or non-linear Winkler foundations (Fig. T7.9(b)). or a frame
structure having 'equivalent fixity' piles (Fig. T7.9(c)). The lengths of 'equivalent
fixity' piles from the ground or (virtual) embankment surface is typically given by
11[3 (Ministry of Transport, Japan, 1989), where
(T7.25)
For a wharf with steel piles. an approximate elastic limit (i.e. threshold level). ?,v.
at which plastic hinges occur at the pile head in about half of all piles may be
obtained based on the equivalent leg lengths of a pile-deck system, 1/{3. as (Yokota
et al., 1(99):
where
~ . =O.82Pu
(T7.26)
(T7.27)
P
u
: the limit lateral load when plastic hinges form at both the top and
embedded portion of all piles
M
p
: pile bending moment at plastic hinge state
l: unsupported length of a pile between the deck and virtual ground
surface
subscript i: i
th
pile.
Sloping ground conditions exist throughout ports as natural and engineered
embankments (e.g. rubble mounds for breakwaters, dikes for pile-supported
wharves, dredged channel slopes, river levees). On-shore and submarine slopes in
ports have been often found to be vulnerable to earthquake-induced deformations.
Two of the primary causes for this are high water levels and weak foundation soils
often encountered in port areas. In addition to waterfront slopes, several recent
cases involving failures of steep, natural slopes along marine terraces located in
backland areas have resulted in damage to coastal ports. Large-scale deformations
of these slopes can impede shipping and damage adjacent foundations and buried
structures, thereby limiting port operations following earthquakes.
In most cases involving soils that do not exhibit considerable strength loss due
to shaking, common pseudo-static rigid body methods assuming slip surfaces will
generally suffice for evaluating the stability of slopes in ports. These methods of
evaluation are well established in the technical literature (Kramer, 1996). The
seismic stability of slopes is evaluated by modifying static limit equilibrium
analyses with the addition of a lateral seismic body force, that is the product of a
seismic coefficient k
h
similar to k; previously defined and the mass of the soil
bounded by the potential slip surface (seeFig. T7.ll). This mass of soil should be
modified with an added mass in order to account for the hydrodynamic force due
to the presence of sea water. This increase can be derived, as a first approxima-
tion, by assuming a Westergaard's pressure distribution along the vertical
face. The seismic coefficient k
h
is specified as a fraction, roughly Y2, of the peak
330 PlANC
Rubble
mound
Fig. n.ll. Schematic figure for pseudo-static analysis of an embankment.
horizontal acceleration ratio Gmulg at the crest of an earth dam because the lateral
inertial force is applied for only a short time interval during transient earthquake
loading (Hynes-Griffin and Franklin, 1984; California Division of Mines and
Geology, 1997).
The pseudo-static methods outlined above are simple to use, but require a high
level of engineering skill and judgement when dealing with the following issues:
(a) the influence of excess pore pressure generation on the strength of the soils,
including cyclic progressive deformation and strain softening; (b) the range or
pattern of slope deformations that may be associated with various factors of
safety; and (c) the effect of the elasticity and permeability of the materials used
for breakwaters and embankments, that produces hydrodynamic pressures deviat-
ing from the classical Westergaard's solution. With the reasonable application of
engineering judgement, the pseudo-static methods are routinely used for screen-
ing purposes, and can be useful in evaluating the approximate threshold level of
slope stability.
T7.2 SIMPLIFIED DYNAMIC ANALYSIS
(1) Sliding block analysis
Simple, straightforward methods have been developed for evaluating the perma-
nent displacements of gravity walls during earthquakes (e.g., Newmark. 1965;
Franklin and Chang. 1977; Richards and Elms. 1979; Whitman and Liao, 1985).
These methods are based on the sliding block model, adopting similar assump-
tions to those made for the pseudo-static analysis outlined in Section T7.1.1.
First, the stability of the wall and the backfill are evaluated using the lateral
earth pressure theory based on Mononobe-Okabe's equations. The threshold
acceleration is determined by the value resulting in a factor of safety of unity for
sliding of the wall-backfill system. For example, the threshold acceleration. a; for
a vertical retaining wall is given by the expression (Richards and Elms. 1979)
(T7.28)
where u; is the coefficient of interface friction between the wall and the found-
ation rubble or soil, P
ae
is the active earth thrust computed using the
Mononobe-Okabe method, 8is the wall-backfill interface friction angle, W
g
is the
weight of the wall per unit width, and g is the acceleration of gravity. It should be
noted that at must be known in order to calculate Pae, therefore an iterative proce-
dure is necessary.
Once the threshold acceleration has been determined. then a set of acceleration
time histories are selected for sliding block analysis. The use of more than one
acceleration time history is necessary because displacements calculated through
sliding block analysis are sensitive to the characteristics of acceleration time
history used in the analysis. The acceleration time histories should be repre-
sentative of the design level ground motions in both duration and frequency
content. When the ground motion acceleration exceeds the threshold acceleration,
at' the wall-backfill system begins to move by translation along the base of the
wall and the failure plane through the backfill. By double integrating the area of
the acceleration time history that exceeds at, and continuing the time integration
until the sliding stops. the displacement of the wall relative to the firm base below
the failure plane can be determined as shown in Fig. T7.12. This computation can
be easily performed using common spreadsheet routines or a simple computer
code.
Numerical studies based on the sliding block method of analysis have led to the
development of simplified relationships between ground motion intensity and
seismically-induced deformations. Based on a large database of nearly 200 earth-
quake records, conservative envelopes for evaluating sliding displacements were
obtained by Franklin and Chang (1977) as shown in Fig. T7.13. These results
were obtained for earthquake motions with a peak acceleration and velocity of
0.5g and 76 cm/s, respectively.
The sliding block theory suggests that the vertical component of input acceler-
ation time histories should affect the sliding displacement. However, a pararne-
teric study, using an earthquake motion recoded at Kobe Port Island site (PGA
H
=
544 cm/s', PGA
v
=200 cm/s'), suggests that the effect of the vertical component
is not significant. and it either increases or decreases the sliding displacement less
than about 10 percent (Nagao et al., 1995).
A simplified expression to approximate these envelopes was proposed by
Richards and Elms (1979) as
d=0.087 v m a x ~
4.
4
(T7.29)
332 PlANe

15
15
-
-
I I
\
, \
10 5
5 I I 10
I I Time (5)
I I
I I
.50
0
L-- - - --l.- - - - --l.- - - - .....J
Time (s)
(a)
- relative displacement (em)
100 - - - relative velocity (cmIs)
I I ,
50
I I I,
I I
I,l I 1
250

-250
5l 500 ;;,cm;.;.;;,,;:/S;.,.2 r- T""'" --,
! tJ:lreshold acceleration : a,
:2
.21
'0
c::
.2
I
1(time)
1-T7.,.,..,-r.t-- - - - - - - --
a
--\--------
V =ato
Vc'v
\ "
Velocity base
1(time)
d
t (time)
(0)
Fig. T7.12. Computing displacement in
the sliding block analysis.
(a) Example of sliding
block analysis.
(b) Acceleration. veloci-
ties and displacement
from sliding block
analysis.
where d is the permanent lateral displacement (em). V
max
is the peak ground veloc-
,
ity (crn/s). and a
max
is the peak ground acceleration (cm/s). This expression is
considered valid for cases where the.ratio a/cz
max
is greater than or equal to 0.3.
As mentioned earlier. the computed displacement from sliding block analysis
is sensitive to a number of key parameters. including those for evaluating thresh-
old acceleration and the nature of the earthquake motion. Indeed. the scatter of
displacements computed by Franklin and Chang (1977) covers a range which
spans to about a factor of ten. The sensitivity is evaluated further by Whitman and
Liao (1985). taking into account the effects of factors such as the dynamic
response of the backfill, kinematics of backfill wedge, tilting of the wall and ver-
tical accelerations on the computed wall deformations. Based on the results of
sliding block analyses of 14 ground motions, the following relation was derived
to define the mean value:
d
- (37;max) ( 9.4
l1t
)
= exp ---
amax amax
(T7.30)
This expression forms the" basis for a statistical analysis of the likely sources of
error, leading to an alternative recommended design line based on a safety factor
of 4, equivalent to a 95% probability of non-exceedance (see Fig. T7.13)
In (T7.31)
a
max
9.4
The design recommendation that has been proposed in the tentative Eurocode
(CEN, 1994) provides the horizontal and vertical seismic coefficients through the
relations
k
- adesign I g
h -
rEC
(T7.32)
where adesign is the design acceleration. Values of the reduction factor r
EC
are
related to the allowable displacements given in Table T7.7.
For the designer, these expressions for evaluating the likely sliding dis-
placement of a retaining wall have considerable merit. However, some cautionary
notes are necessary. Actual seismic performance of gravity quay walls during
earthquakes often does not meet the assumptions inherent in sliding block analy-
sis. Where the movement of the wall is associated with significant deformation in
the foundation soils, the displacements computed by the sliding block approach
were substantially smaller than the displacements observed in the field (Iai,
1998a; Yang, 1999). Where the foundation is firm but rocking type response of the
wall is involved, the design curves discussed earlier are again found to be uncon-
servative (Steedman and Zeng, 1996). For liquefiable backfill, other techniques
334 PlANC
r--,
I
.........
:-.....
... -
I
.
, ,
I;
I ... .. I
..
.
.....
\ \. I
I....
,
. \ '
....
\ \
" "
"'. "

"
<,
..
I
. ....'-
,.,
;
\
.
I
..

I
" \
II r.
II

,___AnCIT records }
\
=-..-Seed Idriss M=8.25
,
- .
I I
_ . San Fernando 1971 Franklinand Chang
I
"
I'
- - - - Other natural records (19n) i I\i :
,-- .
I
_. - Rocksite records I
.\ i
r-_ Richards and Elms (1979) ,
'I
e-- - -rnII<lir
I I
Ii
Ii
I'
11
1 I ,
E
E.
-
c:
(I)
100
o
m
Q.
.!!!
'"C
(I)
Cii
'"C
(I)
N
"'0
m 10
"'0
c:
m
(;)
1000
0.01 0.5 1
Fig. T7.13. Proposed simplified results for evaluating slide displacement (after Franklin
and Chang. Richards and Elsm, 1979; Whitman and Liao. 1985; as
reported by Steedman, 1998).
Table T7.7. Sliding displacement (CEN. 1994).
Type of retaining structure
Free gravity wall that can accept a displacement up to 30(adeshmlg) (ern)
As above with a displacement up to 20(adesign/g) (em) -
Other retaining structures without allowing a displacement
}.5
1
that enhance limit equilibrium-based methods for predicting seismic displace-
ments are recommended (e.g.. Byrne et al., 1994). Consequently. it is important
to verify that the design conditions of interest meet the assumptions inherent in
sliding block analysis, e.g. rigid firm ground, wall sliding without tilt. and rigid
wedge-like backfill movement.
. (2) Simplified chart based on parametric study
In order to enhance the applicability of the simplified dynamic analysis for
gravity quay walls for general geotechnical conditions, seismic performance of
gravity quay walls was studied through effective stress analysis by varying struc-
tural and geotechnical parameters of a quay wall under various levels of shaking
(Iai et al., 1999). Major parameters studied are width to height ratio of a gravity
wall, WIH, thickness of soil deposit below the wall. D1, and geotechnical condi-
tions represented by SPT N-values of subsoil below and behind the wall (see Fig.
T7.14). The effective stress analysis was based on the multiple shear mechanism
using the computer program FLIP (finite Element Liquefaction Program), which
has been demonstrated as applicabile to the seismic analysis of gravity quay walls
during the 1995 Hyogoken-Nambu (Kobe) earthquake (Iai, 1998b).
Some of the details assumed for simplicity in the analysis include:
a) The geotechnical conditions of the soil deposits below and behind the wall
were assumed to be identical, represented by an equivalent SPT N-value N
65
(SPT N-value corrected for the effective vertical stress of 65 kPa). The equiv-
alent SPT N-value has been widely used for assessment of liquefaction poten-
tial in Japanese port areas (e.g. Port and Harbour Research Institute, 1997).
SPT practice in Japan typically corresponds to the energy ratio of 73%, about
20% higher than the practice in the USA, where the SPT N-value is typically
normalized to the energy ratio of 60% as in (N,)60' With the additional differ-
ence in the effective vertical stresses of 65 kPa and 100 kPa for correction,
(N')60 is about 1.5 times the equivalent SPT N-value.
b) Model parameters for the effective stress analysis were determined from the
equivalent SPT N-values based on a simplified procedure (Morita et aI., 1997).
c) The peak acceleration of the input seismic excitation was assigned at the base
layer as an incident wave (as of 2 rather than E + F). The time history of the
earthquake excitation was that of the incident wave at the Port Island (Kobe)
vertical seismic array site at a depth of - 79 m during the 1995 Hyogoken-
Nambu earthquake.
d) For simplicity, the thickness of backfill, D2, was assumed the same as the wall
height, H.
The results of the parametric study are summarized in tenns of residual hori-
zontal displacement of the wall, as shown in Figs. T7.15 through T7.18. The most
sensitive parameter was the SPT N-value of subsoil below and behind the wall.
The second was the thickness of the soil deposit below the wall. Although the
width to height ratio of a gravity wall is a sensitive parameter for a quay wall with
finn competent foundation, the effect of this parameter becomes less obvious
when the soil deposit below the wall becomes thick or soft.
Based on the results of the parametric study, a simplified procedure was devel-
oped for evaluating the residual horizontal displacement of a gravity quay wall
under seismic excitations. In this procedure, the displacement may be evaluated
based on the relevant parameters as follows:
1) Normalized residual horizontal displacement, d/H, is evaluated based on
equivalent SPT N-value for a prescribed level of shaking at the base by refer-
ring to Fig. T7.15 ~ r Fig. T7.16.
336 PlANe
2) Effects of thickness of soil deposit below the wall, Dl/H, is corrected by refer-
ring to Fig. T7.17.
3) Effects of width to height ratio, W/H, is corrected by referring to Fig. T7.18.
The applicability of this procedure was confirmed with the case histories from
the 1993 Kushiro-Oki earthquake and the 1995 Hyogoken-Nambu (Kobe) earth- .
quake (Iai et al., 1999). As is the case with other simplified procedures, the engi-
neer should confirm the applicability of these charts based on relevant case
history data and modify the charts accordingly before use in design practice.
(3) Evaluation of liquefaction remediation based on parametric study
The heightened awareness of the potential impact of liquefaction-induced ground
failures on waterfront retaining structures and the corresponding economic impact
of such damage has led to the increased utilization of soil improvement for miti-
gating liquefaction hazards. All other factors being equal, the effectiveness of the
ground treatment is a function of the level of densification or cementation and the
volume of soil that is treated. Soil improvement techniques have been used at
numerous ports throughout the world (e.g. Iai et aI., 1994). Although fewcase his-
tories exist for the performance of improved soils subjected to design level earth-
quake motions, past experience at waterfront sites has shown that caissons in
improved soils have performed much more favorably than have adjacent caissons
at unimproved sites which experienced widespread damage (e.g. Iai et al., 1994;
Mitchell et aI., 1995).
Design guidelines are available for specifying the extent of soil improvement
adjacent to retaining walls (Port and Harbour Research Institute. 1997). The
method is based on the standard pseudo-static analysis familiar to most engineers
(e.g. Ebeling and Morrison, 1992). In addition to the lateral seismic coefficient, a
dynamic pressure is added to account for earth pressures due to liquefied soil act-
ing at the boundary of the improved soil zone. The improved soil zone can be
Backfill
Soil
Fig. T7.14. Typical cross section of a gravity quay wall for parameter study.
detertnined by accounting for the width of the active and passive failure surfaces
which pass through the backfill or soils below the dredge line, and adding
additional width of improvement as a margin of safety. For structures underlain
by liquefiable soils, the improved zone must extend deep enough to preclude
failures due to loss of soil bearing capacity or deep seated rotational failure.
In order to develop a simplified technique for estimating seismically-induced
deformations of gravity walls (specifically gravity caissons), a parametric study
was conducted using a 2D numerical non-linear effective stress model, validated
with well-documented case histories (Dickenson and Yang, 1998: Yang, 1999).
The results of the study have been synthesized into practice-oriented design charts
for estimating the lateral deformations of gravity quay walls. The walls were
designed with heights ranging from 12 m to 19 m, and height-to-width (H:W)
ratios of 0.7 and 1.2. The H:W ratios correspond to horizontal seismic coefficients
of approximately 0.1 to 0.25, the range of interest for most port engineers. The
vertical seismic coefficient was not incorporated into the analyses. The configu-
ration of the model caisson and the surrounding soil is shown in Fig. T7.19. The
foundation and backfill soils are both cohesionless materials, each layer modelled
with uniform density. The range of densities used can be correlated with the nor-
malized SPT N-values, (N
1
)6Q, of 25 in the foundation soil, 10 to 20 in unimproved
backfill, and 30 in improved backfill.
Five earthquake motions covering the magnitude range of engineering interest
(M 6 to 8) were selected for the parametric study. Each record Was scaled to three
0.1 0.2 0.3 0.4 0.5 0.6 0.7
Input Excitation Level (g)
(b)
0.05
0.1 0.2 0.3 0.4 0.5 0.6 0.7
Input Excitation Level (g)
(a)
0.05
0.35 0.35
-+--N=5
~ N = 8
0.3
......-N=10
0.3
.-.._........................._...._.._....
-Q-N=15
-.--N=20
--6-N=25
0.25
N:Equivatent SPT
0.25
N-value
0.2 O.
:I: :I:
<, <,
'1J '1J
0.15 0.15
0.1 0.1
Fig. T7.l5. Effects of input excitation level (for WIH = 0.9).
(a) DlIH = 0.0.
(b) Dl/H = 1.0'.
338 PlANe

0.3
0.25
:I: 0.2
<,
"0
0.15
0.1
0.05
-+-0.1g
-o-"02g
--'-0.3;
"''V- 0.4g
-'-0.6g
V-'--v"tl
ov .
5 10 15 20 25 30
Equivalent SPT N-value
(a)
0.3
. 0.25
:I: 0.2
<,
'"0
0.15
0.1
0.05
(b)
30
Fig. T7.16.. Effects of equivalent SPT N-value (for WIH=0.9).
(a) DIIH = 0.0.
(b) DlIH = 1.0.
0.2 0.2
r-+-0.19
,'-0"0.29
i-...-o.39
i"-"""0.49
0.15 0.15
:-'-0.6g
1.2
....................-- "" -- _ :
o -- -- -- -.A ------.-.- ---()
o 0.2 0.4 0.6 0.8 1
Dl/H
(h)

J:
::c 0.1
1.2
(al
0.4 0.6 0.8
Dl/H
0.2
__ -0- ... -.--.....<>
.........
............
.....................--.v
sr:
-:
,.'-
-:
.'
0.05 .
J:
::c 0.1
Fig. T7.17. Effects of thickness of soil deposit below the wall (for WIH = 0.9).
(a) Equivalent SPT N-\'alue (l O).
(b) Equivalent SPT N-\'alue (20).
0.1!ir----.-----..--....-----.----. 0.15.--.....---.....--.....------,
-+-O.lg
"00.2g
-'-0.3g
-Q-"0.4g
-.llr-0.6g
0.1 0.1
1.1
"
.............q - I;1
0.8 0.9
W/H
(b)
0.7
<>-....,._..-.'-_....-_..-.-_...-<)-- --<>
.;;:t-_._.
..
oL - ! : : : : : : i : : : : = : : : c = = ~ = = ~ ~
0.6
0.05
1.1 0.8 0.9
W/H
(a)
0.7
..............
............. _- ...... -
'.""""-"7 "'" """"'''7
<>-..-'" - __ .
0.0
Fig. T7.18. Effects of width to height ratio W/H (for equivalent SPT N-value of 15).
(a) Dl/H = 0.0.
(b) Dl/H = l.0.
different maximum acceleration values ranging from O.lg to OAg. The lateral and
vertical deformations of the wall and surrounding soil, as well as the excess pore
pressures in the backfill were monitored during seismic loading. Notable assump-
tions in the parametric study include; (a) no pore pressure generation in the foun-
dation or improved backfill soils with (N\)6Q - 30, and (b) the soil improvement
extends to the base of the backfill (i.e. the same elevation as the dredge line)
depth.
In order to account for the duration of the earthquake motions, a normalized
ground motion intensity has been used. This parameter is defined as the maximum
horizontal acceleration within the backfill at the elevation of the dredge line
(amaxCd)) divided by the magnitude scaling factor (MSF) listed in Table T7.8
(Arango, 1996; Youd and Idriss, 1997). It is recommended that, if a site specific
seismic study is not performed to determine amaxCdl' the peak ground surface accel-
eration be reduced using the reduction factor, r
d
, developed for estimating the
variation of cyclic shear stress (or acceleration) with depth (e.g. Makdisi and
Seed, 1978). The values of r
d
for 15 and 7.5 m walls are approximately 0.78 and
0.95, respectively. It should be noted that the reduction factor was developed
using one-dimensional dynamic soil response methods, which will yield approx-
imate acceleration values for the two-dimensional soil-structure interaction appli-
cations discussed herein. .
The results of the parametric study are shown in Fig. T7.20. The lateral defor-
mations at the top of the wall are plotted versus the width of the improved soil,
340 PIANC
and as functions of backfill density and the H:W ratios of the caissons. This set of
curves can be used for estimating the lateral deformations of gravity caissons in
unimproved soils, as well as at improved sites.
The numbered triangles superimposed on the chart correspond to field case
histories. Information pertaining to the strength of shaking, the wall geometry
and the observed displacements are provided. The relationships provided in
Fig. T7.20 demonstrate the benefit of mitigating liquefaction hazards to gravity
retaining walls. It is also evident that the incremental benefit of a wider zone of
ground treatment begins to decline once the soil improvement extends more than
about two to three times the total height of the wall. At this point, the cost of addi-
tional soil improvement may outweigh the benefits. It is interesting to note that
the soil improvement guidelines prepared by the PHRI (1997) based on shake
table tests, correspond to a normalized width of soil improvement of roughly 1.7
to 2.0.
The recommended procedures for utilizing the results of the parametric study
to estimate the permanent displacement at the top of caissons include:
1) Design the wall using standard pseudo-static limit equilibrium methods to
determine the wall geometry, H:W,
2) Determine Qmax(d) with a site-specific seismic study or approximate with empir-
ical soil amplification factors to yield the peak ground surface acceleration and
the reduction factor, r
d
, recommended by Makdisi and Seed (1978).
3) Determine the magnitude scaling factor for the specified earthquake magni-
tude, and multiply by QmaxCdl to yield the intensity, and
4) Based on the SPT N-values, (N
j
)60, of the backfill soils, the width of the ground
treatment behind the caisson, and the ground motion intensity factor. enter Fig.
T7.20 and obtain the normalized lateral displacement, d/H. From this, the lat-
eral displacement at the top of the wall, d. can be estimated. This lateral defor-
mation includes the contribution of both sliding and tilting on the overall
movement at the top of the wall.
f-w--t--- Lex\ - - -
I
\ / i ~ ' ~
.. ~ . . ..
....
I
. . I ' ~ ~ Improved Soil
H
Caisson f 1_.:.
Loose Soil
I
V\j ,. i RubbleBackfiii""
I
! ........'-""
(Dr =35%)
i ..~
O.2H
,
.,--'0,"" RubbleFoUDdalion i t ~ .
1
H
Dense Soil (Dr =75%)
J
Fig. T7.19. Cross section of model wall.
Table T7.8. Magnitude scaling factors (Arango, 1996).
Magnitude
MSF
5.50
3.00
6.00
2.00
7.00
1.25
7.50
1.00
8.00
0.75
8.25
0.63
(1) Sliding block analysis
In order to evaluate the order-of-magnitude displacement of sheet pile walls based
on sliding block analysis, the procedure used for the gravity quay walls discussed
in T7.2.l poses two difficulties. First, the active earth pressure may not be accu-
rate because of the flexible nature of the sheet pile wall. Second, the inertia force
on and bottom friction at the wall, i.e. two of the major parameters for gravity
quay wall analysis, are not the major parameters in the sheet pile-backfill soil sys-
tem. These problems are solved by idealizing the movement of the sheet pile-
backfill system, assuming united rigid block motion of the wall and backfill
(Towhata and Islam, 1987). As shown in Fig. T7.1 (b), the driving force is the iner-
tia force acting on the soil wedge and the resistance force includes the passive
earth pressures in front of the wall and the shear resistance along the failure sur-
face of the soil wedge.
By expressing the dynamic earth pressure coefficients K
ae
and Kpe and the
ultimate anchor resistance T; as functions of K
a
, K; and T, before the earthquake
(Seed and Whitman, 1970)
(T7.33)
(T7.34)
(T7.35)
the threshold horizontal seismic coefficient is obtained as (Towhata and Islam,
1987)
where
k; = a tan lXae - b +tan( -aae)(1 +b tan aae )
1+c tan os,
a = mJ; + ~ +tYw(Hw+UmbY +L1Up
W
m
(T7.36)
(T7.37)
(N1)60=20
W/H = 0.7
W/H =1.2
r - - -
(N1)60=lO
WIH=O.7
. - - - - WIH= 1.2
0.05 -f='-<=--+----t---\
0.10 h.........,;----4---=-....::---I---::::---=....;..,;-----=.......,---+----i
0.15
0.25 -1-----4-----....1-----+-----1
0.20
2 3 4 5 6
Normalized Improved Distance (Lu/H)
(a)
ClISeNo.
a
lPUDdd
) a-=ftJ/MSF
d/H W/H
I. RokkoIslan
0.54 0.42 0.21 - 0.31
0.7
2. Port Island 0.29 0.22 0.14 -0.19
0.7
3. Akita 0.19
0.]9
0.06 - 0.10 0.75
4. Kushiro
0.]9 (SHAKE 0.19 0.02 -0.04
1.03
0.30
c.
e
f-
CJ
...
-
-
-CIl
342 PLANe
Q.
0
f-
.c
-
-III
-

o.

- C
Q.
o
.!2 ."

_ CIl
f:u

-.c III _
..:l ....
0
III
...
C
0
z
I :' 3 4 5
Normalized Improved Distance(Ln,IH)
000 J..- ...l- ---J
C
ill)
Fig. T7.20. Normalized lateral displacement at the top of the caisson.
(a) Normalized earthquake intensity (arnaxIlJlMSF) for backfill soil SPT
(N
1
)60 = 10.
(b) Normalized earthquake intensity (arna:o.td/MSF) for backfill soil SPT
(N
1
) 60 = 20.
(T7.38)
(17.39)
(T7.40)
where
m =0 for no anchor capacity, I for full anchor capacity (dependant on
the amount of excess pore pressure generation around the anchor)
P
p
=static passive earth thrust
t1U=excess pore water pressure due to cyclic shearing; subscripts a and
p denote pressures within the active and passive soil wedges .
n = 1 when the anchor block is above the water table, and Y!Yb when
the anchor is completely submerged
For definitions of the rest of the parameters, refer to Fig. T7.1(b). Here, it is
assumed that the ground water table behind the wall is the same as that of sea
water-surface (i.e. H; =H
sub
) '
Using the threshold seismic coefficient k, defined for the sheet pile-backfill sys-
tem, displacement of the wall is computed based on Newmark's sliding block
approach.
(2) Simplified chart based on parametric study
In order to enhance current screening tools for deformation-based design of
anchored sheet pile quay walls, a parametric study using a calibrated 2D numeri-
cal effective stress model was conducted to examine the performance of sheet pile
quay walls (McCullough and Dickenson, 1998; McCullough, 1998). In the para-
metric study, sheet pile walls were designed with heights ranging from 7.5 to
15 m, and with horizontal seismic coefficients ranging from 0.1 to 0.16. The
design procedure compiled by Ebeling and Morrison (1992) was used for sheet
pile quay wall design. The coefficients were limited to 0.16, because of the
unpracticable anchor requirements required for the 15 m sheet pile walls with
coefficients of approximately 0.2 or larger. The vertical seismic coefficient was
specified to be zero. The tie-rod length was such that the active failure plane of
the wall did not intersect the passive failure plane of the anchor, with the active
and passive failure planes originating at the base of the wall and anchor, respec-
tively. The anchor was a continuous sheet pile wall, with the same stiffness as the
sheet pile wall. The tie rod spacing was two meters. The distances from the
ground surface to the ground water table and the tie rod were 27 percent of the
wall height, H.
(T7.41)
344 PlANe
The foundation and backfill soils were modeled as cohesionless materials, with
the engineering properties represented by average SPT N-values. A major aspect
of the study was modelling the effectiveness of soil improvement on the computed
wall deformation. The normalized SPT N-values (N)60 for the improved and
dense foundation soils were 30 and 25, respectively, while (N)60 of the backfill
was either 10 or 20.
Five earthquake motions were chosen for use in the parametric study in a sim-
ilar manner as for gravity quay walls, discussed in Section T7.2.1 (3). Some
assumptions made for the parametric study include (a) no pore pressure genera-
tion in the foundation or improved soils, (b) the soil improvement extends to the
depth of the backfill layer, and (c) the tie-rod/whale system will not fail. Based on
approximately 100 parametric models, sensitivity of five design parameters were
evaluated: (a) depth of embedment of the sheet pile wall, D
emb
, (b) stiffness of the
sheet pile wall, EI, (c) length of the tie rod, (d) density of the backfill soil, and (e)
extent of soil improvement, Lex!' A definition sketch of the modeled geometry is
provided in Fig. T7.21.
To increase the applicability of this study, normalized dimensionless factors
were developed. A normalized displacement factor shown in Eqn. (T7.41) was
developed by normalizing the displacements at the top of the wall, d, by the wall
stiffness, EI, total wall height, H + D
emb
, and the buoyant unit weight of the back-
land soil adjacent to the sheet pile wall.
asr
A normalized soil improvement factor, n
e XI
' was also developed by dividing the
extent of soil improvement, Lex!' by the total wall height, H + Demo.
The results of the parametric study are presented in Fig. T7.22. The contour
lines indicate various levels of earthquake intensity for backfill soils with (N)6O of
10 and 20. The normalized earthquake intensity was defined using the same pro-
cedure as that for gravity quay walls in Section T7.2.1(3). The effectiveness of soil
improvement for minimizing quay wall deformations is clearly demonstrated by
the design chart. It is also noted that the incremental benefit of soil improvement
beyond next values of approximately 2.0 decreases considerably. In comparison,
the n
eXI
values as determined from the PHRI (1997) recommendations for the para-
metric study sheet pile quay walls are approximately 1.9 to 2.5.
There are fourteen case histories plotted on the chart, which are arranged
according to the SPT N-values (N\)6o of the backfill soils. Table T7.9 tabulates the
case histories. It should be noted that seven case histories are closely predicted by
the chart, five case histories are significantly over-predicted and only two of the
case histories are significantly under-predicted. These results indicate that the
design chart can be conservatively used as a preliminary design chan or screen-
ing tool.
The recommended procedures for utilizing the results of the parametric study
to estimate the permanent displacement at the top of sheet pile walls include:
l) Design the wall using current pseudo-static methods to determine the wall
geometry (H. Demo. El, and anchor length).
2) Determine amuxidl from a site-specific seismic study or an approximate empiri-
cal relationship.
3) Determine the earthquake intensity factor for the specific earthquake by divid-
ing the magnitude scaling factor into amux<dl'
4) Based on the density of the backfill and the extent of soil improvement.
estimate the permanent lateral displacement at the top of the sheet pile wall, d,
using Fig. T7.22 and the normalized displacement equation (Eqn. T7.41).
The results of the parametric study for non-liquefiable sites can be compared to
the design chart by Gazetas et al. (1990), previously addressed in Section
T7.1.2(l), and shown in Fig. T7.23. The parametric study results for non-liquefi-
able sites are shown by solid stars in this figure, with numbers beside the solid
stars indicating the computed displacements. One group of solid stars (center-
right) is for a 7.5 m wall geometry. Two groups (top-right and bottom-right) are
from the parametric study varying the length of the tie-rod and anchor, and the last
group (left-center) comes from the parametric study varying the depth of embed-
ment. It is evident from Fig. T7.23, that the computed displacements follow a
trend that is roughly consistent with the relationship developped by Gazetas et al.
(1990); i.e. displacement increases for smaller EAI and EPI. This indicates that
the chart by Gazetas et al. (1990) is adequate as a crude, first-order evaluation of
sheet pile walls. However, the parametric study indicates the following limitations
in the chart by Gazetas et al. (1990):
1) Although the chart by Gazetas et al. (1990) indirectly includes effects of earth-
quake motion parameters through EAI and EPI, the sheet pile wall displace-
ment can be more sensitive to the earthquake motion parameters as indicated
by the parametric study.
2) The incremental benefit of sheet pile embedment and tie-rod length beyond the
lengths normally specified in seismic design are not as substantial as indicated
by the chart by Gazetas et al. (1990).
-----L
at
backfill soil
Fig. T7.21. Definition sketch of an anchored sheet pile bulkhead with soil improvement.
346 PlANe
8 2 3 4 5 6 7
Nonnalized Soil Improvement, n exl(l-ext!(H+D
emb
)
1
I
I
I I I. I
Case History
Penetration Resistaf.'ce of the
(N, )IO (b10wsJ30 em)
Unimproved Backfill Soil:
-
o s10
(N, )IO (blowsJ30 em)
0
15
_ 10
20
__ 20
-
U
\
()\
\ \
\
\
\
~ "
\
Ground Motion Intensity Factors
(arnax(dyMSF)
-
\ ~
~ ' "
1/
J
,
I'--...
0.5
.... "
<,
-
0 ~ 3 '
....
~
.... -
0.
11
I'
0.004
0.002
0.010
0.012
0.000
o
"iii
it
II
s:
- lo-
0
C.
0
-
II
-
-
.r; .D
-
?--
16
i ~ 0.008
E .D
II E
u 0-
- +
c.::J:
~ :::::: 0.006
Q ===
~ ~
-
..J
"C
II
l:!
"iii
E
o
z
0.014
Fig. T7.22.. Permanent horizontal displacements at the top of anchored quay walls.
(l) Response spectrum analysis
In simplified dynamic analysis, a pile-supported wharf is idealized by a single
degree of freedom (SDOF) system. A SDOF system consists of a mass, a spring
and a damper as shown in Fig. T7.24. The mass is typically that of the deck and
surcharge. The spring is typically computed based on the equivalent pile-deck
framework assumed for pseudo-static analysis discussed in Section T7.1.3 or
through pushover analysis discussed later.
Dynamic response of a SDOF system is computed using time histories of
design earthquake motions or response spectra. These routines are well estab-
lished in the technical literature (e.g. Kramer. 1996). When the dynamic response
of a structure exceeds the elastic limit, the structure begins to exhibit non-linear
response due to yielding. In simplified dynamic analysis of a pile-supported
wharf, the non-linear response is typically evaluated using linear response spectra
and a ductility factor, which is the maximum displacement devided by yield dis-
placement. This approach is widely used for designing buildings and bridges.
Some of the fundamental issues relevant to this approach will be reviewed below.
Table T7.9. Case histories of sheet pile quay walls.
Earthquake (N
t
)60 amaxld/MSF Displacement
(g) (em)
Normalized
displacement
dElIH+ DemJyo)
1968 Tobchi-Oki
1993 Kushiro-Oki
1964 Niigita
1983 Nihonkai-Chubu
1968 Tokaehi-Oki
1968 Tokaehi-Oki
1968 Tokaehi-Oki
1973 Nemuro-Hanto
1978 Miyagi-Ken-Oki
1968 Tokaehi-Oki
1983 Nihonkai-Chubu
1993 Kushiro-Oki
1993 Guam
1993 Kushiro-Oki
6
6
10
10
15
15
15
15
15
20
20
20
20
30
0.26
0.20
0.14
0.10
0.23
0.23
0.12
0.16
0.10
0.26
0.09
0.16
0.18
0.21
12 to 23 0.0120
19 0.0020
100 0.0031
110 to 160 0.0066
16 to 57 0.0090
12 to 19 0.0029
30 0.0077
30 0.0031
87 to 116 0.0129
-60 0.0049
-5 0.0006
50 to 70 0.0044
61 0.0019
no damage (- 5) 0.0004
In order to approximate the effect of elasto-plastic response, the following
assumption is often made: the energy stored in a SDOF system is the same for lin-
ear and elasto-plastic systems. This assumption, called the constant energy
assumption, is illustrated in Fig. T7.25. Consider a linear system shaken by an
earthquake motion resulting in a response with a maximum displacement d' and
a maximum force F'. Consider a response of an elasto-plastic system with elastic
limit of F; undergoing the same earthquake input motion resulting in a response
with a maximum displacement d
m
With the constant energy assumption, the max-
imum displacement d
m
of the elasto-plastic system is determined in such a way
that the area of the triangle OA'B' is the same as that of OYAB.
Energies stored for linear and elasto-plastic systems are (l/2)F'd' and
(l/2)Fydl2f.1-l), where J1 =dm/d
y
denotes displacement ductility. By introducing
a spring coefficient k for the linear system, it follows that F' =kd' and F; =kd.,
so that the constant energy assumption demands
This equation reduces to the following relationship.
F
y
= ,---tF'
\}2"i=l
(T7.42)
(T7.43)
Since inertia force is proportional to acceleration, Eqn. (TI.43) is converted to a
relation between peak response accelerations of linear and elasto-plastic systems
a' and a
y
as follows.
o 0.5 1
Embedmentparticipationindex(EPI)
*referto Table TI.3 (a) and (d)
348 PIANC
2.5
2
o
-0.5
o
o
o
o
o
o
o
~ 7
Degree of damage"
o 0
[J 1
e 2
3
4
o
Fig. T7.23. Case history data of
sheet pile quay walls,
including data from the
parametric study for
models improved sites
(solid stars with com-
puted lateral deforma-
tions (em.
(T7.44)
A parallel relationship for displacements of yield limit and maximum response
are given by the definition of ductility factor as follows.
d
m
=).U1y (T7.45)
Since residual displacement d, of the elasto-plastic system is given by
d, =d
m
- d., Eqn. (T7.45) demands d, =(j.1- 1)d
y
By introducing a modification
factor c to correct for the difference between dynamic response and the constant
energy assumption, the residual displacement d, is given by
(T7.46)
For example, the correction factor for a bridge supported by concrete piers is
typically c = 0.6 (Japan Road Association, 1996).
The allowable (displacement) ductility factor J1 for use in design is generally
determined from ultimate displacement d; with allowing a safety margin;
typically J1 = 2 to 4 for bending and J1 = 1 for shear. For further discussions on
ductility criteria, see Section 4.3 in the Main Text.
(a)
d
\ . l M - ~ . Amax
L.W.L +0.00
ali inal Ground -4.50 I
Steel Pile D=700mm I
Concrete Pavement
Rubbles
- 2.0 -22.0 ! -22.0
Pile 1 Pile 2 Pile 3
+
(b)
Fig. T7.24. Modelling of a pile-supported wharf for simplified dynamic analysis.
(a) SDOF system.
(b) Pile supported wharf.
The ultimate displacement d
u
is governed by ultimate curvature CPu for piles hav-
ing a plastic hinge length L
p
with a free pile length L as follows (Priestly
et al., 1996):
d; =d, +3(CPu -CPy)4 (L -0.54) (T7.47)
Plastic hinge length L
p
is proportional to a diameter of pile D
p
or given as L
p
=0.1
to 0.5D
p
for a concrete pile (Japan Road Association, 1996).
oJ Equation (T7.47) forms the basis to relate the ductility limit specified in curva-
ture with displacement ductility limit 1.1. The curvature ductility limit is further
350 PlANe
Force
F
Linear system
F'+----+-__e
A
'
N
~
~
~
y ~ ~ (E:sto-PlaSliC system
o
d'
.B
d
m
Displacement
d
Fig. T7.25. Load-displacement response of
elasto-plastic system evaluated
with constant energy assump-
tion.
related to the strain limit through the relationship between the extreme fibre strain
and the curvature as
(T7.48)
Equations (T7.47) and (T7.48) were used as the basis to develop the damage cri-
teria for piles discussed in Section 4.3 in the Main Text. For further details, refer
to Priestly et al. (1996), Ferritto (1997a), Ferritto et al. (1999), and Yokota et al.
(1999).
(2) Pushover analysis
In simplified dynamic analysis. the load-displacement relationship for the pile-
deck system shown in Fig. T7.25, is evaluated by pushover analysis. Pushover
analysis is accomplished by performing a multi-stage pseudo-static analyses with
an increasing level of external load. In this analysis, the structure is generally ide-
alized by a beam-deck framework with or without the subgrade reaction springs
shown in Fig. T7.9 to take into account the reaction on the embedded portion of
the structure. With an increasing level of external load, the sequence of yield in
the structures and a transition from elastic response to the ultimate state of failure
will be identified. The yield generally begins from the pile heads most landward
to those seaward and then moves down to the embedded portion of the piles.
Figure T7.26 shows an example of the results of a pushover analysis.
Simplified dynamic analysis of pile-supported deck may be accomplished
through the following steps:
1) evaluate the elasto-plastic parameters d; and Fy defining the load-displace-
ment relationship shown in Fig. T7.25 based on the results of pushover analy-
sis or other simplified procedures;
2) evaluate the seismic response of SDOF based on non-linear response analysis
using a time integration technique and obtain the ductility factor u through
Eqn. (T7.45). Alternatively, use the linear response spectra to obtain the peak
response acceleration a' and obtain the ductility factor through Eqns. (T7..+3)
or (T7.44);
3) compare the response ductility with the relevant ductility criteria/strain limit.
In simplified dynamic analysis based on response spectra/pushover analysis,
the movement/deformation of the dike and the zround or the soil-structure inter-
...
action is not fully taken into account. The effects of these issues should be con-
sidered in the final evaluation of the results of the simplified dynamic analysis.
(3) Ductility criteria
. Extensive studies have been performed to establish ductility criteria for pile-sup-
ported wharves (Ferritto, 1997a; Ferritto et al., 1999, Yokota et al., 1999). The
damage criteria defined in terms of strain in piles may be given as shown in Table
T7.10. In this table, the design ultimate compression strain of confined concrete
may be taken as
Ccu=0.004 +(1.4psjYh
C
snJ . (T7.49)
where
p; effective volume ratio of confining steel
/yh: yield stress of confining steel
c
sm:
strain at peak stress of confining reinforcement, 0.15 for grade 40
and 0.12 for grade 60
fcc: confined strength of concrete approximated by where t: is
28 days unconfined strength of concrete.
The serviceability limit shown in Table T7.10 corresponds to an essentially elas-
tic state, but it does not imply a rigid requirement that the strains in all parts remain
below yield limit. Concrete structures may be considered serviceable, without any
significant decrement to structural integrity, provided concrete strains reached dur-
ing maximum seismic response do not cause incipient spalling of concrete cover,
and if residual crack widths after the earthquake are sufficiently small to negate the
need for special corrosion protection measures such as crack grouting. This latter
requirement implies that significantly larger crack widths might momentarily exist
during seismic response, but these have no effect on corrosion potential. Thus the
serviceability limits shown in Table T7.10 are based on strain limits sufficiently low
so that spalling of cover concrete will not occur, and any residual cracks will be fine
enough that remedial grouting will not be needed. The specified incremental strain
for prestressing strand is less than for reinforcing steel to ensure that no significant
loss of effective prestress force occurs at the serviceability limit state. For either hol-
low or concrete filled steel shell piles, it is recommended that the maximum tension
strain in the shell be also limited to 0.01 to ensure that residual displacements are
negligible. For the steel shell pile connected to the deck by a dowel reinforcing bar
detail, the limit strains for reinforced concrete apply.
352 PLANe
Collapse Umit
I
d
d3
Displacement at pilecap

\
: pile 1 : in-ground plastic hinge (@)
I pile 2 : in-l"OUnd plastic (@)
N
pile 1 : plastic hinge at pile
: pile 3 : in-ground plastic hince (3 :
I pile 2 : plastic hince at pile cap 2 I
I I I
:i \pile 3:: plastic hinge at pile cap (2) :
I :"'" pile 3
1
: yield fibre stress at pile cap (CD) I
I I I I
I I I I
I I I I
I I I I
I I I I
------------------------
F
F3
"'0
ctl
.9
F2
ca
...
(1)
F,
1ti
...J
(a)
CD yieldfibre stress
at pilecap
d, Bending
M Moment
Stress Distribution
in
tjJl CT y
plastic hinge
at pilecap
d2
@ in-ground plastic hinge
I
I
I c:::>
1r (D " - D '4)
M
- p p CT
y - 320 y
p
D
p
: Outer Diameter
D
p
' : Inner Diameter
CT y : Yieldfibre stress
(0
3
- 0 '3)
M
- p p CT
p- 6 y
(bl
Fig. T7.26. Pushover analysis of a pile-supported wharf.
(a) Load-displacement relationship.
(b) Sequential change in bending moment in piles.
Table T7.lD. Strain ductility criteria for pile-supported wharves (Ferritto et al., 1999; with
modification}.
Extent of
damage
Parameters Upper
limit
Concrete extreme fibre compression strain
Reinforcing steel tension strain
Prestressing strand incremental strain
Structural steel fibre compression strain
(pile and concrete filled pipe)
Hollow steel pipe pile fibre compression strain
Degree I
(Serviceable)
Degree II
(Repairable)
Pile-deck
hinge
Embedded
portion
(in-ground
hinge)
Prestressing strand
Hollow steel pipe pile fibre compression
strain
Prestressing strand
Hollow steel pipe pile fibre compression
strain
0.004-
O.OlD
0.005
0.008
0.008
Eqn. (T7.49),
but <0.025
0.05
0.04
0.035
0.025
Eqn. (T7.49),
but <0.008
O.OlD
0.015
0.035
0.025
For repairable damage state, limited permanent deformation are acceptable.
Distinction needs to be made between the pile/deck hinge and the in-ground hinge
locations, since access to the latter will frequently be impractical after an earth-
quake. The criteria shown in Table T7.10 reflect these considerations. For further
details on strain ductility criteria, refer to Technical Commentary 5.
Displacement ductility is related to curvature/strain through Eqns. (T7.47) and
(T7.48) and is inversely proportional to pile length. Based on the common dimen-
sions of pile-supported wharves, the strain ductility criteria shown in Table T7.10
may be converted to displacement ductility criteria.
In Japanese practice, it is common to use steel piles for pile-supported wharves.
Displacement ductility limits shown in Table T7.11 have been adopted for use
with the simplified analysis discussed in 'Section T7.1.3 for typical pile-supported
wharves with vertical piles (Ministry of Transport, Japan, 1999). In this table, the
displacement ductility criterion for Level 2 earthquake for use with the simplified
analysis is given by
354 PlANC
J1 = 1.25+62.5.!L., but J1 S 2.5
o,
where
'-"
(T7.50)
t
p
: thickness of steel pipe pile
D
p
: pile diameter
For use with the simplified dynamic analysis, including pushover analysis, the
strain ductility limit is given by
t
E
max
=0.44 -p (T7.51)
D
p
The upper limit for Degree II damage is defined when the embedded portion of
the first few piles reach at the strain limit specified by Eqn. (T7.51).
In most applications involving waterfront slopes and embankments, it is necessary
to estimate the permanent deformations that may occur in response to cyclic load-
ing. Allowable deformation limits for these slopes will often reflect the sensitiv-
ity of adjacent structures, foundations and other facilities to these soil movements.
Enhancements to traditional pseudo-static limit equilibrium methods of embank-
ment analysis have been developed to estimate embankment deformations for
soils which do not lose appreciable strength during earthquake shaking
(Ambraseys and Menu, 1988; Makdisi and Seed, 1978; Jibson, 1993).
As dicussed earlier for gravity and sheet pile walls, rigid body, sliding block
analyses, which assume that the soil behaves as a rigid, perfectly plastic material,
can be used to estimate limited earthquake-induced deformations. Numerical
studies based on the sliding block method of analysis have led to the development
of useful relationships between ground motion intensity and seismically-induced
deformations (e.g. Ambraseys and Menu, 1988; Makdisi and Seed, 1978; Jibson,
1993). The relationship proposed by Makdisi and Seed is shown in Fig. T7.27.
Table T7.11. Displacement ductility factor (Ministry of Transport, Japan. 1 9 9 9 ~ with
modification).
Reference earthquake
level
Level 1
Level 2
Importance of structure
(see Table 3.3)
Special class
Class A
Class B
Class C
Special class
Limit displacement
ductility factor
1.0
1.3
1.6
2.3
Eqn. (T7.50)
Given that the sliding block analyses are based on limit equilibrium techniques,
they suffer from many ?f the same limitations previously noted for pseudo-static
analyses. One of the primary limitations with respect to their application for sub-
marine slopes in weak soils is that strain softening behaviour is not directly incor-
porated in the analysis. The sliding block methods have. however. been applied
for liquefiable soils by using the post-liquefaction undrained strengths for sandy
soils. A recent method proposed by Byrne et al. (1994) has been developed for
contractive. collapsible soils that are prone to liquefaction.
Examples of displacement criteria for use with the sliding block analysis of
dikes supporting pile-supported wharves used for Navy Facilities and Oil
Terminals are 7 to 15 em for the serviceablility limit and 15 to 30 em for
repairable state limit (Ferritto, 1997a, 1997b; Ferritto et al., 1999).
T7.3 DYNAMIC ANALYSIS
T7.3.1 Anal ysis procedures
Dynamic analysis, generally using finite element or finite difference techniques,
involves coupled soil-structure interaction, wherein, the response of the founda-
tion and backfill soils is incorporated in the computation of the structural
response. A structure is idealized as either a linear or non-linear model, depend-
ing on the level of earthquake motion relative to the elastic limit of the structure.
The stress-strain behaviour of the soil is commonly idealized with either equiva-
lent linear or effective stress constitutive models, depending on the anticipated
strain level within the soil deposit. Fairly comprehensive results can be obtained
from soil-structure interaction analysis, possibly including failure modes of the
soil-structure systems, extent of displacement, and stress/strain states in soil and
structural components.
Issues regarding structural modelling of port structures, either through linear or
non-linear models, are similar to those for other structures, including buildings,
bridges and offshore structures. For example, modelling of a caisson for gravity
quay walls is generally accomplished using a linear model. Modelling of sheet
pile walls, cellular walls and pile-supported wharves often requires an elasto-plas-
tic type non-linear model to allow ductile response of the structure beyond the
elastic response limit. Special attention is needed for pile-deck connection details
for designing pile-supported wharves. For further details, refer to Technical
Commentary 5.
Geotechnical modelling of foundation and backfill soil, either through linear'
(equivalent linear total stress), total stress non-linear or effective stress models, is
a primary consideration when evaluating the seismic performance of port struc-
tures. Total stress analysis procedures, either equivalent linear or non-linear, do
356 PlANe
1000.00
100.00
-
E
~ 10.00
-C
Q)
E
Q)
o
ll)
0..
(fl 1.00
i5
10.000
3.2
1.000
-
~
E
Q)
0.100
E
Q)
0
~
Q.
(fl
s
0.010
0.10
0.001
. I . I
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
kt/ke
Fig. T7.27. Empirical relationships between permanent displacement of sliding block and
ratios of accelerations (after Makdisi and Seed, 1978).
not include the effects of change in excess pore water pressure or effective stress
during shaking, and therefore changes in the soil stiffness and strength are not
accounted for. Among the total stress analysis procedures, the equivalent linear
procedure has been the most widely used technique in practice for computing the
dynamic response of soil deposits, earth embankments, and soil-structure interac-
tion. This is based on a linear analysis, using strain level dependent shear moduli
and damping. The computer code, FLUSH (Lysmer et aI., 1975), is one of the
most widely used. Limitations in the equivalent linear analysis include:
- residual displacements of soil-structure systems are not computed: and
- applicability beyond the shear strain level of about one percent is questionable.
The shear strain level of one percent corresponds, for example, to a state where
level ground with a 10mdeposit deforms 0.1 m. This deformation is considerably
smaller than those discussed for performance criteria for port structures in
Chapter 4 of the Main Text. Those who use the equivalent linear analysis should
be aware of these limitations.
Modifications to a total stress analysis have been applied to overcome
the limitations in the equivalent linear analysis. These modifications relate
to: (I) incorporating the effect of excess pore water pressure increase in terms of
reduced shear moduli in the equivalent linear analysis (e.g. FLUSH-L (Ozutsumi
et al., 2000)), and (2) using reduced shear strength in the non-linear analysis such
as using easily available computer codes (e.g. FLAC (ITASCA, 1995)). These
total stress computer codes have an improved range of applicability over equiva-
lent linear analysis. Certain limitations still remain because the effects of a pro-
gressive increase in excess pore water pressures are not included.
Non-linear, effective stress analysis methods are the most ideal for analyzing
residual displacements and/or evaluating performance of structures beyond the
strain level of one percent in soil. Many computer codes have been developed and
utilized in practice. However, users should be aware of the fact that most of these
computer codes are still under development and useful only as a research tool. It
is best to consult with specialists in geotechnical earthquake engineering and
effective stress analysis to discuss the stage of development of the computer code
and the effective stress model and then decide whether the computer program is
appropriate for use in practice. Proceedings of the relevant conferences, including
Japanese Geotechnical Society (1989,1991) and National Science Foundation,
USA (Arulanandan and Scott, 1993), might offer relevant information. The final
decision on what computer code should be used for the analysis will depend on
the following issues:
- availability of solid theoretical background outlined in technical literature;
- availability of solid procedures for determining analysis parameters from
commonly used geotechnical investigations;
- applicability of the procedure for well documented case histories of seismic
performance of port structures;
- availability of computer code (preferably widely used and tested by non-spe-
cialists/consulting engineers other than the special group of researchers who
originally developed it).
For more details, refer to Technical Commentary 4.
T7.3.2 Examples
The following are examples of dynamic analyses for typical port structures.
(1) Gravity quay wall
During the 1995 Hyogoken Nambu earthquake in Kobe, Japan, many of the cais-
son walls suffered damage as shown in Fig. T7.28. These caisson walls were con-
structed on a loose saturated backfill foundation of decomposed granite, which
was used for replacing the soft clayey deposit in Kobe Port to attain the required
foundation bearing capacity. Subjected to a strong earthquake motion having peak
accelerations of 0.54g and 0.45g in the horizontal and vertical directions, these
\
358 PIANC
caisson walls were displaced an average of 3 m (maximum displacement - 5 m)
toward the sea, settled 1 to 2 m and tilted about 4
0
toward the sea. Although a slid-
ing mechanism could explain the large horizontal displacement of the
caisson walls, this mechanism did not explain the large settlement and tilt of the
caissons. Reduction in the stiffness of foundation soils due to development of
excess pore water pressure was speculated as a main cause of the observed
caisson damage at Kobe Port. .
This speculation was confirmed by a series of effective stress analyses using the
computer code FLIP. The model parameters were evaluated based on in-situ
velocity logging, the blow counts of Standard Penetration Tests (SPT N-values)
and the results of cyclic triaxial tests. The specimens used for cyclic triaxial tests
were undisturbed samples obtained by an in-situ freezing technique. Input earth-
quake motions were those recorded at the Port Island site about 2 km from
the quay wall. The spatial domain used for the finite element analysis covered a
cross-sectional area of about 220 m by 40 m in the horizontal and vertical
directions.
The effective stress analysis resulted in the residual deformation shown in
Fig. T7.29. As shown in this figure, the mode of deformation of the caisson wall
was to tilt into and push out the foundation soil beneath the caisson. This was con-
sistent with the observed deformation mode of the rubble foundation shown in
Fig. T7.30, which was investigated by divers. The order of wall displacements was
also comparable to that observed and shown in Fig. T7.28.
In order to evaluate the overall effect of geotechnical conditions on the
displacements of a gravity wall, the following three analyses were performed
as a parameter study. The parameter study included a virtual soil model, to
be called non-liquefiable soil, which has the same properties as those used
in the aforementioned analysis but without the effect of excess pore water
Backfill Soil
Sand Drain
Unit(m)
Fig. T7.28. Cross section of gravity quay wall at Kobe Port and defonnationlfailure
during 1995 Great Hanshin earthquake. Japan.
pressures. To distinguish the cases in the parameter study, the case which
dealt with the actual quay wall during the earthquake described earlier is
designated as Case 1. Cases 2 through 4 are defined depending on the
extent of the non-liquefiable soil relative to the caisson wall, as shown in
Fig. T7.3l.
Major results of the analysis are summarized in Table T7.12, including those of
Case 1. These results indicate that the deformation of the gravity wall may be
reduced up to about one half of that actually observed, if the excess pore water
pressure increase was prevented in the subsoil as in Case 2. In particular, the effect
of the pore water pressure increase in the foundation soil beneath the caisson wall
is about twice as much as that of the backfill as understood from the comparison
of Cases 3 and 4. These results were partly confirmed by the seismic performance
of the quay walls at Port Island (Phase II), where a caisson wall had been placed
on a foundation improved by the sand compaction pile technique before the earth-
quake (lai et al., 1996).
-36.0m
+4.0m
Lateral
Oisplacement
3.5m
1-1
Inclination
4.1"
V rtiealOsplac ment 1 5 e I e m :z: -r ...... _

..: : : :
! -
-
I I I
. -
- I
,
I
- -

.
r
III
-
.

- -
-
.
- -
- -
.'
-
-
. .
.
-
.
,
Fig. T7.29. Computed deformation of a gravity quay wall.

ro(tt\e
f
- - - Before Earthquake
- After Earthquake

l>
1.984- :
II
!: Caisson
ited Soil II
.I
II
, I
LW.L:O.O
Cross Section of
theQuay Wall
+4.
.
Fig. T7.30. Deformation of rubble foundation of a quay wall investigated by divers.
360 PlANC
For variations in the peak acceleration of the earthquake motion input at the
base, the horizontal residual displacement at the top of the caisson wall was
computed as shown in Fig. T7.32. These response curves constitute the basis for
performance based design. For example, let us suppose a caisson quay wall is con-
structed at a site where seismic hazard analysis resulted in 0.2g and 0.3g for
Levels 1 and 2 earthquakes, respectively. For a Case 1 quay wall, horizontal dis-
placements at the top of the caisson will be 1.3 and 2.1 m for Levels 1 and 2 earth-
quakes, which do not satisfy the performance criteria discussed in Chapter 4 in the
Main Text. For a Case 2 quay wall with full liquefaction remediation measures,
displacements will be reduced to 0.3 and 0.9 m for Levels 1 and 2 earthquakes,
which satisfy the performance criteria.
The effect of the vertical component of input acceleration time histories was
studied by varying the peak acceleration of the vertical component of acceleration
ranging from zero to 0.40g (!chii et al., 1997). The peak acceleration of the hori-
zontal component was unchanged from the original value of 0.54g. The phase dif-
ference between the vertical and the horizontal components were also varied. This
parametric study suggested that the effect of the vertical component is to increase
the residual displacement less than about 10%.
CASE2
CASE-3
Loose Saunted Sou
I I
! I
CASE4
Fig. n .31. Conditions assumed for parametric study. Case 2 through 4.
Table T7.12. Major results of parametric study for gravity quay wall.
Case
Case I
Case 2
Case 3
Case 4
Residual displacements of caisson
Horizontal Vertical
(m) (m)
3.5 1.5
1.6 0.6
2.1 0.7
2.5 1.1
4.1
2.4
3.1
2.2
100 200 300 400 500 600
Input Max Acceleration (Gal)
-.- Case1
3.5 - 0 Case2
.....- Case3
3 - "0" Case4
, , ,
.... --.-..... --- ... -....... -.....--_..'- ......
, ,
, ,
, ,
,
Fig. T7.32. Effects of input accelera-
tion levels on horizontal
residual displacement.
(2) Sheet pile quay wall
During the 1983 Nihonkai-Chubu earthquake in Japan, many anchored steel sheet
pile walls suffered damage at Akita Port. Most of the walls were constructed by
backfilling clean sand dredged from the nearby seabed with placed SPT N-values
ranging from 20 to 50. During the earthquake, these walls were shaken by an
earthquake motion with peak accelerations of 0.24g and 0.05g in the horizontal
and vertical directions. Figure T7.33 shows a typical cross-section and deforma-
tion of an anchored sheet pile wall. The backfill liquefied and the sheet pile wall
moved 1.1 to 1.8 m towards to the sea as shown by the broken lines. The sheet
pile wall yielded and a crack opened about 6 m below sea level.
An effective stress analysis using FLIP resulted in the deformation shown in
Fig. T7.34 (Iai and Kameoka, 1993). Increases in earth pressures and bending
moments were also computed as shown in Fig. T7.35. In this analysis, the model
parameters were evaluated based on the SPT N-values arid cyclic triaxial test
results. The wall and anchors were modeled using linear beam elements and, thus,
there was a limitation in simulating the seismic response of the wall after yield of
the wall. It is to be noted, however, that the displacement computed at the top of
the wall was 103m, consistent with that observed. The maximum bending moment
362 PIANC
computed in the sheet pile wall exceeded the yield level, also consistent with that
observed.
In order to discuss the overall effect of the geotechnical conditions in the
backfill, a parameter study was performed by varying the SPT N-values of the
backfill. The maximum bending moment in the sheet pile wall was reduced for
higher SPT N-values as shown in Fig. T7.36. In this figure, N
6S
designates SPT
N-values 'corrected to the effective overburden 'pressUre of 65 kPa, which is
widely used in the design practice in Japanese ports (Port and Harbour Research
Institute, 1997). The evidence to support this result was available from the case
history of an undamaged quay wall at Akita Port (lai and Kameoka, 1993).
(3) Pile-supported wharf
During the 1995 Hyogoken Nambu earthquake of 1995 in Japan, a pile-supported
wharf suffered damage at Takahama Wharf in Kobe Port. The horizontal residual
displacement of the wharf ranged from 1.3 to 1.7 m, with a typical example of the
cross-section and deformation of the pile-supported wharf shown in Fig. T7.37.
As shown in this figure, the wharf was constructed on a finn foundation deposit
consisting of alternating layers of Pleistocene clay and sandy gravel. The SPT
N-values ranged from 10 to 25 for the clay and 30 to 50 or higher for the sandy
gravel. The firm deposit was overlain by an alluvial sand layer with SPT N-values
of about IS, the thickness of which was variable, of about two meters on average.
Behind the retaining wall made of concrete cellular blocks was a hydraulic
.-. Before earthQuake
- AfterEarthQuake
20.0m
.,' "I
I
. ::E:: I
: :&(): :
I: .:.
I .It':l.,

::.!!: : -

" .'
...___:i eli

:: II::
.:-:.
::e-::
::e::
::0::

: :-e.= :
:: .:

: :c..::

0-
ir
BackfiR Sand
-16.0
, '
, '
, .
"
" , '
:'

:Cl:
, -
, a>
: a>
, -
:U5
L-;..--.- ------------- ...-.1"--_
l.W.L O.O ::!: ':
. ..........__._--_...._._-..._._.
Opening of
a Crack -
Unit (m)
Fig. 17.33. Cross section of an anchored sheet pile quay wall at Akita Port and
damage/deformation during 1983 Nihonkai-Chubu earthquake. Japan.
AnalysisMethods 363
O.Om
-10.2m
23.0m
C,
E
.....,--,
g
i
17"'i
-
I - ;.r- ____
1/ / ,- j/V/1/ /'/

-t- _Ij..-"

! /1/:// /1/ r> - -- -- -
. J,..-ll :""'"T-
',: '/ 1../:/' /- 1/,"1/:/
.. m
t:==.J:9
I /i /1, /! /'/ /1. /'/' JI
-f1'" I'
., _.\ - I
l :" ./ t,' /,/!-/

I l I I
I/i/
l

I i
!. ' I
I ...J
" , .. __ .=P-,., I! i '
I
!
!
I I -;;. GI! l j I \
i
j,
I
..-
I ,
I i Ii I
I'
t
:
I
'I j;
I \
" \i
II
,
"
,
I
I \
I
\,
"
:
i
!
I i
I
I i
I
\ I \ I i \
,
Scale of Deformation I,Om
. Fig. T7.34. Computed deformation of a sheet pile quay wall.
-10,2m
Os
5s
15s
30s
Tie Rod
Moments (kN' m)
o 1000
I I
Yield Point
-300 150 0 150 300
I I I !
Bending Stresses (MPa)
(b)
Bending
-2000 -1000
I 1
(a)
Os
155
30s
Earth Pressures (kPa)
-200 -100 0 100
I
:O.Om
300

c--
._._._.. _---
Fig. T7.35. Computed earth pressures and bending moments in a sheet pile quay wall.
(a) Earth pressures.
(b) Bending moment.
backfill of decomposed granite with SPT N-values of about 10. The deck of the
quay was made of reinforced concrete slabs and beams supported by steel pipe
piles having a diameter of 700 mm.
The steel piles buckled at the pile heads, except for the piles located most
landward. A crack was found at the pile - concrete beam connection located
most landward. The piles, pulled out after the earthquake for investigation, shown
in Fig. T7.37(b), also showed buckling below the mudline at the depths shown
in Fig. T7.37(a). As shown in this figure, some of the buckling was located
close to the boundary between the layers of alluvial sand and Pleistocene
gravel. Displacements of the rubble dike, measured by divers at five locations 5 m
apart; were about the same as those of the deck as shown in Fig. T7.37(c).
364 PlANe
500
0
I I I
0 2 4 6 8 10 12
14{N)
0.66
I I I I I I J
35 40 45 50 55 60 65
Dr (%)
1000
2500
E
eti
Q.
::.. 2000
'E
CD
E
o
~ 1500
C'l
c
:c
c
CD
co
E
::l
E
'x
co
~
Elastic Limit
Fig. T7.36. Reduction of maximum bending moment with increasing SPT N-valuesl
density.
The backfill behind the retaining structure settled about 1 m. These measure-
ments indicate a somewhat uniform movement of the dike and the retaining wall
toward the sea.
Effective stress analysis using FLIP resulted in the residual deformation
shown in Fig. T7.38 (Iai, 1998b). In this analysis, the model parameters were eval-
uated based on the SPT N-\'alues. The piles were idealized using elasto-
plastic (bi-linear) beam elements. Linear springs were included to account for
the three-dimensional interaction between the piles and the rubble dike. The
stiffness of the springs used for this particular analysis corresponds to the
coefficient of horizontal subgrade reaction k
h
-
sub
= 147 kN/m
3
The computed
deformation of the dike and the retaining wall, shown in Fig. T7.38, is associated
with the significant deformation in the alluvial sand layer, which is consistent
with the observed deformation of the pile-supported dike shown in Fig. T7.37.
The bending moment and curvature of the piles accumulated gradually. reaching
the residual values shown in Fig. T7.39. These residual values were close to
the peak values during shaking. The computed locations at which the piles
yielded (shown in Fig. T7.39) are also consistent with those observed, shown
earlier in Fig. T7.37. In particular. the bending moment at the pile heads located
most landward was not large, contrary to the results obtained by conventional
design practice taking into account only the effect of the inertia force on the
deck. The displacement of the dike obviously redistributed the bending moments
among the three rows of piles.
Pleistocene Cia 1
Pleistocene Gravel (2)
Pleistocene Clay (2)
- After Earthquake
- - - Before Earthquake:
Local Buckling
(Unit in meters)
5.0 1.0,2.0
! I
: I.
(a)
0.566
(b)
TAKAHAMA
WHARF
sz
15m m
H 0-
_. -I iNA
- -21 O.
1.5m m
H 0,
~ _. -I J N C
- -21 o.
1.5m m
tI 01
_ - - '2
1]
NO.D
- I
1.5m m
f--t 0,
I ~ _. -I; No.E
- 21
- - . Before Earthquake
--After Earthquake
(c)
Fig. T7.37. Cross section of a'pile-supported wharf at Kobe Port and deformation/failure
during 1995 Great Hanshin earthquake, Japan.
(a) Cross section and deformation/damage.
(b) Plan of pile-supported wharf.
(c) Displacements in rubble dike.
(4) Breakwater
Similar to the gravity walls. mentioned earlier, the composite breakwaters in
~ - _ . _ - - _ . . - _ ~ ~ -
Kobe Port were constructed on a loose decomposed granite, which was back-
filled into the area excavated from the original alluvial clay layer. The thickness
of the clay layer reaches a maximum of 25 m in Kobe Port as shown in
Fig. T7.40. During the 1995earthquake, the breakwater settled about 1.4 to 2.6 m.
366 PlANC
.. 1.20m .. 1.20m
lO.004m lO.10m
--_..__.._ - - ~ - . - - - - - - - - - - - - - - - - _ . -
~ . ~ .
-
.. .
-
. .
+O.Om
I
v
, ,
. .. ...
- ,
I
- - ,
..
- - - ,
-

... -
- -
.
-
-
_.
-
-
-
-
'.
,
-7.5m
- -
l;4.
-
-25.0m
Fig. T7.38. Computed deformation of a pile-supported wharf.
1- ------- ElasticLimit I 1- -------Elastic Limit I
Elevation(m} Elevation(m}
4
4
2
I
,
I I
2
-, - -
I Q)
-
0
I .= 0
Q)"
Q)I I
Ill.
-2
::-
~
-2 =1 I 1"0
c,
ll.1
I
I ~
~
'"0 0::
~
"01
I ~
...
I
Ctl
...
-s
~ ,
I
'"0
~
s
(l)
ctl,
c:
~ I
I Ie Ctl
-e
I 1
ctl oS (l)
(l)
~ I I I ..J en U
-10 enl I I 10
I
'-
12
-12 I
I
14 I
-14
I
-16
16
I
I
-18.
2
.
5
,
I
18
0 2.5 -2.5 0 2.5 -2.5 0 2.5 -0.1 0 0.1 -0.1 0 01 -0.' 0 0'
-,
Bending Moment (MNm/pile) Curvature (m )
(a)
<h)
Fig. T7.39. Computed residual bending moments and curvatures in piles.
(a) Bending moments.
(b) Curvatures.
The horizontal displacements of the breakwater. however. were less than
tens of centimeters. The mode of deformation suggested that the caisson was
pushed into the rubble foundation and the rubble was also dragged down
and pushed into the loose deposit below' it. Although the caissons used for
the quay walls and the breakwaters were similar to each other. the modes
of deformation were different. The effect of backfill or lack of it was presumed to
be the cause for this fundamental difference.
The effective stress analysis using the computer code FLIP resulted in
the deformation shown in Fig. T7 Al (Iai et al.. 1998). The deformation
mode is consistent with that observed and shown in Fig. T7.40. The settlement at
the top of the breakwater accumulated gradually and reached 1.3 m
while the computed horizontal displacement at the top of the breakwater
remains very small. The results are basically consistent with the observed
performance.
Unit em)
Clay
.__.... Before Earthquake
--After Earthquake
Sl L W L "'000
'J HW.L+1.70
Clay

'a-
c0
<0
N -0.64
8.70 :
. . .
n -9.20.--- i : : _._. 000 1000
... .m_ J:- :::. Lo.k . .!-< I 1 I' I
.. I 11,70 I '.
... .::.>:3>:':-
";$ for Replacing Alluvial Clay Layer ".;,./

I 15.70 I
Fig. T7AO. Cross section of a breakwater at Kobe Port and deformation during 1995
Great Hanshin earthquake, Japan.
ke
-40.00
40.00(m) 30.00 20.00
40
10.00
..
0.00 20.00
--,-, ... -.-, + 0
.......
Before Earthqua
After Earthquake
fill
.
-9.40
.. .
R
I
,
I
,
I I I I
,
I I I I I I I I
,
I I
I

I
.
I
, I I
I
I I I t .
, I
I I
I I I I I I
i
I I
.I I
I
.I
,
.
I I I I
,

I I I I
I
I I
.
, ,
I
.
I I I
,
I
,
I
.
I I J I
. : t I
!
I I I I
i
, I
I I I
, .
: ,
I
I
-30.00
Fig. T7AI. Computed deformation of a breakwater.
368 PlANe
T7.3.3 Remarks on dynamic analysis of slopes and walls
In situations where the movement of a slope has an impact on adjacent structures,
such as pile-supported structures embedded in dikes, buried lifelines and other
soil-structure interaction problems, it is becoming more common to rely on
numerical modelling methods to estimate the range of slope deformations which
may be inducedby design level ground motions (Finn, 1990). The numerical
models used for soil-structure interaction problems can be broadly classified
based on the techniques that are used to account for the deformations of both the
soil and the affected structural elements. In many cases, the movement of the soil
is first computed, then the response of the structure to these deformations
is determined. This type of analysis is termed uncoupled, in that the computed soil
deformations are not affected by the existence of embedded structural
components.
A common enhancement to this type of uncoupled analysis includes the intro-
duction of an iterative solution scheme which modifies the soil deformations
based on the response of the structure so that compatible strains are computed. An
example of this type of analysis is drag loading on piles due to lateral spreading.
In an uncoupled analysis, the ground deformations would be estimated using
either an empirical relationship (e.g. Bartlett and Youd, 1995) or a sliding block
type evaluation (e.g. Byrne et al., 1994). Once the pattern of ground deformations
has been established, a model such as LPILE (Reese and Wang, 1994) can be used
to determine the loads in the deformed piles. In addition, modifications can be
made to the p-y curves to account for the reduced stiffness of the liquefied soil
(O'Rourke and Meyerson, 1994). The lateral spread displacement is forced onto
the p-y spring (i.e., drag loading).
In coupled numerical analysis, the deformations of the soil and structural ele-
ments are solved concurrently. Two-dimensional numerical models such as
FLUSH (Lysmer et al., 1975), FLAC (Itasca, 1995), DYSAC2 (Muraleetharan
et al., 1988), LINOS (Bardet, 1990), and FLIP (lai et al., 1992) have been used
to model the seismic performance of waterfront components at pons (e.g., Finn,
1990; Roth et al., 1992; Werner, 1986; Wittkop, 1994, Iai, 1998b). The primary
differences in these numerical analyses include (a) the numerical formulation
employed (e.g. finite element method, finite difference method), (b) the constitu-
tive model used for the soils, and (c) the ability to model large, permanent defor-
mations. Each of the methods listed, have been useful in evaluating various
aspects of dynamic soil-structure interaction.
Advanced numerical modelling techniques are recommended for soil-structure
interaction applications, such as estimating permanent displacements of slopes
and embankments with pile-supported wharves. The primary advantages of these
models are as follows: (a) complex embankment geometries can be evaluated. (b)
sensitivity studies can be done to determine the influence of various parameters
on the seismic stability of the structure, (c) dynamic soil behaviour is much more
realistically reproduced. (d) coupled analyses which allow for such factors as
excess pore pressure generation in contractive soils during ground shaking and the
associated reduction of soil stiffness and strength can be used. (e) soil-structure
interaction and permanent deformations can be evaluated.
These advantages may warrant the use of the dynamic analysis for designing
high performance grade structures despite the following disadvantages: (a) the
engineering time required to construct the numerical model can be extensive, (b)
numerous soil parameters are often required, thereby increasing laboratory testing
costs (the number of soil properties required is a function of the constitutive soil
model employed in the model), (c) very few of the available models have been
validated with well documented case studies of the seismic performance of actual
retaining structures, therefore the level of uncertainty in the analysis is often
unknown. As is the case with other analytical procedures, it may be advisable
for the user to confirm the applicability of the analytical procedure based on the
relevant well documented case histories before using them for design
purposes.
T7.4 INPUT PARAMETERS FOR ANALYSIS
This section further elaborates the input parameters for analysis discussed in
Section 5.4 in the Main Text. As mentioned in the Main Text, the reliability of the
results depends not only on the type of analysis, but also on the reliability of input
parameters. The input parameters elaborated in Tables T7.13 through T7.15 in this
section, may be useful in identifying possible sources of uncertainties in input
parameters, and setting up an engineering plan to improve the reliability.
Port suuctures/coudit ions
Gravity
quay wall
Sheet pile
quay wall
Pile-
supported
wharf
Earthquake
Structural
Geotechnical
Earthquake
Structural
Geotechnical
Earthquake
Structural
Design puramctcrs
k
c
: equivalent seismic coefficient
k,: threshold seismic coefficient
(Geometrical extent of liquefiable soils
relative to the position and dimensions
of a wall for a liquefiable site)
k
c
: equivalent seismic coefficient
k,: threshold seismic coefficient
relative to the position and dimensions of
a wall for a liquefiable site) ,
Pile-deck structure: design response spectra
Dike/retaining wall: k.: equivalent seismic
coefficicnt
resistance at clastic limit
I/{J: depth, from mudlinc, at the bottom
end of an equivalent unsupported pile
(
(3 = 4 [)" 1
4El
Input parameters
Regional PGA at 'bedrock
Site Category (site amplification factor)
W, H & water level: Cross-sectional dimensions of a
gravity wall
c, tfJ: cohesion and internal friction angle of soils
p". 8: friction angles at the bottom and back face of the
wall
ground water level (SPT/CPT for a liquefiable site)
Regional PGA at bedrock
Sheet pile wall: Dimensions (H. D
cm
" & water level)
& material parameters (e.g. E, I, A, yield stress)
Tie-rod: Length L and material parameters
(e.g. E, A, yield stress)
Anchor: Type (vertical or batter pile or deadman),
Dimensions and material parameters
(e.g. E, I, yield stress)
8: friction angle at sheet pile/soil interface
Regional PGA at bedrock
Site Category (site dependent response spectra)
Piles: Dimensions (L, Dr & water level) & material
parameters (c.g, E, I, A, Dr' yield/ultimate
stress/strain)
Deck: Dimensions & material parameters
relevant structural parameters for gravity/
sheet pile walls
W
-.J
o
""
o
Geotechnical Dike/retaining wall: Pile-deck structure:
k,: threshold seismic coefficient k
h
-
suh
: coefficient of lateral subgrade reaction
(Geometrical extent of liquefiable soils (or p-y curve)
relative to the position and dimensions of Dike/retaining wall:
a wall for a liquefiable site) relevant geotechnical parameters 1'01' dike,
gravity/sheet pile walls
Cellular Earthquake k,,: equivalent seismic coefficient {llIIax: Regional PGA at bedrock
quay wall Site Category (site umplification factor)
Structural for overall stability: Cell: Dimensions (H, W, D"1II1> & water level)
k,: threshold seismic coefficient & material parameters (e.g. E, I, A, yield stress)
Geotechnical
C, l/>: cohesion and internal friction angle of soils
relative to the position and dimensions
0: friction angle at the cell/soil interface
of a wall for a liquefiable site)
for structural performance:
elastic limit
limits
Crane Earthquake design response spectra {I
max
: Regional PGA at bedrock
Site Category (site dependent response spectra)
Structural resistance at elastic limit Dimensions and material parameters
(e.g. E, l, A, yield stress)
Geotechnical N/A N/A
Break-water Earthquake k,,: equivalent seismic coefficient a
max
: Regional PGA at bedrock
Structural k,: threshold seismic coefficient W, H & water level: Cross-sectional dimensions .
(Geometrical extent of liquefiable soils of breakwater
Geotechnical
relative to the position and dimensions of
c, l/>: cohesion and internal friction angle or rubble
a breakwater for a liquefiable site)
mound and foundation soils
JI.. : friction angles at the bottom of a caisson for a
composite type break water
(SPT/CPT for a liquefiable site)
Tablc T7. 14. Input parameters for simplified dynamic analysis.
Design parameters Input parameters
Gravity Earthquake empirical equations: bedrock earthquake motions
quay wall peak acceleration
peak velocity
time history analysis:
time histories of earthquake motions
Structural sliding block analysis: W, H & water level: Cross-sectional dimensions of a gravity wall
Geotechnical
lI,: threshold acceleration
G-y& D-ycurves (for site response analysis)
simplified chart:
c, /1: cohesion mH.1 intcrnal friction angle of soils
W, II & water Icvel: Cross-sectional
Jlh' 8: friction angles at the bottom and hack face of the wHII
dimensions of a gravity wall
ground water level
SPT N-valucs
undrained cyclic properties and/or SPT/CPT data
. (extent of soil improvement)
Sheet pile Earthquake empirical equations: bedrock earthquake motions
quay wall peak acceleration
1'''':lX: peak velocity
Structural sliding block analysis: Sheet pile wall: Dimensions (H, D
emb
& water level) & material
lit: threshold acceleration parameters (e.g. E, I, A, yield/ultimate stress/strain limits, ductility
simplified chart: factor)
W. 1I & water level: Cross-sectional Tie-rod: Length L and material parameters (e.g. E, A, yield/ultimate
dimensions of a gravity wall stress/strain limits)
SPT N-values Anchor: Type (vertical or batter pile or deadman), Dimensions
(extent of soil improvement) and material parameters (e.g. E. I. yield/ultimate stress/strain limits)
--_... __ ._-_._-- _._-- ..
GcotcchnicaI G-y& D-ycurves (for site response aimlysis)
c, /1: cohesion and internal friction angle of soils
8: friction angle at wall/soil interface
ground water level
W
-..J
N
z
o
Pile- Earthquake Pile-deck structure: bedrock earthquake mouons
supported standard procedure:
wharf design response spectra
site specific study:
empirical equations:
a
m3
l\ : peak acceleration
"max: peak velocity
time histories of earthquake
motions
Structural Pile-deck structure: Pile-deck structure:
resistance at elastic limit/ultimate Piles: Dimensions (L, & water level) & material parameters
states (e.g. E, J, A, D", yield/ultimate bending moment, shear fracture
ductility factor/strain limit criteria, ductility factor, M-t/> curve, strain limit)
1/fJ: depth, from mudline, at the Deck: Dimensions & material parameters & material parameters
bottom end of an equivalent (e.g. E, dimensions, shear fracture criteria, strain limit, pile-deck
unsupported pile connect ion)
[f3 )
relevant structurul parameters for gravity/sheet pile walls
._----
_._-
Geotechnical Dike/retaining wall: G-y & D-ycurves (for site response analysis)
{II: threshold acceleration undrained cyclic properties and/or SPT /CPT datu
k
h
-
sub:
coefficient of lateral subgrade reaction (or p-y curve)
relevant geotechnical parameters for dike, gravity/sheet pile walls
Cellular Earthquake empirical equations: bedrock earthquake motions
quay wall 111ll3l\: peak acceleration
"max: peak velocity
Table T7. 14, Continued.
--------------_.- ,...-
----------- ---------_.._ ~ - - ._--
Structural for overall stability: Cell: Dimensions (1/, W & water level) & material parameters
lit: threshold acceleration (e.g. E, I, A, yield/ultimate stress/strain limits)
Geotechnical
for structural performance:
G-y& D-ycurves (for sire response analysis)
yield/ultimate stress/strain limits
c, f/J: cohesion and internal friction angle of soils
8: friction angles at cell/soil interface
ground water level
undrained cyclic properties anti/or SPT /CPT data
. -
-
Crane Earthquake standard procedure: bedrock earthquake motions
design response spectra
site specific study:
Structural resistance at elastic limit/ultimate Dimensions and material parameters (e.g, E, I, A, B, yield and
strain limit ultimate bending moment, shear fracture criteria, ductility factor,
ductility factor M-f/J curve, strain limit)
Geotechnical N/A N/A
Break- Earthquake empirical equations: bedrock earthquake motions
water a l l l a ~ : peak acceleration
"max: peak velocity
----- . ~ . __.._-_... ---- _.._-------
Structural a: threshold acceleration W, /I & water level: Cross-scctionals dimensions of breakwater
_._-----
Geotechnical G-y& IJ-ycurves (for site response antilysis)
r, q,: cohesion and internal triction angle of rubble mound and
foundation soiIs
)I,,: friction tingles at the bottom of n caisson for a composite
type break water
undrained cyclic properties and/or SPT /CPT data
--
w
~
"'t1
-
z
o
Table T7.15. Input parameters for dynamic analysis"
Gravity
quay 'wall
Earthquake time histories of earthquake motions at the
bottom boundary of analysis domain
If the bottom boundary of anal ysis domai n di Hers from
the bedrock;
for moderate shaking:
G-y & D-ycurves
bedrock earthquake motions
for strong shaking:
the same as geotechnical parameters for non-linear
analysis (below)
Structural
Geotechnical
W, H & water level: Cross-sectional dimensions of a gravity wall
equivalent linear analysis:
G-y& D-ycurves
linear analysis:
Sheet pile wall: Dimensions (H, D ~ I l l " & water level) & material parameters (E, J, A, yield stress)
Tie-rod: Length L and materiul parameter (EA, yield stress)
Anchor: Type (vertical or batter pile or deadman), Dimensions and material parameters (E, J, yield stress)
non-linear analysis:
in addition to those for linear analysis, M-J curve, yield/ultimate stress/strain limits
Sheet pile
quay wall
Earthquake
Structural
parameters for elastic properties
parameters for plastic properties: for shear
and dilatancy
time histories of earthquake motions at the
G, K: shear and bulk modulus
C, : cohesion and internal friction angle of soils
I'b' 0: friction angles at the bottom and back face of the wall
ground water level
If the bottom boundary of analysis domain differs from the
bedrock;
G-r & D-ycurves
for strong shaking:
analysis (below)
;p
~
I"
~
C/> _.
C/>
~
~
g-
P-
C/>
\.;.)
-.)
lJl
---'-._" -.-,--.--.------- ----
Design parameters I Input parameters
(il'olcchnic:tl I equivalent linear analysis:
G--y & O-y curves
uon-Iinear analysis: G, K: shear and bulk modulus
parameters for clastic properties r, tfJ: cohesion and internal friction angle of soils
parameters for plastic properties: for shear 0: friction angle at sheet pile/soil interface
and dilatancy ground water level
w
-.)
0\
"'tl
-
z
()
Pile-
supported
wharf
Earthquake I time histories of earthquake motions at the I If the bottom boundary of analysis domain differs from the
bottom boundary of analysis domain bedrock;
G-y& D-ycurves
for strong shaking:
analysis (below)
Structural linear analysis:
pile-deck structure:
Piles: Dimensions (L, D
clIl h
& water level) & linear material parameters (E, I, A, DJ" yield stress)
Deck: Dimensions & material parameters
dike/retaining wall:
relevant structural parameters for gravity/sheet pile walls
in addition to those for linear analysis, M-tfJ curve, shear fracture criteria, yield/ultimate stress/strain
limits
Geotechnical I equivalent linear analysis:
G- y & D-y curves
I
parameters for clastic properties
parameters for plastic properties: for shear
and dilatancy
G, K: shear and bulk modulus
Pb' 0: friction angles at retaining wall
ground water level
Cellular
quay wall
Earthquake time histories of earthquake motions at the
If the bottom boundary of analysis domain differs from the
bedrock;
G-y& D-ycurves
for strong shaking:
analysis (below)
Crane
----_.. _-.-
Structural I linear analysis:
Cell: Dimensions (II, W & water level) & material parameters (e.g. E, I, A, yield stress)
in addition to those for linear analysis, M-lP curve. yield/ultimate stress/strain limits
Geotechnical I equivalent linear analysis:
G-y& D-ycurves
non-linear analysis: G, K: shear and bulk modulus
parameters for elastic properties c, lP: cohesion and internal friction angle of soils
parameters for plastic properties: for shear 8: friction angle at cell/soil interface
and dilatancy ground water level
Earthquake I time histories of earthquake motions at the Ill' the bottom boundary of analysis domain differs from the
bottom boundary of analysis domain bedrock;
G-y& D-ycurves
for strong shaking:
analysis (below)
Structural I linear analysis:
Dimensions and material parameters for crane (and quay wall, if crane-quay wall interaction is analyzed)
Non-linear analysis:
in addition to those for linear 'lIlalysis, M-lP curve, yield/ultimate strcss/strain limits, shear fracture criteria
Geotechnical I The same as those for quay wall analysis if crane-quay wall interaction is analyzed
E
'.
,
r
r
(
c
(,
t
TableT7.15. Continued.
Break- Earthquake time histories of earthquake motions at the If the bottom boundary of analysis domain differs from the
water bottom boundary of analysis domain bedrock;
G-r& D-rcurves
for strong shaking:
analysis (below)
Structural W. H & water level: Cross-sectional dimensions of breakwater
Geotechnical equivalent linear analysis:
G-r& D-rcurves
:
non-linear analysis: G, K: shear and bulk modulus of soils/rubbles
parameters for elastic properties c, : cohesion and internal friction angle of soils/rubbles
parameters for plastic properties: for shear Jlb: friction angles at the bottom of a caisson for a
and dilatancy composite type break water
w
-.....]
00
-e
-
~
o
TC8: Examples of Seismic Performance Evaluation
T8..1 Gravity Quay Wall (Grade A)
T8.1.1 Performance requirements and design conditions
T8.1.2 Simplified analysis
T8.1.3 Simplified dynamic analysis
T8.1.4 Simplified dynamic analysis for evaluation of liquefaction
remediation
T8.1.5 Dynamic analysis
T8.2 Sheet Pile Quay Wall (Grade B)
T8.2.2 Simplified analysis (North American method)
T8.2.3 Simplified analysis (Japanese method)
T8.2.4 Simplified dynamic analysis (deformations)
T8.3 Sheet Pile Quay Wall (Grade S)
T8.4 Pile-Supported Wharf (Grade B)
T8.5 Pile-Supported Wharf (Grade S)
T8.5.1 Performance requirements and wharf conditions
Examples of Seismic Performance Evaluation
This technical commentary presents examples of seismic performance evaluations
for five port structures and illustrates the basic procedures employed. These exam-
ples are based on field case studies, modified slightly to fit, where necessary, to
the seismic guidelines addressed in the Main Text. Thus, the examples shown are
hypothetical. These design examples are intended to illustrate the application of
the various analysis procedures outlined in this report.
T8.1 GRAVITY QUAY WALL (GRADE A)
A gravity quay wall with a water depth of 14.5 m was proposed for construction.
The performance grade of this quay wall was determined to be Grade A over
the life span of 50 years (see Section 3.1 in the Main Text). Based on seismic
hazard analysis for the quay wall site, the PGA of L1 and L2 earthquake motions
with a probability of exceedance of 50% and 10% were 0.15g and 0.3g at the
bedrock, respectively. One of the simplified dynamic analysis methods requires
input data on the earthquake moment magnitudes for the L1 and L2 earthquakes.
In this example, these magnitudes were 6.0 and 7.0, respectively. Damage criteria
for this quay wall were determined as shown in Table T8. L These criteria were
established based on the criteria discussed in Section 4.1 in the Main Text, with
modification for tilting and differential settlement due to the specific performance
requirements of the quay wall.
The proposed cross-section of the gravity quay wall is shown in Fig. T8.1.
It was proposed to remove the clay layer below the caisson wall and replace
it with sand backfill, in order to attain sufficient bearing capacity. The four
possible geotechnical cross sections with respect to liquefaction remediation
of backfill sand were considered for seismic performance evaluation as shown
in Fig. T8.2.
382 PlANe
Table T8.1. Damage criteria for a gravity quay wall.*
Extent of damage Degree I Degree II
Gravity Normalized residual Less than 1.5-5%
wall horizontal displacement 1.5%
(dlH)**
Residual tilting towards Less than 2-5
0
the sea 2
0
Apron Differential settlement
between apron and
non-apron areas
Less than
0.3 m
N/A
Degree III Degree IV
5-10% Larger
than 10%
5-8
0
Larger
than 8
0
N/A N/A
* Modified from the general criteria shown in Table 4.1 in the Main Text, due to the
specific requirements of the quay wall.
** d: Residual horizontal displacement; H: height of gravity wall.
VH.W.L.+'.7m
+ VL.W.L.:tO.Om
y
Alluvial Clay layer
Fig. T8.1. Cross section of a gravity quay wall.
Backfill Soil
SandDrain
Unit (m)
Since the required performance was Grade A, simplified and simplified dynam-
ic analyses were performed in the preliminary stage, and dynamic analysis was
performed in the final stage (see Section 5.1 in the Main Text). The simplified
analysis was based on pseudo-static analysis and the simplified dynamic analyses
were based on parametric studies (see Technical Commentary 7). The dynamic
analysis utilized the effective stress based finite element method program FLIP
(see Section 5.3 in the Main Text and Technical Commentary 7). Input data nec-
essary for the analysis were acquired from geotechnical investigations performed
at a nearby site, including SPT, PS-Iogging, and laboratory cyclic tests of in-situ
frozen samples (see Section 5.4 in the Main Text and Technical Commentary 4).
SPT N-values and the location of in-situ freezing sampling are shown in Fig. T8.3.
CASE-1
Non-Liquefiable Soil
CASE2
Loose Saturated Soil
CASE-3
Non-Liquefiable Soil
CASE4
Loose Saturated Soil
Fig. T8.2. Four design alternatives in geotechnical conditions.
The structural parameters of the caisson and geotechnical parameters for backfill
soils and rubble are shown in Tables T8.2 and T8.3, respectively
For the seismic performance evaluation, the limiting curves were defined as
shown in Fig. T8.4 for a wall height of 18.5 m based on the damage criteria shown
in TableT8.I.
Pseudo-static analysis of the gravity quay wall was performed for L1 earthquake
. motion. Using the Ll acceleration of O.15g at bedrock, site response was per-
formed, and a ground surface acceleration of O.25g was obtained. A seismic
coefficient of 0.15 used in the pseudo-static analysis was determined using
Eq. (TI.ll) as follows:
k
h
=0.6 x 0.25 =0.15
384 PlANe
After the earthquake-..--.!
11.50
In- Situ Freezing
Sampling
\ I
SPT N-vaw Na6 - 8
..,
.: '.: .:
:: .....: .' .' ';
': .: :18ackfill S a n ~ : ::' '
.' : for Replacing' ...
: Clay Layer '.'
I
I
I
I
: -28.00
. ' .
....
o'
i-l.W.L:0,00
:& H.W.L+ 1.70
--jQ..GQ
!lQQ___ .. ' _ " .' '.,
.. .;'.1- .
.."''Foundation Rubblel0....200k /Stone ': .
(Unit: m)
Fig. T8.3. SPT N-values and location of in-situ freezing sampling.
Cross-sections of the structure with all relevant and simplified parameters
required for the analysis are shown in Figs. T8.5 and T8.6.
(1) Active earth pressures and thrust
Thus, the active soil wedge is defined by the angle of failure measured from the
vertical direction as
90
0
- 63
0
= 27
0
The numbers with the square marks shown in Fig. T8.6 within the soil wedge
are areas in m
2
of various portions used in the analysis. Using these areas, a
representative unit weight of the material above the water table. including the
Examples of Seismic Performance Evaluation 385
Table T8.2. Parameters for concrete caisson and joint elements.
Parameters density Young's modulus (kPa) Shear modulus (kPa) Poisson's ratio
Caisson 3.0x 10
7
1.3x 10
7
0.20
Joint elements: Friction angle 8 = 31 (bottom of caisson): 15 (behind caisson)
Table T8.3. Parameters for soils and rubble.
Parameters Density Initial shear Poisson's Internal Phase
(t/rrr') modulus (kPa) ratio friction transformation
angle (0) angle (0)
Foundation soil* 1.8 5660 x
0.3 37 28
Backfill soil* 1.8
10000 x (a' )0.5
0.3 36 28
mo _
Clay 1.7 6270 x (a' )0.'
0.3 30
Foundation rubble 2.0 18200 x (j"::)0.5 0.3 40
and rubble backfill
* Refer to Iai et al. (1998b) on the parameters for dilatancy.
capping material and backfill soil, is (1 kN = 0.1 tf)
(20 x 0.19 + IS x 2S.92)129.1 =IS kN/m
3
Then, Eqn. (T7.6) gives for the equivalent unit weight of the backfill
Ye =lS[1_(lS.1 )1]+10x(IS.I)2 =12.7 kN/m
3
IS.S IS.S "
A representative internal friction angle for the backfill can be detennlned by direct
averaging based on the corresponding cross-sectional areas (see Fig. TS.6)
= (0.19x40 +2S.92x36 + 14.4x36+43.69x40)/S7.2
= (7.6+ 1041.12+S1S.4+ 1747.6)/S7.2 = 3So
Assuming that half the operational surcharge is present during the earthquake, the
modified seismic coefficient is derived by Eqn. (T7.7)
k
' 10x1S,S+1xlSx3.402 +lSx3.40x1S.10+1x20x1S.12
It = kit . -----=-------------..--:-------:-
lOx lS.5 + 1x lSx3.40
2
+ lSx 3.40x lS.10 + txlOx 15.1
2
= 0.15x ISS+ 104+ 924.1+ 2280.1 = 0.15x 1.48=0.22
18S+104+924.1 +1140.1
The alternative procedure using Eqn. (T7.7') will be shown in the example of
Section T8.2.3. Use of Eqn. (T7.1) now gives
386 PlANe
IV:Collap.
II:He.-cou.pH
E _------+------.........,I---____.f---- :- 1.S
l:
C>
E
i
o
c

I
c II : Repn-eble
c::
0'-=-----.....-------:'--------
0.15 0.30
Design Earthquake Motions Ill)
IV: Collapse
St-------t--------;-------
i II : He.- collapse
e;,
'"
0>

i=
II: Repairable

Olooc:::;;.....-----'------......II.------_
0.15 0.30
Design Earthquake Motions (g)
Fig. T8.4. Limiting curve for defining wall displacement.
1JI = =12.4
and the dynamic active earth pressure coefficient of the Mononobe-Okabe method
is found by Eqn. (T7.2)
cos?(38 -12.4)
Kae = = 0.356
1
') 4 (15 1') 40)[1' sin(38 +15)sin(38 - 12.40) ]-
cos _. cos + _. -r-
cos(15 +12.4)
The total earth thrust is now
r; )XI8.50
2
)x. 18.50
2
_ H _ 18.)0
=806.6 kN/m
This gives a horizontal force
Paecoso = 806.6xcosI5 = 779.1 kN/m
\7 O.OO
-14.50
+4.00
Y =18 kN/m31 cb =36'
y=20 kN/m
3
cb =40'
Fig. T8.5. Simplified cross-section for pseudo-static analysis.
+4.00
-14.5
-14.5
+4.00
Y eq=12.7 kNlm
3
cb eq=3S'
r wet=18 kN/m
3
__ ~ + o . 6 0 .
(b)
(a)
+2.00
+117
(c)
Fig. T8.6. Diagram for evaluating average values.
(a) Soil wedge geometry,
(b) Original parameters.
(c) Equivalent parameters.
388 PlANe
at about 0045 x 18.50 = 8.33 m above the - 14.50 m elevation,
and a vertical contribution
P
ae
sin8 = 806.6xsinI5 = 208.8 kN/m
at the interface between structure and soil.
(2) Hydrodynamic force
The hydrodynamic force in front of the wall gives a horizontal force according to
Westergaard's expression
e; =2.xO.15XI0.2xI4.52 =187.6 kN/m
12
at OAxI4.5=5.8 m above the-14.50 m elevation.
(3) Inertia and other driving forces
The wall inertia forces are estimated as follows:
cap 11.60x2xI8xO.15 = 62.6 kN/m at 17.50 m above bed (-14.5 m)
caisson 11.60xI6.00x21xO.15 = 584.6 kN/m at 8.50 m above bed
footing 15.60xO.5x21 xO.15 = 24.6 kN/m at 0.25 m above bed
The backfill inertia force
2.0xI8.0xI8xO.15 = 97.2 kN/m at 9.50 m above -14.50 m
The static bollard pull can be taken into account by reducing its value by 50%
50%x20=10 kN/mat 19.0 mabove -14.50 m
(4) Vertical forces
The resisting forces and their points of application are determined as follows:
cap 11.60x2.0xI8=417.6 kN/m
at 7.8 m from A (see Fig. T8.5)
caisson 11.60x1.70x21 =414.1 kN/m
11.6 2x 1.40+ 2 ,... 7 46 f A
at --x +L.=. m rom
3 1.40+2.0
11.60x 14.30x 11 = 1824.7 kN/m
at 7.8 m from A
footing 15.60xO.50x 11 = 85.8 kN/m
at 7.8 m from A
Using the backfill stone area above the water table =+xO.83(1 +2) +2 xO.57
= 2.39 rn ', the backfill equivalent unit weight above water table is determined
as
{20x2.39 + 18x (2x 3.40 - 2.39)}/ 6.80 = 18.7 kN/m
3
2.0x3.40x 18.7= 127.2 lul\Um at 14.60 m from A
2.0x 13.40x 10= 268 kN/m at 14.60 m from A
The hydrostatic pressure due to the variation of water table gives the following
forces:
+xO.6!xlO.2=1.8 kN/m at txO.60+14.50=14.70 m above -14.50 m
14.50xO.60xlO.2 = 88.7 kN/m at 7.25 m above -14.50 m
Uplift +xO.60xIO.2x15.60=47.7 kN/m at tx15.60=10.40 mfromA
It is noted that the hydrostatic pressure distribution depends on the permeability
of the foundation and backfill rubble. If the latter has a much lower permeability
than the former, then no uplift force is generated and the pressure diagram at the
back face dies off linearly to the foundation level AB.
(5) Force and moment balance
The various forces mentioned previously act on the structure as shown in Fig. T8.7.
Now, the stabilizing moments with respect to point A are calculated as follows:
stabilizing 395.2 x 14.60
-
5769.9 kNmlm
208.8 x 13.60
=
2839.7
414.1 x 7.46
-
3089.2
2,328.1 x 7.80
-
18,159.2
3,346.2 29,858.0 kNmlm
-47.7
3,298.5 kN/m
The overturning moments with respect to the same point are
10.0 x 19.00 = 190.0
62.6 x 17.50 - 1,095.5
584.6 x 8.50 - 4,969.1
24.6 x 0.25 - 6.2
187.6 x 5.80 - 1,088.1
(negative) 47.7 x 10.40 = 496.1
779.1 x 8.33 = 6,489.9
1.8 x 14.70 - 26.5
88.7 x 7.25 = 643.1
97.2 x 9.50 - 923.4
1,836.2 kN/m 15,927.9 kNmlm
390 PlANe
4' , , + f
10.40 .1
10
A
I.
-
41f:, 62.6
I I
I
I
I
7.46 1414.1
I
!_--
---------

- - -
--
---
I
I-' 1
,
1.8
f-
I
f-
I
I f-
0

1
-
,
,..:
I
-
I
...
-
I
7.80
20B.8
1
- I.
- ,
0

- I
en

:- 0
4
7
.
6
I
...
f-- "":i
,
"'
. r "'-- :-

-,
, f-
I
,
M
I
f-
I I
M
f-
I
cD
I
f-
I
I
0
-
It)
I
, It)

-
I
I
a:i
-
"
, (7)
-
I
Iri ,
I
-
i
I
-
;
24.61 I
-
I
I
f--
-e-O.2S-
-
-
I
:-
i
I
,
Fig. T8.7. Loads acting on structure during earthquake.

The factor of safety against overturning is
F!<J = 29,858.0/15,927.9 =1.87
Depending on the definition of this factor of safety and when a non-zero angle of
friction between the backfill and the wall back face is assumed, one can take the
overturning moment of the earth thrust before analysing it into vertical and hori-
zontal components.
The factor of safety against sliding is
F; = 3,298.5x .LLb/ 1,836.2
where .LLb == tan31 = 0.6
Thus, r; = 1.08
Thus, the proposed cross section is likely to satisfy the seismic performance
requirements at Ll earthquake motion.
The pressures at foundation are derived as follows:
Eccentricity
eo> =(15,927.9+ 414.lxO.34- 208.8x5.80 - 395.2x6.80)/1298.5
= 3.69 m> 15.60/6 = 2.60 m
Therefore. the vertical stress applied at the both ends of footing A (sea side) and
B (land side) are given as
c, =0
0".. = 2x(total verticalload)/(3x(width))
=2x 3.298.5/(3x(7.80 - 3.69)) =535 k N / m ~
add surcharge at half its value 535++x20 =545 kN/m~
The preliminary analysis includes the use of a simplified method to predict the
approximate range of deformations expected. This method utilizes non-dimen-
sional parameters with respect to caisson geometry, thickness of soil deposit
below the caisson, and geotechnical conditions represented by SPT N-values
of the subsoil below and behind the wall (see Section T7.2.1(2) in Technical
Commentary 7). The displacement at the top of the caisson under the prescribed
earthquake motion was estimated.
Among the four alternatives shown in Fig. T8.2, Cases 1 and 4 were chosen as
examples. Case 1 represents a quay wall with non-liquefiable soil deposits and, in
this example, equivalent SPT N-value, N
65
, was assumed to be 14 and 20. The N
65
is an SPT N-value measured with a Japanese SPT energy efficiency of 72%
and normalized to an overburden pressure of 65 kPa (T4.4.2 in Technical
Commentary 4). Case 4 represents a quay wall with liquefiable soil deposits and
in this example, equivalent SPT N-value, N
65
, was assumed to be 7.
For a wall height, H, of 18.5 m, and a width, W, of 15.6 m, the normalized
caisson geometry is
W = 15.6m = 0.84
H 18.5m
The normalized thickness of soil deposit below the caisson is (see Fig. T7.14 for
definitions)
Dl = 18.5m = 1.0
H 18.5m
For the Ll earthquake motion (a
max
= 0.15g at the bedrock), the residual lateral
displacement was estimated through Fig. T7.16(b), referring to W/H =0.9. This
approximation (W/H = 0.84 == 0.9) was quite reasonable as can be inferred by the
graph of Fig. T7.18(b).
Thus, we have
Case I (Non-liquefiable):
d
For N
6S
=14: -:: 0.005, hence, d == 0.005x 18.5=0.09 m
H
d
For N
6S
=20: - =0, hence, d == 0 m
H
392 PIANC
Case 4 (Liquefiable):
For N", =7: ~ =0.04, hence, d:: 0.04xI8.50 =0.74 m
<>.J H .
For theL2 earthquake motion (a
max
= 0.30g at the bedrock), the displacements become
Case 1 (Non-liquefiable): .
d
For N65 = 14: - =0.03, hence, d:: 0.03x 18.5= 0.56 m
H
d
For N65 = 20: - =0.0075, hence, d 0.0075x 18.5= 0.14 m
H
Case 4 (Liquefiable):
d
For N
65
= 7: - =0.18, hence, d O.l8x18.50 =3.33 m
H
The requirements for Grade A performance were that the residual horizontal
displacements at the top of the wall are less than 0.3 m and 0.9 rn, for the LI and
L2 earthquake motions, respectively (Fig. T8.4). Table T8.4 compares the esti-
mated displacements with the allowable displacements. Based on the simplified
dynamic analysis, it was seen that if the soil deposit was non-liquefiable with N
65
larger than about 10 (i.e. Case I), this quay wall satisfied the required perform-
ance criteria.
T8.1.4 Simplified dynamic analysis for evaluation of liquefaction remediation
The preliminary analysis includes the use of a simplified method to predict the
approximate range of deformations expected with respect to the area of soil
improvement as remediation of liquefiable soils. This method utilizes non-dimen-
sional parameters with respect to caisson geometry, and area of soil improvement
to estimate displacement at the top of the caisson under the prescribed earthquake
motion (see Section T7.2.I(3) in Technical Commentary 7).
Among the four alternatives shown in Fig. T8.2, Case 2 was chosen as the
example. Using the above parameters, it is possible to enter the design chart and
obtain estimates for the pennanent lateral deformation at the top of the caisson.
Case 1 Case 4
Table T8.4. Comparison of estimated and allowable displacement for a caisson wall.
Earthquake levels Estimated displacement (m) Allowable
displacement (rn)
Levell
Level 2
0.74
3.33
0.09
0.56
o
0.14
0.3
0.9
MSF
MSF
Similarly to Section T8.l.3, the normalized caisson geometry is
W =l5.6m =0.84
H l8.5m
Assuming that the rubble fill is non-liquefiable. it is treated as a zone of soil
improvement. with an equivalent width (L) equal to approximately half of the
rubble width (9.9 m). Therefore, the normalized soil improvement is
!::- = 9.9m =0.54
H 18.5m
For the L I earthquake motion (M; 6.0 and amax(rockl of 0.15g), the site response
study indicated that the maximum acceleration at the elevation of the dredge line
within the backfill (amax(d) is equal to 0.25g. The magnitude scaling factor
(MSF) was determined as 2.0 from Table T7.8 in Technical Commentary 7. The
normalized earthquake motion is
___ 0.25g =0.13
2.0 g
The loose material has SPT (Nl)60-values approximately equal to 10. Therefore,
Fig. T7.20a in Technical Commentary 7 was used to determine the normalized
and dimensional lateral displacements at the top of the caisson as
d
-:0.03
H
d = (0.03)(18.5 m) = 0.56 m
If the soil in the back land was improved; so that the SPT (Nt)60-values were
approximately 20, Fig. T7.20b in Technical Commentary 7 would be used and the
displacement would be estimated as follows:
d
-:0.01
H
d=(0.01)(18.5 m)= 0.19 m
For the L2 earthquake motion (M
w
7.0 and amax(rock) of 0.30g), the site response
study indicated that the maximum acceleration at the elevation of the dredge line
within the backfill (amax(d) is equal to 0.35g. The magnitude scaling factor (MSF)
was determined as 1.25 from Table T7.8 in Technical Commentary 7. The nor-
malized earthquake motion is
_a
max
_
f
_
dl
= 0.35g = 0.28 .
1.25 g
The loose material has SPT (N\)60-values that are approximately equal to 10, and
therefore, Fig. T7.20a inTechnical Commentary 7 was used to determine that the
394 PlANe
lateral displacement at the top of the caisson was
d .
- ='0.08 to 0.09
H
d = (0.08 to 0.09)(18.5 m) = 1.5 to 1.7 m
If the soil in the back land was improved, so that the SPT (Nl)60-values were
approximately 20, Fig. T7.20b in Technical Commentary 7 would be used and the
displacement would be estimated as follows:
d
-::0.04
H
d =(0.04)(18.5 m) =0.74 m
As mentioned earlier, the requirements for Grade A performance are that
the residual horizontal displacements at the top of the wall are less than 0.3 m
and 0.9 m, for the L1 and L2 earthquake motions, respectively (Fig. T8.4).
Based on the simplified analysis, it is necessary to improve the backfill soil
sufficiently to preclude liquefaction from occurring. It should be noted
that this simplified analysis method is not able to estimate the tilting of the
caisson.
As mentioned earlier, the limiting curves for seismic performance evaluation were
defined, as shown in Fig. T8.4, for a wall height of 18.5 m based the damage cri-
teria shown in Table T8.1.
The cross section of the gravity quay wall was idealized into plane strain
finite elements shown in Fiz. T8.8. Element size was defined small enouzh
~ . ~
to allow seismic wave propagation of up to 2 Hz throughout the analysis.
including the liquefaction state. In order to simulate the incoming and out-
going waves through the boundaries of the analysis domain, equivalent viscous
dampers were used at the boundaries. The effect of the free field motions
was also taken into account by performing one dimensional response analysis
I
-50.Om
I
O.Om
I
... SO.Om
I
+100.0m
, '
...0.6m
13.5m
-36.0m
Fig. T8.8. Finite element mesh for analysis.
at the outside fields. and assigning the free field motion through the viscous
dampers. Before the earthquake response analysis. a static analysis was
performed with gravity under drained conditions to simulate the stress
conditions before the earthquake. The results of the static analyses were used
for the initial conditions for the earthquake response analyses. The input earth-
quake motions were those recorded at the Port Island site during the Great
Hanshin earthquake.
Concrete caisson was modelled by linear elements. with soil-structure inter-
faces modelled by joint elements, as shown in Fig. T8.9. The parameters are
shown in Table T8.2. The seawater in front of the caisson was modelled as an
incompressible fluid.
Parameters for soil elements were calibrated based on the results of the geo-
technical investigation as shown in Table T8.3. In particular, undrained cyclic
properties of sand were calibrated as shown in Fig. T8.l O. It should be noted that
appropriate characterization of pore water pressure build-up and the increase in
shear strain amplitude are important for parameter calibration, since these
undrained cyclic properties' significantly affect the degree of deformation of the
soil-structure system.
Seismic analysis was performed for the four alternatives shown in
Fig. T8.2 over Ll and L2 earthquake motions. Computed residual deformations
of the quay wall for Cases 1 and 4 for L2 earthquake motion are shown
in Fig. T8.ll. The results of the analysis are summarized in Fig. T8.l2.
Although all the alternatives satisfy the criteria for tilting as shown in
Fig. T8.l2(b), only Case 1 satisfied the horizontal displacement criteria as
shown in Fig. T8.12(a). The remaining criteria with for differential settlement
between apron and non-apron areas for L1, as shown in Table T8.1, are also
satisfied by Case I. Consequently, Case I was proposed as the recommended
design of a Grade A gravity quay wall.
Concrete cap
\
Joint elements behind caisson
Joint elements in bottom of caisson
Fig. T8.9. Modelling of caisson structure.
396 PlANe
Measured
time

Measured
G.L -20.1am - -20.7Om

.J
-1 .0 8.0 -6.0 -4.0 2.0 0.0 2.0 4.0 6.0 8.0 10.0
E a(")
100
G.L 20.1am - -20.70m

0.1..-0<0'-=-----------
-1
. -
.$l .....
'i!'g
0"
b.,
10]
0 et':._ 000
0
0 0 0 0
10
Computed
Computed
10 5 -5
200-.,-----r------;-----,------/
-10
10
i! .e
m
2
III Of---<A::FH-H+ttt++tttl-t+lH+1H+H+t-
.. 'U
(b)
-:--------'-----,
100
100
200
0 50 100 150 200
CTm'(kPa)
2
Computed
100
2c.=5%
Measured
0.6
2t..=1;o
'" 21:,"'2"..
III
t:>
2&.",2"/0
0.5,
2...=5
c
..o
III ...
!5'r 0
...

0.4 2&.=1
%
.,
8
t> 0:
0.3
..
..
100
f
0.2
iii
c
0.1
0
-200 0
0 50 100 150 200 0.1 1 10 100
(el CTm'(kPa) (d) Nunber of Load Cycles N,
+4.0m
Inclination
. 2.6
0
1 Lateral
Displacement
2.1m
Vert' cal Displaceme t 0 9m t-1 I n :::
-..
,...... .,..
.:
.
, ,
- - - - -
.
,
.,
..
-
. .
--
- - -- - -
- -- --
-
- - - -- -
- - - -
I
-
36.0m
(a)
-36.0m
+4.0m
Lateral
Displacement
0.9m
H
03
Inclination
0.5
0
\1
v . I D'
ertica ISp acement m::
,
:
,
:
,
:
' ,
: ,
. . ..
,
I
I
(b)
Fig. T8.ll. Computed residual deformation.
(a) Alternative case 4.
(b) Alternative case 1.
T8.2 SHEET PILE QUAY WALL (GRADE B)
A new sheet pile quay wall with a 6 m water depth was being constructed for a
small harbour in a seismically active region. The seismic performance grade of
this relatively small quay wall was designated as Grade B. Simplified analysis was
performed as a preliminary to determine the depth of sheet pile embedment
and the structural parameters of the sheet pile wall and tie rod. Based on Grade B
requirements, a simplified dynamic analysis was also performed to estimate the
Fig. T8.1O. Undrained cyclic properties: laboratory test and computed results.
(a) Time histories.
(b) Stress-strain relationships.
(c) Stress paths.
(d) Cyclic resistance.
398 PlANe
0.1 0.2 0.3 0.4 0.5 0.6
Input PeakAcceleration (g) (b)
-
0.6 0.1 0.2 0.3 0.4 0.5
InputPeakAcceleration (g) (a)
Fig. T8.12. Seismic response and limiting curve for acceptable damage.
(a) Residual horizontal displacement.
(b) Tilting.
deformations of the sheet pile quay walls and the required extent of soil improve-
ment to limit the permanent deformations.
The simplified analysis was based on the pseudo-static analysis using
Mononobe-Okabe's method (see Section T7.1.1 in Technical Commentary 7).
When applied to sheet pile wall analysis, the pseudo-static analysis differs
depending on the regional design practice. The North American practice (Ebeling
and Morrison, 1992) and the Japanese practice (Ministry of Transport, Japan,
1999) were used to illustrate the differences. The simplified dynamic analysis to
estimate the wall deformations was based on the method discussed in Section
T7.2.2 (McCullough, 1998: McCullough and Dickenson, 1998).
The proposed geometry of the sheet pile quay wall is shown in Fig. T8.13.
The wall was backfilled with hydraulically placed granular materials that
were improved adjacent to the wall due to soil liquefaction concerns. which
have been noted to cause serious damage to sheet pile quay walls (Kitajima
and Uwabe, 1979; Gazetas. et al., 1990). The foundation soils were fairly
dense granular materials, with relatively high SPT (NI)6o-values. The anchor
system for the wall consisted of a continuous sheet pile wall. It was
assumed during this analysis. that no excess pore pressures were generated
adjacent to the wall, a reasonable assumption since the SPT (NJ)6o-values
in the foundation and improved backfill are approximately 20. The Mononobe-
Okabe method requires an initial estimate of the depth of embedment
(DemrJ From several iterations, this was chosen as 3.75 meters. For simplicity.
the water table behind the wall was assumed to be the same as that of the
sea water level. Surcharge on the quay wall was not considered in this example
problem.
1----L.:,,---1
-..---"...... - " " " : > ' O ' O ~ - - - -
Backfill soil
Y,," = 18 kN/m'
Y..,-, =16 kN/m'
Y,. =8 kN/m'
ts,; valuce 10
t!J=35"
sheet pile wall
Fig. T8.13. A proposed sheet pile quay wall.
Foundation soil
r_= 19 kN/m
3
J'l,=9 kN/m'
(Nr)(f(J value= 20
'=40"
Table T8.5. Qualitative and quantitative description of the reported degrees of damage
(after Kitajima and Uwabe, 1979).
Degree of damage
(description of damage)
No damage
Negligible damage to the wall itself;
noticeable damage to related
structures (i.e. concrete apron)
Noticeable damage to the wall itself
General shape of anchored sheet
pile preserved, but significantly
damaged
Complete destruction, no
recognizable shape of wall
remaining
Maximum permanent
sheet pile (ern)
o
Less than 30
30 to 100
100 to 400
Larger than 400
Average permanent
sheet pile (ern)
o
Less than 10
10 to 60
60 to 120
Larger than 120
Grade B performance requires that, for an LI earthquake motion, the structure
will be serviceable, and for an L2 earthquake motion, the structure will not col-
lapse (Fig. 3.2 in the Main Text). For a sheet pile quay wall, these damage crite-
ria can be approximately quantified by the relationships proposed by Kitajima and
Uwabe (1979), and presented in Table T8.5. These relationships indicate that for
the wall to be serviceable, the average deformations should be approximately 10
em or less and that to prevent collapse of the wall, the average deformations
should be approximately 120 em or less. The specific Grade B requirement is that
the deformation factor (d/H) be less than 1.5% (Table 4.2 in the Main Text), which
for this 9 m high wall, is 13.5 em.
400 PlANe
A site response analysis was performed at the quay wall site for the Ll and L2
motions of M; = 6.5 and 7.5 earthquakes. The peak ground accelerations at the
ground surface for Ll and L2 motions were obtained as amax(surfaceJ = 0.45g and
O.60g, respectively.
T8.2.2 Simplified analysis (North American method)
The simplified analysis based on the North American practice (Ebeling and
Morrison, 1992) is shown in this section.
.(1) Active earth pressures and thrust
The soils adjacent to the wall have varying densities, yet the Mononobe-Okabe
analysis method requires that the densities be uniform. It was therefore necessary
to estimate equivalent densities assuming the wedge failure plane (Fig T8.14),
where Ywet is the unit weight of the improved backfill above the water table, 'Ysat is
the unit weight of the improved backfill beneath the water level and the unit
weight of the foundation soil (since they are the same). The equivalent buoyant
unit weight of the backfill soil is equal to
YweI H+D
m
} - (H
SUb
+ D
emb
)2) + Yh (H
SUb
+ D
emb
y.
Ye = (H + Demhi
(17 kN/m
3)(12.75
mi -(9.75 m)2)+(9 kN/m
3)(9.75
mY
-
(12.75 m)'
= 12.3 kN/m
3
The equivalent saturated unit weight is similarly obtained as
(17 kN/m
3)12.75
m)2 -(9.75 m)'l)+(19 kN/m
3)(9.75
m)2
YC.sal = (12.75 m")
=18.2 kN/m
3
T .r
Fig. T8.14. Diagram used in estimating an equivalent
soil unit weight.
The vertical seismic coefficient (kJ was assumed to be zero and the horizontal
seismic coefficient (k
h
) was calculated as (see Section T7.1.1 in Technical
Commentary 7)
k; a;, r=HO.:5
g
r=0.26
The Mononobe-Okabe equations were originally derived for dry backfill. With the
inclusion of water in this example, the horizontal seismic coefficient is modified
using Eqn. (T7.5) as
k' = k . Yeo,al = 0.26. 18.2 kN/m
J
= 0.38
hI h 1'/ 3 kN/ J
Yo _. m
The above equation is essentially the same as Eqn. (T7.7). The seismic inertia
angle (Eqn. (T7.I) in Technical Commentary 7) is then
lfIl = tan" =tan" (0.38) =20.6
The friction angle between the wall and soil (8) was assumed to be 20, half of
the internal friction angle of the soil ().
For level backfill, as shown in the definition sketch, the coefficient of active
earth pressure is calculated using Eqn. (T7.2) in Technical Commentary 7
K = cos
2
( - 1Ift)
.., cos v, cos(1If1 +S)[1 + sine + S) sine -11ft) ]2
cos(8+ lfI, )
The horizontal dynamic active thrust acting on the wall is calculated using Eqn.
(T7.3) in Technical Commentary 7
(P ) =K . .l.y (R+D )! -cos S
xx ae2 e cmb.
=0.48t12.3 kN/m
J
(12.75 m)' . cos 20
= 452 kN per meter of wall
This force includes both the static, (PJx' and incremental dynamic, (L1Pae)x. forces
(P
a
,,) .
402 PlANe
It is now necessary to determine the point of action of the dynamic force. It is
known that the incremental dynamic portion acts at an elevation above the
pile toe that is 60% of (H+D
emb
) Using this information, and the calculation of
the static component force (Pa)xand its location it is possible to estimate the
location of the dynamic earth pressure force as
;: = +(.dP.) .
(P
a
.).
In order to find the static earth pressure force, Coulomb's earth pressure theory
is used, with the coefficient of active earth pressure (K
a
) being
K = cos
2
a cos 8[1 + sine + 0) sin ]2

cos(O)
_ cos
2
40 , = 0.20
COS200 [1+ sine40 +20) sin40]-
cos(20)
Using the active earth pressure coefficient and the areas represented by 1, 2 and 3
in Fig. T8.15, the horizontal components of the static active earth pressure forces
and the location of these forces (above the pile toe) is given by
= K, . [* 17 (3 m)"]. cos8
= 0.20 117 kN/m
3
(3 m)" . cos20
= 14.3 kN per meter of wall
=3.75m +6m +t3m =IO.75m
(Pll2). = K. -[17 kN/m
3
3 m- 9.75 m} cos8
= 0.20 17 kN/m
3
3 rn 9.75 m cos20
;pa2 =19.75 m =4.88 m
(Pll.1 ) . =K. '[1.9 kN/m
l
9.75 m9.75 m}cos8
= 0.20 *9 kN/m
l
9.75 m 9.75 m cos 20
PbollOlft
....-----1 Dynamic Incremental
Active Earth
Pressure Force, (,jp r
I-.I----+--r-
P,op -
Tie Rod
Force, T
FES
(
t:.Pat! )
PIO/ =1.6' H+D
t!mb
(
t:.Pat! )
POOI/O"'=0.4 H+D
emb
Fig. T8.15. Resultant earth pressure forces.
=r9.75m =3.25m
Therefore, the total static active earth pressure is

= 14.3 kN+93.2 kN+80.2 kN = 188 kN per meter of wall
The total static active earth pressure force acts at the following distance above the
pile toe:
"

= 14.3 kN10.75 m + 93.2 kN 4.88 m+ 80.2 kN 3.25 m = 4.6 m
188 kN
The incremental dynamic active earth pressure force is then equal to
(L1P
ae)x
= (Pae)x =452 kN-I88 kN=264 kNpermeterofwall
The location of the. incremental active earth pressure force is
;,1P. = 0.6 (3.75 m+6 m+ 3 rn) = 7.65 m
ae
404 PLANe
Therefore, the dynamic active earth pressure acts at
= . +(.6Pac )x.
Pac (Pac)x
188 kN 4.6 m+265 kN 7.7 m 64 b th il
- = . m a ove e p e toe
452kN
i.e. at about 50% of the total wall height.
(2) Passive earth pressures and thrust
Following the same procedures as those used in calculating the active earth pres-
sures behind the wall, the equivalent horizontal seismic coefficient for use in esti-
mating the passive earth pressure in front of the wall is
= .Ysat =0.26. 19 kN/m
3
=0.55
2 Y
b
9 kN/m
3
The seismic inertia angle (Eqn, T7.1) is then
l/I2 =tan" =tan-
leO.55)
=28.8
A factor of safety of 1.2 is typically applied to the passive soil strengths to take
into account the partial mobilization of passive resistance at the design condition
(Ebeling and Morrison, 1992). Factoring the angle of internal friction for the soil
strength
</1* =arctan ( tan;) )=arctan( tan(40) )= 35.00
1._ 1.2
The factored angle of friction between the wall and soil is
(
tan(8) ) (tan(200) )
8* =arctan "). =arctan . ") =16.9
1._ 1._
The passive earth pressure coefficient is calculated as
cos:! ( *-'II., )
K = -
pe sin( *+8*) sin(*
cos 'II., cos 'II., +u. - -
- - cos(8 *+'II:J
cos' (35.0 - 28.8)
== , = 3.9
cos28.8 cos(28.8 +16.9
0)[1-
sin(35.0 +16.9) sin(35.0 - 28.8) J
cos(16.9 +28.8)
Examples of Seismic Performance Evaluation -l-05
The horizontal passive earth pressure force is then given by
(Ppe \ =Kpe . +Yb . (D
emb
i .cos S'
I , '
=3.9 9 kN/m- (3.75 mr .cos 16.9=234 kN per meter of wall
The horizontal passive earth pressure force acts at approximately one-third the
depth of embedment from the pile toe (Ebeling and Morrison, 1992)
== t 3.75m = 1.25m above the pile toe
(3) Hydrodynamic force
Since the fluid elevation is equal on both sides of the wall, the net hydrostatic fluid
pressure is zero, but there is a hydrodynamic fluid force (Pwd)' that is estimated
using the seismic coefficient, and the depth of water in front of the wall (H
w
) '
using Eqn. (T7.8)
p =.1. . 1.- r .(H )2 = .1. . 0 26 . 9 81 kN/m
3
. (6 m)'
wd 12 "h w w 12'
The hydrodynamic force acts at
= 0.4 H; + DerOO = 0.46 m+ 3.75 m = 6.15 m above the pile toe
...s
(4) Moment balance about the tie-rod
The depth of embedment is calculated by summing the moments about the eleva-
tion of the tie-rod/anchor, with the distance from the pile toe to the tie rod being
10.75 m.
:L.M
tie
_
rod=(P",,)x(10.75
)-(Pae)x(10.75 )-P
wd(l0.75

r x
=(234 kN)(10.75 m-1.25 m)-(452 kN)
(10.75 m-6.40 m)-(52.6 kN)(10.75 m-6.15 m)
=16.2 kNm out of balance
Using an embedment depth of 3.75 meters, results in a relatively small conserva-
tive moment imbalance, and therefore, will be used as the seismic design depth of
embedment. It should also be noted that the depth of embedment may need to be
increased depending on environmental and operational conditions (such as scour,
propeller wash, etc.).
406 PlANe
(5) Tie-rod force (TFES)
The tie-rod force (TFES) is calculated by summing the forces in the horizontal direction
TfES = (P.)x +P
wd
-(Ppe)x
= 452 kN + 52.6 kN - 234 kN = 270 kN per meter of wall
(6) Maximum Bending Moment (MFF.S and M
design
)
The maximum bending moment occurs at the same elevation as zero shear with-
in the sheet pile wall. The zero shear is calculated at elevation 'beneath the fluid
level (Fig. 8.15). Since the dynamic active earth pressure distribution is unknown,
it is separated into two parts, the static active earth pressure and the incremental
dynamic active earth pressure. Above the dredge level, the contribution due to the
static active earth pressure is (refer to Fig. T8.15)
=[[t17 kN/m
3
(3 m)2]+[17 kN/m
3
03 m']+[t9 kN/m
3
or]]
-K; . costS
=[76.5 kN/m+51 kN/m
2.'
+4.5 kN/m
3
.'2]0.20cos20
= 14.4 kN + 9.6 kN/m' + 0.84 kN/m
2
. '2 per meter of wall
The incremental dynamic active earth pressures is shown as area 5 in Fig. T8.15
(Ebeling and Morrison, 1992).
PlOP =1.6.( (L1P
ae
)x )=1.6.( 265kN )
H + D
emb
9 m+3.75 m
= 33.2 kN/m per meter of wall
-0 4.( )-04.( 265 kN )
Pbollom - -
H + D
emb
9 m+3075 m
= 8.3 kN/m per meter of wall
= t .(3 m+')[2PlOP - [PlOP - jJoonom ] (3 m+,)]
. H + Demb
PlOp - Pbollom =] 095 kN/m:! per meter of wall
H + D
emb
(
1.95 kN/m:: ) r: .,
= - 2 .. -+(-3 m1.95 kN/m- +33.2 kN/m)'
+(3 m33.2 kN/m-9 rrr' .1.95 J
=- 0.98 kN/m2. t; + 27.3 kN/m . ( +90.8 kl'\f per meter of wall
The distribution of the hydrodynamic fluid pressure is given by
Pwd =('2 k
h
Yw . (2 =1
72
0.269.81 kN/m
3
(2
= 1.46 kN/m
2
. (2 per meter of wall
At the point of zero shear, the summation of the forces in the horizontal direction
is equal to zero
Substituting in the values and solving for (
(0.84 - 0.98 + 1.46) kN/m
2
t; + (9.6 + 27.3) kN/m (
+ (14.4 + 90.8) kN - 270 kN = 0
(=3.93 m beneath the fluid level
It is now necessary to sum the moments an elevation of (. This will provide the
maximum moment in the sheet pile wall. Table T8.6 provides the data for sum-
ming the moments about the tie rod. From the table, the maximum moment in the
...
sheet pile wall (M
FEs
) is approximately 450 kN per meter of wall, which occurs at
approximately 3.93 m beneath the water table.
The maximum moment is reduced using Rowe's (1952) moment reduction pro-
cedure (Fig. T8.16). This procedure utilizes a flexibility number (p), and estimates
the reduction in moment for various sheet pile sections.
(T8.1)
Table T8.6. Forces and lever arms for calculating the maximum moment.
Horizontal Horizontal force Lever arm about. Moment about
force (kN per meter of wall) tie rod (m) tie rod (kN'm
designation per meter of wall)
o:
(Pa2)x
(Pa3)x
(Ll?:ac)x
PWd
14.4
9.6'= 37.6
0.84.,2= 13.02
-0.98,2+ 27.3' + 90.8 = 183
1.49.,2= 22.6
o
'/2 + 1 = 3.0
213' + 1 = 3.62
1.169*
, + 1 - 0.4' = 3.36
o
113
47.1
214
75.8
450
*Using the stress distribution from Fig. T8.15.
408 PlANe
150 200
-, 'I\..
'\
-,
~
r-,
r-,
1/
I-- .- Penetration In mediumcompact andcompact coarse grained soils
r-,
......
.......
r-,
N
r-,
.....
......
r-....
<,
t"-
r-,
......
--......
r""'--o
~
rr--
t"'-
r--
r-.
r--
Penetration in verycompact coarse --t--
--
f---I r--
i""'-
~
grained soils
r---
~
-
0.3
0.2
5 6 7 8 10 12 14 16 18 20 30 40 50 60 70 80 100
(H+O.".J I1n
2
]
Value of P = EI in Ii) per running foot of wall
0.4
1.0
O.B
~ t 0.7
-
.s-
f 0.6
~
o 0.5
~
0.9
Fig. T8.16. Moment reductioncurves (afterRowe, 1952; as reponed in U.S. Navy, 1982).
where
E: Young's modulus (200 GPa)
I: moment of inertia (m4)
The flexibility number is then calculated by (and has been converted to English
units for the use of Rowe's chan)
(6m+lm+2m+3.75m)4 . 26427m
4
2189in
4
p= =-
200xl0
9
Pa I 200xH1 Pa I psi I
The allowable stress (O'allow) of the sheet pile is commonly taken as approximately
87% of the yield stress (O'y) for seismic design, where the yield stress for steel is
taken as 250 MPa.
O'allow =0.87O'y =0.87250xl0
6
Pa=218xl0
6
Pa
The allowable moment (Mallow) can be calculated as
Mallow = Z; . O'allow per meter of wall
where Z, is the elastic section modulus of the sheet pile wall (rn")
The sheet pile properties, and the design and allowable moments are shown in
Table T8.7 for several standard Bethlehem Steel sheet pile sections. From this
table, it can be seen that section PZ22 is the sheet pile wall section best suited for
this specific design scenario. The design bending moment is
M design = 203 kNm per meter of wall
Examples of Seismic Performance Evatuauon '+U,;/
Table T8.7. Calculations of the design and allowable sheet pile wall moments.
Sheet pile I
Ze p*
Mdesign/ MFES Mdesign 1 V / ~ l 1 o w
(Bethlehem Steel) (in" per ft (em) per
(l/psi per (Fig. T8.16) (kNm (kl"lm
of wall) m of wall) ft of wall) per m) per m:
PZ22 8.+.4 973 25.93 0.45 203 212
PZ27 18.+.2 1624 11.88 0.61 275 35..J.
PZ35 361.2 2607 6.06 0.86 387 568
PZ40 490.8 3263 4.46 1.0 450 711
* Rowe's flexibility number.
(7) Design of the tie rods (Tdcsi<>n)
"
The seismic design of the tie rods normally utilize a factor of safety of 1.3, there-
fore the design tie-rod force is
..
Tctesign = 1.3 .TFES = 1.3270 kl'\llm = 351 kN per meter of wall
Assuming a 2 meter spacing of the tie-rods, each tie-rod has a design load of
72msign = 2 m- 351 kN/m = 702 kN
Assuming a standard allowable stress in the tie-rods of 60% of the yield stress, the
minimum diameter of the tie-rod is calculated as
"( 72 m-design )
0.6 a,
Diameter = 4 . =
TC
(8) Design of the anchor
(
702 kN )
O.6250xHf Pa
4 =7.7 em
3.14159
The anchor design follows the same procedures as outlined for the design of the
sheet pile wall, with the seismic active pressures behind the anchor and TFES being
balanced by the seismic passive resistance in front of the anchor. An anchor
design has not been performed for this example, and the reader is referred to
Ebeling and Morrison (1992) for a complete design example.
(9) Location of the anchor
It is commonly recommended (Ebeling and Morrison, 1992) that the location of
the anchor be such that the active failure plane behind the sheet pile wall does not
intersect the passive failure plane in front of the anchor, as shown in Fig. T8.17.
The seismic active (lXae). and passive (llpe) angles of failure for a level surface
-410 PlANe
;: ... ~ : . ."... ~
Fig. T8.17. Recommended anchor location (North American Practice).
condition are calculated using
[
-tan(</>-lJI) + q ]
aae = </> - lJI + arctan
1+[tan(S+lJI)(tan(</>-ljf) + cot(</>-lJI))]
[
tan(</> -ljf) + C., ]
ape = -</> + ljf +arctan -
1+[tan(S+ljf)(tan(</> -ljf) +cot( </> -lJI))]
where
q = ~ t a n ( -o/)[tan(-0/) +cottc/> -0/)][1 + tan(S +0/) cot(</> -ljI)]
C2 = Jtan( -lJI)[tan( -lp) +cot(</> -lJI)][1 + tan(S + lp)cot(q>-ljI)]
T8.2.3 Simplified analysis (Japanese method)
(T8.2a)
(T8.2b)
(T8.2c)
(T8.2d)
The method adopted in Japanese design practice is based on the analysis of a
beam embedded in Winkler foundation (Takahashi et al., 1993: Ministry of
Transport, Japan, 1999). This method is essentially the same as Rowe's approach
to sheet pile wall (1952, 1955). In the following example, the first trial for embed-
ment depth was 4.0 m.
(1) Active earth pressures and thrust
The case of partially submerged soil is treated by idealizing the effect of overlay-
ing soil layer as a surcharge to the soil layer of interest: Eqn. (T7.7) is used
instead of Eqn. (T7.7). For simplicity, design surcharge qsur will be omitted in this
example (i.e. qsur =0). Active earth pressures at designated elevations are
Table T8.8. Seismic active earth pressures at designated elevations.
Elevation Thickness hi
X
8
l.Xhi
k' K
accos8 r;
h
(rn) (rn) (kN/m
J
)
(0) (0) (kN/m:!)
+ 3.0 3.0 17 40 15 0 0.26 0.35 0
-0.0 51 18
0.0 6.0 9 40 15 51 0.41 0.49 25
- 6.0 105 51
-6.0 4.0 9 40 15 105 0.46 0.55 58
- 10.0 141- 78
computed as shown in Table T8.8 using the equation
Pae = KII':cosoCi. Yihi+ qsur)
where Ii is (Ywel)i and (11,)i above and below the water table, respectively.
The active earth thrust is (see Table T8.9)
Pae =I. +{( Pae )IOp +(Pae )bollom}i hi =527 kN per meter of wall
(2) Passive earth pressures and thrust
(T8.3)
Passive earth pressures and thrust are calculated through a procedure similar to
that shown in Section T.8.2.2. Using typical wall friction angle in Japanese design
practice (i.e. 0=15), the passive earth pressure coefficient is calculated as
follows:
Ysal = 0.26. 19 = 0.55
Yb 9
K
_ cos
2
( - '112)
pe - z
r sin( +8) sin(- 'l'Z)]
cos'l'z cos('l'z +8) ll- .
cos(0+'l'z)
cos' (40 - 28.8)
= ------r--;:::==================:;2
cos 28.80cos(28.80 +15.00)[1- sin(40 +15.0) sin(40 - 28.8
0
)]Z
cos(15.0 +28.8)
=5.43
The horizontal passive earth pressure at the wall toe is given by
Ppe = KpecosoyiD
emb
=5.43xcos15.00x9.0x4.0=189 kN/m
z
4]2 PlANe
The horizontal passive earth thrust is then given by
Ppc =*189x4.0 =378 kN per meter of wall
acting at
z-; == 7.0+t4.O =9.7 m below the tie-rod level
(3) Hydrodynamic forces
The hydrodynamic forces are the same as shown in Section T8.2.2 and given as
P
wd
= 1
72'
k
h
Yw (Hw)2 = }7
2
0.26 9.81 kN/m 3(6 m) 2
=52.6 kN per meter of wall
The hydrodynamic force acts at
ZPWd = 7.0 - 0.4 . Hw = 7.0 - 0.4 6 m =4.6 m below the tie-rod
(4) Moment balance about the tie-rod
The moment balance about the tie-rod is
LMtie-rod = Ppc . Zppc - It{(ppe )lOp (z )lOp +(ppe )bonom (z )bollom } i h - Pwd z-;
= (378 kN) (9.7 m) - (see Table T8.9 = 3466 kN m) - (52.6 kN) (4.6 m)
= 41 kN . m out of balance
The depth of embedment of 4 meters results in a small conservative moment
imbalance. In the Japanese design practice, a safety factor of 1.2 is multiplied for
driving moment. If this practice is followed, the embedment depth of 4 meters is
unconservative.
(5) Simple beam analysis (Msimple and Tsimple)
Bending moments in the sheet pile and forces in the tie-rods are initially evaluat-
ed using the simplifying assumption that the contra flexure point is at the
mudline level (Tschebotarioff, 1949). Therefore, the sheet pile wall is assumed to
be a simple beam spanning between the tie-rod and the mudline levels for the pur-
pose of calculating both bending moments and the tie-rod force. These results are
later corrected for sheet pile flexure and foundation rigidity based on a beam
analysis in Winkler foundation. This correction is essentially the same as the use
of Rowe's reduction factor (Rowe, 1952, 1955).
For the simple beam supported at the tie-rod and the mudline levels, active
earth thrust above the mudline and driving moment about the tie-rod is computed
as (see Table T8.9)
( = I {(Ppe )top + (Ppe )hOllOln } i hi
above mud line
=255 kN per meter of wall
1
I'vEmudline = I 2 {(Ppe )top ( Z)top + (Ppe )bonom ( z)botlom }ihi + PWd . :Pwd
above mud line
=990+242=1232 kN'm
The equivalent reaction force exerted by the foundation (i.e. mudline support) is
Roundation = Mmudline / Hsimple = 1232/7.0 = 176 kN per meter of wall
where Hsimple: a span of the simple beam (m) (i.e. sheet pile wall span between tie-
rod and mudline levels)
The tie-rod force Tsimple is obtained by summing the forces in the horizontal
direction
I:imple = (Pac )mudline + Pwd - Rounda.tion
=255 kN + 53 kN -176 kN == 132 kN per meter of wall
The maximum moment occurs at the same elevation as zero shear within the sheet
pile. The zero shear is calculated as elevation z beneath the tie-rod level. Using the
tie-rod force Tsimple, earth thrusts in Table T8.9, and hydrodynamic pressures, zero
shear is found to occur between the elevation 0.0 and - 6.0 m. At the zero shear
location z:
(Pac )above y +(PWd )above: - 1';imple = 0
Table T8.9. Seismic active earth pressures at designated elevations along full height of
wall.
Elevation Thickness hi
(Pae)top (l/2)(Pae)top h, Distance below
(m) (m)
(Pa.:)bollom (l/2)(Pae)bonom hj
tie-rod elevation z,
(kN/m
2
) (kN/m
2
) (m)
+3.0 3.0 0 0 1.0 - 2/3x 3.0 = - 1.0
0.0 18 27 1.0 - II3 x 3.0 = 0.0
0.0 6.0 25 75 1.0 + II3 x 6.0 = 3.0
--6.0 51 153 1.0 +2/3 x 6.0 = 5.0
-6.0 4.0 58 116 7.0 + 1/3 x4.0 = 8.3
-10.0 78 156 7.0 + 2/3 x 4.0 =9.7
414 PlANe
=27+ 25+{25+(51-25) z }Z-1)+3.7(Z-W-132
= 2.17(z -Ii +3.7(z _1)1.5 +25(z -1) -105
=0
z = 3.9 m beneath the tie-rod level
It is now necessary to sum the moments about the tie rod, at an elevation of z. This
will provide the maximum moment in the sheet pile wall. Table T8.10 provides
the data for summing the moments about z. By multiplying the data shown on the
second and fourth columns in this table, and summing these moments, the maxi-
mum moment in the sheet pile wall is
Msimple= 269 kNm per meter of wall,
which occurs at 3.9 m beneath the tie-rod level.
(6) Corrections by Winkler beam analysis (Mdesign and Tdesign)
The results of the simple beam analysis are corrected based on Winkler beam
analysis (Takahashi et aI., 1993; Ministry of Transport, Japan, 1999). This
procedure is analogous to the use of Rowe's reduction factor and utilizes a
non-dimensionalized Rowe's flexibility number, called similarity number,
defined by
(T8.4)
Table T8.1O. Seismic active earth pressures at designated elevations.
Elevation Thickness hi (l/2)(Pae)lOp hi Distance below the Distance above the
(m) (m) (l/2)(P-ae)hollom hi tie-rod level z, elevation at :=3.9 m
(m)
2.9
3.0 + 3.0
0.0
0.0
- 2.9
+ 1.0
o -1.0
27 0.0
36 1.0 + 1/3 x 2.9 =2.0
54 1.0+ 2/3 x2.9 = 2.9
Tie rod Tsimple =-r- 133 0.0
Fluid Pressure Pwd= 18 1.0 + 2.9 x 0.6 = 2.7
4.9
3.9
1.9
1.0
3.9
1.2
where
p:
kh.,ul'l:
Hsimp\c:
Rowe's flexibility number (see Eqn. (T8.1))
coefficient of subgrade reaction (MFa/m)
a span of the simple beam (m) (i.e. sheet pile wall span between
tie-rod and mudline levels)
E: Young's modulus (MPa)
I: moment of inertia (m
4/m)
The Winkler beam analysis results in the following equations:
At/design = 4.5647w-
O
.
2
+0.1329
At/simple
(T8.5a)
'ldesign = +0.5514
'Tsimple
(T8.5b)
D
emb
=5.0916w-
o.2
-0.2591
Hsimple
(T8.5c)
The coefficient of subgrade reaction is determined based on Fig. T8.18. For SPT
N-value of 20, k
h
-
sub
=:: 30 Mpa/m. The similarity number is then calculated by
7
4
x30 0.36
W= =--
I2xl0
5
I
The Japanese steels of SY295 and SY390 have O"allow =180x 10
3
kPa and
235 x 10
3
kPa, respectively. The allowable bending moment can be calculated as
Mallow = Z: .O"allow per meter of wall
where Z, is the section modulus of the sheet pile wall.
c
80
.Q
t)
ca
(1)
ex:
60
(1)
"C
ca-
E

40
:Ja.

-- 0
-c:
20 (Takahashi et al, 1993)
(1)
'(3
:E
(1)
0
0
o
Fig. T8.18. Coefficient of subgrade reaction.
416 PlANe
The relevant parameters and design bending moments computed are shown in
Table T8.11 for typical sheet pile sections. From this table, it can be seen that SP-
III with ~ Y 3 9 0 steel or SP-IV with SY295 steel is the sheet pile wall section that
is suited for this specific design scenario. Design bending moments are
Mdesign = 300 or 339 kNm per meter of wall
The design tie-rod force is also computed using similarity number, as shown in
Table T8.12. The design tie-rod force is
T
design
= 140 or 149 kN per meter of wall
(7) Embedment depth and anchor location
Design embedment depth should be deep enough to satisfy the fixed earth support
condition. Using Eqn. (T8.5c) for sheet pile SP-III with SY295 steel, the design
embedment depth is determined as
D
emb
= 7.0xO.833 = 5.8 m
The recommended location of anchor is determined as shown in Fig. T8.19. The
active failure plane behind the sheet pile wall is assumed to originate from the
mudline level because the design/analysis is performed for fixed earth support
condition.
Table T8.11. Sheet piles and design bending moment.
Sheet pile type J (m
4/m)
Z, (m
3/m)
Similarity Mdesig/ Mdesign
(Japanese) number Msimple kNmlm
OJ Eqn. (T8.5a)
Mallow
kNmlm
SY295 SY390
SP-II
SP-III
SP-IV
0.87xlO-4 0.87xl0-
3
4.1x10
3
1.00 269
1.64xlO-4 1.31x10-
3
2.2x10
3
1.11 300
3.19xI0-4 2.06x10-
3
1.1x10
3
1.26 339
157 204
235 308
371 484
Table T8.12. Design tie-rod forces.
Sheet pile J (m
4/m)
Zc (m
3/m)
Similarity
Tdesig/ Tdesign
(Japanese) number
Tsimple
k.N/m
OJ Eqn. (T8.5b)
SP-II 0.87 x 10-4 0.87 X 10-
3
4.1 X 10
3
0.99
1... ..,
,)-
SP-III 1.64 x 10-4 1.31 x 10-
3
2.2 X 10
3
1.05 140
SP-IV 3.19xlO-4 2.06 x 10-
3
1.1 X 10
3
1.12 149
Examples of Seismic Pcrfonnance Evaluation 417
J2
m1
: contra flexure point
Fig. T8.19. Location of anchor (Japanese practice).
(8) Comparison of North American and Japanese methods
Major design parameters obtained through the simplified analyses in
Sections T8.2.2 and T8.2.3 are shown in Table T8.13. These results indicate the
following:
a) Sheet pile bending moment are computed in both design practices based
on Winkler foundation analysis using Rowe's reduction (correction) factor
but differences are noted because of the different earth pressure compu-
tations.
b) Design tie-rod forces significantly differ from each other because of the dif-
ferent conditions adopted for design (i.e. free earth support condition for North
American practice and Winkler foundation (Rowe's) condition for Japanese
practice).
c) Embedment depth and location of the anchor is also different because of the
different design conditions mentioned above.
The reader should be aware of these differences in the pseudo-static analysis of
sheet pile walls.
T8.2.4 Simplified dynamic analysis (deformations)
The sheet pile wall determined in Section T8.2.2 was used for the simplified
dynamic analysis. The simplified dynamic analysis was performed to determine
the extent of soil improvement necessary to meet Grade B performance require-
ments stated in Section T8.2.1.
This example utilizes the empirically developed design chart proposed by
McCullough (1998) and McCullough and Dickenson (1998) (see Section T7.2 in
Technical Commentary 7). The design chart shows non-dimensional parameters
with respect to a wall displacement and a soil improvement area for a quay wall
under prescribed earthquake motions. The backfill conditions are specified by the
SPT N-values equal to 10 or 20.
418 PIANC
TableT8.13. Comparison of simplified analysis results.
Analysis method Mdesign Tdesign D
emb
(kNmlm) (kN/m) (m)
North American practice 203 351 3.75
Japanese practice 300 to 339 140 to 149 5.8
By referring to the sheet pile wall section shown in Fig. T8.13, the sheet pile
wall parameters for a PZ27 sheet pile section are
E= 2xl0
8kPa
1= 2.51 X 10-4 m
4/m
The wall height (H) is 9 m, and the wall depth (D
emb
) is 3.75 m. The soil buoyant
unit weight (Yb) directly adjacent to the wall is 9 kPa.
Substituting these values into the normalized lateral displacement factor yields
d- EI d (2xlOS kPa) (2.51xl(j4 m
4/m)
d
------- ---
(H +Demb)sYb (9 m+3.75 m)' (9 kPa) 60.4
where d is the displacement at the top of the wall.
The ground motion intensity factor is determined for the earthquake motions
using Table T7.8 in Technical Commentary 7 to determine the magnitude scaling
factor (MSF) and Fig. T8.20 to determine the depth reduction factor (r
d
) , using the
following equation:
llmax(dl llmax . 1d
MSF MSF
The magnitude scaling factor values for both earthquake motions (LI and L2) are
tabulated in Table T8.14.
The normalized lateral displacement factor is then determined as shown in
Table T8.15 as a function of soil improvement (next) using Fig. T7.22 in Technical
Commentary 7 for uniform normalized blowcounts of 20.
By plotting the values tabulated above, it is possible to develop a figure that
relates the extent of soil improvement to lateral displacements (Fig. T8.21). It is
interesting to note that for this example, it is practically impossible to limit the
deformations to 10.3 em for the Ll motion and 13.9 em for the L2 motion,
utilizing densification of the sandy backfill. Other forms of mitigation (e.g. soil
treatment by cementation/grouting or strengthening, and structural mitigation
measures) may be necessary to limit earthquake-induced deformations to
smaller values. It can also be noted that soil improvement beyond approximately
30 meters has little effect on the earthquake-induced lateral displacements.
Based on Grade B requirements, and the estimated displacements highlighted
above, the length of soil improvement needs to be approximately 30 meters to meet
o 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0 r - - - , - - _ r _ - - , - - , - - , . - - - - r - - ~ ~ - _ r _ _ _ .
AVERAGE VALUES
RANGE FOR DIFFERENT
SOIL PROFILES
6
9
:[
12
...
s:
15
C.
CIl
c 18
21
24
27
30
3
Fig. T8.20. Range of r
d
values for different soil profiles (after Seed and Idriss, 1982).
Table T8.14. Magnitude scaling factors.
Qmax(surface)
r
d
(at 9 m)
Qmax(dl
MSF Qmax(d/MSF
Ll 0.45 0.92 0.41 1.63 0.25
L2 0.60 0.92 0.55 1.00 0.55
Table T8.15. Lateral displacements.
Normalized soil Length of soil
improvement, neXl improvement, L
exl
(m)
Normalized lateral
displacement, dJH
Lateral
displacement, d
(m)
Exceeds limits of design chart
0.014 0.85
0.006 0.36
0.0035 0.21
0.003 0.18
0.0025 0.15
0.0023 0.14
o 0
1 15
Ll 2 30
3 45
8 120
o 0
0.5 7.5
1 15
L2 1.5 22.5
2 30
3 45
8 120
0.011
0.0035
0.0021
0.0019
0.0017
0.66
0.21
0.13
0.12
0.10
420 PlANe
80
70
-
E
~ 60
.:
c: 50
Q)
E
Q)
~ 40
c..
:6 30
e
Q) 20
~
10
\
\
.
,
L1 motion
~
,
- - - .. L2 motion
\

,
,
\
,
,
\
\
\
\
,
\
,
,
"-
--..
a. ____
------ -----
-----
~
20 40 60 80
Soil improvement (rn)
100 120
Fig. T8.21. Lateral displacement for the design problem.
Grade B requirements of 1.5% deformation (d/H) during the L1 event (deforma-
tion of 13.5 em). Note that these same criteria cannot be met for the L2 event.
T8.3 SHEET PILE QUAYWALL (GRADE S)
An existing sheet pile quay wall with a water depth of - 11.5 m was proposed for
upgrading into a quay wall with a water depth of - 16.6 m. The performance
grade of this new quay wall was designated as Grade S '(see Section 3.1 in the
Main Text). Based on seismic hazard analysis, the earthquake motions from the
active seismic fault located nearby was dominant and determined as L2 motion
with an estimated earthquake magnitude of M
JMA
= 7.5. PGA at the bedrock level
was evaluated as O.53g. Based on the needs of the owners/users of the quay wall,
extremely rigorous damage criteria were determined as shown in Table T8.l6. In
particular, 'elastic' in this table was defined as the state that the extreme fibre
stress is less than the allowable stress.
Proposed cross section of the sheet pile quay wall is shown in Fig. T8.22.
Two different types of anchorage, i.e. battered and vertical pile types, were
installed with two sets of tie rods connected separately. Profiles of natural ground
soil layers are summarized in Fig. T8.23. The bottom boundary for the analysis
was chosen at - 46 m where the bedrock layer was encountered.
Table T8.16. Damage criteria for a sheet pile quay wall.
Extent of damage Degree I
Residual
displacements
Peak response
stresses
Normalized residual horizontal
displacement (d//{)*
Sheet pile wall
Tie rod
Anchor
Less than 1.5%
Elastic**
Elastic**
Elastic**
*d: Residual horizontal displacement; H: height of sheet pile wall from mudline.
**Less than allowable stress.
Sheet ile
Design water level
.16.6m
H.W.L. +2.0m
iL.W.L. to.Om
Fig. T8.22. An example of sheet pile type quay wall.
D.L-11.Sm
Soil
Yb=10 kN/m
3
-19.5m
=3S"
Clay
Yb=7.4 kN/m
3
-28.Sm
c=515kPa
Clay
Yb=5.1 kN/m
3
-33.0m
c=390kPa
Clay
Yb=6.5 kN/m
3
-46...... S2m
c=475kPa
Fig. T8.23. Natural ground soil layers.
422 PlANC
Since the required performance was Grade S, dynamic analysis was performed for
seismic performance evaluation for L2 earthquake motion (see Section 5.1 in the
Main Text). The method used and the evaluation of input parameters were similar
to those for the gravity quay wall analysis shown in Section T8.1.5.
Cross section of the sheet pile quay wall was idealized into a set of finite ele-
ment mesh shown in Figs. T8.24 and T8.25. Parameters for structural materials
are summarized in Table T8.17.
Computed deformation' and computed distribution of excess pore water pres-
sure ratios are shown in Figs. T8.26 and T8.27. The computed displacement at the
top of the sheet pile was 13 em, and was confmned to satisfy the displacement
criterion of less than 30 em (equal to 1.5% of the wall height of 20.4 m).
Computed bending moments of the sheet pile wall and vertical pile anchor are
shown in Fig. T8.28. Computed stresses of structural components are summarized
in TableT8.I8. Only the tie-rod force between the sheet pile wall and battered pile
anchor was beyond the allowable tensile force. Consequently, it was recommend-
ed to replace the existing tie-rods with higher strength tie-rods.
10
. Structure scale (m) 0,,-._--,
-28.5m
-33.0m
Backfill sand (Not improved)
Backfil sand (Improved) J
\ +3.8m
:.um
Groundwatar
level
Concrete cap
<,
)7
+ - : ~ Brokenrubble
II
ackhll
Concreie blocks
rubble
.
I
Ag+Dsl
~ . 6 m
Dcl+Dscl
~
t I I
-33.0ft0
Dc2
I I I I
I
Dc2+D8C2
II 1
~
I I II I
Fig. T8.24. An example of soil layers of sheet pile type quay wall.
10
,
Structure scale (rn) 0,--,_--,
sz
-e-
,
I
,
Fig. T8.25. An example of FEM mesh for sheet pile case.
Table T8.17. Parameters for structural materials.
Shear Poisson's Density Cross sectional Geometrical
modulus ratio (t/rn') area per unit moment of
(kPa)
~
breadth (rnvm) inertia per unit
breadth (rn'zrn)
Sheet pile 8.1 x 10
7
0.30 7.85 8.14 x 10-
1
7.31 x 1 O - ~
Coupled pile 8.1 x 10
7
0.30 7.85
~
1.89 x 10-
3
1.55 x 10--
Vertical pile 8.lxI0
7
0.30 7.85 2.66 x 10-
2
3.18 X 10-
3
Tie rod between 6.2 x 10
7
0.30 7.85 9.65 x 10-4 0.0
sheet pile and
coupled pile
Tie rod between 6.2 x 10
7
0.30 7.85 1.75 x 10-
3
0.0
sheet pile and
vertical pile
10
, Structure scale (m) L-I__
"
..
SZ
-
Fig. T8.26. Computed deformation of sheet pile quay wall.
T8.4 PILE-SUPPORTED WHARF (GRADE B)
A pile-supported wharf with a water depth of 12 m was proposed for construction.
The performance grade of this new wharf was designated as Grade B over the
life span of 50 years (set? Section 3.1 in the Main Text). A seismic hazard
and response spectrum analysis estimated that the maximum deck level accelera-
tions are 0.30g and 0.57g for the 50% (L1) and 10% (L2) probability
of exceedance, respectively. Damage criteria were determined for the pile-deck
structure, as shown in Table T8. 19. Damage criteria for Degree II were not defined
because the performance of a Grade B structure is specified only by damage
degrees l-and III (see Section 3. I in the Main Text).
424 PlANe
10
,
Strudure scale (m) 0,-,_---'
Pore pressure ratio
UI a
mo
' :
80.2-0.4
m:304-0.6
_06-08
.0.8-0.9
.09-1.0
Fig. T8.27. Distribution of excess pore water pressures after shaking.
The proposed cross section and plan of the pile-supported wharf are shown in
Fig. T8.29. Geotechnical parameters, including the coefficient of subgrade reac-
tion, were determined from a geotechnical investigation and are given in Table
T8.20. The wharf was supported by four rows of 1.2 m diameter steel pipe piles.
Piles in rows 1 through 3 have a wall thickness of 12 mm, and the piles in row 4
have a wall thickness of 14 rom. Structural parameters for these piles are given in
Table T8.21. Loads considered in the desizn include a 10 kN/m:! surcharze on the
deck, 30 kN/m
2
deadweight of the d e c k , ~ and crane loads of 2400 k.N~ per unit
framework of the pile-deck system. The total load on the deck, including the deck
deadweight for a unit framework with a breadth of 5.0 m, is 6600 kN. The unit
framework considered for design is indicated by hatching in Fig. T8.29(b).
The Grade B example analyses presented in this section include a simplified
analysis used for the preliminary design and a simplified dynamic analysis based
on the pushover method to satisfy Grade B design requirements (see Section 5.1
in the Main Text).
(T8.6)
The simplified analysis method is being used to estimate the elastic performance
of the structure. For this simplified analysis, the natural period of the pile-deck
system is computed by
t=2n: ~
. ~ ~
5
5
a ... -.. .. . ... ,. , . ,
E -5
.......
c
.2-10
...
co
>
~ - l S
UJ
-20 ..... - .. . ... - ...
-25
-1000 -500 a 500 1000
Sheet pile after shaking
(kNm/m)
a . .. . .
E -5
.......
c:
.2 -10
....
~
>
(1)
ijj-15
-20 .
-25
-1 000 -500 a 500 1000
Vertical pile after shaking
(kNm/m)
5
o
E -5
.......
c
0-10
.,;j
co
~ -15
ijj
-20
...... "." ......
.......'" ........
5
o
E -5
.......
c
o -10
',;j
~
~ -15
jjJ
-20
... ':-' - -, ... :....
......." .. " .... .. .... 1 ........
II
.. "" ..
......." .. .. .... .. ... 1 .......
-25
-1000 -500 a 500 1000
Sheet pile at maximum
(kNm/m)
-25
-1000 -500 a 500 1000
Vertical pile at maximum
(kNm/m)
(TS.7)
Fig. T8.28. Computed bending moment of structures.
where
T; natural period of a pile-deck system (s)
~ : surcharge and deadweight of deck (kN)
g: acceleration of gravity (=9.8 m/s')
KHj : equivalent spring stiffness for lateral reaction by r pile (kN/m)
The equivalent spring stiffness in Eqn. (T8.6) was computed based on an equiv-
alent fixity pile by
K
Hj
= 12El
j
. (
4
+ ~ i J
where
E: Young's modulus of elasticity (kN/m
2
)
l: moment of inertia of i-th pile (m")
426 PlANC
,:.
Table T8.18. Computed stresses of structures.
Tie rod between sheet pile and coupled pile
Allowable tensile force 1100 (kN)
Tie rod between sheet pile and vertical pile
Allowable tensile force l247 (kN)
Structure
Sheet pile
Section modulus
Allowable stress
Vertical pile
Section modulus
Allowable stress
Coupled pile
Sheet pile
Allowable stress
1.22 X 10-
1
(m
3/m)
181 (N/mrn
2
)
6.36 X 10-
3
(m
3/m)
279 (Nzmm'')
3.78 X 10-
3
(m
3/m)
206 (Nzmm')
Unit Maximum value
Bending moment 6170
(kNmlm)
Fiber stress 50.6
(N/mm
2
)
Bending moment 1320
(kNmlm)
Fiber stress 208
(Nzmm'')
Bending moment 350
(kNmlm)
Fiber stress 93.3
(Nzmm'')
Axial force 1686
(kN)
Axial force 1186
(kN)
Table T8.I9. Damage criteriafor pile-supported wharf (Grade B).*
Extent of damage Degree I Degree II Degree III Degree IV
N/A
N/A
N/A N/A
Residual Differential Less than N/A N/A
displace- settlement 0.1-0.3 m
ments between deck
and land behind
Residual tilting Less than
towards the sea 2-3
0
Peak
response
Piles** Essentially N/A
elastic response
with minor or
no residual
deformation
Ductile response Beyond
near collapse the state
(double plastic of Degree
hinges are allowed III
to occur at only
one pile)
*Modified from the criteria shown in Table 4.3 in the Main Text, due to the specific per-
formance requirements of the pile-supported wharf.
**Bending failure should precede shear failure in structural components.
Ii: unsupported length of i-th pile between the deck and slope surface (rn)
1/f3i: length of equivalent fixity pile below the slope surface (m) (see Eqn.
(T7.25) in Technical Commentary 7)
For obtaining the unsupported lengths of piles l.; a virtual slope surface
was considered in design as indicated by the broken line in Fig. T8.29(a)
21.0
30 5.0 5.0 5.0 3.0
~ 1 ~ 5 1 I i Q ~ ~ ~ ~ ~
+3.00 lU..l.lJ.i
. L.w.L. +0.00
-1200
-2700
-32.00
Clay
Sand
';1.2
Unit (m)
Sand 34.00
Pile-1 Pile-2 Pile-3 Pile-4
(a)
21.0
3@S=15.0
o
o
N 0
II liS
l() N
@
~
Fig. T8.29. An example of pile-supported wharf (Grade B).
(a) Cross section. .
(b) Plan.
(see Section T7.3 in Technical Commentary 7). From the geotechnical and
structural parameters in Tables T8.20 and T8.21, the parameters needed in
Eqns. (T8.6) and (T8.7) were calculated, and are shown in Table T8.22. In
particular, f3i for pile rows 1 through 3 was computed using Eqn. (T7.25) in
Technical Commentary 7 as
29000 x 1.2 = 0.28 (m-I )
4x 2.06x Iff x 0.00723
(T8.8)
428 PIANC
Table T8.20. Major geotechnical parameters for pile-supported wharf (Grade B).
Soil layers Density Coefficient of subgrade Internal friction angle t/J
(t/m") reaction k
h
sub
(kN/m
3
) (0) or unconfined
compressive strength qu
(kN/m
2
)
Rubble 1.9
Soil layer 1 (Clay) 1.6
Soil layer 2 (Sand) 2.0
Soi11ayer 3 (Sand) 2.0
29,000
29,000
117,000
290,000
Table T8.21. Major pile parameters.
Type of parameter
Diameter (m)
Thickness (m)*
Cross sectional area (m')
Moment of inertia (m4)
Elastic section modulus (m')
Yield stress (kN/m
2
)**
Young's modulus (kN/m
2
)
Pile parameters
Piles 1 through 3
1.2
0.011
0.0410
0.00723
0.0120
315.000
2.06xI0
8
Pile 4
1.2
0.013
0.0484
0.00850
0.0142
315,000
2.06x10
8
(T8.9)
*Cross-section area and moment of inertia are computed by allowing loss of cross section
in 1 mm thickness due to corrosion.
**Steel used was SKK490 in JIS-A-5525.
Table T8.22. Equivalent spring coefficients and relevant parameters of piles.
Parameters Pile 1 Pile 2 Pile 3 Pile 4
1; (m) 14.4 13.1 11.9 11.0
1/f3; (m) 3.6 3.6 3.6 3.8
1; + 1/f3i (m) 18.0 16.7 15.5 14.8
Eli (kNm
2
) 1,490,000 1,490,000 1,490,000 1,750.000
K
Hi
(kN/m) 3,070 3,840 4.800 6..+80
Thus the natural period of the pile-deck system was computed for pile rows 1
through 4 by Eqn. (T8.6) as
t; = 21C 6600 = 1.2 s
9.8 x (3070 + 3840 +4800 + 6480)
The approximate elastic limit (i.e. threshold level) of this pile-deck system was
evaluated based on Eqns. (T7.26) and (T7.27) in Technical Commentary 7 as
follows.
Examples of Seismic Performance Evaluation 42.9
The plastic hinge moment withoutaxialload MrAj is given by
MpO =Zp(}y
where
(T8.l0)
Z, = (4/3){rp3 -(rp _t
p
)3} : sectional modulus at plastic hinge (ru')
r
p
: pile radius (m)
t
p
: wall thickness of the pile (m)
O'y: yield stress (kN/m
2
)
The computed parameters shown in Tables T8.21 and T8.22 resulted in the
plastic hinge moment for pile rows 1 through 3 being
MpO = 1{0.6cY -(0.60-0.011)3}x315000=4900 (kNm)
and for pile row 4 as
MpO = 1{0.6cY -(0.60-0.013)3}x315000=5770 (kNm)
(T8.11a)
(T8.11b)
The plastic hinge moment with axial load N is computed by (e.g. Yokota, 1999)
Mp = MpO cos(l!- ;r;1 (T8.12)
NyO 2
where
N: axial load (kN)
N
yo
: yield axial load without bending moment (kN)
The axial loads due to the surcharge, deadweight, and crane, as well as the cal-
culated plastic moments are tabulated in Table T8.23.
The limit lateral load Pu' when plastic hinges form at pile top and in the embed-
ded portion' of the piles was computed using Eqns. (T7.27) in Technical
Commentary 7 as
P: = L 2Mpi = 2x3530 + 2x4860 + 2x4860 + 2x5730
I, + ~ 18.0 16.7 15.5 14.8
'i j3j (T8.13)
= 2380 (kN)
and the approximate elastic limit P; was computed using Eqn. (T7.26) in
Technical Commentary 7 as
Py = 0.82Pu = 0.82 x 2380 = 1950 (kN)
(T8.14)
Table T8.23. Plastic hinge moment of piles.
Parameters Pile 1 Pile 2 Pile 3 Pile 4
MpOj (leNm) 4900 4900 4900 5770
Nj (leN) 6300 1000 1000 1100
;(leN) 12900 12900 12900 15200
M
pj
(kNm) 3530 4860 4860 5730
430 PlANe
Using the peak deck response of 0.30g for the Ll earthquake motion, as
determined from the seismic hazard and site response analysis, the seismic load
during Ll earthquake motion was estimated as
F = 0.30 x6600 =1980 (leN)
(T8.15)
It can be noted that F is nearly equal to Py in this preliminary study, indicating that
the proposed design pile-deck system may satisfy the required performance crite-
ria for Levell earthquake shown in Table T8.19.
As mentioned earlier, the seismic hazard and site response analyses resulted in
peak deck responses of O.30g and O.57g for the Ll and L2 earthquake motions,
respectively. The subsoil, including the slope under the deck, was considered a
non-liquefiable granular material.
A pushover analysis was performed using the .geotecbnical and structural
parameters in Tables T8.20 and T8.21. In this analysis, bi-linear (elasto-plastic)
springs were used to represent the subgrade and axial reactions of the subsoil. The
deck was idealized as a non-linear beam having-a tri-linear moment-curvature
relationship. The piles were idealized as non-linear beams having a bi-linear
moment-curvature relationship, as shown in Fig. T8.30, and specified by the fol-
lowing equations:
M=
EZer
MpO cos(l!- 1t)
NyO 2
(T8.l6)
where
M
E: Young's modulus of elasticity (kN/m
2
)
Ze: elastic sectional modulus (rrr')
r: pile radius (m)
~ - - - - - - - - - - - - - - - - - - - - ~ ; - - - - - - - - - - - -
cP
v
cP
P
Pile Curvature
Fig. T8.30. Moment-curvature rela-
tionship of piles for push-
over analysis.
(T8.17)
: pile curvature (m-
I
)
p: pile curvature at the onset of a plastic hinge (rn')
MpD: plastic hinge moment without axial load (kNm) (Eqn. (T8.10
N: axial load (kN)
Nyu: yield axial load without bending moment (kN)
The yield bending moment (My), shown in Fig. T8.30. is given by
M
v
=(a
v
- N )Ze
- . A)
where A: pile cross-sectional area (m").
The computer program N-Pier (Yokota et al., 1999) was used for the analysis.
This program has been made available (downloadable) free of charge from the rel-
evant web site (http://www.phri.go.jp). although the current version is limited for
use on personal computers with a Japanese fonts/characters.
The pushover analysis resulted in the load-displacement curve shown in
Fig. T8.31. The P; value was determined as the 'break-point' in the curve and the.
P
u
value was determined as the point at which a pile develops a double plastic
hinge (at the pile cap and in the embedded portion of the pile). It was determined
from the curve that points P, and P
u
in this figure. were the upper limits for
'damageDegrees I and III, as defined in Table T8.19.
The seismic loads due to the peak deck responses for the LI and L2 earthquake
motions were compared to the limiting loads for Damage Degrees I and III in
Table T8.24. In this comparison, the equivalent elastic limit was chosen at the load
level P; and a displacement ductility factor J1 was defined using this equivalent
elastic limit as J1 =d/d
u
Equation (T7.43) in Technical Commentary 7 was used
to compute the seismic loads for a ductile response during a L2 earthquake
motion. The results tabulated in Table T8.24 indicate that the proposed design sat-
isfies the required performance criterion for peak responses during both the L1
and L2 earthquake motions.
The pushover analysis resulted in tilting of less than 1
0
, less than the 2-3%
required for the Degree I performance. Differential settlement between the deck
and back land was inferred from the horizontal displacement of the deck. Since
the peak horizontal displacement of the deck during L1 earthquake motion was
estimated as 0.2 m, the differential settlement was considered to be less than this
value. This is a crude estimate, and a separate analysis should be performed to
evaluate the settlement at the crest of the slope if this is a primary concern for this
specific structure. Instead of performing a separate analysis, a relevant earthquake
case history was investigated, and it was concluded that the proposed design
structure satisfies the criteria with respect to the differential settlement shown in
Table T8.19.
432 PlANe
3000 r---------------------------,
P
u
2500
- P
y
:
2000 .- plastichinge
c, pile2: plastic hingeat piletop
] pile-1: plastichingeat piletop
.3 1500 pile-4: plastichingeat piletop
It) pile3: plasti.c hingeat piletop
- I CI) I
. 1000
500
0
0 50 100 150 200 250 300
d
350 400
d
y u
(a) Lateral Displacement d(mm)
10000
P
y
1000
Z
cr
"0
It)
100 0
...J
It)
-
CI)
-
It)
...J
10
1000
10 100 d,
Lateral Displacement d(mm)
__-l. __'___'__ ---.J
1
(b)
Fig. T8.31. Results of pushover analysis.
(a) Plotting onregular scale.
(b) Plotting on log-log scale.
T8.5 PILE-SUPPORTED WHARF (GRADE S)
An existing bulkhead with a water depth of - 7.0 m was proposed for upgrading
by constructing a new pile-supported wharf at a water depth of - 14.6 m in front
Table T8.24. Seismic loads and limit loads based on the pushover analysis and the
criteria.
Earthquake level
PGA at deck level
Ductility factor ,u
Seismic loads F (kN)
Limit loads P, and P; (kN)
Limit displacements d; and d; (m)
Tilt (0)
Level I
0.30g
1.0
1,980
2,380
d; =0.21
< 1
Level 2
0.57g
1.5
2.660
2.700
d; =0.31
< 1
Table T8.25. Damage criteria for pile-supported wharf (Grade-S).
Extent of damage Degree I
Residual
displacements
Peak response stresses
*Less than yield stress.
Deck
Piles
Retaining
structure
Differential settlement between
deck and land behind
Pile top
Sheet pile wall
Tie rod
Anchor
Less than 0.3 m
Elastic*
Elastic*
Elastic*
Elastic*
of the existing sheet pile wall. The performance grade of this new wharf was
designated as Grade S over the life span of 50 years (see Section 3.1 in the Main
Text). Based on seismic hazard analysis, the L2 earthquake motion with a
probability of exceedance of 10% was 0.39g at the bedrock. Damage criteria were
determined for pile-deck structure and sheet pile retaining structure, as shown in
Table T8.25 .. In particular, 'elastic' in this table was defined as the state where the
extreme fibre stress is less than the yield stress.
A proposed cross-section of the pile-supported wharf is shown in Fig. T8.32.
Geotechnical investigations shown in Fig. T8.33 revealed a liquefiable layer
2.3 m thick below the elevation of -17.3 m. The effect of this liquefiable layer on
the performance of the pile-supported wharf was one of the primary issues inves-
tigated during the seismic performance evaluation. Initially, the proposed cross
section shown in Fig. T8.32 had partially improved portion of the liquefiable layer
by a cement deep mixing stabilization method (CDM). Rubble materials were
used above the CDM improved area to form a dike in front of the sheet pile retain-
ing wall. The combined resistance by the sheet pile wall and the CDM portion was
the main issue studied in the seismic performance evaluation.
Since the required performance of this pile supported wharf was Grade S, dynam-
ic analysis was performed for seismic performance evaluation for L2 earthquake
434 PlANe
~ . L.I.75
vCt. Loll
..~ . _ __._._ _ _ .
Fig. T8.32. An example of vertical pile supported wharf.
(a)
0.. .. _ .. .. .. ..
(h)
Fig. T8.33. Boring logs.
(a) Parallel to the face line of the wharf.
(b) Transverse to the face line of the wharf.
10
, Structure scale (m) 9
'"--_......
Vertical pile ancnorage
Sheel pile
l
Piled pier
1
,-
,
Clay
". r .
Backfill
Sandy gravel
rubble
Sandy clay
Y.:
" ' ~
~
ubble L.-
Rubble Gravel
... Slllv ciav "Siltv ciav
Sandl!lQuenablel COM SandCliquetiable)
I--
-
Grave Gravel
Sandy silt
Sandy silt
Silly clay Silly clay
Sand Sand
Silly clay
11111
looo.
Gravel
I""" r-
Silty clay
Clay
Sand
Sandy silt
Sand
III
Silly clay
1111
Fig. T8.34. An example of soil layers of vertical pile supported wharf.
Structure scale (m) L-J0
It
I
Fig. T8.35. An example of FEM mesh for vertical pile supported wharf.
436 PIANC
Linear plane element
--- --
-- ---
COM
Table T8.26. Parameters for CDM.
Fig. T8.36. Modelling of piles in CDM improved
layers.
Bulk modulus Internal friction Cohesion
x, (kPa) angle tP( (0) c (kPa)
Unconfined Young's
compressive modulus
strength qu (kPa) E (kPa)
1320 6.6 x io'
Soil Sheet pile Soil
t t t
Initial shear
modulus
Go (kPa)
2.5 X 10
5
6.5 x io' o 662
\ V
Fig. T8.37. Modelling of sheet pile.
50,..-----------------------,
40 -
-o
-
30 I-
20 -
*
0 \
.a-

: ..
10 I-
Alluvial
Pleistocene
... Pleistocene(tensile)
I
20
I I I I
30 40 50 60
Plasticity index Ip
I
70 80
Fig. T8.38. Friction angle and plasticity index for normally consolidated clay.
Table T8.27. Parameters for structures.
Vertical Sheetpile Vertical Tie rod Pier slab
pile of steel pile
pipe piles anchorage
Outside diameter (rn) 1.500 1.100 0.800 0.065
Thickness (rn) 0.019 0.016 0.009
Cross sectional area (rn'') 8.840E - 2 5.449E - 2 2.236E - 2 3.320E - 3 3.30*
Geometrical moment of 2.420E - 2 8.00E - 3 1.75E - 3 (Assumed 0.958*
inertia (rn') to be 0)
Section modulus (m') 3.232E - 2 1.460E - 2 4.37E - 2
Shear modulus (kPa) 7.9E + 7 7.9E + 7 7.9E + 7 7.9E + 7 1.27E+7
Poisson's ratio 0.3 0.3 0.3 0.3 0.20
Density (t/rn') 7.85 7.85 7.85 (Assumed 2.87**
to be 0)
Effective shear area ratio 1/2 1/2 1/2 (Assumed 1/1.2
to be 0)
*Per 6 m breadth.
** Including the weight of pavement.
Deformation at the top of pier
Residual Maximum
Horizontal
51 66
Vertical 4 4
Deformation at the top of sheet pile
Residual Maximum
Horizontal 55 76
Vertical 11 13
Structure scale (m) ~ O
Displacement (m) - 0.50
""
.. . .
, "' I.' I
I , I , , ""
: : : : : : : ::::. . ..
I t.
o , , I , ,
/
(ern)
Fig. T8.39. Computed deformation of vertical pile supported wharf.
motion (see Section 5.1 in the Main Text). The method used and the evaluation of
input parameters were similar to those for the gravity and sheet pile quay wall
analyses discussed in Sections T8.1.5 and T8.3.
The cross section of the pile-supported wharf was idealized into finite elements
shown in Figs. T8.34 and T8.35. Modelling of the piles in the CDM improved area
is shown in Fig. T8.36. Piles and CDM elements were not connected directly but
438 PIANC
Strudure scale (m) L-2,0
Porepressure ratio
U100mo':
803-06
.0.6-0e
.o.e-o&
.0.&-'.0
S7
mdll
Fig. T8.40. Computed maximum excess pore water pressure ratio of vertical pile
supported wharf.
Table T8.28. Computed maximum stresses of structures and yield point.
Vertical pile of pier Sheet Vertical Tie
Total maximum
pile pile rod
Bending moment Axial force related
ancho-
related stress stress stress rage
Sea Middle Land Sea Middle Land Sea Middle Land
side side side side side side
Yield
315 315 315 315 315 440
stress
Computed 284 204
.,.,.,
-22 23 -4 306 228 227 123
229
-_...
stress
connected through linear plane elements to allow free response of CD!\.1 portions
between the piles. Parameters for CDM elements based on the unconfined com-
pression strength qu are summarized in Table T8.26. Modelling of sheet pile at the
bulkhead is illustrated in Fig. T8.37. Joint elements are used to simulate the fric-
tion at the soil-structure interface. The angle of internal friction for clay layers
was determined based on the existing laboratory data shown in Fig. T8.38 (after
Tsuchida. 1990). Parameters for structural components are summarized in
Table T8.27.
The computed deformation and computed maximum excess pore water pres-
sure ratio are shown in Figs. T8.39 and T8.40. Computed deformation resulted in
Examples of Seismic Performance Evaluation .+39
the differential settlement of 0.5 m between the deck and the land behind. This
differential settlement was considered not acceptable based on the damage crite-
ria set forth in Table T8.25. Liquefiable layers were completely liquefied as shown
in Fig. T8AO. and it was speculated that liquefaction remediation measures for
wider portions of the liquefiable layer could be effective in reducing deformation.
Computed maximum stresses and yield stresses of structural components are
summarized in Table T8.28. The performance of all the structural components sat-
isfies the damage criteria except for that of the vertical pile anchor. Consequently,
liquefaction remediation measures for wider portions of the liquefiable layer were
recommended for the next cycle of seismic performance evaluation. The sequence
of seismic performance evaluation with a modified cross section was repeated
until the damage criteria were satisfied with optimum area of soil improvement.
List of Symbols
A
Acps(f)
A
e
Agross
A
h
A
hs
Age
Asp
At
a,b
,
a
adesign
amax(surface)
amax(rock)
at
aCt)
cross sectional area of a pile
amplitude of each frequency component (f) of the CPS
effective shear area (= O.8A
gross
)
uncracked section area of a pile
total area of transverse reinforcement, parallel to direction of
applied shear cut by an inclined shear crack
area of stabilized hysteresis loop
total area of dowel bars in the connection
spiral or hoop cross section area
total required area of the headed bond bars
empirical constants
maximum acceleration of a linear SDOF system
design acceleration
equivalent acceleration at the layer depth
peak acceleration
peak acceleration within the backfill at the elevation of the dredge-
line
peak acceleration at ground surface
peak acceleration at bedrock
threshold acceleration
time history of acceleration
yield acceleration of elasto-plastic SDOF system
correction factor of measured SPT blowcount for borehole diameter
correction factor of measured SPT blow count for hammer energy
ratio ER
m
correction factor for N
65
, based on the fines content
R
max
correction factor, for earthquake motion irregularity
scaling factor for earthquake moment magnitude
correction factor of measured SPT blowcount for overburden stress
correction factor for CPT tip resistance
correction factor of measured SPT blow count for rod length
442 PlANe
C
s
cc.;
co.u,
C
,
C
C
ex
Co
C
u
Cv
D
D
D
D'
D*
Dane
D
emb
Dr
D
p
D'
p
D(y>
[D(Y)]lab
Do
Dl
d
d'
-
d
db
d
r
d,
d
m
d
p
d
r
d
u
d
y
E
E
Ed
En
correction factor of measured SPT blow count for samplers with
or without liners
maximum value of Cross Correlation function
coherence function of signals x(t) and yet)
cohesion of soil
effective stress cohesion
depth from extreme compression fiber to neutral axis at flexural
strength (see Fig. T5.l3)
concrete cover to center of hoop or spiral bar
undrained shear strength of soil
consolidation coefficient
dead load vector
equivalent damping ratio of soil
damping ratio of critical of SOOF system
diameter of the connection core, measured to the centerline of the
spiral confinement
reference diameter of a pile
distance of anchor from tie-rod level of active soil wedge
depth of wall embedment
depth of the effective point of rotation of the sheet pilewall below
the mudline
pile (outer) diameter, or equivalent pile width, or gross depth
(in the case of a rectangular pile with Spiral confinement in TC5)
inner pile diameter
strain dependent damping ratio of soil
laboratory measured damping ratio of soil scaled to Do
small strain damping ratio of soil
thickness of soil deposit below a gravity wall
displacement at the top of the wall
maximum displacement of a linear SOOF system
average sliding displacement
diameter of dowel reinforcement
depth to fixity of a pile
leg expansion of a crane
maximum displacement of elasto-plastic SOOF system.
nominal diameter of the prestressing strand
residual displacement of SDOF system ( = d - d
y
)
ultimate displacement of a elasto-plastic SDOF system
yield displacement of an elasto-plastic SOOF system
earthquake load vector
Young's modulus
dowel modulus of elasticity
seismic energy
e.
s; e,
ER
ER
m
e
e.;
F
F'
Fe
F
L
F
m
F
p
F
s
r;
r;
F
T
F
y
f
fa
f ~
te
te
tm
fpu
fpue
fpum
!r
is
fy
he
fye
fyh
fym
t:
G
G(y)
G(y), GIGo
Go (or G
max
)
g
H
He
List of Symbols 443
equivalent subgrade elastic modulus ( = kh-sUbDp)
earthquake loads (E) in the principal directions x and y
energy efficiency for SPT procedure
energy efficiency of a particular SPT procedure
void ratio
eccentricity
seismic load
maximum force of a linear SDOF system
tines content
factor of safety against liquefaction
maximum load on inelastic structural response
axial prestress force in pile
factor of safety during an earthquake
factor of safety against overturning
factor of safety against sliding
a set of discrete variables describing the fault type
elastic limit of elasto-plastic SDOF system
frequency
average compression stress at the joint center caused by the pile
axial compression force N
ultimate (unconfined) compressive strength of concrete at 28 days
confined strength of concrete approximated by 1 . 5 f ~
concrete compression strength with earthquake increase factor
concrete compression strength with capacity protection
ultimate strength of prestressing strand
prestress strand ultimate strength with earthquake increase factor
prestress strand ultimate strength with capacity protection
resonant frequency
=O.OOI5Ed
nominal yield strength of reinforcement
dowel overstrength bar capacity ( = I.4fy)
reinforcement yield strength with earthquake increase factor
yield stress of confining steel, transverse spiral or hoop reinforce-
ment
reinforcement yield strength with capacity protection
half-power frequencies _
equivalent (secant) shear modulus of soil
strain dependent shear modulus
normalised equivalent shear modulus
small strain shear modulus
acceleration of gravity ( = 9.8 mls
2
)
height of wall
hinge height of a crane leg
444 PlANe
Hsimple
n.:
n.;
HI
Hw
H
o
h
d
h
o
I
Ie .
I
eff
Ii
1
0
or t,
I p
]
K
x,
x;
x.
K
Hi
K
L
K
p
Kpe
K
s
Kit
K
o
.1K
ae
sx;
k
k
c
k ' - I ~ T > e
k
d
k
DC
k
c
k
h
k ~
kh-sun
k
l
k
P1
a span of a simple beam (i.e. sheet pile wall span between tie-rod
and mud line levels)
thickness of backfill between the mudline and the backfill water
table
thickness of backfill soil above the water table
height of a tsunami at the shore . .
water depth
height of a tsunami at an offshore site
depth of the pile cap
water depth at the offshore site
moment of inertia of piles
soil behaviour type index
effective moment of inertia
moment of inertia of i-th pile
maximum intensity of earthquake
plasticity index
an empirical power index to define small strain damping of soil
bulk modulus of soil
static active earth pressure coefficient
dynamic active earth pressure coefficient
correction factor for obtaining (qelN)cs
equivalent spring stiffness for lateral reaction by i-th pile
total stiffness of the tributary length of wharf
static passive earth pressure coefficient
dynamic passive earth pressure coefficient
correction factor for obtaining (N
1
)6Ocs
rebound bulk modulus.
earth pressure coefficient at rest
dynamic component of active earth pressure coefficient
dynamic component of passive earth pressure coefficient
coefficient of permeability
factor dependent on the curvature ductility within the plastic hinge
region, given by Fig. T5.12
parameter for specifying subgrade reaction for C-rype subsoil
deck stiffness for force-displacement response of a wharf
stiffness of wharf for damage control limit
equivalent seismic coefficient
horizontal seismic coefficient
equivalent horizontal seismic coefficient for submerged condition
coefficient of lateral subgrade reaction
initial stiffness of inelastic structural response
additional spring to be added at the deck/pile interface to represent
the inelastic stiffness of the top plastic hinge
k
s
ks-type
k
sub
k
t
k
u
k"
k ~ .
L
L
i;
L
span
L
T
Lw
Lx
1
Mdesign
M
FES
M
J
M
L
Mmudline
M
N
M
p
MpO
M
s
Msimple
Mtie-rod
M
u
M
w
My
List of Symbols 445
stiffness of wharf for serviceability limit
parameter for specifying subgrade reaction for S-type subsoil
initial modulus of subzrade reaction
....
threshold seismic coefficient
unloading stiffness of inelastic structural response
vertical seismic coefficient
equivalent vertical seismic coefficient for submerged condition
design live load vector
a free pile length/distance from the pile/deck intersection to the
point of contraflexure in the pile
extent of soil improvement
longitudinal length of wharf segment
seismic stress ratio
plastic hinge length of piles
rail span for cranes
transverse length of wharf segment
travel path of the waveform
linear spectrum of a signal x(t)
unsupported length of a pile between the deck and virtual ground
surface
actual embedment length provided
embedment length of pile in deck
unsupported length of i-th pile between the deck and slope surface
depth of contraflexure point
depth to fixity for pile analysis
bending moment
Richter magnitude
overstrength moment (i.e. maximum possible moment) of the
plastic hinge
design bending moment for sheet pile wall
maximum bending moment under free earth support condition
Japan Meteorological Agency (JMA) magnitude of earthquake
local magnitude
driving moment above the mudline level
nominal flexural strength
pile bending moment at plastic hinge state
plastic hinge moment without axial load
surface wave magnitude of earthquake
bending moment of sheet pile based on a simple beam analysis
moment balance about the tie-rod
ultimate bending moment of a pile
moment magnitude of earthquake
yield bending moment
446 PlANe
m
N
i
N
1
N
m
.
Nw(f)
Nyo
N
65
11
n
P
P
a
r;
(Pae)mudline
Pfoundation
an empirical power index to define small strain shear modulus of
soil
long-wave body wave magnitude
short-wave body wave magnitude
SPT penetration resistance (blow count)
axial load
number of cycles
equivalent number of uniform cycles
SPT reference penetration resistance (for a particular energy
efficiency)
axial load for the i-th pile
number of load cycles to cause liquefaction
measured SPT blow count
axial compression load on pile, including seismic load
number of wavelengths of a sinusoidal signal (f) while propagating
yield axial load without bending moment
effective SPT N-value (= Japanese SPT N-value with average
SPT energy efficiency ratio of 72% corrected to the effective over-
burden pressure of 65 kPa)
SPT N-value normalized to the effective overburden pressure of 1
kgf/m'' ( = 98kPa) with SPT energy efficiency ratio of 60%
equivalent value of (N
1
) 60 for clean sands
equivalent N
65
value, for Ip < 10 or 5 < F; 15%
equivalent N
65
value, for lv> 20 or Fe> 15%
increment of N, for calculating N;'
mean annual number of events that exceeds magnitude M
an empirical power index to define small strain shear modulus of
soil
normalized soil improvement factor ( = LexJ(H + D
emt
,) )
number of piles in row i in the tributary width
subgrade reaction per unit length of a pile at the depth;
static active earth thrust
dynamic active earth thrust
dynamic active earth thrust above the mudline level
equivalent reaction force exerted by the foundation (i.e. mudline
support)
static passive earth thrust
dynamic passive earth thrust
probability of the acceleration G
max
being exceeded in the exposure
time T
L
the limit lateral load when plastic hinges form at top and embed-
ded portion of all piles
hydrodynamic force
P
y
Mae
Mpc
P
,
P
pa
pae
Pdw
Ppc
Pt
pu
pWd
Q
q
q/o;
qc
qcl
qclN
(qcIN)cs
qsur
R
SA
SA(T, D)
So(T, D)
ST
Sv(T, D)
[SIR]xy
s
T
List of Symbols 447
approximate elastic limit
dynamic component of active earth thrust
dynamic component of passive earth thrust
(=PIDp) Subgrade reaction per unit area of a pile at the depth x
mean effective stress
atmospheric pressure
dynamic active earth pressure
dynamic fluid pressure
dynamic passive earth pressure
nominal principal tension stress in the joint region
ultimate resistance of soil
hydrodynamic pressure
an empirical parameter to define small strain damping of soil
deviatoric stress
deviatoric/radial effective stress ratio
CPT tip penetration resistance
reference value of CPT tip resistance
normalised and corrected CPT tip resistance
equivalent value of q-, for clean sands
uniformly distributed surcharge
a measure of distance
in-situ liquefaction resistance
live load reduction factor
reduction factor for estimating the variation of cyclic shear stress
(or acceleration) with depth
a reduction factor for specifying the effective seismic coefficient
as a fraction of adesignlg
pile radius
pore water pressure ratio
distance between two point, during wave propagation
an empirical dimensionless parameter to define small strain shear
modulus of soil
acceleration response spectra
acceleration response spectrum
displacement response spectrum
a set of discrete variables describing the site foundation conditions
velocity response spectrum
signal to noise ratio of signal x(t) and yet)
spacing of hoops or spiral along the pile axis.
natural period of SDOF system
design tie-rod force per meter
ultimate anchor resistance
tie-rod force under free earth support condition
448 PlANe
T
L
T
R
T ~
Tsimple
Tw
t
tG
t
p
1.'0'
U
c
.1U
U
Ud
U / ( J ~
V
C
Vi
V
N
V
p
V
ph
Vph(f)
V
R
V
s
V
shell
Vs\
V
sk
V
t
V.n
\'max
x(t)
X
r
y(t)
Yg
Ze
z,
exposure time, life span of structure
return period
Natural period of a pile-deck system
tie-rod force based on a simple beam analysis
travel time of the waveform
time
degradation parameter
thickness of steel shell pile
time delay between signals x(t) and y(t)
coefficient of uniformity
excess pore water pressure due to cyclic shearing; subscripts a and
p denote pressure within the active and passive soil wedges
excess pore water pressure
pore water pressure in a drained layer
excess pore water pressure ratio
shear strength from concrete
pile shear force at the limit state deflection ..1\s
shear strength of pile
shear strength from axial load
Fourier component phase velocity
phase velocity of the frequency component (j) of a signal
Rayleigh wave velocity
shear wave velocity
contribution of the shell to the shear strength
shear wave velocity normalised to a reference effective stress
shear key force
shear strength from transverse reinforcement
total segment lateral force at displacement ..1
T
for pure transla-
tional response
peak velocity
nominal shear stress in the joint region corresponding to the pile
plastic moment capacity
width of a gravity wall
energy damped in the stress-strain loop of soil
weight of a gravity wall per unit width
surcharge and deadweight of deck
energy stored in the stress-strain loop of soil
a signal for wave propagation analysis
center of rigidity
a signal for wave propagation analysis
ground motion parameter or response spectrum ordinate
elastic sectional modulus
sectional modulus at plastic hinge
subscript i
subscript x
subscript Y
a
1/f3
Y
Yb
Ycyc
Yd
Ye
Ye.sat
YI
Ysat
Yv
Yw
Ywet
L1
L10c
L1g
List of Symbols 449
elevation below the tie-rod level
elevation of passive earth thrust below the tie-rod level
elevation of hydrodynamic force below the tie-rod level
.Ih
l
.th 1
l pi e or layer
horizontal component
horizontal component normal to x-direction
a factor for specifying the effective seismic coefficient as a frac-
tion of Qma../g
seismic active angle of failure
'-'
angle between line joining the centers of flexural compression in the
deck/pile and in-ground hinges, and the pile axis (see Fig. T5.14)
seismic passive angles of failure
angle of critical crack to the pile axis for transverse shear (see Fig.
T5.13)
ratio of existing construction exposure time over new construction
a parameter for a beam analysis in Winkler foundation
(
= ~ kh-subDp I
4EI J
length of equivalent fixity pile below the slope surface
shear strain amplitude
buoyant unit weight of soil (= Ysat- Yw)
cyclic strain amplitude
unit weight of dry soil
equivalent unit weight of partially submerged soil wedge
equivalent total unit weight for submerged condition
linear threshold shear strain
saturated total unit weight of soil
volumetric threshold shear strain
unit weight of seawater
unit weight of soil above the water table
displacement of piles
displacement of a wharf at damage control limit
initial gap between the deck segments of pile-supported wharves
and piers
limit state deflection
upper bound displacement of deck
residual displacement
displacement of a wharf at serviceability limit
displacement of deck under pure translational response
friction angle at the wall-soil interface (for sign convention, refer
to Section T7.1.1 and Fig. T7.1)
450 PlANe
&, do
8*
E
e;
Ecu
max
Ef
Esm
E.,.
Evd
Ee
E
a
~
11
B
p
A.
~ f )
}I"R
)1
)1b
) 1 ~
)1.1
c;
C;P
a
C;P
ae
C;.1P
ae
o,
O"allow
,
O"e
O " ~
,
O"mO
cyclic degradation indexes
factored wall interfa:ce friction angle for passive earth pressure
computation
strain in piles
strain at 50% peak shear resistance
ultimate compressive concrete strain
strain limit for embedded portion of steel piles
irrecoverable (or plastic) distortional strain
strain at peak stress of confining reinforcement
volumetric strain
inelastic volumetric strain
irrecoverable (or plastic) volumetric strain
a random error term with zero mean and o standard deviation
elevation beneath the fluid level in front of the sheet pile wall
equivalent viscous damping of inelastic response as a percentage
of critical damping
inelastic rotation at plastic hinge
wave length
wavelength of a sinusoidal signal (f) while propagating
wave length of Rayleigh waves
ductility factor
friction angle at bottom of wall and/or at toe of a crane
curvature ductility ( = tP /f/Jy)
displacement ductility factor (= dm/d
y
)
depth from the top of the pile
elevation of static active earth thrust from the toe of the sheet pile wall
elevation of dynamic active earth thrust from the toe of the sheet
pile wall
elevation of dynamic component of active earth thrust from the toe
of the sheet pile
elevation of dynamic passive earth thrust from the toe of the sheet
pile wall
elevation of hydrodynamic force from the toe of the sheet pile wall
bulk soil density
Rowe's flexibility number for sheet pile design
longitudinal reinforcement ratio, including prestressing steel
effective volume ratio of confining steel
vertical stress on shear plane
axial stress in soil specimen
allowable stress in piles
effective confining stress; current consolidation pressure in soil
mean effective stress in subsoil
initial mean effective vertical stress in subsoil
T
av
. Tee
Tdmax
T
max
rla:o
(Ti/aDNI=20
(Ti/aD
NC
(T/acJOC
<P
<Pcps(f)
<Pp
t1(f)
'
*
List of Symbols ~ 5 1
maximum consolidation pressure that the soil sample has been
subjected to
radial stress in soil specimen
vertical effective stress in subsoil
initial effective vertical stress in subsoil
in situ total vertical stress in subsoil
yield stress in piles
initial effective stress in soil
shear stress
average shear stress
waveform travel time (with respect to CC
max
)
maximum shear stress, from site response analysis
cyclic shear stress amplitude
maximum shear stress
cyclic shear stress ratio
cyclic resistance for liquefaction potential assessment or cyclic
stress ratio at 20 cycles of loading
liquefaction strength for normally consolidated state
liquefaction strength for overconsolidated state
phase of a sinusoidal function
phase of each frequency component of the CPS
1.0 for assessment, and 0.85 for new pile design
phase delay of a sinusoidal signal
internal friction angle of soil
pile curvature
effective stress friction angle of soil
factored internal friction angle for passive earth pressure compu-
tation
limit state pile curvature for Level 2
plastic limit state pile curvature ( =(>s or Ls) - Y)
limit state pile curvature for Levell
ultimate pile curvature
yield pile curvature
seismic inertia angle
similarity number (i.e. a non-dimensional Rowe's flexibility num-
ber) (see Eqn. (T8.4))
load combination symbol for vector summation
Abbreviations:
APSxx auto power spectrum function of signal x(t)
BE bender elements
CCxy cross correlation function of signals x(t) and yet)
CH cross hole test
452 PlANe
CPSX)'
CPT
CQC
CRR
CSR
CSS
CTS
CTX
DFf
DH
EAI
EPI
El
E2
FDM
FEM
IFf
LDT
Ll
L2
MDOF
MSF
OCR
PGA
PGA
H
PGA
v
PGV
PGV
H
PGV
v
RC
SASW
SC
SDOF
SH
SPT
SPT N-value
SSI
IDI2D/3D
cross power spectrum function of signals x(t) and y(t)
cone penetration test
Complete Quadratic Combination rule .
cyclic resistance ratio (liquefaction resistance)
cyclic stress ratio (seismic load)
cyclic simple shear test
cyclic torsional shear test
cyclic triaxial test
direct Fourier transform
down hole test
effective anchor index for sheet pile design
embedment participation index for sheet pile design
earthquake motion equivalent to the design seismic coefficient
earthquake motion about 1.5 to 2.0 times larger than EI
finite difference method
finite element method
inverse Fourier transform
local displacement transducers
level 1 earthquake motion
level 2 earthquake motion
multi degree of freedom
magnitude scaling factor
overconsolidation ratio ( = a;Ja:)
peak ground acceleration
peak ground horizontal acceleration
peak ground vertical acceleration
peak ground velocity
peak ground horizontal velocity
peak ground vertical velocity
resonant column test
spectral analysis of surface waves
seismic cone technique
single degree of freedom
... ...
shear wave horizontal components
standard penetration test
standard penetration test blow counts
soil structure interaction
one, two, and three dimensions
References
Abrahamson. N.A. & Silva, W.J. 1997. Empirical response spectral attenuation relations
for shallow crustal earthquakes, Seismological Research Letters 68(1).
ACI Committee 116 2000. Cement and Concrete Terminology. 73 p.
Alarcon, A., Charneau, J.L. & Leonards, O.A. 1986. A new apparatus for investigating
the stress-strain characteristics of sands, Geotechnical Testing Journal, ASTM 9(4):
204-212.
Arnano, R., Azuma, H. & Ishii, Y. 1956. Aseismic design of walls in Japan, Proc. 1st
World Conference on Earthquake Engineering, San Francisco, pp. 32-1-32-17.
Ambraseys, N.N. & Menu. 1.M. 1988. Earthquake-induced ground displacements,
Earthquake Engineering and Structural Dynamics 16: 985-1006.
American Petroleum Institute (API) 1993. Recommended Practice for Planning,
Designing and Constructing Fixed Offshore Platforms - Working Stress Design, API
Recommended Practice 2A-WSD (RP 2A-WSD), 20th edition, 191 p.
Anderson, lO. 1991. Guerrero accelerograph array: seismological and geotechnical
lessons, Geotechnical News 9(1): 34-37.
Andrus, R.D. & Stokoe, K.H. 1997. Liquefaction resistance based on shear wave
velocity. Proc. NCEER Workshop 011 Evaluation of Liquefaction Resistance of Soils,
Technical Report NCEER-97-0022. National Center for Earthquake Engineering
Research, pp. 89-128.
Annaki, M. & Lee, K.L. 1977. Equivalent uniform cycle concept of soil dynamics,
Journal of Geotechnical Engineering Division, ASCE 103(OT6): 549-564.
Arango, 1. 1996. Magnitude scaling factors for soil liquefaction evaluations, Journal of
Geotechnical Engineering, ASCE 122(GTl1): 929-936.
Arulanandan, K. & Scott, R.F. (eds.) 1993. VELACS - Verification of Numerical
Procedures for the Analysis ofSoil Liquefaction Problems, Balkerna, 1801 p.
Aubry, D., Chouvet, D. Modaressi, A. & Modaressi, H. 1985. GEFDYN_5: Logiciel
d'analyse du comportement statique et dynamique des sols par elements finis avec
prise en compte du coup1age sol-eau-air, Rapport Scientifique Ecole Centrale de Paris
(in French).
Baldi, G., Ghionna, V.N., Jamiolkowski, M. & Lo Presti, D.C.F. 1988. Parametri di prog-
etto per il ca1colo di cedimenti di fondazioni in sabbia da prove penetrometriche, Proc.
. / Conv. Naz. G.N.C.S./.G., CNR 'Deformazioni dei terreni ed interazione terreno-strut-
tura in condizioni di esercizio,' - SOE, Padova, Vol. 1, pp. 3-18 (in Italian).
Bardet, J.P. 1990. LINOS -A Non-Linear Finite Element Programfor Geomechanics and
Geotechnical Engineering, Civil Engineering Department, University of Southern
California.
Bartlett. S.F. & Youd, T.L. 1995. Empirical prediction of liquefaction-induced lateral
spread. Journal of Geotechnical Engineering, ASCE 121(4): 316-329.
454 PlANe
B.C. Hydro 1992. Seismic withstand of timber piles, Detailed Test Procedures and Test
Results, Vol. 2: Appendices, B. C. Hydro, Hydroelectric Engineering Division,
Geotechnical Department.
Bea, R.G. 1997. Background for the proposed International Standards Organization
reliability based seismic design guidelines for offshore platforms, Proc. Earthquake
Criteria Workshop - Recent Developments in SeismicHazard and Risk Assessments
for Port, Harbor; and Offshore Structures, Port and Harbour Research Institute, Japan
and Univ. of California, Berkeley, USA, pp. 40-67.
Benuska, L. (ed.) 1990. Lorna Prieta Earthquake Reconnaissance Report. Eanhquake
Spectra, Supplement to Vol. 6, EERI, 448 p.
Bishop, A.W. 1955. The use of the slip circle in the stability analysis of slopes,
Geotechnique 5(1): 7-17.
Blackford, M.E. 1998. International responses to Pacific Tsunami Warnings and Watches,
Proc. UJNR, 30th Joint Meeting of U.S.-Japan Panel on Wind and Seismic Effects,
Gaithersburg, NIST.
Blazquez, R. 1995. Liquefaction modelling in engineering practice, Proc. 10th European
Conference on Earthquake Engineering, Wien, Vol. 1, Balkema, Rotterdam,
pp.501-513.
Bolton, M.D. & Wilson, J.M.R. 1989. An experimental and theoretical comparison
between static and dynamic torsional soil tests, Geotechnique 39(4): 585-599.
Boore, D., Joyner, W. & Fumal, 1. 1997. Equations for estimating horizontal response
spectra and peak acceleration from Western North-America earthquakes: A summary
of recent work, Seismological Research Letters, 68( 1).
Budek, A.M., Benzoni, G. & Priestley, MJ.N. 1997. Experimental investigation
of ductility of in-ground hinges in solid and hollow prestressed piles, Div. of
Structural Engineering Report SSRP 97-17, University of California, San Diego,
108 p.
Buslov, V.M., Rowghani, M. & Weismair, M. 1996. Evaluating earthquake damage to
concrete wharves, Concrete International. pp. 50-54.
Buslov, V.M. 1996. Surveys in Port of Eilat, VBC report.
Byrne. P.M., Jitno, H., Anderson, D.L. & Haile, J. 1994. A procedure for predicting
seismic displacements of earth dams, Proc. 13th International Conference on Soil
Mechanics and Foundation Engineering, New Delhi, India, pp. 1047-1051.
California Division of Mines and Geology 1997. Guidelines for Evaluating and
Mitigating Seismic Hazards in California, Special Publication, 117 p.
California State Lands Commission, USA 2000. California Marine Oil Terminal
Standard (under development as of May 2000).
CEN (European Committee for Standardization) 1994. Eurocode 8: Design Provisions
for Earthquake Resistance of Structures. Part J-l: General Rules - Seismic Actions
and General Requirements for Structures Part 5: Foundations.
Retaining STructures and Geotechnical AspeCTS (ENV 1998-5).
CEN (European Committee for Standardization) 1997. Eurocode 7: Geotechnical design.
Part 3: Geotechnical design assisted by field tests. (ENV 1997-3).
Chang. Yl., 1937. Lateral pile loading tests, Transaction ofASCE 102: 173-176.
Comanin. C.D. (ed.) 1995. Guam Earthquake of August 8, 1993 Reconnaissance Report,
Eanhquake Spectra, Supplement B to Vol. 11, EERI, 175 p.
Cornell. C.A. 1968. Engineering seismic risk analysis, Bull. Seism. Soc. Am. 58:
1583-1606.
Cornell, C.A. & Winterstein, S.R. 1988. Temporal and magnitude dependence in earth-
quake recurrence models, Bull. Seism. Soc. Am. 78: 1522-1537.
References 455
Coulomb. CA. 1776. Essai sur une application des regles des rnaxirms et rrunirrus
quelques problemes de statique relatifs a l' architecture, Memoires de l'Academie
Royale pres Divers Savants, Vol. 7.
De Alba. P.. Seed. H.B. & Chan, C.K. 1976. Sand liquefaction in large-scale simple shear
tests. Journal of Geotechnical Engineering Division, ASCE lO2(GT9): 909-917.
Dickenson. S.E. & Yang, D.S. 1998. Seismically-induced deformations of caisson retain-
ing walls in improved soils, Geotechnical Earthquake Engineering and Soil Dynamics
Ill. Geotechnical Special Publication Vol. No. 75, ASCE, pp. 1071-1082.
Dobry, R.. Yokel. EY, Ladd, R.S. & Powell, DJ. 1979. Prediction of pore pressure build-
up and liquefaction of sands during earthquakes by the cyclic strain method, Research
Report. National Bureau of Standards, U.S. Department of Commerce, Washington
D.C.
Dobry. R.. Powell. D.J., Yokel, F.Y. & Ladd, R.S. 1980. Liquefaction potential of satu-
rated sand: the stiffness method, Proc. 7th WCEE, Istanbul, Vol. 3, pp. 25-32.
Dobry, R. 1991. Soil properties and earthquake ground response, Proc. IOth European
Conference on Soil Mechanics and Foundation Engineering, Florence, Vol. 4,
pp.1171-1187.
d'Onofrio. A., Silvestri, F. & Vinale, F. 1999a. Strain rate dependent behaviour of a nat-
ural stiff clay, Soils and Foundations 39(2): 69-82.
d'Onofrio. A.. Silvestri, F. & Vinale, F. 1999b. A 'new torsional shear device,
Geotechnical Testing Journal, ASTM 22(2): 101-111.
d'Onofrio, A. & Silvestri, F. 2001. Influence of micro-structure on small-strain stiffness
and damping of fine grained soils and effects on local site. response, Proc. 4th
International Conference on Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics, San Diego.
Dmevich, VP.. Hardin, B.O. & Shippy, DJ. 1978. Modulus and damping of soils by res-
onant column method, Dynamic Geotechnical Testing, ASTM STP 654, American
Society for Testing and Materials, pp. 91-125.
Dyvik, R. & Madshus, C. 1985. Lab measurement of G
max
using bender elements,
Advances in the Art of Testing Soils Under Cyclic Conditions, ASCE, New York,
pp. 186-196.
EAU 1996. Recommendations of the Committee for Wateifront Structures, Harbours and
Watenvays, 7th English edition (English Translation of the 9th German edition), Ernst
& Sohn, Berlin, 599 p.
Ebeling, R.M. & Morrison, E.E. 1992. The Seismic Design of Wateifront Retaining
Structures, U.S. Army Corps of Engineers, Technical Report ITL-92-llINCEL TR-
939,329 p.
Egan, J.A., Hayden, R.F., Scheibel, L.L., Otus, M. & Serventi, G.M. 1992. Seismic repair
at Seventh Street Marine Terminal, Grouting. Soil Improvement and Geosynthetics,
Geotechnical Special Publication No. 30, ASCE, pp. 867-878.
Elrnrabet, T., Levret, A., Ramdani, M. & Tadili, B. 1991. Historical seismicity in
Morocco: Methodological aspects and cases of multidisciplinary evaluation,
Seismicity, Seismotectonic and Seismic Risk of the Ibero-Maghrebian Region.
Monografia Num. 8 Institute Geographic National. Spain.
EPRI 1993. Guidelines for Determining Design Basis- Ground Motions, Report No. TR-
102293. Electric Power Research Institute, Palo Alto (California).
Esteva, L. 1 9 7 0 ~ Seismic risk and seismic design decisions. In RJ. Hansen (ed.), Seismic
Design of Nuclear Power plants, MIT Press, Cambridge, Massachusetts. i
Falconer, TJ. & Park, R. 1983. Ductility of prestressed concrete piles under seismic load- I
ing, PCI Journal 28(5): 112-114. I
I
I
I
1
456 PIANC
Ferritto, J.M. 1997a. Design criteria for earthquake hazard mitigation of navy piers and
wharves, Technical Report TR-2069-SHR, Naval Facilities Engineering Service
Center, Port Hueneme, ) 80 p.
Ferritto, J.M. 1997b. Seismic design criteria for soil liquefaction, Technical Report
TR-2077-SHR, Naval Facilities Engineering Service Center, Port Hueneme, 58 p.
Ferritto, J.M., Dickenson, S.E., Priestley, MJ.N., Werner, $.D. & Taylor, C.E. 1999.
Seismic criteria for California marine oil terminals, Technical Report TR-2103-SHR,
Naval Facilities Engineering Service Center, Port Hueneme.
Finn, W.D.L., Lee, K.W. & Martin, G.R. 1977. An effective stress model for liquefaction,
Journal of Geotechnical Engineering Division, ASCE 103(GT6): 517-533.
Finn, W.D.L., Yogendrakumar, M., Yoshida, N. & Yoshida, H. 1986. TARA-3: A program
to compute the response of 2-D embankments and soil-structure interaction to seismic
. loadings, University of British Columbia, Canada.
Finn, W.D.L. 1988. Dynamic analysis in geotechnical engineering, Earthquake
Engineering and Soil Dynamics II: Recent Advances in Ground-Motion Evaluation,
Finn, W.D.L. 1990. Analysis of deformations, stability and pore water pressures in port
structures, Proc. POLA Seismic Workshop on Seismic Engineering, Port of Los
Angeles, pp. 369-391.
Finn, W.D.L. 1.991. Assessment ofliquefaction potential and post-liquefaction behaviour
of earth structures: Developments 1981-1991 (State-of-the-art paper), Proc. 2nd
International Conference on Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics, St. Louis, Vol. 3, pp. 1833-1850.
Franklin, A.G. & Chang, EK. 1977. Eanhquake resistance of earth and rockfill
dams. Report 5: Permanent displacements of earth dams by Newmark analysis.
US Army Corps of Engineers, Watenllays Experiment Station, Miscellaneous Paper
2-71-17.
Fukushima, Y. & Tanaka, T. 1990. A new attenuation relation for peak horizontal accel-
eration of strong earthquake ground motion in Japan, Bull. Seism. Soc. Am. 80:
757-783.
Gazetas. G.. Dakoulas, P. & Dennehy. K. 1990. Empirical seismic design method for
waterfront anchored sheetpile walls. Proc. ASCE Specialty Conference on Design and
Performance of Earth Retaining Structures. ASCE Geotechnical Special Publication
N
"') - ' ) ~ " ')50 o. _). pp. -_ .
Goio. S.. Tatsuoka. E, Shibuya, S.. Kim. Y.S. & Sato, T. 1991. A simple gauge for local
small strain measurements in the laboratory, Soils and Foundations 31(1): 169-180.
Grunthal, G. (ed.) 1998. European Macroseismic Scale 1998. Cahier du Centre Europeen
de Geodynamique et de Seismologie. Luxembourg. Vol. 15, 99 p.
GSHAP 1999. Global Seismic Hazard Assessment Program. hnp://seismo.ethz.ch/
GSHAP.
Gulkan, P. & Sozen, M. 1974. Inelastic response of reinforced concrete structures to
earthquake motion. ACI Journal. pp. 604-610.
Gutenberg. B. & Richter. C.E 1944. Frequency of earthquakes in California. Bull. Seism.
Soc. Am. 34: 1985-1988.
Hall. IF. (ed.) 1995. Northridge Earthquake ofJanuary 17. 1994 Reconnaissance Report
- Vol. 1. Earthquake Spectra, Supplement C to Vol. 11.
Hardin. B.O. & Black, \V.L. 1968. Vibration modulus of normally consolidated clay,
Journal of Soil Mechanics and Foundations Division. ASCE 94(SM2): 353-369.
Hardin. B.O. & Black, W.L. 1969. Closure to Vibration modulus of normally consoli-
dated clay. Journal of Soil Mechanics and Foundations Division. ASCE 95(SM6):
1531-1537.
References .+57
Hardin. B.a. & Drnevich, VP, 1972. Shear modulus and damping in soils: design equa-
tions and curves. Journal of Soil Mechanics ami Foundations Division. ASCE
98(SM7): 667-692.
Hardin. B.a. 1978. The nature of stress-strain behaviour for soils - State of the Art. Proc.
Earthquake Engineering and Soil Dynamics. ASCE. Pasadena (California). VoL 1. pp. 3-89.
Heaton. TH.. Tajima. F. & Mori, A. W. 1982. Estimating ground motions using recorded
accelerograrns. Report by Dames & Moore to Exxon Production Res. Co.: Houston.
Hoar. RJ. 1982. Field Measurements of Seismic Wave Velocity and Attenuation for
Dynamic Analvsis, Ph.D. Diss., The University of Texas at Austin.
Hynes-Griffin. M.E. & Franklin, A.G. 1984. Rationalizing the seismic coefficient
method, US Army Corps of Engineers. Waterways Experiment Station, Miscellaneous
paper GL-84-13, 21 p.
Iai, S. & Koizumi. K. 1986. Estimation of earthquake induced excess pore pressure in
gravel drains. Proc. 7th Japanese Earthquake Engineering Symposium. pp. 679-684.
Iai, S., Koizumi. K. & Kurata, E. 1987. Basic consideration for designing the area of the
ground compaction as a remedial measure against liquefaction, Technical Note of Port
and Harbour Research Institute, No. 590, pp. 1-66 (in Japanese).
lai, S., Koizumi, K. & Kurata, E. 1991. Ground compaction area as a remedial measure
against liquefaction, Tsuchi-to-Kiso, Japanese Society of Soil Mechanics and
Foundation Engineering 39(2): 35-40 (in Japanese).
lai, S., Matsunaga. Y & Kameoka, T 1992. Strain space plasticity model for cyclic
mobility, Soils and Foundations 32(2): 1-15.
Iai, S. & Finn, W.D.L. 1993. Evaluation of seismic sheet pile wall design, Proc. 4th
Canadian Conference on Marine Geotechnical Engineering, Vol. 1: St. Jone 's.
Newfoundland, pp. 293-310.
lai, S. & Kameoka, T. 1993. Finite element analysis of earthquake induced damage to
anchored sheet pile quay walls, Soils and Foundations 33(1): 71-91.
lai, S., Matsunaga, Y, Morita, T., Miyata, M., Sakurai, H., Oishi, H., Ogura, H., Ando,
Y, Tanaka, Y. & Kato, M. 1994. Effects of remedial measures against liquefaction at
1993 Kushiro-Oki earthquake, Proc. 5th US-Japan Workshop on Earthquake Resistant
Design of Lifeline Facilities and Countermeasures against Soil Liquefaction. NCEER-
94-0026, National Center for Earthquake Engineering Research, pp. 135-152.
S., Sugano, T, Ichii, K., Morita, T., lnagaki, H. & Inatorni, T 1996. Performance of
caisson type quay walls, The 1995 Hyogoken-Nanbu Earthquake, -Investigation into
Damage to Civil Engineering Structures, Japan Society of Civil Engineers, pp. 181-207.
lai, S. & Ichii, K. 1997. Excess pore water pressures behind quay walls, Observation and
Modeling in Numerical Analysis and Model Tests in Dynamic Soil-Structure
Interaction Problems, Geotechnical Special Publication No. 64, ASCE, pp. 11-25.
lai, S. & Tsuchida, H. 1997. Damage to ports and airports, Reconnaissance Report on the
Northridge, California, Earthquake of January /7, 1994, Japan Society of Civil
Engineers, pp. 157-168 (in Japanese). ,
/ Iai, S. 1998a. Rigid and flexible retaining walls during Kobe earthquake, Proc. 4th
Conference on Case Histories in Geotechnical Engineering, St. Louis,
pp. 108-127.
lai, S. 1998b. Seismic analysis and performance of retaining structures, Geotechnical
Earthquake Engineering and Soil Dynamics Ill, Geotechnical Special Publication No.
75, ASCE, pp. 1020-1044.
lai, S. & Ichii, K. 1998. Performance based design for port structures, Proc. UJNR 30th
Joint Meeting of United States-Japan Panel on Wind and Seismic Effects, Gaithersburg,
NlST (3-5), pp. 1-13.
458 PlANC
Iai, S., Ichii, K., Liu, H. & Morita, T. 1998. Effective stress analyses of port structures,
Soils and Foundations, Special Issue on Geotechnical Aspects of the January 17, 1995
Hyogoken-Nambu Earthquake, No.2, pp.97-114.
Iai, S., Ichii, K., Sato, Y. & Liu, H. 1999. Residual displacement of gravity quaywalls -
parameter study through effective stress analysis, Proc. 7th U.S.-Japan Workshop on
Earthquake Resistant Design of Lifeline Facilities and Countermeasures against Soil
Liquefaction, Seattle, MCEER-99-0019, pp. 549-563.
Ichii, K., lai, S., & Morita, T. 1997. Effective stress analyses on the performance of cais-
son type quay walls during 1995 Hyogokne-Nanbu earthquake, Report of Port and
Harbour Research Institute 36(2): 41-86 (in Japanese).
Ichii, K., Sato, Yu., Sato, Yo., Hoshino, Y. & Iai, S. 1999. Site characteristics of strong-
motion earthquake stations in ports and harbours in Japan (Part VI), Technical Note of
Port and Harbour Research Institute, No. 935, 696 p.
ldriss, I.M., Dobry, R. & Singh, R.D. 1978. Nonlinear behaviour of soft clays during
cyclic loading, Journal of Geotechnical Engineering Division, ASCE 104(12):
353-369.
Idriss, I.M. 1985. Evaluating seismic risk in engineering practice, Proc. 11th
International Conference on Soil Mechanics and Foundation 'Engineering, San
Francisco, Vol. 1, Balkema, pp. 255-320.
Idriss, I.M. 1991. Earthquake. ground motions at soft soil sites, Proc. 2nd International
Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil
Dynamics, St. Louis, Vol. III, pp. 2265-2272.
lmai, T. 1977. P and S wave velocities of the ground in Japan, Proc. 9th International
Conference on Soil Mechanics and Foundation Engineering, Tokyo, Vol. 2,
pp. 257-260.
Inagaki, H., Iai, S., Sugano, T., Yamazaki, H. & lnatomi, T. 1996. Performance of cais-
son type quay walls at Kobe port, Soils and Foundations, Special Issue on
Geotechnical Aspects of the January 17 1995 Hyogoken-Nambu Earthquake,
pp. 119-136.
Inatorni, T., Uwabe, T., Iai, S., Kazama, M., Yamazaki, H., Matsunaga, Y.. Sekiguchi, S.,
Mizuno, Y. & Fujimoto, Y. 1994. Damage to port facilities by the 1993 Hokkaido-
Nansei-Oki earthquake, Technical Note of the Port and Harbour Research Institute,
No. 791,449 p. (in Japanese).
Isenhower, W.M. 1979. Torsional simple shear/Resonant column properties of San
Francisco Bay mud, Ph.D. Thesis, The University of Texas at Austin.
Ishihara, K. 1976. Foundation of Soil Dynamics, Kajima Publication, pp. 258-260
(in Japanese).
Ishihara, K. & Takatsu. H. 1979. Effects of overconsolidation and K, consolidation on the
liquefaction characteristics of sand, Soils and Foundations 19(4): 59-68.
Ishihara, K. & Towhata. 1. 1980. One-dimensional soil response analysis during earth-
quakes based on effective stress method, Journal of the Faculty of Engineering,
University of Tokyo. Vol. 35. No.4.
Ishihara. K., Yasuda. S. & Yokota. K. 1981. Cyclic strength of undisturbed mine tailings.
Proc. International Conference 011 Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics. St. Louis, Vol. 1, pp. 53-58.
Ishihara. K. 1982. Evaluation of soil properties for use in earthquake response analysis,
Geomechanical Modelling in Engineering Practice, Balkema. Rotterdam. pp. 241-276.
Ishihara, K. & Towhata, 1. 1982. Dynamic response analysis of level ground based on the
effective stress method, Soil Mechanics - Transient and cyclic loads. John Wiley,
pp. 133-172.
Ishihara. K. 1985. Stability of natural deposits during earthquakes. Proc. 11th
International Conference on Soil Mechanics and Foundation Engineering. San
Francisco. Vol. 1. Balkerna, pp. 321-376.
Ishikawa. Y. & Kameda. H. 1991. Probability-based determination of specific scenario
earthquakes. Proc. 4th International Conference on Seismic Zonation. Vol. II. pp. 3-10.
ISSMGE ( T C ~ ) 1999. Manual for Zonation on Seismic Geotechnical Hazards (Revised
Version). Technical Committee for Earthquake Geotechnical Engineering (T(.+).
International Society of Soil Mechanics and Geotechnical Engineering, Japanese
Geotechnical Society, 209 p.
ITASCA 1995. FLAC - Fast Lagrangian Analysis of Continua, User's Manual. Itasca
Consulting Group. Inc .. Minneapolis. Minnesota.
Japan Cargo Handling Mechanization Association 1996. Maintenance Manual for
Container Cranes, 158 p. (in Japanese).
Japan Road Association (ed.) 1996. Specification for Highway Bridge, Pan V. Earthquake
Resistant Design. 228 p. (in Japanese).
Japan Society of Civil Engineers 1990. Preliminary Report of JSCE Reconnaissance
Team on July 16, 1990 Luzon Earthquake. 166 p.
Japanese Geotechnical Society 1989. Effective stress analysis of ground and soil struc-
tures, Proc. Symposium on Seismic Performance of Ground and Soil Structures,
pp. 50-136 (in Japanese).
Japanese Geotechnical Society 1991. Comparative liquefaction analyses, Proc.
Symposium on Remediation of Soil Liquefaction, pp. 77-198 (in Japanese).
Japanese Society of Soil Mechanics and Foundation Engineering 1982. Handbook of Soil
Mechanics and foundation Engineering, pp. 1000-1002 (in Japanese).
Japanese Society of Soil Mechanics and Foundation Engineering 1988. Soft Ground
Improvement Method - From Investigation, Design to Construction, pp. 59-83,
Japanese Society of Soil Mechanics and Foundation Engineering 1991. Symposium on
Liquefaction Remediation of Ground, pp. 15-40, pp. 215-290 (in Japanese).
Jibson, R.W. 1993. Predicting earthquake-induced landslide displacements using
Newmark's sliding block analysis, Transportation Research Record No. 1411,
Earthquake-Induced' Ground Failure Hazards, Transportation Research Board,
National Academy Press, Washington, D.C., pp. 9-17.
Joen, P.H. & Park, R. 1990. Flexural strength and ductility analysis of spiral reinforced
concrete piles, PCI Journal 35(4): 42-61.
Jones, N.P., Thorvaldsd6ttir, S., Liu, A., Narayan, P. & Warthen, T. 1995. Evaluation of a
loss estimation procedure based on data from the Lorna Prieta Earthquake, Earthquake
Spectra 11(1): 37-61.
Johnson, R.K., Riffenburgh, R., Hodali, R., Moriwaki, Y. & Tan, P. 1998. Analysis and
design of a container terminal wharf at the Port of Long Beach, PORTS '98, ASCE,
pp. 436-444.
Kameda, H. & Ishikawa, Y. 1988. Extended sesmic risk analysis by hazard-consistent
magnitude and distance, Journal of Structural Mechanics and Earthquake
Engineering, Japan Society of Civil Engineers (JSCE), No.. 392/1-9: 395-402
(in Japanese).
Kameoka, T. & Iai, S. 1993. Numerical analyses on the effects of initial stress conditions
on seismic performance of sheet pile quay walls, Technical Note of Port and Harbour
Research Institute, No. 751, 29 p. (in Japanese).
Kim. D.S.1991. Deformation Characteristics of Soils at Small to 'Intermediate Strains
from Cyclic Tests, Ph.D. Thesis, The University of Texas at Austin.
460 PIANC
Kinoshita, S. 1998. Kyoshin Net (K-NET), Seismological Research Letters 69(4):
309-332.
Kishida, T. 1983. On the soil improvement work on second plant in Nagoya port, Nisshin
Seifun K.K., (Liquefaction Remediation Measure for Existing Wharf by Gravel
Compaction Method), Proc. 29th National Port and Harbor Construction Conference,
Japanese Association of Pons and Harbors, pp. 82-94 (in Japanese).
Kitajirna, S. & Uwabe, T. 1979. Analysis of seismic damage to anchored sheet pile bulk-
heads, Report of Port and Harbour Research Institute 18(1): 67-127 (in Japanese).
Kowalski, MJ. & Priestley, MJ.N. 1998. Shear strength of ductile bridge columns, Proc.
5th Caltrans Seismic Design Workshop, Sacramento.
Kramer, S.L. 1996. Geotechnical Earthquake Engineering, Prentice Hall. 653 p.
Kubo, K. 1962. Experimental study on the lateral subgrade reaction of piles - Part 2,
Monthly Report of Transportation Technical Research Institute 11(12): 533-559
(in Japanese).
Liao, S.S.C. & Whitman, R.V. 1986. Overburden correction factors for SPT in sand,
Journal of Geotechnical Engineering, ASCE 112(3): 373-377.
Luke, B.A. & Stokoe, K.H. 1998. Application of the SASW Method Underwater, Journal
of Geotechnical and Geoenvironmental Engineering, ASCE 124(6): 523-531.
/ Lysmer, J., Udaka, T., Tsai, c.r & Seed, H.B. 1975. FLUSH: a computer program for
.c _
approximate 3-D analysis of soil-structure interaction problems, Report EERC 7530,
Earthquake Engineering Research Center, University of California, Berkeley. 83 p.
Makdisi, F.I. & Seed, H.B. 1978. Simplified procedure for estimating darn and embank-
ment earthquake-induced deformations, Journal of Geotechnical Engineering
Division, ASCE 104(7): 849-867.
Mancuso, C; Silvestri, F. & Vinale, F. 1995. Italian experiences on dynamic in situ and
in laboratory measurements of soil properties, Proc. 3rd Turkish Earthquake
Engineering Conference, Istanbul, lnstanbul Technical University and Chamber of
Civil Engineers.
Mancuso, C; Silvestri, F. & Vinale, F. 1997. Soil properties relevant to seismic micro-
zonation, Proc. 1st Japanese-Turkish Conference on Earthquake Engineering,
Istanbul. Istanbul Technical University and Chamber of Civil Engineers.
Mander, J.B. Priestley. MJ.N. & Park. R. 1988. Theoretical stress-strain model for con-
fined concrete. Journal of Structural Engineering, ASCE 114(8): 1805-1826.
Manja. D. 1999. Personal communication.
Martin, G.R., Finn, W.D.L. & Seed. H.B. 1975. Fundamentals of liquefaction under
cyclic loading, Journal of Geotechnical Engineering Division, ASCE 101.(GT5):
423-438.
Martin. G.R.. Finn. W.D.L. & Seed. H.B. 1978. Effects of system compliance on lique-
faction tests, Journal of Geotechnical Engineering Division. ASCE 104(GT4):
463-479. .
Matsuoka. H. 1985. Liquefaction resistance of overconsolidated sand and simple lique-
faction prevention method. Proc. 43rd National Conference of Japan Society ~ f Civil
Engineers. Vol. Ill, pp. 15-16 (in Japanese).
Matsuzawa. H.. Ishibashi, 1. & Kawamura, M. 1985. Dynamic soil and water pressures
of submerged soils. Journal of Geotechnical Engineering. ASCE 111(10): 1161-1176.
Mayne. P.W. & Rix, G. 1993. Gmax-qc relationship for clays. Geotechnical Testing
Journal. ASTM 16(1): 54-60.
McCullouQ:h. N.J. & Dickenson, S.E. 1998. Estimation of seismicallv induced lateral
... .
deformations for anchored sheetpile bulkheads, Geotechnical Earthquake Engineering
and Soil Dynamics Ill, Geotechnical Special Publication No. 75. ASCE 1095-1106.
References 461
McCullough. N.J. 1998. The Seismic Vulnerability of Sheet Pile Walls. Master's Thesis
Dissertation. Dept. of Civil. Construction and Environmental Engineering. Oregon
State University.
Mcfluire. R.K. 1995. Probabilistic seismic hazard analysis and design earthquakes: clos-
ing the loop. Bull. Seism. Soc. Am. 85: 1275-1284.
Medvedev. S.Y. & Sponheuer. Y. 1969. Scale of seismic intensity, Proc. 4th World
Conference on Earthquake Engineering. Santiago, Chile. pp. 143-153.
Mejia. L.H. & Yeung. M.R. 1995. Liquefaction of coralline soils during the 1993 Guam
earthquake, Earthquake-Induced Movements and Seismic Remediation of Existing
Foundations and Abutments. Geotechnical Special Publication No. 55, ASCE, pp. 33-48.
Memos. CD. & Protonotarios, J.N. 1992. Patras breakwater failure due to seismic load-
ing, Proc. 23rd Conference on Coastal Engineering. Venice. Ch. 255,
pp. 3343-3356.
Meyerhof. G.G. 1957. Discussion, Proc. 4th International Conference on Soil Mechanics
and Foundation Engineering, Vol. 3, p. 110
Ministry of Transport, Japan, First District Port and Harbour Construction Bureau 1984.
Report on Damage to Akita Port and Restoration - Nihonkai-Chubu Earthquake
(in Japanese).
Ministry of Transport, Japan (ed.) 1989. Technical Standards for Port and Harbour
Facilitiesand Commentaries, Japan Port and Harbour Association (in Japanese); English
excerpt (1991) by the Overseas Coastal Area Development Institute of Japan, 438 p.
Ministry of Transport, Japan (ed) 1999. Design Standardfor Port and Harbour Facilities
and Japan Port and Harbour Association, 1181 p. (in Japanese).
English edition (2001) by the Overseas Coastal Area Development Institute of Japan.
Mitchell. J.K. 1981. State of the art on soil improvement, Proc. 10th ICSMFE, Vol. 4,
1981, pp. 510-520.
Mitchell, J.K., Baxter, C.D.P. & Munson, T.C. 1995. Performance of improved ground
during earthquakes, Soil Improvement for Earthquake Hazard Mitigation,
Mizuno, Y, Suernatsu, N., & Okuyama, K. 1987. Design method of sand compaction pile
for sandy soils containing fines, Tsuchi-to-Kiso, Japanese Society of Soil Mechanics
and Foundation Engineering 35(5): 21-26 (in Japanese).
Modoni, G., Flora, A., Mancuso, C., Viggiani, C. and Tatsuoka, F. 2000. Pulse wave
transmission tests on gravels, Geotechnical Testing Journal, ASTM (accepted for
publication).
Mok, Y-J., Sanchez-Salinero, 1., Stokoe, K.H. & Roesset, J.M. 1988. In situ damping
measurements by cross-hole seismic method, Proc. Earthquake Engineering and Soil
Dynamics Il: Recent advances in ground motion evaluation, ASCE Geotechnical
Special Publication, Vol. 20, pp. 305-320.
Mononobe, N. 1924. Considerations on vertical earthquake motion and relevant vibration
problems, Journal ofJapan Society of Civil Engineers 10(5): 1063-1094 (in Japanese).
Mori, K., Seed, H.B. & Chan, C.K. 1978. Influence of sample disturbance on sand
response to cyclic loading, Journal of the Geotechnical Engineering Division, ASCE
104(GT3): 323-339.
Morita, T., Iai, S., Liu, H., Ichii, K. & Sato, Y. 1997. Simplified method to determine
parameter of FLIP, Technical Note of Port and Harbour Research Institute, No. 869,
pp. 36 (in Japanese).
Mulilis, J.P., Chan, C.K. & Seed, H.B. 1975. The effects of method of sample prepara-
tion on the cyclic stress-strain behaviour of sands, Report No. EERC 75-18,
Earthquake Engineering Research Center, University of California, Berkeley.
462 PIANC
Muraleetharan, K.K., Mish, c., Yogachandran, K. & Arulanandan, K. 1988. DYSAC2:
Dynamic Soil Analysis Code for 2-dimensional Problems, Department of Civil
Engineering, University of California, Davis.
Nasu, M. 1984. Liquefaction remedial measures for Tokaido bullet train embankment,
Foundation Engineering 12(7): 64-68 (in Japanese)
Nagao, T., Koizumi, T., Kisaka, T., Terauchi, K., Hosokawa, K., Kadowaki, Y., & Uno,
K. 1995. Evaluation of stability of caisson type quaywalls based on sliding block
analysis, Technical Note ofPortand Harbour Research Institute, No. 813, pp. 302-336
(in Japanese).
Neelakantan, G., Budhu, M. & Richards, R. Jr. 1992. Balanced seismic design of
anchored retaining walls. Journal of Geotechnical Engineering, ASCE 118(6):
873-888.
Newmark, N.M. 1965. Effects of earthquakes on dams and embankments, 5th Rankine
lecture, Geotechnique 15(2): 139-160.
New Zealand Standards 1992-1997. NZS 4203 (1992) General Structural Design
Loadings for Buildings; NZS 3101 Part 1 (1995) The Design of Concrete Structures;
NZS 3403 Part 1 (1997) Steel Structures Standard; NZS 3403 Part 2 (1997)
Commentary to the Steel Structures Standard; Transit New Zealand (TNZ) Bridge
Design Manual.
Nishizawa, S., Hashimoto, M., Sakata, Y. & Sonoi, K. 1998. Investigation and analysis of
a landing pier of steel pipe piles damaged by the 1995 Hyogoken-Nambu earthquake,
Soils and Foundations, Special Issue on Geotechnical Aspects of the January 17 1995
Hyogoken-Nambu Earthquake, No.2, pp. 133-145.
Noda, 5., Uwabe, T. & Chiba. T. 1975. Relation between seismic coefficient and ground
acceleration for gravity quay wall, Report of Port and Harbour Research Institute
14(4): 67-111 (in Japanese).
Noda, S., Iida, T., Kida, H. & Saimura, Y. 1988. Shaking table tests on a countermeasure
for liquefaction by drainage piles to quaywalls, Proc. 23rd National Conference on
Soil Mechanics and Foundation Engineering, pp. 1015-1016 (in Japanese).
Noda, S., Kida, H. & Iida, T. 1990. Experimental study on attachment of drain function
to steel members as liquefaction remediation measures, Proc. 8th Japan Earthquake
_ Engineering Symposium. pp. 885-890 (in Japanese).
"'::?'Ohmachi. T. 1999. Formulation of Level 2 earthquake motions for civil engineering
structures. Proc. 7th U.S.-Japan Workshop 011 Earthquake Resistant Design of Lifeline
Facilities and Countermeasures against Soil Liquefaction, Technical Report MCEER-
99-0019, pp. 497-506.
Ohta, Y. & Goto, N. 1978. Empirical shear wave velocity equations in terms of charac-
teristic soil indexes, Earthquake Engineering and Structural Dynamics 6: 167-187.
Okabe, N. 1924. General theory on earth pressure and seismic stability of retaining wall
and dam, Journal of Japan Society of Civil Engineers 10(6): 1277-1323.
Okamoto, S 1973. Introduction to Earthquake Engineering, University of Tokyo Press.
Olivares, L. 1996. Caratterizzatione dell'urgilla di Bisaccia in condizioni monotone,
cicliche e dinamiche e riflessi sui comportamento del colle a seguito del terremoto del
J980. Doctoral Thesis, Universita degli Studi di Napoli "Federico II" (in Italian).
Omori, H. 1988. Design/construction example for soft soil, Tank Foundation (Soil
Strengthening Measure by Pore Water Pressure Dissipation Method). Foundation
Engineering 16(12): 122-129 (in Japanese).
O'Rourke, T.D., Meyersohn, W.D., Shiba, Y. & Chaudhuri, D. 1994. Evaluation of pile
response to liquefaction-induced lateral spreading, Proc. 5th US-Japan Workshop on
Earthquake Resistant Design of Lifeline Facilities and Countermeasures against
References "+bj
Liquefaction. Technical Report NCEER-94-0026. National Center for Earthquake
. Engineering Research. pp.457-4-79.
JOzutsumi. 0 .. Yuu. K.. Kiyarna. M. & Iai, S. 2000. Residual deformation analysis of
:; sheetpik quay wall and backfill ground at Showa-Ohashi site by simplified method.
Proc. 12Th V/orld Conference on Earthquake Engineering. Auckland.
Pam. H.J.. Park. R. & Priestley. MJ.N. 1988. Seismic performance of prestressed con-
crete piles and pile/pile cap connections. Research Report No. 88-3, Dept. of Civil
Engineering. University of Canterbury. 319 p.
Pam. H.J. & Park. R. 1990a. Flexural strength and ductility analysis of spiral reinforced
prestressed concrete piles. PCI Journal 35(4): 64-83.
Pam. H.J. & Park. R. 1990b. Simulated seismic load test on prestressed concrete piles and
pile-pile cap connections. PCI Journal 35(6): 42-61.
Park, R.. Priestley. MJ.N. & Walpole. W.R. 1983. The seismic performance of steel
encased reinforced concrete piles. Bulletin, NZNSEE 16(2): 124-140.
Pitilakis, K. & Moutsakis, A. 1989. Seismic analysis and behaviour of gravity retaining
walls - the case of Kalamata harbour quay wall, Soils and Foundations 29(1): 1-17.
Port and Harbour Research Institute, Japan (ed.) 1997. Handbook on Liquefaction
Remediation of Reclaimed Land. (translation by US Army Corps of Engineers,
Waterways Experiment Station), Balkerna, 312 p.
Prakash, S.. Wu. Y. & Rafnsson, E.A. 1995. On seismic design displacements of rigid
retaining walls, Proc. 3rd International Conference on Recent Advances in
Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis; pp. 1183-1192.
Prevost. J.H. 1981. DYNAFLOW: A nonlinear transient finite element analysis program,
Princeton University.
Priestley. MJ.N., Seible, F. & Chai, Y.H. 1992. Design guidelines for assessment retrofit
and repair of bridges for seismic performance, University of California, San Diego,
SSRP-92/01, 266 p.
Priestley, MJ.N., Seible, F. &Calvi, G.M. 1996. Seismic Design and Retrofit of Bridges,
John Wiley and Sons, New York. 686 p.
. Priestley, MJ.N. 1997. Comments on the ATC32 seismic design recommendations for
new bridges, Proc. 4th Caltrans Seismic Design Workshop, Sacramento.
Priestley, MJ.N. & Seible, F. 1997. Bridge Foundations under Seismic Loads - Capacity
and Design Issues, Sequad Consulting Engineers, Solana Beach, Report 97/10.
Priestley, MJ.N. 1999. POLA Berth 400. Dynamic Analyses of Linked Wharf Segments,
Seqad Consulting Engineers, Solana Beach, Report 99/01.
Puertos del Estado, Madrid (Spain) 2000. ROM 0.6, Acciones y Efectos Sismicos en las
Obras Maritimes y Portuarias. (English translation will be available).
Pyke R. 1979. Nonlinear soil models for irregular cyclic loadings, Journal of
Geotechnical Engineering Division, ASCE 105(GT6): 715-726.
Rampello, S., Silvestri, F. & Viggiani, G. 1994. The dependence of small strain stiffness
on stress state and history for fine-grained soils: the example of Vallericca clay, Proc.
1st International Symposium on Pre-failure Deformations of Geomaterials, Sapporo,
Vol. 1, Balkerna, Rotterdam. pp. 273-279.
Reese, LiC; Cox, W.R. & Koop, ED. 1974. Analysis of laterally loaded piles in sand,
Proc. 6th Annual Offshore Technology Conference, Vol. 2, Paper No. 2080, Houston,
Texas, pp. 473-483. .-
Reese, LC. & Wang, S. 1994. Documentation of Computer Program LPILE, Ensoft, Inc.,
Vers 4.0, Austin, TX, 366 p.
Richards, R. Jr. & Elms, D. 1979. Seismic behavior of gravity retaining walls, Journal of
Geotechnical Engineering Division, ASCE 105(GT4): 449-464.
,
,
,
~ .
I
i
464 PIANC
Richart, EE., Hall, J.R. & Woods, R.D. 1970. Vibrations of Soils and Foundations,
Prentice-Hall Inc., Englewood Cliffs (New Jersey).
Richter, CF, 1958. Elementary Seismology, W.H. Feeman and Company, San Francisco.
Rix, G. & Stokoe, K.H. 1991. Correlation of initial tangent modulus and cone penetra-
tion resistance, Proc. International Symposium on Calibration Chamber Testing,
Elsevier Publishing, New York, pp. 351-362.
Robertson, P.K. & Wride, C.E. 1997. Cyclic liquefaction and its evaluation based on the
SPT and CPT, Proc. NCEER Workshop on Evaluation of Liquefaction Resistance of
Soils, Technical Report NCEER-97-0022, National Center for Earthquake Engineering
Research, pp. 41-88.
Robertson, P.K., Campanella, R.G., Gillespie, D. & Rice, A. 1985. Seismic CPT to
measure in-situ shear wave velocity, Measurement and Use ofShear Wave Velocityfor
. Evaluating Dynamic Soil Properties, ASCE, New York, pp. 18-34.
Robertson, P.K., Woeler, DJ. & Finn, W.D.L. 1992. Seismic cone penetration test for
evaluating liquefaction potential under cyclic loading, Canadian Geotechnical Journal
29(4): 686-695.
Roth, W., Fong, H. & de Rubertis, C. 1992. Batter piles and the seismic performance of
pile supported wharves, PORTS '92, ASCE, pp. 336-349.
Rowe, P.W. 1952. Anchored sheet pile walls, Proc. ICE, Vol. 1, Part 1, pp. 27-70.
Rowe, P.W. 1955. A theoretical and experimental analysis of sheet pile walls, Proc. ICE,
Vol. 4, Part 1.
Sabetta, E & Pugliese, A. 1996. Estimation of response spectra and simulation of non-
stationary earthquake ground motions, Bull. Seism. Soc. Am. 86: 353-362.
Sagaseta, C., Cuellar, V. & Pastor, M. 1991. Modelling stress-strain-time behaviour
of natural soils - cyclic loading, Proc. 10th European Conference on Soil Mechanics
and Foundation Engineering, Florence (Italy), Vol. 3, Balkema, Rotterdam,
pp.981-999.
Sasaki, Y. & Taniguchi, E. 1982. Shaking table tests on gravel drains to prevent liquefac-
tion of sand deposits, Soils and Foundations 22(4): 1-14.
Sate, S., Mori, H. & Okuyama, K. 1988. Consideration on improvement effect of
sandy soil by the vibro rod method, Proc. 43rd National Conference of JSCE. III.
Sato, Yu., Ichii. K., Hoshino, Y.. Sato, Yo., Iai. S. & Nagao, T. 1999. Annual report on
strong-motion earthquake records in Japanese ports (1998). Technical Note of Pan and
Harbour Research Institute, No. 942. 230 p.
Schnabel, P.B.. Lysmer, J. & Seed, H.B., ]972. SHAKE: a Computer Program for
Earthquake Response Analysis of Horizontally Layered Sites, Report No. EERC
72-12, Earthquake Engineering Research Center, University of California. Berkeley.
Scan. R.E 1997. Crane response in 1995 Hyogoken Nanbu earthquake. Soils and
Foundations 37(2): 8]-87.
SEAOC ]995. Performance Based Seismic Engineering of Buildings. Structural
Engineers Association of California. Sacramento. California.
Seed. H.B. & Whitman. R.Y. 1970. Design of earth retaining structures for dynamic
loads. ASCE Specialty Conference on Lateral Stresses in the Ground and Design of
Earth Retaining Structures. Ithaca, pp. 103-]47.
Seed. H.B. & Idriss, I.M. 1971. Simplified procedure for evaluating liquefaction poten-
tial. Journal of Soil Mechanics and Foundations Division, ASCE 97(SM9):
]249-]273. .
Seed. H.B., ldriss, I.M., Makdisi, F. & Banerjee, N. 1975. Representation of irregular
stress time histories by equivalent uniform stress series in liquefaction analyses. Report
References 465
No. EERC 75-29. Earthquake Engineering Research Center. University of California.
Berkeley.
Seed. H.B. & Booker. J.R. 1977. Stabilization of potentially liquefiable sand deposits using
gravel drains. Journal ofGeotechnical Engineering Division. ASCE 103(GT7): 757-768.
Seed. H.B. 1979. Soil liquefaction and cyclic mobility evaluation for level ground during
earthquakes. Journal of Geotechnical Engineering. ASCE 105(GT2): 201-255.
Seed. RB. & Idriss. I.M. 1982. Ground motions and soil liquefaction during earthquakes.
Earthquake Engineering Research Institute Monograph Series, Oakland. California.
134 p.
Seed. H.B.. Idriss. LM. & Arango, L 1983. Evaluation of liquefaction potential using
field performance data. Journal of Geotechnical Engineering, ASCE 109(3): 458-482.
Seed. RB.. Tokirnatsu, K., Harder. L.F. & Chung, R.M. 1985. The influence of SPT pro-
cedures in soil liquefaction resistance evaluations, Journal of Geotechnical
Engineering. ASCE 111(12): 1425-1445.
Sekita, Y. Tatehata, H., Anami, T. & Kadowaki, T 1999. Outline and performance of new
tsunami forecast service in the Japan Meteorological Agency, Proc. 31st Joint Meeting
of U.S.-Japan Panel on Wind and Seismic Effects, UJNR, Tsukuba.
Shibuya, S.. Mitachi, T, Fukuda, F. & Degoshi, T 1995. Strain rate effects on shear mod-
ulus and damping of normally consolidated clay, Geotechnical Testing Journal. ASTM
18(3): 365-375.
Shimizu.T, & Takano, Y 1979. Design and construction of LNG tank foundation on
loose sand - Indonesia, Tsuchi-to-Kiso, Japanese Society of Soil Mechanics and
Simonelli, A.L. & Mancuso, C. 1999. Prove dinamiche in sito per la misura delle propri-
eta meccaniche dei terreni (in italian), Advanced Professional Training Course on
Problemi di ingegneria geotecnica nelle aree sismiche, CISM (International Centre for
Mechanical Sciences), Udine, Italy.
Skernpton, A.W. 1986. Standard penetration test procedures and the effects in sands of
overburden pressure, relative density, particle size, aging and overconsolidation,
Geotechnique 36(3): 425-447.
Sritharan, S. & Priestley, MJ.N. 1998. Seismic testing of a full scale pile-deck connec-
tion utilizing headed reinforcement, Div. of Str. Eng., TR981l4, University of
California, San Diego, 34 p.
Steedman, R.S. 1984. Modelling the behaviour of retaining walls in earthquakes, PhD
Thesis, Cambridge University.
Steedman, R.S. & Zeng, X. 1990. The seismic response of waterfront retaining walls,
Design and Performance of Earth Retaining Structures, Geotechnical Special
Publication No. 25, ASCE, pp. 872-886.
Steedman, R.S. & Zeng, X. 1996. Rotation of large gravity walls on rigid foundations
under seismic loading, Proc. ASCE Annual Convention, Washington.
Steedman, R.S. 1998. Seismic design of retaining walls, Geotechnical Engineering, Proc.
Institution of Civil Engineers, Vol. 131, pp. 12-22.
Stark, T.O. & Olson, S.M. 1995. Liquefaction resistance using CPT and filed case histo-
ries, Journal of Geotechnical Engineering, ASCE 121(GTI2): 856-869.
Stokoe, K.R, Wright, S.G., Bay, I.A. & Roesset, I.M. 1994. Characterisation of geo-
technical sites by SASW method, Geophysical characterisation of sites, R.D. Woods
(ed.), Balkema, Rotterdam, pp. 1 5 - 2 5 . - ~
Sugano, T, Kitamura, T., Morita, T. & Yui, Y. 1998. Study on the behavior of steel plate I,'
cellular bulkheads during earthquake, Proc. 10th Japan Earthquake Engineering
Symposium, Vol. 2, pp. 1867-1872. t
466 PIANC
Sugano, T. & Iai, S. 1999. Damage to port facilities, The 1999 Kocaeli Earthquake,
Turkey - Investigation into Damage to Civil Engineering Structures, Earthquake
Engineering Committee, Japan Society of Civil Engineers, pp. 6-1-6-14.
Sugano, T., Kaneko, H. & Yamamoto, S. 1999. Damage to port facilities, The 1999 Ji-Ji
Earthquake, Taiwan - Investigation into Damage to Civil Engineering Structures,
Earthquake Engineering Committee, Japan Society of Civil Engineers, pp. 5-1- 5-7.
Suzuki, Y., Tokito, K., Suzuki, Y. & Babasaki, J. 1989. Applied examples of antiliquefac-
tion foundation ground improvement method by solidification method, Foundation
Engineering 17(9): 87-95 (in Japanese).
Suzuki, Y., Tokimatsu, K., Koyamada, K., Taya, Y., & Kubota, Y. 1995. Filed correlation
of soil liuqefaction based on CPT data, Proc. International Symposium on Cone
Penetration Testing, CPT'95, Linkoping, Sweden, Vol. 2, pp. 583-588.
Takahashi, K., Kikuchi, Y. & Asaki, Y. 1993. Analysis of flexural behavior of anchored
sheet pile walls, Technical Note of Port and Harbour Research Institute, No. 756,
32 p. (in Japanese).
Takeda, T., Sozen, M.A. & Nielsen, N.N. 1970. Reinforced concrete response to simu-
lated earthquakes, Journal of Structural Division, ASCE 96(ST12): 2557-2573.
Talaganov, K.V. 1986. Determination of liquefaction potential of level site by cyclic
strain, Report IZIIS 86-129, Institute of Earthquake Engineering and Engineering
Seismology, Univ, Kiril and Metodij, Skopje, Yugoslavia.
Tanaka, S. & Inatomi, T. 1996. Seismic response analysis of gantry cranes on container
quay walls due to the Great Hanshin earthquake, Proc. 11th World Conference on
Earthquake Engineering, Acapulco, Paper No. 1250.
Tanaka, S. & Sasaki, C. 1989. Sandy ground improvement against liquefaction at Noshiro
thermal power station, Tsuchi-to-Kiso, Japanese Society of Soil Mechanics and
Tatsuoka, E, Jardine, R.I., Lo Presti, D., Dr Benedetto, H. & Kodaka, T. 1997.
Characterising the pre-failure deformation properties of geomaterials, Proc. 14th
ICSMFE, Hamburg, Theme Lecture, Vol. 4, pp. 2129-2164.
Tchebotarioff, G.P. 1949. Large Scale Earth Pressure Tests with Model Flexible
Bulkheads, Princeton Univ.
Terzaghi. K. 1955. Evaluation of coefficients of subgrade reaction. Geotechnique 5:
197-326.
Tokimatsu, K., Kuwayama, S. & Tamura, S. 1991. Liquefaction potential evaluation
based on Rayleigh wave investigation and its comparison with field behaviour, Proc.
Lnd International Conference on Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics, St. Louis (Missouri), Vol. 1. pp. 357-364.
Towhata. I. & Ishihara. K. 1985. Modelling soil behaviour under principal stress axes
rotation, Proc. 5th International Conference on Numerical Methods in Geomechanics.
Nagoya, Balkema, Rotterdam, pp. 523-530.
Towhata. 1. & Islam. S. 1987. Prediction of lateral movement of anchored bulkheads
induced by seismic liquefaction. Soils and Foundations 27(4): 137-147.
Toyota. Y.. Nakazato. S., & Nakajima, Y. 1988. Consideration on behavior of existing
steel pipe pile wall wharf during installation of vibro rod method, Proc. 43rd National
Conference of JSCE, Ill, pp. 756-757 (in Japanese).
Tsinker, G.P. 1997. Handbook of Port and Harbor Engineering. Geotechnical and
Structural Aspects. Chapman & Hall. 1054 p.
Tsuchida. H.. Noda. S., Inatorni, T., Uwabe, T., Iai, S.. Ohneda. H. & Toyama. S. 1985.
Damage to port structures by the 1983 Nipponkai-Chubu earthquake. Technical Note
oj Port and Habour Research Institute, No. 511,447 p. (in Japanese).
References 467
Tsuchida, H.. de La Fuente. R.H.. Noda. S. & Valenzuela. M.G. 1986. Damage to
port facilities in Port of Valparaiso and Port of San Antonio by the March 3.
1985 earthquake in middle of Chile. 4as Jordanadas Chileanas de Sismologia e
lntenieria Antisismica & International Seiminar Oil the Chilean March 3. 1985
earthquake.
Tsuchida, H. 1990. Japanese experience with seismic design and response of port and
harbor structures, Proc. POLA Seismic Workshop on Seismic Engineering, Port of Los
Angeles. pp. 138-164.
Tsuchida, T. 1990. Study on deformation of undrained strength of clayey ground by
means of triaxial tests, Technical Note of Port and Harbour Research Institute, No.
688, 199 p. (in Japanese).
U.S. Navy 1982. DM7.2, Foundation and Earth Structures, US Dept. of the Navy, Naval
Facilities Engineering Command, pp. 7.2-209.
U.S. Navy 1987. Naval Facilities Engineering Command Military Handbook MIL-
HDBK-1002/3 'Structural Engineering Steel Structures'.
Uwabe, T. 1983. Estimation of earthquake damage deformation and cost of quaywalls
based on earthquake damage records, Technical Note of Port and Harbour Research
Institute, No. 473, 197 p. (in Japanese).
Vazifdar. ER. & Kaldveer, P. 1995. The great Guam earthquake of 1993 port disaster and
recovery, PORTS '95, ASCE, pp. 84-94.
Viggiani, G. & Atkinson, J.H. 1995. Interpretation of bender element tests, Geotechnique
45(1): 149-154.
Vucetic, M. 1992. Soil Properties and Seismic Response, Proceedings of the X World
Conference on Earthquake Engineering, Madrid, Balkerna, Vol. 3, pp. 1199-1204.
Vucetic, M. 1994. Cyclic threshold shear strains in soils, Journal of Geotechnical
Engineering Division, ASCE 120(12): 2208-2228.
Vucetic, M. & Dobry, R. 1991. Effects of the soil plasticity on cyclic response, Journal
of Geotechnical Engineering Division, ASCE 117(1): 89-107.
Werner, S.D. 1986. A case study of the behavior of seaport facilities during the 1978
Miyagi-ken Oki earthquake, Proc. Lifeline Seismic Risk Analysis-Case Studies, ASCE,
pp. 113-130.
Werner, S.D. (ed.) 1998. Seismic Guidelines for Ports, Technical Council on Lifeline
Earthquake Engineering, Monograph No. 12. ASCE.
Westergaard, H.M. 1933. Water pressure on dams during earthquakes, Transactions of
ASCE 98: 418-472.
Whitman, R.Y. & Liao, S. 1984. Seismic design' of gravity retaining walls, Proc. 8th
World Conference on Earthquake Engineering, San Francisco, Vol. III, pp. 533-540.
Whitman, R.Y. & Liao, S. 1985. Seismic design of retaining walls, US Anny Corps of
Engineers, Waterways Experiment Station, Miscellaneous Paper GL-85-1.
Whitman, R.Y. & Christian, J.T. 1990. Seismic response of retaining structures, Proc.
POLA Seismic Workshop on Seismic Engineering, Port of Los Angeles, pp. 427-452.
Wilson, E.L., Der Kiureghian, A. & Bayp, E.P. 1981. A replacement of the SRSS methods in
seismic analysis, Earthquake Engineering and Structural Dynamics, Vol. 9, pp. 187-194.
Wittkop, R. 1994. Application of VELACS philosophy to Port of Los Angeles Pier
400 Project, in VELACS - Verification ofNumerical Procedures for the Analysis ofSoil
Liquefaction Problems, K. Arulanadan and R.E Scott (eds), Balkema, pp. 1647-1655.
Wood, H.O. & Neumann. E 1931. Modified Mercalli intensity scale of 1931, Bull. Seism.
Soc. Am. 21: 277-283.
Wu, S.c., Cornell, C.A. & Winterstein, S.R. 1995. A hybrid recurrence model and its
implication on seismic hazard results, Bull. Seism. Soc. Am. 85: 1-16.
468 PIANC
Wyllie, L.A., Abrahamson, N., Bolt, B., Castro, G., Durkin, M.E., Escalante, L., Gates,
H.J., Luft, R., McConnick, D., Olson, R.S., Smith, P.D. & ValJenas, J. 1986. The Chile
Earthquake of March 3, 1985, Earthquake Spectra, 2(2): EERI, 513.
Wyss, M. 1979. Estimating maximum expectable magnitude of earthquakes from fault
dimensions, Geology 7: 336-340.
Yamanouchi, K., Kumagai, T., Suzuki, Y., Miyasaki, Y. & Ishii, M. 1990. Liqi.efaction
remedial measures using grid-shaped continuous underground wall, Proc. 25th
National Conference on Japanese Society of Soil Mechanics and Foundation
Engineering, pp. 1037-1038 (in Japanese).
Yamazaki, H., Zen, K. & Kagaya, H. 1992. Liquefaction characteristics of overconsoli-
dated sand, Proc. 27th National Conference on Soil Mechanics and Foundation
Engineering, pp. 845-846 (in Japanese).
Yamazaki, H., Zen, K. & Koike, F. 1998. Study of the liquefaction prediction based on
the grain size distribution and the SPT N-value, Technical Note of Port and Harbour
Research Institute, No. 914 (in Japanese).
Yang, D.S. 1999. Deformation Based Seismic Design Models for Waterfront Structures,
Ph.D. dissertation, Department of Civil, Construction and Environmental Engineering,
Oregon State University, Corvallis, 260 p.
Yokota, H., Kawabata, N., Akutagawa, H., Kurosaki, K., Tsushima, T., Harada, N. &
Yato, A. 1999. Verification of seismic performance of an open piled pier by an elasto-
plastic method and a simplified method, Technical Note ofPort and Harbour Research
Institute, No. 943,40 p. (in Japanese).
Yoshimi, Y. & Tokimatsu, K. 1991. Ductility criterion for evaluating remedial measures
to increase liquefaction resistance of sands, Soils and Foundations 31(l): ] 62-168.
Youd, T.L. & Idriss, I.M. (eds.) ] 997. NCEER Workshop on Evaluation of Liquefaction
Resistance of Soils, Technical Report NCEER-97-0022, National Center for
Earthquake Engineering Research, pp. ]-40.
Zen, K., Yamazaki H. & Umehara, Y. ]987. Experimental study on shear modulus and
damping ratio of natural deposits for seismic response analysis. Report of Port and _
Harbour Research Institute 26(1): 41-] 13 (in Japanese).
Zen, K. 1990. Development of premix method as a measure to construct a liquefaction-
free reclaimed land. Tsuchi-to-Kiso, Japanese Society of Soil Mechanics and
Zen. K.. Yamazaki. H. & Sato. Y. ]990. Strength and deformation characteristics of
cement treated sands used for premixing method, Report of Port and Harbour
Research Institute 29(2): 85-1] 8 (in Japanese).
Zienkiewicz. O.e. & Bettess, P. 1982. Soils and other saturated media under transient.
dynamic conditions: General formulation and the validity of various simplifying
assumptions, Soil Mechanics - Transient and Cyclic Loads. John Wiley & Sons,
pp.1-16.
Zienkiewicz, O.e.. Chan. A.H.e.. Pastor. M.. Schrefler, B.A. & Shiomi. T. 1999.
Computational Geomechanics with Special Reference to Earthquake Engineering,
John Wiley & Sons. 383 p.
Zmuda. R., Weismair, M. & Caspe. M. ] 995. Base isolating a wharf using sliding friction
isolators at the Port of Los Angeles. PORTS '95. ASCE. pp. ] 263-1274.
Zwamborn, J.A. & Phe1p. D. 1995. When must breakwaters be rehabilitated/repaired?
PORTS '95. ASCE, pp. 1183-] ]94.
Index
-Accelerauon Response Spectrum. 132
Accelerograrn, 132
Acceptable damage levels. 24
Acceptable level of damage, 24
Active angle of failure. 384
Active earth pressures. 310, 410
Active soil wedge. 38...
Allowable bending moment, 415
Allowable deformation limits, 354
Allowable moment. 408
Allowable stress, 408
American Petroleum Institute (API), 323
Anchor design. 409
Anchored steel sheet pile walls. 361
Angle of internal friction, 328
Area of soil improvement. 299
ASCETCLEE guidelines. 86
Attenuation relationship. 146
Axial Load Mechanism. 249
Backfill inertia force. 388
Basic strategies for liquefaction remediation. 273
Batter pile-deck system. 95
Batter piles. 38, 251
Bearing capacity. 309
Bedrock,55
Bedrock motions. 7
Bender elements (BE). 178
Bending moments, 361
Bending stress in the wall. 300
Bi-Iinear moment-curvature relationship. 430
Bi-linear springs. 430
Borehole tests. 173
Bottom boundary. 420
Bracketed Duration; 133
Breakwater. 52. 153
Brittle fracture of steel piles. 41
Buckling. 363
Buoyant unit weight, 312
Caisson walls. 357
Cambridge type. 176
Capacity design checks, 229
Capacity protected members. 247
Capacity protection. 247
Cellular quay wall. 43
Cement deep mixing stabilization method (COM). 433
Cementation. 280
Centre of rigidity. 228
Chang's method. 328
Characteristic earthquakes, 145
Coarse-grained fills. 313
Coefficient of active earth pressure. 40 I. 402
Coefficient of horizontal subgrade reaction. 364
Coefficient of subgrade reaction. 415
Coherence function. 212
Compaction. 275. 282
Composite Action. 265
Composite breakwaters. 365
Concrete infill, 236
Concrete Mechanism Strength. 248
Concrete strain. 233. 234
Concrete Structures. 266. 267
Concrete-filled steel shell piles, 236
Cone penetration tests (CPT). 60
Confinement ratio. 244
Connection detail, 253. 256
Consolidation history. 292
Constant energy assumption. 347
Conventional design concept. 90
Corrosion protection measures. 351
Coulomb's earth pressure theory. 402
CPT tip resistance. 196
Crane, 45
Crane loads. 424
Crane rails. 47,96
Cross Correlation. 209
Cross Correlation function. 214
Cross Power Spectrum. 210. 214
Cross-Hole (CH) test. 173
C-type subsoil. 328
Curvature ductility limit. 349
Curvatures. 235
Cyclic resistance. 10
-----------_._----
470 PlANe
Cyclic simple shear test (CSS), 176
Cyclic stress ratio, 192, 291
Cyclic tests, 175
Cyclic torsional shear test (CTS), 176
Cyclic triaxial test (CTX). 175, 203
Damage Control Limit State, 218
Damage criteria. 25, 31
Damage to a composite breakwater. 21
Damage to a crane. 18
Damage to a gravity quay wall, 15
Damage to a pile-supported wharf, 16, 95
Damage to a rubble mound breakwater, 2I
Damage to a sheet pile quay wall, 16
Damage to breakwaters. 96
Damage to Cellular Quay Walls, 95
Damage to Cranes, 95
Damage to gravity quay walls. 93
Damage to sheet pile quay walls, 94
Damage-control Limit State, 234, 238
Damage-control limit-state strains, 236
Damping, 228
Damping ratio, 180
Deadweight, 424
Degradation indexes, 161
Degradation parameter, 161
Design accelerograms, 149
Design bending moment, 408
Design earthquake levels. 24
Design earthquake motions. 141. 150
Design embedment depth. 416
Design procedure for a soil improvement area. 295
Design procedure for drainage method. 286
Design procedure of premix method. 289
Design strength for deck members. 251
Design tie-rod force. 409. 416
Design ultimate compression strain. 351
Deterioration. 264, 266
Deterministic method. 141
Diameter of the tie-rod. 409
Differential settlement. 431
Dimensionless empirical constant. 326
Direct Fourier Transform. 210
Direct wave velocities, 208
Displacement ductility, 347
Displacement ductility criteria. 353
Displacement ductility factor. 431
Displacement ratio, 317
Double plastic hinge, 431
Down-Hole (DH) technique. 173
Drag loading. 368
Drainage. 166
Drainage method, 284
Dual level approach. 26
Ductility criteria. 351
Ductility factor. 41, 346. 348
Duration. 133
Dynamic (SPT). 168
Dynamic active earth pressure coefficient, 311. 341. 386
Dynamic active earththrust, 311
Dynamic analysis. 56. 205. 223
Dynamic earthpressure force. 402
Dynamic passive earthpressure coefficient, 312
Dynamic passive earththrust, 312
Dynamic response. 297
Dynamic tests. 177
Earth pressure increased by compaction, 302
Earth pressures. 296. 361
Earthquake moment magnitudes. 381
Earthquakes. 129
Eccentricity. 390
Effective anchor index, 318
Effective point of rotation. 318
Effective seismic coefficient., 315
Effective stiffness. 227
Effective stress analysis, 335
Effective stress approach. 161
Effective stress model. 208
Elastic limit, 428
Elastic response, 233
Elastic stiffness. 227. 240
Elasto-plastic model, 167
Element size, 394
Embankments, 153
Embedment length of the sheet pile below the mud- . ~
line. 319
Embedment panicipation index. 318
Empirical correlations. 182
Energy efficiency. 193
Engineering judgement. 67
Equivalent acceleration. 201
Equivalent buoyant unit weight. 400
Equivalent damping ratio. 158
Equivalent elastic limit, 431
Equivalent fixity pile. 221, 328.425
Equivalent leg lengths of a pile-deck system. 329
Equivalent linear analysis, 356
Equivalent linear model. 161
Equivalent number of uniform cycles. 134
Equivalent N-value. 200
Equivalent reaction force. 413
Equivalent saturated unit weight. 400
Equivalent seismic coefficient. 315
Equivalent shear modulus. 158
Equivalent single mode analysis. 224
Equivalent spring stiffness. 425
Equivalent SPT N-value, 335. 391
Equivalent subgrade elastic modulus. 328
Equivalent unit weight. 313. 385
Equivalent viscous dampers. 394
Equivalent viscous damping. 229
Equivalent-depth-to-fixity model. Z42
Eurocode design. 87
Excess pore pressure generation. 369
Excess pore water pressures. 295. 30 I. 314
Existing construction. Z62
Existing piers. 260
Existing structures. Z38. 265. 299
Exposure time. 263. 264
Extreme fibre strain. 350
Factor of safety. 64. 90
Factor of safety against overturning. 390
Factor of safety against sliding. 390
Failures of gravity quay walls. 93
Fines content. 197. 202. 292
Finite difference methods (FDM). 60. 368
Finite element mesh. 422
Finite element methods (FEM). 60. 368
Fixed earth support condition. 416
FLAC. 357. 368
Flexibility number, 408
Flexural Reinforcement. 265
Flexural rigidity. 322
Flexural strength reduction factors, 243
F L I ~ 335. 358. 361. 364. 366. 368
FLUSH, 356. 368
FLUSH-L. 357
Force-displacement curve, 229
Force-displacement response. 255
Foundation deformation analysis. 232
Fourier analysis. 210
Fourier Spectrum. 133
Free earth support condition, 417
Free field motion, 395
Fuse. 39
Fuse-link. 39
Geophysical tests, 172
Geotechnical modelling of foundation, 355
Geotechnical parameters', 383
German design. 84
Grain size distribution, 200
Gravity quay wall. 31
Gravity quay walls, 151
Ground motion intensity factor, 418
Gutenberg-Richter relationship. 144
Hardening spring type, 168
Hazard-consistent magnitude and distance. 150
Headed Rebar Detail, 259
Hinge formation. 227
Hollow prestressed piles, 238
Hollow steel shell pile. 236
Hooks. 256
Horizontal passive earth thrust. 412
Horizontal seismic coefficient, 40 I
Index -+71
Hydrodynamic force. 388. 412
Hydrodynamic pressure. 314
Hydrostatic tluid pressure. 405
Hysteresis loop. 168
Impermeable underground wall. 283
Incident wave. 335
Incremental dynamic active earth pressure. 403
Inelastic rotation. 227
Inelastic stiffness. 243
Inelastic time-history analysis. 224. 230
Inflow of pore water. 296
In-ground hinge. 234
In-ground plastic hinge. 241
Initial modulus of subgrade reaction. 328
Initial shear stiffness. 171
Initial stiffness. 229
Input parameters. 369
In-situ freezing sampling. 382
In-situ freezing technique. 358
In-situ geophysical tests. 171. 179
In-situ penetration tests. 168
Intensity. 129
Interplate earthquakes, 134
Intraplate earthquakes. 136
Inverse Fourier Transform, 210
Japan Meteorological Agency (JMA). 13
Japanese design, 80
Joint elements, 395. 438
Joint failure. 261
Joint shear stress force, 261
Laboratory cyclic tests, 382
Laboratory tests, 174. 179
Large strain, [58. 162
Lateral seismic body force. 329
Lateral spreading, II, 368
Level I earthquake motion (L1), 23. 25. 218
Level 2 earthquake motion (L2), 24, 25. ISO, 218
Limit lateral load. 429
Limit states. 233
Limiting curves. 383
Linear plane elements, 438
Linear threshold strain, 159. 187
LINGS. 368
Liquefaction. 9, 163
Liquefaction charts, 195
Liquefaction potential, 60, 201
Liquefaction potential assessment, 56, 191. 199
Liquefaction remediation of backfill sand, 381
Liquefaction remediation measures. 439
Liquefaction resistance, 10, 170
Load-displacement curve, 431
Load-displacement relationship, 350
Local displacement transducers (LOT). 176
472 PlANe
Local site effects. 8.56. 138.204,205
Location of the anchor, 409
Loss of shear strength. 296
Lowering of ground water level, 281, 294
LPILE,368
Magnitude, 130
Magnitude scaling factor. 339
Maximum bending moment, 406
Maximum deck level accelerations, 423
Medium strains, 158, 159
Methods of analysis for pile-deck structures. 222
Mild steel dowels. 245
Modelling of a caisson, 355
Modelling of sheet pile walls, 355
Modification factor. 348
Modified horizontal seismic coefficient. 312
Modified seismic coefficient, 385
Moment at the pile/deck interface, 261
Moment balance about the tie-rod, 412
Moment-curvature, 234
Moment-curvature characteristics, 243. 244
Moment-curvature curves, 236
Moment-curvature response, 235
Moment-resisting connections, 253
Mononobe-Okabe equation, 310
Mononobe-Okabe method, 386, 398
Movement joints, 221
Multi-degree of freedom (MDOF), 65
Natural period, 424
New Zealand design, 88
Newmark's sliding block approach, 343
NGI and SGI types, 176
Nominal principal tension stress. 261
Nominal shear stress. 261
Non-linear. effective stress analysis methods, 357
Non-liquefiable soil. 358
Normal fault. -136
Normalized caisson geometry, 391
Normalized displacement. 320
Normalized displacement factor, 344
Normalized earthquake intensity. 344
Normalized earthquake motion, 393
Normalized ground motion intensity. 339
Normalized lateral displacement factor. 418
Nonnalized soil improvement, 393
Normalized soil improvement factor. 344
Nonnalized SPT N-values. (N
1
loo. 337
Nonnalized width of soil improvement. 340
N-Pier.431
Onset of liquefaction, 164
Operational damage criteria. 31
Overconsolidation. 290
Overconsolidation ratio. 290
Overturning, 309
Overturning limit for the crane, 51
Pacific Tsunami Warning Centre (PTWC), 12
Passive earth pressure, 312, 404, 411
Passive earth pressure coefficient, 404, 411
Peak deck response, 430
Peak ground acceleration (PGA), 7
Peak Ground Displacement, 132
Peak Ground Horizontal Acceleration, 131
Peak Ground Horizontal Velocity, 132
Peak ground velocity (PGV), 7
Peak Ground Vertical Acceleration, 131
- Peak Ground Vertical Velocity, 132
Peak response accelerations, 347
Penetration tests, 168
Performance evaluation, 27
Performance goals, 217, 262
Performance grades, 24
Performance-based design, 23,90
Permeability, 313
Phase difference, 360
Phase velocity, 214 _
Pile cap, 40
Pile failures, 95
Pile plastic hinges, 232
Pile top, 40
Pile walls, 355
Pile/Column Reinforcement, 265
Pile-deck structures, 219
Pile-supported wharf, 36, 152, 362
Plastic hinge formation, 322
Plastic hinge length, 240, 349
Plastic hinge moment with axial load, 429
Plastic hinge moment without axial load. 429
Plastic hinges, 234. 243
Plastic rotation. 240
Plasticity index. 187.202
Plating, 265
Pore pressure, 163
Pore pressure dissipation, 166
Pore water pressure dissipation. 280
Pore water pressures, 9
Pore-water pressure build-up. 160
Port structures. 13
Potential slip surface, 329
Predominant period, 139
Pre-failure, 158
Pre-failure soil properties. 182
Preload, 281
Preload method. 290
Premix method, 287
Prestressed piles. 236, 254
Prestressing, 265
Principal tension ratio. 261
Probabilistic method. 141
Procedure for liquefaction remediation. 274-
Pseudo-static analysis. 383
Pseudo-static approaches. 309
PS-Iogging. 3 8 ~
Pushover analysis. 66. 226. 350. 430
p-y curves. 323
Rate-dependency. 159
Recorded time histories. 150
Reduced shear moduli. 357
Reduced shear strength. 357
Reduction factor. 333. 339
Refraction surveys. 172
Reliability of input parameters. 66
Repairable. 218
Replacement. 281
Representative internal friction angle. 385
Residual displacement. 229
Residual volumetric strains. 160
Resonant column test (RC). 177
Response spectrum. 66. 346
Response spectrum analysis. 423
Return period. 7
Reverse fault. 136
Rigid block analysis. 64
Rowe's moment reduction procedure. 407
Rowe's approach. 410
Rowe's chan. 408
Safety factor. 317
Sand boils. 11
Sand compaction pile technique. 359
Sea water level. 315
Section modulus. 415
Seismic analysis. 55
Seismic coefficient. 383
Seismic Cone <SC). 174
Seismic Design Codes and Guidelines. 79
Seismic Hazard. 141. 147
Seismic hazard analysis. 7
Seismic inertia angle. 310. 401
Seismic inertia force. 322
Seismic input. 151
Seismic intensity scales. 129
Seismic load combinations. 218
Seismic performance of a cellular quay wall. 43
Seismic performance of a deck/pile system. 39
Seismic performance of a gravity quay wall. 32
Seismic performance of a sheet pile quay wall. 34
Seismic performance of quay walls with cranes. 47
Seismic reflection. 172
Seismic response curve. 28
Seismic source activity. 143
Seismic stability of slopes. 329
Seismotectonic approach. 141
Serviceability limit. 35l
Index 473
Serviceability Limit State. 218. 233
Serviceable. 217
Shear failure line. 168
Shear keys. 225
Shear modulus. 180
Shear Reinforcement. 265
Shear strain level. 356
Shear strain restraint. 28l
Shear strength of concrete piles. 247
Shear strength of concrete-tilled steel shell piles. 25l
Shear strength reduction factors. 247
. Shear stress. l57
Shear-wave velocity. 171. 195. 196
Shear-wave velocity profiles, l70
Sheet pile quay. 33
Sheet pile quay walls. 15l
Significant Duration. 133
Similarity number. 414
Simplified analysis. 56. 204. 223
Simplified dynamic analysis. 56. 205. 223
Simultaneous seismic excitation. 219
Single degree of freedom (SDOF). 65
Single degree of freedom (SDOF) system. 346
Single degree-of-freedom (SDOF) oscillator, 132
Smgle level approach. 26
Site amplification factors. 9
Site classification. 56
Site response analysis. 56
Slidin.,g. 309
Sliding block analysis. 65. 330. 331. 341
Sliding block model. 330
Slope deformations, 368
Sloping embankments. 328
Small strains. 158
Small-strain damping. 183
Small-strain shear modulus. 182
Soil improvement. 336
Soil improvement area. 294
Soil improvement guidelines. 340
Soil-structure interaction. 22. 60. 220. 355. 368
Solidification. 280
Spanish design. 81
Spectral analysis. 150
Spectral Analysis of Surface Waves (SASW) method.
172
Spectral modal analysis. 225
SPT blow count. 196
SPT N-value. 282
SPT procedures. 193
Stability of anchored sheet pile walls. 310
Stand-alone analyses. 226
Standard penetration tests (SPT). 60
Static analyses. 395
Static bollard pull. 388
Static earth pressure force, 402
Steel H-Section Piles. 260
474 PlANe
Steel shell piles. 253
Steel strain. 234
Steel Structures. 266. 267
Straight-bar Embedment Detail. 257
Strain amplitude. 158
. Strain dependent stiffness and damping. 188
Strain ductility criteria. 353
Strain level, 157
Strain-controlled tests. 175
Strength parameters, 168
Stress correction factor. 194
Stress path. l64
Stress reduction factor. 192
Stress-controlled tests. 175
Stress-strain curve. 234
Strike-slip fault, 136
Structural damage criteria. 31
structural parameters. 383
Structural remedial measures. 282
S-type subsoil. 3,28
Subgrade reaction. 350
Subsoil, 157
Static (CPT). l68
Subsoil characterisation. 204
Super piles. 225. 230
Surcharge, 424
Surcharge approach. 294
Temporal model. 145
Threshold acceleration, 64, 330, 33 1
Threshold acceleration for sliding, 65
Threshold horizontal seismic coefficient, 341
Threshold level. 428
Threshold seismic coefficient. 64. 315
Tie-rod force. 413
Timber piles. 252
Timber Structures. 267. 269
Time domain analysis. 208
Time histories. 149
_4
..
"
Total static active earth pressure. 403
Total stiffness. 228
Transverse reinforcement. 245
Transverse Reinforcement (truss) Mechanism. 248
Tri-linear moment-curvature relationship. 430
Tsunami. 12.96
Tsunarnieanhquakes.13
Tsunami Warning System in the Pacific (1WSP). 13
Two-phase medium. 167
Type of faults. 136
Types of cement. 268
Ultimate anchor resistance. 341
Ultimate curvature t/J". 349
Ultimate displacement. 349
Ultimate resistance. 326
Unconfined compression strength, 438
Undrained cyclic properties, 395
Undrained cyclic resistance, 191
Uplift force, 389
US Navy/California State design, 86
Vertical component of input acceleration. 360
Vertical effective stress, 164
Vertical seismic coefficient, 401
Vibration resulting from compaction. 301
Virtual (or a hypothetical) embankment surface, 328
Virtual slope surface. 426
Volumetric threshold. 161, 162. I88
Westergaard dynamic water loading. 3I4
Westergaard's expression. 388
Winkler beam analysis. 414
Winkler foundation. 322.410
Winkler springs. 230
Yield bending moment. 431
Yield curvature. 236
Yield stress, 408

PIANC Seismic Structures

Uploaded by

Copyright:

Available Formats

PIANC Seismic Structures

Uploaded by

Document Information

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

PIANC Seismic Structures

Uploaded by

Copyright:

Available Formats

INTERNATIONAL NAVIGATION

E :. Sand como.clion oile

Steel sheet pile

~ silu and clays

\:1 -=- -.5 .

Gravelly Sands Sands Sud

'1. " -, . '\-". ,.i... 'fl.' _

2.0 iii ii ii' i' i iiii

V rtiealOsplac ment 1 5 e I e m :z: -r ...... _

Fig. T8.7. Loads acting on structure during earthquake.

a cos 8[1 + sine + 0) sin ]2

You might also like