INTERNATIONAL NAVIGATIONASSOCIATIONASSOCIATION INTERNATIONALEDE NAVIGATIONSEISMIC DESIGN GUIDELINESFOR PORT STRUCTURESWorking Group No. 34 of the MaritimeNavigationCommissionInternational NavigationAssociation- 915 A.A. BALKEMA PUBLISHERS / LISSE/ ABINGDON / EXTON(PA) / TOKYOLibrary of Congress Cataloging-in-Publication DataApplied forCover design: Studio Jan de Boer. Amsterdam. The Netherlands.Typesetting: Macmillan India Ltd.. Bangalore. India.Printed by: GrafischProduktiebedrijf Gorter. Steenwijk. TheNetherlands.CD ~ O O I Swets & Zeitlinger B.Y.. LisseTechnical Commentary 5is exempt fromthe copyright because the original material\ \ ' J ~ producedby theUSNavy forthe stateof California. USA. Noonecanexercisethecopyright.The figuresandtablein TechnicalCommentary 6 have beenreprintedfrom:Handbook onliquefaction remediation of reclaimed land. Port and Harbour ResearchInstitute(ed.). ISBN 90 54 j() 653 O. 1997. 25 em. 324 pr..EL"R 82.50/USS97.00/GBP5SAll rights reserved. Nopart ( ~ f this publicationmayhe reproduced. storedinaretrieve!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 2651g18 8 (hardback)ContentsPREFACE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IXMembers of PIANClMarComIWorking Group 34 XIList of Tables and Figures: Main Text XIIIMainTextChapter 1. INTRODUCTION ..... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 3Chapter 2. EARTHQUAKES AND PORT STRUCTURES .2.1 Earthquake Motion .?? L' f: ._._ Iqueaction .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 PileQuay Walls .4.3 Pile-Supported Wharves .4.4 Cellular Quay Walls .4.5 Quay Wallswith Cranes ~ .4.6 Breakwaters ~ .Chapter 5. SEISMIC ANALYSIS .5.1 Types of Analysis .5.2 Site ResponselLiquefaction Analysis .7791213152323?--)2731313336434552555556v r t'JAJ\lL5.3 Analysis of Port Structures .5.4 Inputs andOutputs of Analysis .Technical CommentariesTC1: EXISTING CODESAl\.TD GUIDELINES .Tl .1 List of Seismic Design Codes andGuidelines for PortStructures .TI.2 Reviewed Aspects of Codes and Guidelines .Tl.3 Seismic Design Practice for Port Structures around the WorldTC2: CASE HISTORIEST2.1 Damage to Gravity Quay Walls .T2.2 Damage toSheet PileQuay Walls .T2.3 Damage to Pile-Supported Wharves .T2.4 Damage to Cellular Quay Walls .T2.5 Damage to Cranes .T2.6 Damage toBreakwaters .TC3: EARTHQUAKE MOTION .T3.1 Sizeof Earthquakes .T3.2 Strong Ground MotionParameters < T3.3 Seismic Source and Travel Path Effects .T3.4 Local SiteEffects .T3.5 Seismic Hazard and Design Earthquake Motion .T3.6 GroundMotion Input for Seismic Analysis of Port StructuresTC4: GEOTECHNICALCHARACTERISATION .T4.1 Mechanical Behaviour of Soil under Cyclic Loads .T4.2 Measurement of SoilParameters .T4.3 Evaluation of Pre-Failure Soil Properties throughEmpiricalCorrelations .T4.4 Assessment of Liquefaction Potential .T4.5 GeotechnicalCharacterisation Strategy withReferenceto theAnalysis Requirements .Appendix: Interpretation(>f GeophysicalTeSTS .TC5: STRUCTURAL DESIGN ASPECTS OF PILE-DECKSYSTE!\1S .T5.1 DesignPerformance and Earthquake Levels .T5.2 Modelling Aspects .T5.3 Methods of Analysisfor SeismicResponse .T5.4 StructuralCriteria forPier and Wharf Piles .6066-\'{7779808991939494959596127129130]34]38141]51]55,]57]68]82]91204:!08215217:!]911JI"',.:s:Contents VIIT5.5 PilelDeck Connection Details. . . . . . . . . . . . . . . . . . . . . . .. 253T5.6 Existing Construction 262TC6: REMEDIATION OF LIQUEFIABLE SOILS 271T6.1 Overview....................................... 273T6.2 Outline of Remedial Measures against Liquefaction . . . . . .. 275T6.3 Compaction Method:Design and Installation 282T6..+ 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 294T6.8 Influence on Existing Structures duringSoil Improvement . 299TC7: ANALYSIS METHODS 307T7.1 Simplified Analysis 309T7.2 Simplified Dynamic Analysis . . . . . . . . . . . . . . . . . . .. 330T7.3 Dynamic Analysis 355T7.4 Input Parameters for Analysis. . . . . . . . . . . . . . . . . . . . . . .. 369TC8: EXAMPLES OF SEISMIC PERFORMANCE EVALUATION '" 379T8.1 Gravity Quay Wall (Grade A) 381T8.2 Sheet Pile Quay Wall (Grade B) 397T8.3 Sheet Pile Quay Wall (Grade S) 420T8.4 Pile-Supported Wharf (Grade B) 423T8.5 Pile-Supported Wharf (Grade S) 432LIST OF SYMBOLS 441REFERENCES 453INDEX..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 469PrefaceAlthough the damaging effects of earthquakes have beenknown forcenturies. itis only since the mid-twentieth century that seismic provisions for port structureshavebeen adopted in design practice. In 1997, the International NavigationAssociation (PIANC; formerly the Permanent International Association forNavigationCongresses) formeda working group, PIANClMarComIWG34, tofocus international attentionon thedevastatingeffects of earthquakeson portfacilities. This book. entitled'Seismic Design Guidelinesfor PortStructures: isthe culmination of the efforts of this workinggroup.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. Althoughconstructedinthe marineenvironment, the port facilities areassociatedwithextensive waterfrontdevelopment, and providemultipleland-sea transport con-nections. The port must accommodate small to very largevessels, aswell as spe-cial facilities for handling potentially hazardous materials and critical emergencyfacilities that must be operational immediately after a devastating earthquake.The primary goal of the working group was the development of a consistent setof seismicdesignguidelinesthat wouldhave broadinternational support. Thediverse characteristics of port structures led the working group to adopt an evolu-tionary design strategy based on seismic response and performance requirements.Provensimplifiedmethods andstate-of-the-art analysis procedures have beencarefully selected and integratedin the guidelinesinorder to provide a flexibleand consistent methodology for the seismicdesign of port facilities.This book consists of a main text and eight technical commentaries. The maintext introduces the reader to basic earthquake engineering concepts and a strategyfor performance-based seismic design. The technical commentaries illustrate spe-cific aspectsof seismicanalysis and design, andprovideexamplesofvariousapplications of the guidelines.The working group members,technical editors, and graphicsexperts are listedbelow. The working group would like to express their sincere gratitude to a groupof NewZealandwharf designers/earthquakeengineeringexperts, includingMr.DickCarter, Prof. BobPark, Dr. RobPark, Mr. Grant Pearce andMr. StuartPalmer, who are notapart of theworkinggroup,fortheir contributions on NewZealand design codeand practice. ManycolleaguesinGreece, Japan and USAX PlANeprovidedconstructive critical reviews of thebook, including Mr. Hisao Oouchi,Prof. Kyriazis Pitilakis, Dr. Craig Taylor, Dr. Tadahiko Yagyu, Dr. ShujiYamamoto, andDr. Hiroshi Yokota. Sincere thanksaredueto themfor sugges-tions that improved the quality of the book. The working group would also like toexpress their sincere appreciation to the Coastal -Development Institute ofTechnology, Japan, for fundingthe publication, to the PlANC regionalsectionsand organizations forsponsoring theworking groupmeetings and activities, andmany 'of themember'sorganizationsfor partiallyfundingactivitiesassociatedwith the development of these guidelines.Theworkinggrouphopesthattherecommendedseismicguidelinesfor portfacilities will make a significant contribution towards mitigating seismic disastersin port areas around the world.December 15, 2000SusumuIaiChairmanPlANCIMarComIWG34Members of PIANCMarComiWorking Group 34Chairman:.Susumu Iai, Port and Harbour Research Institute, JapanSecretaries:Takahiro Sugano, Port and Harbour Research Institute, JapanKoji Ichii, Port and Harbour Research Institute, JapanPrimary Authors:Alberto Bernal, ByA Estudio de Ingenieria, SpainRafael Blazquez, Universidad Politecnica deMadrid, SpainHans F. Burcharth, Aalborg University, DenmarkStephen E. Dickenson, Oregon State University, USAJohn Ferritto, Consulting Engineer, USAW.D. LiamFinn, University of British Columbia, Canada/KagawaUniversity, JapanSusumu Iai, Port and Harbour Research Institute, JapanKoji Ichii, Port and Harbour Research Institute, JapanNason J. Mcflullough, Oregon State University, USAPiet W.H. Meeuwissen, DeltaMarineConsultants bv, Netherlands (fromMay 1998)Constantine D. Memos, National Technical University of Athens, GreeceMJ.N. Priestley, University of California, San Diego, USAFrancesco Silvestri, Universita della Calabria, ItalyArmando L. Simonelli, Consiglio Nazionale delle Ricerche, ItalyR. Scott Steedman, Whitby Bird &Partners Ltd., UKTakahiro Sugano, Port and Harbour Research Institute, JapanContributing Working GroupMembers:Steve J. Bowring, Delta MarineConsultants bv, Netherlands (through May1998)Valery M. Buslov, Han-Padron Associates, USABrad P. Erickson, TranSystems Co., USA.x11 PlANeD. Mandja, Ministere de I'Equipement et de I' Amenagement duTerritoireIDIMA, AlgeriaWerner Pall oks, Bundesanstalt fur Wasserbau, GermanyFilippo Vinale, Universita degli Studi di Napoli Federico II, ItalyMentors:Thorndike Saville, Jr., USA(through Oct. 1998)ThomasH. Wakeman, Port AuthorityofNew York&NewJersey, USA(fromOct. 1998)Technical Editors:Elizabeth Hausler, University of California, Berkeley, USAJohn Stebila, Berkeley, USAGraphics:Yoko Sato, Port and Harbour Research Institute, JapanList of Tables and Figures: Main TextTablesTable 3.1. Acceptablelevel 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 portstructures.Table 4.1. Proposed damage criteria for gravity quay walls.Table 4.2. Proposed damage criteria for sheet pilequay walls (whenan anchor ismoredifficult torestore than a wall).Table 4.3. Proposed damage criteria for pile-supported wharves.Table 4.4. Proposed damage criteria for cellular quaywalls.Table 4.5. Proposed damage criteria for cranes.Table 4.6. Tolerance for ordinarymaintenance of container cranes.Table 5.1. Typesof analysis related with performance grades.Table 5.2. Methods for site response analysis and liquefaction potentialassessment.Table 5.3. Analysis methods for port structures.Table 5.4. List of analysis methods and references.Table 5.5. Major input parametersforanalysis.Table 5.6. Analysis output.Table 5.7. Outputs from dynamicanalysis.FiguresFig.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 fromBea,1997; GSHAP, 1999).Schematic figure of propagation of seismic waves.Peak horizontal bedrock accelerations over various returnperiods.Mechanism of liquefaction.Evidenceofsandboilsduetoliquefactionat HakodatePort, Japan,during the Hokkaido-Nansei-oki earthquake of 1993.Damage toa breakwater at Okushiri Port, Japan, during theHokkaido-Nansei-oki earthquake of 1993.Typical port structures.Fig. 3.1.Fig. 3.2.Fig. 3.3.Fig. 4.1.XIV PlANeFig. 2.7. Damage toacaissonquay wall at KobePort, Japan, during the GreatHanshin earthquake of 1995. .Fig. 2.8. Crosssection of a caisson quay wall at Kobe Port.Fig. 2.9. Damage toa sheet pile quay wall at Ohama Wharf, Akita Port, Japan,during the Nihonkai-chubu earthquake of 1983.Fig. 2.10. Crosssection of a sheet pile quaywallat Ohama Wharf, Akita Port.Fig. 2.11. Damagetoasheet pilequaywall at Shimohama Wharf, AkitaPort,Japan, during the Nihonkai-chubu earthquake of 1983.Fig. 2.12. Cross sectionofasheetpilequay wall at Shimohama Wharf, AkitaPort.Fig. 2.13. Cross sectionof apile-supportedwharfat TakahamaWharf, KobePort, Japan, and damage during the Great Hanshin earthquake of 1995.Fig. 2.14. Damagetopiles at Takahama Wharf, KobePort (afterextraction forinspection.Fig. 2.15. Damageto a crane at KobePort, Japan, duringtheGreat Hanshin. earthquake of 1995.(a) Overview.(b) Close-up.Fig. 2.16. Damagetoacompositebreakwaterat Kobe Port, Japan, duringtheGreatHanshin earthquake of 1995.Fig. 2.17. Damage toarubblemound breakwater at Patras Port, Greece.afteraseries of earthquakes in1984.(a) Typical cross section.(b) Crosssection before andafter the failure.Flowchart forseismic performance evaluation.Schematic figure of performance grades S, A, Band C.Examples of seismic performance evaluation.Deformation/failure modes of gravity quaywall.(a) On firmfoundation.(b) On loose sandyfoundation.Fig. 4.2. Parameters for specifying damage criteria forgravity quay wall.Fig. 4.3. Deformation/failure modes of sheet pile quaywall.(a) Deformation/failure at anchor.(b) Failure at sheet pile wall/tie-rod.(c) Failure atembedment.Fig. 4.4. Parameters for specifying damage criteria forsheet pile quaywall.(a) With respect to displacements.(b) With respect to stresses.Fig. 4.5. Preferred sequence for yieldof sheet pile quaywall.Fig. 4.6. Deformation/failure modes of pile-supported wharf.(a) Deformation due to inertia forceat deck.(b) Deformation due tohorizontalforce fromretaining wall.(c) Deformation due to lateral displacement of loosesubsoil.Fig. 4.8."-Fig. 4.9.Fig. 4.14.Fig. 4.15.Fig. 4.16.LlSlOl l a O l ( ; ~ dllU1 15UI\o , , ~ . .Fig. 4.7. Parameters for specifying 'damage criteria for pile-supported wharf.(a)With respect todisplacements.(b)Withrespect tostresses.Preferred sequence for yielding of pile-supported wharf.Deformation/failure modes of cellular quay wall.(a)On finn foundation.(b) On loose sandy foundation.Fig. 4.10. Deformation/failure modes of cellular quay wall involving crossdeformation.Fig. 4.11. Parameters for specifying damage criteria for cellular quaywall.(a) With respect to displacements.(bjWith respect tostresses.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 todifferential 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)Rubblemound breakwater.Fig. 4.17. Examples of berthing behind breakwater.Fig. 4.7. Parameters for specifying damage criteria for pile-supported wharf.(a) With respect to displacements.(b) With respect to stresses.Fig. 4.8. Preferred sequence foryielding of pile-supported wharf.Fig. 4.9. Deformation/failure modes of cellular quay wall.(a) On firmfoundation.(b) On loose sandy foundation.Fig. 4.10. Deformation/failure modes of cellular quay wall involving crossdeformation.Fig. 4.11. Parameters for specifying damagecriteria for cellular quay wall.(a) With respect to displacements.(b) With respect to stresses.- Fig. 4.12. Schematic figure of gantry crane.Fig. 4.13. Deformation modes of gantry crane.(a) Widening of span between thelegs.(b) Narrowing span between the legs due to rocking motion.(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.(a) Caisson resting onsea bed.(b)'Vertically composite' caisson breakwater.(c)'Horizontally composite' caisson breakwater.(d) Rubble mound breakwater.Fig. 4.17. Examples of berthing behind breakwater.MAINTEXTCHAPTERIIntroductionTheoccurrence of a large earthquakenear amajorcitymaybearare event. butitssocietal andeconomic impact canbe sodevastatingthat it is amatter ofnational interest. The earthquakedisasters inLosAngeles, USA, in 1994(61fatalitiesand-Wbillion USdollarsin losses); Kobe, Japan, in1995(over 6.400fatalities and100billionUSdollarsinlosses); Kocaeli, Turkey, in 1999 (over15,000 fatalities and 20 billion US dollars in losses); Athens, Greece, in 1999 (143fatalities and 2 billion US dollars in losses); and Taiwan in1999 (over 2,300 fatal-ities and 9 billionUS dollars in losses) arerecent examples. Although seismicityvaries regionallyas reflected inFig. 1.1, earthquakedisasters haverepeatedlyoccurred not onlyinthe seismically active regionsintheworldbut alsoin areaswithin low seismicity regions, such as in Zones1or 2 in the figure. Mitigating theoutcome of earthquake disasters is a matter of worldwide interest.Inordertomitigatehazards andlosses due toearthquakes, seismicdesignmethodologies havebeen developed andimplemented in design practice in manyregions since the early twentieth century, often in the formof codes andstandards. Most of these methodologies arebased on a force-balance approach, inwhich structures are designed to resist a prescribed level of seismic forcespecifiedasafractionofgravity. These methodologieshavecontributedtotheacceptableseismicperformanceofport structures, particularlywhenthe earth-quakemotions aremoreor lesswithinthe prescribeddesign level. Earthquakedisasters, however, havecontinuedtooccur. Thesedisasters arecausedeitherby strong earthquakemotions, oftenin thenear fieldof seismic source areas, orby moderate earthquake motions in the regions where the damage due to groundfailureshas not been anticipated or considered in theseismic design.The seismic design guidelines for port structures presented in this book addressthe limitations inherent if) conventional design, and establish theframework for anew design approach. In particular, the guidelines areintended tobe:- performance-based, allowing a certain degree of damage depending onthe specific functions and response characteristics ofa port structure andprobability of earthquake occurrence inthe region;- user-friendly, offeringdesignengineers achoiceof analysis methods, whichrange from simple to sophisticated, for evaluating theseismicperformance ofstructures; and -4 Y1ANCEARTHQUAKEINTENSITYZONEO=O.OO-O.05gZONE1=O.OS-O.15gZONE2=O.1S-0.25gZONE3=O.2S-0.35gZONE4=O.3S-0.45gZONE5=O.4S-0.55g':,SEISMOTECTONIC TYPEA=Shaliow crustal fault zonesB=Deepsubduction zonesC=Mixedshallow crustal faultand deep subduction zonesD=lntraplate zones13NOTE: Values of acceleration corresponding to a return period of475 years.Some areas of low average seismichazard have historically experienced major destructiveearthquakes.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 functionsof port structures, economic and social environment, and seismic activities maydiffer from regionto region.The expected users of the guidelines are design engineers, port authorities, andspecialists inearthquakeengineering. The applicabilityof theguidelines willreflect regional standards of practice. If a region has no seismic codes or standardsfor designing port structures, theguidelinesmay be usedas a basisto .developeanew seismic designmethodology, or codes applicabletothat particular region. Ifa regionhasalready developed seismiccodes, standards, or establisheddesignpractice, then the guidelines maybe used to supplement these designand analy-sis procedures (see TechnicalCommentaryI for existing codes and guidelines). Itis not the intent of the authors to claim that these guidelines should be used insteadof the existingcodesorstandardsor establisheddesignpracticein the region ofIntroduction 5interest. It is anticipated, however. that the guidelines will,with continual modifi-cationand upgrading, berecognizedasanewanduseful basisfor mitigatingseismicdisasters inport areas. It ishopedthat theguidelines mayeventuallybe accepted worldwide as recommended seismic designprovisions.Earthquake engineering demands background knowledge in several disci-plines. Althoughthis background knowledgeis not a pre-requisite to understand-ing the guidelines, readers may findit useful to have reference textbooks readilyavailable. Pertinent examples includeKramer (1996)on geotechnical earthquakeengineering and Tsinker (1997)ondesign practice for port structures.ThisMain Text provides an overview of theseismicdesignguidelines. More.details inthe particular aspectsof the seismicdesignguidelines canbe found inthe following Technical Commentaries (TC):TCI:Existing Codes and GuidelinesTC2: Case HistoriesTC3: Earthquake MotionTC4: Geotechnical CharacterisationTC5: Structural Design Aspects of Pile-Deck SystemsTC6: Remediation of Liquefiable SoilsTC7: Analysis MethodsTC8: Examples of Seismic Performance EvaluationsCHAPTER 2Earthquakes and Port StructuresThischapteraddressesissues fundamental tounderstandingseismiceffects onportstructures. As illustratedinFig. 2.1, seismic wavesare generatedalongacrustal fault andthey propagate throughupper crustal rock, travelling tothe sur-faceof thebedrockat a siteof interest. Thegroundmotions thenpropagatethrough the localsoil deposits, reaching the ground surface and impacting struc-tures. Dependingon theintensityof shakingandsoil conditions, liquefaction ofnear-surfacesoilsand associatedgroundfailures mayoccurandcouldsignifi-cantlyaffect the portstructures. If anoffshore fault motion .involves verticaltectonic displacement of the seabed, tsunamis maybe generated. The engineeringaspectsofthesephenomenaarecollectivelyimportant intheevaluationof theseismic effects on port structures.2.1 EARTHQUAKE MOTION(l) Bedrock MotionThe bedrock motions used for seismicanalysis and design at a particular site arecharacterized through seismic hazard analysis. If a specific earthquake scenario isassumed in theseismic hazard analysis, the bedrock motion is defineddetennin-istically based on the earthquake sourceparameters and wave propagation effectsalong the source-to-site path. Most often, however, the bedrock motion is definedprobabilistically through theseismichazardanalysis,takinginto accountuncer-tainties in frequency of occurrence and location of earthquakes.Oneof thekeyparametersin engineeringdesignpracticeistheintensityofbedrock motiondefined in terms of peakgroundacceleration (PGA), or in somecasespeak ground velocity(PGV). This parameter is usedeither byitself or toscale relevant ground motion characteristics, including response spectra andtimehistories. In the probabilistic seismic hazard analysis, the level of bedrock motionis defined as a function of a returnperiod, or a probability of exceedance over aprescribedexposuretime.Anexampleis shown- inFig. 2.2, inwhichPGAs atbedrock are obtainedbased on geologic, tectonicandhistorical seismicactivity8 PlANe }SOil Layers/. BedrockFig. 2.1. Schematic figureof propagation ofseismic waves.Fig. 2.2. Peakhorizontal bedrockaccelerations over van-OllSreturn periods.1000500Return Period (years)SanFrancisCO//.

.::t.cg 0.0 o 100Oi 1.0-r----------------,-co;;cg 0.8Q)oo dJHe, whereJ..4,: coefficient of friction; d\: expansion of leg span (m), He: height of the hingefrom the apron (m) (refer to Fig. 4.15).4)Damage Degree IV of the quay wall should be the state beyond the upper limitof the damage Degree III.A set of tolerancesfor ordinary maintenance of cranes is shown in Table 4.6(Japan Cargo Handling Mechanization Association, 1996). Thisset of tolerancesFig. 4.15. Parametersforevaluationoverturning of crane.N: Axial load from a crane legTable 4.6. Tolerance for ordinary maintenance of container cranes.Parameters ToleranceRail SpanLspan ( ~ < 25 m)(25 m~ u: ~ 40m)Leveldifference between seaand land siderailsCurving in vertical directionCurving in horizontal directionInclinationRail joint Differential displacements (vertical and horizontal)Gap*Relative to original layout.10mm15mm~ 1 0 0 05mrn per 10m5 mrn per 10m1/500Imm5mm*52 PlANecouldbe used asoneofthereferences to establishmorerestrictivedamagecriteria than discussed earlier.4.6 BREAKWATERSA breakwater is usuallymade of a rubble mound, a massive structure such as acaisson, 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 bottomof the caisson, -and with associated resistanceto overturning and bearing capacityfailure. Typical failure modes expected during earthquakes are shown in Fig. 4.16.Breakwaters are generally designed to limit wave penetration and waveovertopping during specific design storms, and at the same time designed to resistthe relatedwave 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 andan earthquake can be treated astwoindependent loadsituations. Onlywaveactionsfromamoderatesea stateshould beconsidered together withthe designearthquakes. Decision on this seastate has to be made based on the site-specific long-term statistics of the storm.Selectionof theappropriatedesigncriteriadepends onthe functions ofthebreakwaterandthe typeof earthquake-inducedfailuremodes. However, for allbreakwaters, themain criterionis the allowable settlement ofthe crest levelbecauseit determines the amount ofovertoppingandwavetransmission. Forbreakwaterscarrying roads and installations, additional criteria for allowable dif-ferentialsettlement, tilting and displacement of superstructures and caissons areneeded.Shaking ofthebreakwater may cause breakageof 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.ZwambomandPhelp, 1995). Thesamecriteriamaybeadoptedfor theearth-quake-related damage.Sincethe damagecriteriaforbreakwatershavenot beenfullydevelopedatpresent, the performance grade is being described in order to indicate the relativedegree 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 dependingon the level of acceptable human life safety);- provisionof berthingonthe port side of thebreakwater, and relatedaccessroads (Grade B) (see Fig. 4.17);- provisionof cargohandlingfacilitiesonthebreakwater, including conveyorbelts (Grade B), andpipelines for oil and liquid gas (Grade A or S. dependingon the level of threatof explosion).Damage Criteria 53-Crest lowering due to settlement ofsubsoil.Oifferentialsettlementandtilt ofcaissons.Possible damage tocaissonshearkeys Slip failures andsubsequentcrest loweringdue to liquefactionsubsoil-. Differential settlementandtilt ofcaissons Damage111 caisson shear keys iftheyexist(a)Rubble Mound Crest lowering due to shakedawn of rubble foundationDifferential settlement ofcaissonsCaisson.Crest lowering andlateral spreading due to settlemenl/liquefactionof subsoil Differential settlementofcaissons-Slip failures and subsequentcrest loweringdue to liquefaction ofsubsoil Differential tilt andsettlement ofcaissons Damage111 caiSSOll shearkeysif they exist(b)11-----c..sscn III of slender types ofconcrete armour units.Subsequent increasedovertoppinlt and risk ofdisplacement ofcaisson(c) .. "'. ......::::{'t:.::;-===- 2 ;/ ., ',RUllllleMound "-- ------------- ----- Crest loweringdue toshake downof rubblematerial Differenrial settlement of superstructure elements'5?

-Crest loweringandlateralspreadingdue 111setdemenl/liquefaction of subsoil-Differential settlement of superstructure elements Failures due to liquefaction ofsubsoil.Subsequent lowering ofcrest-Possibletilt anddisplacement ofsuperstructure elements(d)Fig. 4.16. Deformation/failure modes of breakwater.(a) Caisson resting on sea bed.(b) 'Vertically composite'caisson breakwater.(c) 'Horizontallycomposite' caisson breakwater.(d) Rubble mound breakwater.Damage Criteria 54Fig. 4.17. Examplesof berthingbehind breakwater.CHAPTER 5Seismic AnalysisSeismicanalysis ofport structures isaccomplishedinthreestepsthat includeassessment of the regional seismicity, the geotechnical hazards, and soil-structureinteractionanalysis. The first step is todefine theearthquakemotions at thebedrock (see Fig. 2.1for the relativelocation of the bedrock with respect to theseismic source, groundsurface andport structures). This is typicallyaccom-plished by seismic hazard analysis based on geologic, tectonic and historical seis-micitydata available for theregionofinterest. Thesecondstepinvolves thefollowingtwo interrelated aspects of dynamicsoil response: (1) an evaluation oflocal site effects for obtainingthe earthquake motions at or near the groundsurface; and(2) an assessment of theliquefactionresistance of thenear surfacesandysoils andtheassociatedpotential for groundfailures. Once the groundmotion and geotechnical parameters have been established, then seismic analysisof 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 oftheresults of the analysisprocedures.Thisis particularlyimportant inseismicdesign, given the multidisciplinary input that is required for these evaiuations, andtheinfluenceofthis input onthe final designrecommendations. Thischapterprovides general recommendations for the type and level of analysisrequired forvarious port structures.5.1 TYPES OF ANALYSISThe objective of analysisin performance-based designistoevaluatetheseismicresponse of the port structure with respect to allowable limits (e.g. dis-placement, stress, ductility/strain). Higher capabilityin analysis is generallyrequiredfora higherperformancegradefacility.Theselected analysismethodsshould reflect the analytical capability required in the seismic performanceevaluation.A variety of analysis methods are available for evaluating the localsite effects,liquefactionpotential and the seismic response of port structures. These analysis56 PlANemethods are broadly categorized based on a level of sophistication and capabilityas follows:1) Simplifiedanalysis: Appropriateforevaluating approximatethresholdlimitfor displacements and/or elastic response limit and an order-of-magnitude esti-mate for permanent due to seismic loading. .2) Simplifieddynamic analysis: Possible to evaluate extent of displacementlstress/ductility/strain based on assumed failure modes.3) Dynamicanalysis: Possible to evaluate bothmodesandthe extent ofthe displacementlstress/ductility/strain.Table5.1showsthetypeof analysisthat maybemost appropriate for eachperformancegrade. Theprinciple appliedhereisthat thestructuresofhigherperformancegradeshouldbeevaluatedusingmoresophisticatedmethods.Asshown intheindex inTable 5.1, lesssophisticated methods maybeallowed forscreeniJ}gor...Wi\excitation.In thepresentstateof practice, it is desirableto confirmtheapplicabilityofmethodsfor analysisof port structuresby using suitable casehistoriesor modeltest results before accomplishing the seismic performance evaluation.5.2 SITE RESPONSEILIQUEFACTION ANALYSISSite response analysis and liquefaction potential assessments are typically accom-plished using the methods outlinedin Table 5.2. Useful references that highlighttheapplicabilityofthe various modeling procedures, have beenpreparedbyseveral investigators (e.g.. Finn, 1988; Idriss, 1991).(l) Site response analysisIn simplified analysis, local site effects are evaluated based on the thickness of thedepositsandthe averagestiffness toa specified depth (generally30 m), or overthe entiredeposit abovethe bedrock. Thisinformationisthenusedtoestablishthe site classification, leadingto the use of specifiedsite amplificationfactors orsite dependent response spectra. This type of procedure is common in codes andstandards.In simplified dynamic analysis. local site effects are evaluated numerically withmodelssuchas commonequivalentlinear. total stressformulations. Soil layersare idealizedas horizontallayersof infinitelateral extent (i.e. one-dimensional(ID. Thesemethods areusedtogeneratetimehistoriesof acceleration, shearstress, andshear strainat specified locations in the soil profile.In bothof thesecategoriesofanalyses, thecomputedgroundsurfaceearth-quakemotionparametersare usedasinput for subsequentsimplifiedstructuralanalysisasdiscussedin Section 5.3.Table 5.1. Types of analysis related to performance grades.Type of analysis Performance gradeGrade C Grade B Grade A Grade S'Simplified analysis:Appropriate for evaluating approximate threshold level and/or elastic limitand order-of-magnitude displacementsSimplified dynamic analysis:Of broader scope and more reliable. Possible to evaluate extent ofdisplacement/stress/ductility/strain bused on assumed failure modesDynamic analysis:Most sophisricated. Possible to evaluate both failure modes and extent ordisplacement/stress/ductiIity/strainIndex: Standard/final design Preliminary design or low level of excitations.IINU ..,!t.)e""; .'N. . .. :.;. .. ,." ' ",' "." '. .,: :''l!. Li1I..... ,.,,, .1 ..L.cJl ..'.; 1",/.1..:11: t. .\..14 i ....C/J(\I_.'Jl:;jo'

::Jeo'- What factor is adoptedfor specifying the effective seismiccoefficientas a fraction of peak ground acceleration for retaining structuredesign?>- What types ofresponsespectraare usedfor pile-supportedstructuresindesign?An overview of the existing codesand guidelinesreviewedwith respectto theaforementionedaspectsisshown in Table T1.1. Additional specifics. tobackupthe descriptions in Table T1.1, are presented below.(1) Japanese design(Port andHarbourResearchInstitute, 1997: MinistryofTransport, Japan, 1999).Thedual level approachhasbeen adoptedfor structures ofSpecial Class ofImportance. However, the single level approach is adopted for structures of ClassA, 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 lossof serviceability.For L2:Minor orlittle damagellittle or short-term loss of serviceability.rn.i )Existing Codes and Guidelines 81For retaining structures of Special Class of Importance. the criteria for struc-tural damageand criteria regarding serviceability are shownin Tables Tl.2 andT1.3.The seismic coefficient for usein retaining structuresis defined as follows forSpecial Class structures.tk: - Q max( O? )h - -- a max:S; ._gg( )/ 3I am:tllkh =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 byEqn. (T1.1).For a pile-supported wharf with vertical piles, analysisis performed based ona simplified procedureand pushover method. The ductility limitsforuse in thesimplifiedprocedure(see TechnicalCommentary7) for Ll earthquakemotionare specified as shown in Table TIA. Pushover analysis is performed for SpecialClass structuresandthe strain limitsprescribed are:Level I motion: equivalent elastic,Level2 motion: Cmax =O.44tp/Dp for theembedded portion.Pile-supported wharves with.vertical steel piles are designed using theresponse spectra shown in Fig. T1.1 for L1 motion. These spectra have been com-puted based on 2Dsoil-structureinteractionanalysisfor typical pile-supportedwharf cross sections. For L2 motion, time history analysis shouldbe performedand theresults should meet theductility limits for L2 earthquake motion shownin Table T1.5.Comprehensive guidelinesareshown onliquefactionpotential assessment andimplementation of remedial measures (Port and Harbour Research Institute, 1997).(2) Spanish design(ROMO.6, 2000)Dual earthquake levelsmust be considered in the design, but only Level 2 earth-quake motion is considered for low seismicdesign acceleration.Analysis proceduresspecifiedin ROMrangesfromthesimpler methodsforthe lower design accelerationtothemoresophisticatedmethodsfor the higherdesign 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: structuralbehaviour.Pile-supportedwharves/pierswithvertical steel pilesare designedusing theductility factors given as follows:Wharves: Pile cap:J1 = up to 5.0 Embedded portion: J1 = 2.0Piers: Pile cap:J1 =up to 4.0 Embedded portion: J1 =1.5Table '1'1./. Overview ofexisting codesandguidelines for port structures.Japanese Dual level for For Special Class:(1997, 1999) Special Class No damage for LI;Single level for Minor damage andClass A, B, C lillIe or short-termloss of serviceabilityfor L2Codes/ McthodologyGuidelines S' I I I, IIIgC (II'uualevel approachDamage criteria00NAnalytical procedure Remarks'i:l.....The factor afor Response spectrazspecifying the(Jeffective seismiccoefficient as afraction of anlulgSee Eqn. (TI.I) LI: See Fig. Tl. I These codesL2: Time history andanalysis [no guidelinesresponse spectra have beenspecified] specificallyestablishedfor designinga=0.7 See Eqn. (Tl.2)I port structuresDuctility/strainlimit for pile-supportedwharf/pierSimplifiedprocedure usesdisplacementductility factors.Pushoveranalysis usesstrain limitDuctility factorsfor pile headand embeddedportions.Different forwharf and pierDisplacementfor retainingstructuresNut SpecifiedSee TablesTI.2& TI.3Performancelevel(Acceptabledamage level)Little or short-termloss of serviceabilityfor L I: No collapsefor L2Dual levelapproach:Seismicinputin each leveldepends on theimportanceand life timeof structureSpanish(2000)German(1t)9(j)Single level Not specifiedapproach.More detailedanalysisisrequired ifdamage canendangerhumane liveNot specified Not specified Not specified;a to be takenfrom nationalcodeNot specifieduUS DUlll Method Serviceable Soil limits Ductility limits N/A Linear orS Navy LI 75 year under LI and specified for specified for NonlinearA(.1997) L2 500 year repairable wharf dikes piles DynamicEvents under L2 AnalysisASCE- Dual level Critical structures: Site-specific Ductility limits See Eqn. (T1.1) a) responseTCLEE approach: Serviceable under based on specified for spectrum( 1998) exposure time LI and repairable serviceability piles methodsand ground under L2 and effect on b) time historymotion are Other structures: adjacent analysisbased on the Repairable for LI structures c) pushoverimportance. analysisand design lifeCalifornia Dual Method Serviceable Soil Strain limits for N/A A) Single ModeMarine Ll 75 year under LI and deformation piles under 8) Multi-modeOil L2 500 year repairable under L2 limits for soil limit conditions C) PushoverTerminals Events structure D) Inelastic(2000) interaction Time HistoryEurocode Dual level No reference to Not specified Not specified. Design ground See Eqn. (TI.3) These codestT1( 1(94) approach Port Structures A guidance is acceleration refer to xVi'given for the fraction for buildings...._.::;)design of { l J c s i ~ nand otherOQplastic hinging is specified civil()0engineenng0-New Zealand Dual method The Ultimate Limit Not specitied Both are Design Spectra Based on seismicI1lworks withtil(1992-19(7) state peak response specified are used zone, subsoilno direct15spectra is given condition,0-referenceC)structure periodc10 port_.and ductility0-I1lstructures-factor5'I1ltil00VJDeeper than 7.5 m84 PlANCTable Tl.2. Structural damage criteria in Japanese standard.Type of retaining walls Water depth------------------Shallower than 7.5mHorizontal displacement Horizontal displacementoto 0.2m 0 to 0.3moto 0.3m 0.3 to 0.5 m_ Gravity quay wall No repair needed for operation Partial operationallowedSheet pile quay wall No repair needed for operation Partial operation allowedHorizontal displacementoto 0.2 m0.2 to 0.5 mHorizontal displacementoto 0.3 m0.3to1.0 mTable T1.3. Serviceability criteria in Japanese ports.Main body Upper limit of settlementof retaining Upper limit of tiltingstructure Upper limit of differential horizontal displacementApron Upper limit of differential settlement on apronUpper limit of differential settlement betweenapron and backgroundUpper limit of tilting0.2 to 0.3 m3-500.2 to 0.3 m0.3 to 1.0m0.3 to 0.7 m3-5% [towards sea]5%[towards land)Table TIA. Ductility factor for Ll motion in the simplified analysis.Special Class J.l = 1.0Class A u = 1.3Class B u = 1.6Class C J.l= 2.3Design response spectra are specified as follows:For T =0 to TA: SAladesign =1 + 1.5 TITAFor T =TA to TB: SA..!adesign =2.5For T> TB: SAladesign = KCITwhereadesign = 0.04 to 0.26g,TA =KC/lO, TB =KC/2.5K=1.0 to 1.5 (depends on type of seismicity)C=1.0 to 2.0 (dependson ground conditions)(T1.2)(3) German design (EAU, 1996)TheGermanrecommendations forquaywallsdescribea single level approachwith a conventional seismic earth pressure based onthe Mononobe-Okabe equa-tion (Mononobe, 1924; Okabe, 1924). Pressures are assumedto increase linearlywith depth, although it is accepted[R125] that tests have shown this not to be thecase. The soil-water system is assumed to move as one body. and no water hydro-dynamic suction in frontof the wall is taken into account. The requirements foraccuracy of the calculations are correspondingly more stringent when earthquake ._._ .... 1 _."'- _ , .'ExistingCodes and Guidelines 85Region A1.00Region B1.0010 1.0Natural Period (s)/.\ I I: ,6-,I ",'@ i;~! IID ~, ! ~,, IIIIl!Ii :, IiII! :CQl~o0.10,!olEIII'ijjen0.0110 0.1 1.0Natural Period (5).,2': I I ' ~ I, ,(3)I :I1 iI~ ~I'~II j , ,I I III1,0.00.1C- no damage or serviceable under L1 motion;>- repairable or short-term loss of serviceability under L2motion;- damagecriteriatospecifylimit conditions inengineeringterms, includingwall displacement andpile ductilitydemand, have beenunder developmentand further studiesmay be needed to establish them;- analyticalproceduresadoptedfordesign, includingthefactorforspecifyingthe 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 designpractice maybe 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 PlANeIt must be emphasized that there is a significant difference between the designapproach recommended in these guidelines and the conventional design concept.In the conventionaldesign concept, especially when referring tothesimplifiedanalysis, an equivalent seismic coefficient isusedas aninput parameter repre-senting adequatelytheensembleof groundmotions, andafactor of safetyisapplied to detemiine the dimensionsand" properties of the 'structure.In the pro-posed approach, the design is based on the seismic performance of the structureevaluated appropriatelythroughresponseanalysis for avarietyofinput earth-quake motions. The ensemble of the seismic responses, rather than the ensembleof input motions, is used as a basis for accomplishing thedesign in the proposedguidelines. Foreachresponseanalysis, input parametersmost appropriatearethose well defined in terms of appliedmechanics, such asa peak ground accel-eration for the simplified analysis and/or an equivalent parameter clearly definedin terms of peak ground acceleration. Consequently, no factor of safety should beapplied to input data used in seismic analysis for evaluating the threshold level ofthe structure. This important distinctionbetweenconventional designand per-formance-based design shouldalso beborne inmind when interpreting thedesign guidelines included in the various seismiccodes.Te2: Case HistoriesT2.1T1.2T2.3T2.4T') -_.)T2.6Damageto GravityQuay WallsDamage to SheetPileQuay WallsDamage to Pile-Supported WharvesDamage to Cellular Quay WallsDamageto CranesDamageto BreakwatersTECHNICAL COMMENTARY 2Case HistoriesSelection 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-bers1 through 29 shownin this tablemay be usedas indexes in referring to therelevant case histories shown in Tables T2.2 through T2.30 in this technical com-....mentary. These case histories illustratevarious seismic effects on port structures.T2.1 DAMAGE TO GRAVITY QUAY WALLSAs discussedinSection4.1in the Main Text, agravity quaywall is made of acaissonor other gravityretainingstructureplacedonseabed. Stabilityagainstearthpressures from thebackfillsoilbehind thewallis maintained by the massof the wall and friction at the bottom of the wall. For gravity quay walls on firmfoundations, typical failure modes during earthquakes are seaward displacementsand tilting (Tables T2.3 through T2.5). For a loose backfill or natural loose sandyfoundation, the failure modes involve overall deformationofthefoundationbeneath the wall, resulting in large seaward displacement, tilting and settlements(TableT2.7). Awall with arelatively small widthtoheight ratio, typicallyless thanabout 0.75, exhibits apredominant tiltingfailure moderather thanhorizontaldisplacements, resulting, inthe extremecase, inthecollapseof thewall (Table T2.2).Theevidenceof damagetogravityquaywalls (Tables T2.2throughT2.9)suggests that:I) most damage to gravity quay walls is often associated with significant defor-mation of asoft orliquefiablesoildeposit, and, hence, if liquefactionis anissue,implementingappropriateremediationmeasuresagainst liquefactionmaybe "aneffective approach toattainingsignificantlybetterseismicper-formance;2) most failures of gravity quay walls in practice result from excessive deforma-tions, not catastrophiccollapses, and, therefore, designmethods basedondisplacements and ultimate stress states are desirable for defining thecomprehensive seismic performance; and94 PlANe3) overturning/collapse of concrete block type walls could occur whentilting isexcessive, andthis typeofwall needs careful considerationinspecifyingdamage criteria regarding the overturning/collapse mode.T2.2 DAMAGE TO SHEET PILE QUAY WALLSAs discussed in Section 4.2 in the Main Text, a sheet pile quay wall is composedof interlockingsheetpiles, tie-rods, andanchors. Thewall issupportedat theupperpart byanchors andthelowerpart by embedment infoundation soil.Typical failure modes during earthquakes depend on structural andgeotechnicalconditions, as shown in Tables TI.IO through T2.15. The evidence of damage tosheet pile quay walls suggests that:1) most damage to sheet pile quay walls is often associated with significant earthpressures fromasoft orliquefiable backfill soil (TablesT2.10andT2.12),and, hence, if liquefaction isan issue,implementing appropriate remediationmeasures against liquefaction may be an effective approach to attainingsignificantly better seismic performance (Tables T2.11and T2.13);2) if the anchor isembedded in competent soil, most failures of sheetpile quaywallsoccuratthesheet pilewall duetoexcessive bending moment(TablesT2.10 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 ofthe sheet pile wall and anchor system; and3) structural damage toa sheet pile quay wall is governed by stress states ratherthan displacements. It is therefore important to determine the preferredsequence and degrees of ultimate states that may occur in the composite sheetpile quay wallsystem.T2.3 DAMAGE TO PILE-SUPPORTED WHARVESAs discussed in Section-l.S in the Main Text, a pile-supported wharf is composedof a decksupportedbya substructure consisting of pilesanda rockdike/slope.The unsupported pile length above the dike/slope surface is variable. When rock-fill suitable for construction of thedike is uneconomical. agravityor sheet pileretaining structure is also constructed to replace a portion of the dike. Theseismic responseof pile-supportedwharves is influencedtobyinteraCliQ!l.Q!!ringgr91!D.c!Typical failuremodesduring earthquakesdependonthemagnitudeof theinertia forcerelativeto thegrounddisplacement. Inparticular. thefailure associatedwiththegrounddis-placementshown in Tables T2.16. T2.] 7, TI.20, and T2.22 suggeststhat specificdesign consideration is required in the geotechnical aspect of seismicdesign.Case Histories 95Amongvarious structuralcomponentsinthepile-supportedwharves. specialattention should bepaid to the application of batter piles. Batter piles remain themost efficient structural component for resistinglateral loads duetomooring.berthingand craneoperation. However, thebatter pile-decksystem resultsinamuchmore rigidframethanone with vertical piles. Asa result. large stress con-centrations andshear failuresof concrete batter piles have been observed duringearthquakes(Tables T2.16and T2.18). Thismodeof pilefailuredemonstratesthat thestructural design of batter piles used inregionsof high seismicity mustaccount 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 preferredsequence and degrees of ultimate statesto occur in thecomposite pile-supportedwharf system. Most pilefailuresareassociatedwithliquefactionof soil whichcan result in buckling of thepile, loss of pile friction capacity, or development ofpile cracking and hinging. Hinging may be at the cap location or at aninterface between soil layers of differing lateral stiffness (Table T2.20). Damagetopiles in thesoil mayalsooccurabout Ito3pilediametersbelowgradeinnon-liquefyingsoilsdueto thedynamicresponseof thewharf(seeTechnicalCommentary 5).T2.4 DAMAGE TO CELLULAR QUAY WALLSAsdiscussedin Section 4.4in the Main Text, a cellular quaywall is made of asteel plateor sheet pilecell withsandorother fill. Resistanceagainst inertiaforces and earth pressures isprovided by the fill friction at the bottom surface ofthecell for a non-embedded cell, or bytheresistance of the foundation subsoil atthe cell embedment.Typical failuremodesduringearthquakesdependoncellembedment and geotechnical conditions, resulting in varying degrees of seawarddisplacement, settlement and tilting.Inaddition tothegross deformation, warp-ingof the cellular cross section canoccur (Table T2.23). Suchperformanceofcellular quaywalls should beconsidered in the seismic design criteria.T2.5 DAMAGE TO CRANESAsdiscussed in Section 4.5in the Main Text, a crane consists of an upper struc-ture ~ o r handlingcargoes, anda supportingstructurefor holdinginplaceandtransportingtheupper structure. Thecraneisgenerallymadeof asteel frame.Thesupportingstructureiseither arigidframe typeorahingedlegtype,andrests on railsthrough thewheels.96 PlANeA craneat restis fixedtorails or toa quaywall withclamps or anchors. Thestrengths of theseclampsoranchorsprovidestheupperlimitresistanceagainstexternal forces acting on the crane. Acrane in operation, however, is not supportedwith clamps or anchors. The lateral resistance of the craneagainst external forcesduring operation is only due to friction and resistance fromthe wheel flanges. ..Typical failuremodesduring. earthquakesarederailment of wheels, detach-ment orpull-out of thevehicle, ruptureof clampsandanchors, buckling, andoverturning. Widening of a span between thelegs duetothe deformation of thequay wall results in derailment or buckling of the legs (Table T2.26). Conversely,narrowing of a leg span can also occur due tothe rocking response of thecrane(Table T2.24). This is due to an alternating action of the horizontal component ofresistingforces fromthequaywall duringarockingtyperesponseinvolvingupliftingof onesideofthelegs. Therockingresponsecanalsoresult inthederailment, leading totilting. If the tilting becomesexcessively large, overturn-ing of the crane mayresult (Tables T2.25and T2.27). The overturning typicallyoccurs toward land because of the location of thecenter of gravity of thecranesat rest, anddifferential settlement on thecranefoundationassociatedwiththedeformation of quaywalls.Crane rails are often directly supported either by a portion of a retaining wall,orbythedeck of a pile-supportedwharf. Whenthewidth of thegravitywall issmall, or the quay wall is a sheet pile or cellular type, a separate foundation oftenconsisting of piles is provided to support therails. In order toachievedesirableseismic performance of quay walls with cranes, special consideration is requiredfor the rail foundation. Onesuch considerationcould includeprovidinga dedi-cated and cross-tied upper structure tosupport therails.T2.6 DAMAGE TO BREAKWATERSAsdiscussedinSection 4.6 in the Main Text. a breakwater isusuallymadeof a. Jrubble mound, a caisson, or a combination of both placed onthe seabed.Stabilityagainst a horizontal external load is maintained byshear resistance of the rubble.friction at the bottom of the caisson, andwithassociatedresistance tooverturn-ingandbearingcapacityfailure. Typical failuremodes duringearthquakesaresettlement associated with significant foundation deformation beneath thecaisson/rubble foundation(Table T2.28). It should be noted that even withmoderate earthquakes, failure can happen due to extremely soft clay deposits andlarge seismic amplificationwithin andbeneath therubblemound (Table T2.29).Damage to breakwaters froma tsunami typically resultsinthe caissons beingwashedoffortiltingintotherubblemound(TableT2.30). If the performanceobjective of a breakwater includes preventive measures against a tsunami. appro-priate designcriteria needtobe establishedwithrespect toboth the earthquakemotion andtsunamis.Case Histories 97Table T2.1. Selected casehistories of damagetoport structures (1980-1999).No. Port structure Earthquake Portyear (magnitude")1 Gravity Block 1985(A,t =7.8) San AntonioPort Chile")quay 1986(.H\ =6.2) Kalarnata Port Greece3 wall 1989 (M =6.0) Port of Algiers Alseriac4 Caisson 1993UII!l =7.8) Kushiro Port Japan5 1993 (Ml=7.8) Kushiro Port Japan6 1995 (Ml=7.2) KobePort Japan.7 Block 1999 (Mw=7.4) DerincePort Turkey8 Caisson 1999 (Ms=7.7) Taichung Port Taiwan9 Sheet pile quay wall 1983 (Ml=7.7) Akita Port Japan10 1983 (MI=7.7) Akita Port Japan11 1993 (Ml=7.8) Kushiro Port Japan12 1993 (Ml=7.8) Kushiro Port Japan13 1993 (Ml=7.8) Hakodate Port Japan14 1993(Ms=8.1) Commercial Port, GuamUSA15 Pile- RCpile**1989(.Ws = 7.1) Port of Oakland USA16 supported 1990(Ms=7.8) Port of San Fernando Philippines17 wharf PCpile** 1994(Ms = 6.8) Port of LosAngeles USA18 1994 (Ms=6.8) _Port of Los Angeles USA19 Steel pile 1995 (Ml =7.2) Kobe Port Japan20 PCpile 1995(Ms = 7.2) Port of Eilat Israel21 Dolphin Steel pile 1995 (lV!l =7.2) Kobe. Sumiyoshihama Japan22 Cellular quay wall 1995 (Ml=7.2) KobePort Japan23 Crane 1983 (Ml =7.7) Akita Port Japan24 . 1985(Ms=7.8) San Antonio Port Chile25 1995 (Ml =7.2) KobePort Japan26 1999 (Mw =7.4) Derince Port Turkey27 Break- Composite 1995 (Ml =7.2) Kobe Port Japan28 water Rubble 1984(M =4.5) Patras Port Greece29 Composite 1993*** (M, =7.8) Okushiri Port Japan* For definition of magnitude. refer to Technical Commentary 3.**RC: Reinforced concrete; PC: Prestressed concrete.***Damage dueto Tsunami.98 PlANeTable T2.2. Case history No.1: San Antonio Port, Chile. (Ms =-7.8)Structural Structure and location: Berths1 &2and seismic Height: + 4.9 mconditions Water depth: - 9.6 mConstruction (completed): 1918-1935Earthquake: March 3, 1985 Chile earthquakeMagnitude: M, =7.8Peak ground surface acceleration (PGA): 0.67 gPGA evaluated basedon strongmotion earthquake records.Cross section: Block quay wall6.0'9'41815Scal. level I, T1m) 1m) Soiltype ISPNvalue.47ft t, or. >n,n4n .. no - Sandand I"':1 gravel fill :,!:2 2.S! : : : i :3 1.'0 Coarse sane ;::.:::: : ,If' I5 Finesand :,:, 1 '.: i, ;6 '.20 1'9: :Mediumto ::: Icoarse sane ' , 8 3.65 r:::9 '1iClay 10",10 I: :l':::12 .7.50 witnsilly1m... 5: ,'o?F'lne sane I13 8.65 "" h stone ::: "PO''l II.... I. .I I I Fmeto :: :mediumsand iii: :16 -12.00 : : I ';'0'17 I II : I :Finesand ' .,, : : ,pO;(Unit: m)Hydraulic fill9.03.84 1.0I..0.161.0oN-+4.9';7 000-9.6Concrete Backfillblocks --+I..r-.-..........,---t (5-100kg stoneswith gravel)+1.68MS.l +0.9_ ...... Boring log at the backfillFeatures of damagelcommentsCollapse of block typequay wall over 60% of wharf length(452 m) due to strong earth-quake motion and backfillliquefaction.References: Tsuchidaet al. (1986): Wyllie et al. (1986)StructuralandseismicconditionsCase Histories 99Table T2.3. Case history No.2:Kalamata Port. Greece.Height: + 2.1mWater depth:- 9.5mEarthquake: September13, 1986Kalarnata earthquakeMagnitude: M, = 6.2Peak groundsurface acceleration (PGA): 0.2 to 0.3 gPeak groundsurfacevelocity (PGV): 0.4 mlsPGA/PGV evaluated based on1D & 2Dequivalent linear analysis.Seismic Horizontal displacement: 0.15 0.05mperformance Tilt: 4 to 5 degreesCross section: Block quay wall000--------- .... -_... -....... ------_.. ----3.0m~SG(Unit:m)SPTN-value10 20 30 40 50FillIndex: SM=silty sand, SG=gravely sandFeatures of damage/commentsSeaward displacement withtilt of block typequaywallconstructed on firmfoundation.The backfill settled 20 cmbehindthewall withnosettlement atadistance of 30 to' 40meters away. The quay wall remained serviceable.Reference:Pitilakis and Moutsakis (1989)100 PIANCTable T2.4. Case history No.3: Port of Algiers, Algeria.. Structural Structure and'location: QuayNo. 34and seismic Height: + 2.2 mconditions Water depth: - 10.0 mEarthquake: October 29, 1989Chenoua earthquakeMagnitude: M =6.0, Epicenter located 60 kmto the west of AlgiersSeismic Horizontal displacement: 0.5 mperformance Vertical displacement: 0.3 mCrosssection:Block quay wall-10.00sz L.W.L. 0.00""F"o"- interplate> intraplateInterplate earthquakes originate in the great stress accumulated in contactsbetween tectonicplates (theprincipal fragments intowhichtheearthcrust isdivided). Mostdestructive earthquakes, with magnitude above 7 to 8. belong toEarthquakeMotion 1350.5.,.--------------------------.,------,..L.-villi"IJ '" .. I'Ci 0.3Cg0.1to.. -0.1 +-:..I.a0.4~ l - - - - __-..J_0~::l< 0.24 6.5r.20.8,..--------,-------,Stress-Strain Curve0.60.4o 0.2bE......>-....>< 0.20.40.6t1AZ1/fliWIVVlrVf

6.5(d)0.8 Computedfor0.6 Fuji river sand0.4 Or=75%- 0 0.2bE0.0

300 0oIIIIII200ll>.E.....;:III1.2 r---'"-------------------,50 40gravelsa ~ d sandslsiltsandclays io i'-------'!l! !20 30plasticity index. Ip10,cx~ 0.8ofCIlCIl.E 0.6....-;:;CIl(b)0.7 r----------------------,0.6E0.5~c~ 0.4o~ 0.3ll>5 0.2oo. ,(c)20 40 60plasticity index, Ip80 100Fig. T4.18. Variationofstiffnessparameters with plasticityindexIpfornatural soils(after Mancuso et al., 1997).(a) Parameter S.(b) Parameter11.(c) Parameter m.Geotechnical Characterisation 18550 40!i... :I!20 30plasticityindex, I p10!'-..------".c----;-------------15001250(/)'EC1l1000'0cQi0750cIIIVIC1lc:500~iii25000(a),Ii,; I!_.,1--.---1......I, I! ...I tjI......II -...,IiA.........!... :...I...,I! ,IIIit iIf: I,!0.8c:"IiCD0.6"0.::CIlCIlC1l0.4c:;;:iii0.20(b)o 10 20 30plasticity index, Ip40 50Fig. T4.19. Soil stiffnessparameter variation with plasticityindex 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 decreasewith stress level become higher going from stiff to soft clays; also, at the samestress state and history, Do increases with plasticity index; thevalues of Doof medium/fine-grained compacted materials are often high-er thanthosetypical of natural clays, dueto therelativelyyoung age of theman-made soil deposits.For thesame normally consolidated marine soilsanalysed to obtain the correla-tion in Fig. T4.19, d'Onofrio and Silvestri (2001) used the simple power functionrelationshipDo=Q(:J186 PlANe5r------r----r---,.-----,-----,50 40 20 30plasticity index, I p10(a)- - "0ECl.loC,) 2 ..........'i__---':::_""'::::::..---1Clc:'5.E o60 50 10 20 30 40plasticity index.I pIII!,I-IAA IAAAII-AI IiA II.-i, j(b)oo0.50.10.4x 0.3Clc:c.. 0.2Eco"CFig. T4.20. Soil damping parameters variation with plasticity index I p for normally con-solidated marine clays (after d'Onofrio andSilvestri, 2001).(a) Parameter Q.(b) Parameter j .to relate thedecrease of the small-strain damping to the increase of meaneffec-tive stress. The correlation with the plasticity index lp of the damping parametersQ (representing the value of Do. expressed in %. at a p' =pa)and j, shownin Fig.T4.20 was obtained. The plots confirmthat both thevalues of Do and itsrateofdecreasewith 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(J1using the plasticity indexI.Geotechnical Characterisation 187TableT4.7. Typical ranges of soil density andshear wavevelocity.Soils Typical ranges---:--------------p Organic soils. peatsVery soft claysLoosevolcanicsilts/sandsSoft clays. loose sandsDense/cemented volcanic silts/sandsStiff clays. densesandsCemented sandsandgravelsIntactrocksl.0 to1.31.4 to1.71.0 to1.71.6 to1.91.2 to1.91.8 to2.11.9 to 2.22.0 to 3.0800Based on thetest data found in literatures, their own laboratorytestdata, Vucetic andDobry (1991)report the average G(y) and D("/) curves showninFig. T4.21. Bothplots showthat thedegree ofnon-linearity(hence, theenergy dissipation)increaseswith decreasing plasticity. Thegranular soilswith[p = 0, shownon-linearity over a linear threshold strainy, of about 0.001%. Notethat the curvesrelevant tothedampingratio arenot scaledtothe small-strainvalue, Do, and that, according to the observationsin Section T4.3.2, suchsmall-strainvalue should increase with I p and decrease with p',The strain dependent curves for Japanese marine clays have been proposed byZen et al. (1987) as shownin Figs. T4.22 and T4.23. The effects of variations inthe state variables (e, pi,OCR) on the shape of nonnalisedcurves were found tobe negligible forfine-grained high 'plasticity soils(Fig. T4.22(a. On the otherhand, the increaseinstresslevel cansignificantlyshift totheright the G(y)relationship for a medium to low plasticity soil (Fig. T4.22(b, c)). In contrast tothe soilscompiledbyVucetic andDobry(1991) (Fig. T4.2l), theJapanesemarine clays do not appear to exhibit dependency on plasticity for [p > 30. Again,thedegree of non-linearityisseentoincreasewithdecreasingplasticity. Theinfluence of stateand consolidationvariables ontheincrease of dampingratiowith shear strain was not clearly established, andtheaverage D(i? curves givenin Fig. T4.23 were suggested.The linear threshold strain ri, beyond which soil non-linearity is observed, canbe alsocorrelatedwith the plasticity index[P. Figure T4.24(a) shows that,depending on particle grading and microstructure, Yl may vary from 0.0001 % tomore than0.01%. The values ofYI for granular soilsincrease with decreasing par-ticle size; those for volcanic sands and silts are higher with respect to thosetyp-ical of other non-plastic soils, due tothe stronger inter-particle bonds providedby cementation and/or particle interlocking. The degree of non-linearity of com-pactedfine-grainedsoils maybemoderatelyaffectedbyplasticity(Mancusoet al., 1995), whereas that of natural clays can be strongly affected by]88 PlANe0.001 0.01 0.1CYCLIC SHEAR STRAIN.'"(%)0.8c.s-C) a.'/'0.2 (a)IOCR =1-151Ip =2003 10101530501000.001 0.01 0.1CYCLIC SHEAR STRAIN .., (%)IOCR=1-S!aL..-__--"__ a.OOO1(b)25520

eo615.-1.0Sampling method Standard sampler Cs1.0Sampler without liners 1.1 to1.3Thecurve indicatedby Seedet at (1983) for CqisshowninFig. T4.28(b);NCEERworkshop(YoudandIdriss, 1997) suggestedto usethe generalizedexpreSSIOnc = q a'.va(T4.20)with n dependentongraincharacteristics of thesoil, andrangingbetween 0.5(clean sands) and1.0 (clayey soils).Themeasured shear-wave velocityVs canbe normalised toa reference effec-tive overburden stress through theexpression(T4.21)wherencanbe3(Tokimatsuet al., 1991)or 4(Finn, 1991;Robertsonetal., (3)Use of liquefaction chartsAliquefaction chart isusually expressed asacurve inaplane definedwiththecyclicstress ratio on the ordinate and the selected field estimator of liquefactionresistanceontheabscissa. Thecurvelimits theboundarybetweentheexperi-mental points representing observed liquefaction cases (above) and non-liquefiedcases (below). Thereforecit empirically represents the locus oftheeRR, i.e. the196 PlANeminimum CSR required to produce liquefaction of a soillayer characterised bythe reference field property.FiguresT4.29 throughT4.31show thecriteriaobtained as theconsensus atthe NCEERworkshop(YoudandIdriss, 1997)for evaluationofliquefactionresistance, respectively, based onnormalisedand corrected values of SPT blowcount (N1) 60, CPT tip resistance qa, and shear-wave velocity VS 1 The curves inthese figuresaregivenfor earthquakeswithmoment magnitudeMw = 7.5, ongently slopingground (slopes lower than about 6%) withlow overburden pres-sures (depths less thanabout15m). The values of CRR for momentmagnitudesother than7.5 need to be corrected by a scaling factor, CM, decreasing with thedesign earthquakemagnitude. Table T4.9shows the values originally suggestedby Seed and Idriss (1982), together with theupper andlower bounds suggested

25'Percent Fines =35 15 s 5II0.51------4-----fl'----t"--i-----+------lCRR curves for 5.15. and35 percent fines. respectivelyIIII, AIOI20AdjustmentRecommendedByWorlcshop".50+FINES CONTENT 5%Modified ChineseCodeProposaJ (claycontent== 5%)@0.1Marginal NoLiquefactionLiquefactionLiquefactionPan American data eJapanese data 0 eChinese data. A 0 O.4I------+-----f-:::+--++----+------lt{

N".2-Cl:l0::: (/)(/)E....CI:l

U10 20 30 40Corrected BlowCount, (NJ)6O50. Youdand ldriss (1997)Fig. Evaluation ,ofCRR fromSPT N-value(Youd and Idriss,1997, modified afterSeed et al., 1985).Geotechnical Characterisation 197Fig. T4.30. Evaluation of eRR fromCPT data (Robertsonand Wride, 1997).3001:--/0 Li4uelaction I6. 6.Bo 0 Oqjo6.t:Ol'ico I 0.5-zc:: =n..gCl)"0Q)cr:Geotechnical Characterisation 203ThetwocorrectedequivalentN-valuesare plottedinFig.T4.34against theequivalent acceleration, andthezone towhich a soil layerbelongsis deter-mined as follows:,. WhenN ~ ; is in Zone1. liquefaction potential is evaluated by Zone I.,. WhenN ~ ; is in Zone II, liquefaction potential is evaluated by Zone II.,. When N ~ ; is inZoneIII or IV, andwhenN;5 is in Zone 1. II, orIIIliquefactionpotential is evaluatedby Zone III.,. WhenN ~ ; isin Zone III orIV, andwhenN;s isinZone IV, liquefactionpotential is evaluated by ZoneIV.Case 3: The plasticity index lp >20, and the fines contentFe >15%.Acorrected equivalent N-value is calculated using Eqns. (T4.29) and (T4.30),and thenliquefaction potential is evaluated using Fig.T4.34.(2) Cyclictriaxial test (2nd step)When theliquefaction potential cannot be determined from the grain size distri-butionand SPTN-value (i.e. for zones II andII!), it is evaluated basedonundrainedcyclictriaxial tests usingundisturbedsoil samples. Basedontherelationshipbetweenthe cyclicstress ratio and number of cycles to cause lique-factioninthelaboratorytests, thecyclicresistance('l",1( ) ~ ) N l = 20 for usein theliquefaction potential assessment is obtained by reading off the cyclic stress ratioat 20cycles ofloading(Fig. 1'4.36). Usingthiscyclic resistance, thein-situliquefaction resistance Rmax is obtained using the following equation:(T4.31)0.50.4b(J

Ql"0 Ol-:l--+--""""----------@c..E-cFig. A1. Example of time domain analysis of a CH test with two receivers.The waveform velocity between two receiverslocated along a straight line canbedeterminedbythe CrossCorrelationfunction, providedthatthesignalsareacquired in digital form, The CrossCorrelation [CCt).('r)] of two signals x(t) andy(t) is defined as+-CC.t:'.(r)=Jx(t)y(t+r)dt(A.2)wherer is the variable time delaybetween signals x(t) and yet).Hence, a Cross Correlation valueisgiven bythesum of theproducts of thetwo signal amplitudes, after having shifted thesecond record, yet), withrespectto the first, x(t), by a time interval 1:. The CrossCorrelation functionyields dif-ferent values by varying thetime delay1:, and reaches its maximumCCm:u whenthedegree of correlationbetweenthesignalsx(t) andy(t+r) ismaximum (i.e.when 'the waveformsarepracticallysuperimposed). Thetimedelay 1:max corre-sponding to CCmax is thebest estimate of the travel time of thewholewaveformbetween the two receivers,The Cross Correlation shownin Fig. A2 is obtained using two receiver signals(5waves) of Fig. AI. The CrossCorrelation valueshave beennormalisedwith., .._".' re210 PlANe50 100 150 't'(ms)200Fig. A2. Cross-correlation function of the receiver signals in Fig. AI.respect to the maximum CCrnax,hence the function varies between -1and 1. Thetravel time is approximately 62 ms.A.2 Frequency domainanalysisBy meansof the Fourier analysis, a generic signal x(t) can be decomposed into asum of a series of sinusoidalfunctions, each one characterised by its amplitude(A), phase(4)) andfrequency (f)(seeFig. A3). Thesine 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 thecorrespondent Linear Spectrum canbeperformedby means of the algorithm Direct Fourier Transform(DIT); inverse-ly, it is possible to convert a Linear Spectrum into thecorrespondent time recordby meansof the algorithm Inverse Fourier Transform (Iff). The frequency con-tent of a signal iseffectively represented bythe amplitude versus frequency plotof theLinearSpectrum; inFig. A4, suchdiagrams areplotted forthereceiver(RIand R2) signals in Fig. AI.Eachsine functionof a Linear Spectrum can beeffectively represented byavector (with length proportional to the amplitude, and direction equal to the phaseof the sine function),and therefore, by a complex number. The rules and tools ofcomplex algebra are applicable to signal analysis in the frequency domain. In thefollowingparagraphs, the most usefulfunctions for frequencydomain analysisare reviewed.The Cross Power SpectrumCPSxy of two signals x(t) and yet) is the product of thecorrespondent Linear Spectra L, and 4(A.3)Eachfrequencycomponent (f)of theCPS has anamplitude[AcpsCfl]equaltothe product of theamplitudesof thesamefrequencycomponentsof the linearspectra. and phase[4>cps(f)] equalto their phase difference. TheAcps(j) allowsGeotechnical Characterisation 211A++AAAtO==C>- ~=t-It+A+Fig. A3. Transformationof an amplitude-time signal to an amplitude.frequencysignalby means of theOFf.>E.....IV'"0:;,.-Q.EE.Q.Eo Io100100F" A7 Phase velocity diazram from a CH test.19. . 214 PlANeA.3 Spectral analysis and propagation velocityThe Cross Correlation function(previouslydefinedinthe time domainanalysisparagraph), and hence the travel time andthe correspondent waveformvelocity,canbe immediately determinedutilisingthefrequencydomain analysis. TheCross Correlationoftwo signal x(t) and y(t) is given bytheInverseFourierTransform of the Cross Power Spectrum of the two signalsCC.l).(r) =Jx(t)y(t+ T)dt =IFT[cPSyx(!)](A.5).The Cross Power Spectrum allowsfor thedetermination of the velocity of eachof 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 propagationbetweentwopointsat knowndistance is representedbythecorresponding number of wavelengths N: given byNJ!) =- ~ ~ )(A.6)If rl2 is the distance between thetwo points, the wavelength A(J) and the veloci-tyVph(f), defined as phase velocity, of the component (f) are givenby(A.?)The phase velocity diagramVph(f), calculated for a series of signalsx(t) and yet)from a CH test is shownin Fig. A7. together with therelative Acps(f), COxy(f)and(JJcps(j). It is clear that, for frequencies greater than25 Hz. the Vph(!)dia-gram is practically horizontal, slightlyvarying aroundthe 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 SystemsT5.1T5.2T5.3T5.4T5.5T5.6Design Performance and Earthquake LevelsT5.1. 1 Performance goalT5.1.2 Design earthquakelevelsT5.1.3 Limit statesT5.1"+ Seismic load combinationsModelling AspectsT5.2.1 Soil-structure interactionT5.2.2 Movement jointsMethods of Analysis for Seismic ResponseT5.3.1 Method A:Equivalent single mode analysisT5.3.2 Method B: Multi-mode spectral analysisT5.3.3 Method C:Pushover analysisT5.3..+ Method D: Inelastic time-history analysisT5.3.5 Method E:Gross foundation deformation analysisStructural Criteria forPier and Wharf PilesT5.4.l Deformation capacity of pile plastic hingesT5.4.2 Implication of limit statesT5.4.3 Moment-curvature characteristics of pilesT5.4.4 Primary structural parameters and response of pilesT5.4.5 PlastichingesT5.4.6 Shear strength of pilesT5.4.7 Design strength for deck membersT5.4.8 Batter and timber pilesPilelDeck Connection DetailsT5.5.l Steel-shell pilesT5.5.2 Prestressed pilesT5.5.3 Practical connection considerationsT5.5.4 Capacity of existing substandard connection detailsExistinz Construction....T5.6.1 Structural criteria for existing constructionT5.6.2 Strengthening of an existing structureT5.6.3 Deterioration of waterfront structuresTECHNICAL CONIMENTARY 5Structural Design Aspects of Pile-Deck SystemsThistechnical commentarywill further illustratethegeneral design methodo-logies developed in the Main Text and present some numerical guidelines for theperformance of pile-deck structures. Piles are a common and significant elementof waterfrontconstruction. Thistechnical commentarywill provide guidanceinpile design.The followingguidance is directly basedon TechnicalReport TR 2103-SHR'Seismic Criteria For California Marine Oil Terminals' by J. Ferri tto,S. Dickenson, N. Priestley, S. Werner andC. Taylor, July 1999. This reportwassponsoredbythe CaliforniaState Lands Commission, Marine FacilitiesDivision. Thestructural criteriadefinedhereinarebased largely ontheworkof: Nigel Priestley. Thesecriteriaaredevelopedfromacompilationof currentpractice in North America, with state-of-the-art technology for evaluatingseismicdamage potential. Althoughtheguidanceaddresses structural designof concrete, steel and wooden piles, the focus of attention is directed toconcrete piles, reflecting the practice in North America. Design criteria adoptedin other regions, where steel pileshavebeentypicallyused, notablyinJapan,might be different fromthis guidance as briefly discussed in TechnicalCommentary 7.T5.1 DESIGN PERFORMANCE AND EARTHQUAKE LEVELST5.1.1 Performance goalThe general performancegoalswerediscussedin Chapter3 in theMainText.This technical commentary focuses on pile-deck structures,specifically piers andwharves. The reader is referred to Tables 3.1 to 3.3 in the Main Text. In this tech-nical commentary, piersand wharvesareas Grade A structures andare expected to perform as follows:( 1) Toresistearthquakesofmoderatesizethat canbeexpectedtooccuroneor moretimesduringthelifeof the-structure(Level 1)without structuraldamage of significance (Degree I: Serviceable).........,218 PlANe(2) To resist major earthquakes that are considered as infrequent events (Level 2),andto maintainenvironmental. protection and life safety, precludingtotalcollapsebut allowingameasureof controlledinelasticbehaviourthat willrequire repair (Degree II: Repairable).T5.1.2 Design earthquake levelsThis section utilizes the following earthquake levels presented in Chapter 3 in theMain Text asdefined events.. Level1- (Ll) Groundmotions with a50%probabilityof exceedancein50 yearsexposure. This eventhasareturntime of 75 years and is considered amoderate event,likely tooccur one or moretimes during the life of the facility.Such an event is considered a strength event.Level 2- (L2) Groundmotions with a 10%probabilityofexceedancein50 years exposure. This ~ v e n t hasa return time of 475 years and is considered amajor event. Such an event is considered a strength and ductility event.T5.1.3 Limit statesServiceabilityLimit State(DegreeI: Serviceable) All structures and theirfoundationsshould be capable of resisting theLevel I earthquakewithout sus-taining damage requiring post-earthquake remedial action.DamageControl Limit State (DegreeII: Repairable) Structures and theirfoundations should be capable of resisting a Level 2 earthquake. without collapseand with repairable damage, whilemaintaining life safety. Repairabledamage tostructures and/or foundations, andlimitedpermanent deformationare expectedunder this levelof earthquake.T5.IA Seismic load combinationsWharves and -piers may be checked for the following seismic load combinations,applicable to bothLevel 1 and Level 2 earthquakes:(l + kJ(D ffi rL) ffi E (T5.1)(1 - kJDffi E (T5.2)where D = DeadLoadL'= Design LiveLoadr = LiveLoad reduction factor (dependson expected L present in actualcase; typically 0.2 but could be higher)Structural Design Aspects of Pile-Deck Systems 219E =Levell or Level 2 earthquakeload. as appropriatek, =Verticalseismic coefficient (typically onehalf of theeffectivepeakhorizontal ground accelerationing)EB = Loadcombination symbol for vector summationNote: seismicmass forEshouldinclude anallowanceforrL, but neednotinclude an allowancefor themass of flexiblecrane structures.Effects of simultaneous seismic excitation in orthogonal horizontal directions.....indicated by x and yin Fig. T5.1.maybe considered in design and assessment ofwharvesandpiers. For thispurpose, it will besufficient toconsider twocharac-teristic cases: .1000/c E/B30% e,30% Ef $ 100% s,(T5.3)(T5.4)where Ex and E; aretheearthquake loads (E) in the principal directions x and y,respectively.Whereinelastic time historyanalysesinaccordancewith therequirements ofMethodDin Section T5.3arecarried out, .theaboveloading combinationmaybe replacedby analyses under the simultaneous actionofx andy directioncomponentsof groundmotion. Suchmotions shouldrecognize the direction-dependency of fault-normal and fault-parallel motions with respect to thestructure principal axes, where appropriate.T5.2 MODELLING ASPECTSThissection further illustrates the modelling aspects of pile-deck structures sum-marizedin Technical Commentary 7.ILandward F.dL'C ,. x...... (. (Seaward Edge(a)1--------IIIIIIII(b)1--------------...I II .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,yexcitation....' ....220 PlANeT5.2.1 Soil-structure interactionFigure T5.2(a)represents a typical transverse section of awharf supported on asoil foundationcomprisedof different materials, includingalluvial sandandgravel,perhaps withclaylenses, r i p ~ r a p , and otherfoundationimprovementmaterials. Themost precisemodellingof thissituationwouldinvolve inelasticfinite element modelling of the foundationmaterial to a sufficient depthbelow,andtoasufficient distance oneachside, such that strains inthefoundationmaterial at the boundaries wouldnot be influencedby the response of thewharf structure. Thefoundationwouldbeconnectedtothe pilesbyinelasticWinkler springs at sufficientlyclose spacing so that adequaterepresentationof the pile deformation relative to the foundation material,and precise definitionof the in-ground plastic hinging,would be provided. Close to the ground surface,where soil springstiffness hasthe greatest influenceon structural response,the springs should have different stiffnesses and strengths inthe seaward,landward, and longitudinal (parallel to shore) directions, as a consequence of thedike slope.The pile elements .wouldberepresented by inelastic properties basedonmoment-curvature analyses. Since thekey properties of piles(namely strengthand stiffness) depend on the a ~ i a l load level which varies during seismicresponse, sophisticated interaction modelling would be necessary.It is apparentthat structural modellingoftheaccuracyanddetail suggestedabove, thoughpossible, isat theupperlimit of engineeringpractice. andmaybeincompatiblewiththeconsiderableuncertaintyassociated with theseismicinput. In many cases, it is thus appropriate to adopt simplified modellingtechniques. TwopossibilitiesareillustratedinFig. T5.2. InFig. T5.2(a), thecomplexitymaybe reducedby assuming that thepilesarefixed at theirbasesF4ASF54 3 2........f2A2[ 800 700g600 500400300200100 __.,..-0.00Fig. T5.8. Continued.efficiency. For assessment of existingstructures, theconcrete coremaximumstrain corresponding to the damage-control limit state shouldnot be takenlargerthan 0.007atthe pile/deck hinge. Higher maximum concretestrains. as given byEqn. (T5.l2), areappropriate if thecore of therectangular pileis confinedby acircular spiral and the longitudinal reinforcement or prestressing is also circularlydisposed. However, suchsections typically havelow ratios of concrete core areatogrOSS section area, resultinz in flexural strength of the confined core being less- ......... - . than that of theunconfined gross section. .Hollowprestressedpiles aresometimes used for marine structures. These,however. haveatendencytoimplodewhenlongitudinal compressionstrains atStructural Design Aspects of Pile-Deck Systems 2391200N=O---f-E N=1500 kNz

...800c

Ec..

4000.12 0.10, .,0.04 0.06 0.08Curvature(11m)0.02 0.001500,..eZ

... 1000c

,..c::0

50(f)0.00 0.04 0.08 0.12Curvature(11m)0.16 0.20Fig. T5.8. Continued.the inside surfaceexceed0.005. Consequently, the damagecontrol limit stateshould haveanadditional requirementtothe strain limitsdefinedearlierthatinside surface compression strains must not exceed 0.005. Note that to check forthis condition, the moment-curvature analysis must be able to model spalling ofthe outside cover concrete when strains exceed about 0.004 or 0.005. The outsidespalling cancause asuddenshift of the neutral axis towardsthecentre of thesection resulting in strains on theinside surface reachingthe critical level soonafter initiation of outside surface spalling. More information on pilesis availablein Priestley and Seible(1997).(T5.l3)(T5.l4)240 PlANeT5.4.4 Primary structural parameters andresponse of piles(l) Elastic stiffnessTheeffective elasticstiffnessmaybecalculated fromthe slopeof the'elastic'.portion of the bi-linearapproximatiori to the moment-curvature curve (e.g.Fig. T5.7), asEleff =MN1ywhere leff effective moment of inertiaMN nominal flexural strengthThevalue of theyield curvature, y' for a reinforced concrete pile or pile/deckconnection is rather insensitive toaxialload level orlongitudinal reinforcementratio. Results of analyses of alarge number of cases indicate that Eqn. (T5.13)may be approximately expressed as a fraction of the gross sectionstiffness asEleff/EIgross =0.3+ Ntt];AgrosJwhere N is the axial load level, and Agross is the uncracked section area.For prestressed piles, the effective stiffness is higher than for reinforcedconcrete piles, andvalues in the range 0.6< EleiE1gross < 0.75 are appropriate. forprestressedpiles withreinforceddowel connections to thedeck, the effectivestiffnessshouldbeanaverage of that fora reinforcedand a prestressed connec-tion, orashort length, approximately2Dp long of reducedstiffnessappropriatefor reinforced pile, should be located at thetop of theprestressed pile.(2) Plastic rotationThe plastic rotation capacity of a plastichinge at a givenlimit state dependsonthe yield curvature, Y'the limit-state curvature. s for Levellor LS for Level 2,and the plastic hinge length Lpand is given byGp=pLp= s orLSJ - )Lp(T5.l5)(3) Plastic hinge lengthThe plastic hingelength for piles depends onwhether the hinge is located at thepile/deckinterfaceor isanin-groundhinge. Becauseof thereducedmomentgradient in thevicinity of thein-groundhinge. the plastic hinge length is signi-.ficantly longer there. For pile/deckhinge locations withreinforcedconcretedetails.the plastic hinge length canbe approximated bySI units 4= 0.08L + 0.022f.dh >O.044hdh (MPa, mm)US units L; = 0.08L + 0.15fdh> 0.30h.do (ksi, in.) .(T5.l6a)(T5.16b)where I, is the yield strength of the dowel reinforcement. of diameter do. and Listhe distance from the pile/deck intersection to the point of contraflexure in the pile.Structural Design Aspects of Pile-Deck Systems 1For prestressed piles where thesolid pileis embedded in the deck (an unusualdetail in the USA). the plastic hinge length at the pile/deck interface can be takenas(T5.17)For in-ground hinges, the plastichinge length depends onthe relativestiffness ofthepileandthe foundationmaterial.ThecurvesofFig. T5.9relatetheplastichinge length of the in-ground hinge to the pile diameter,Dp,a reference diameterD* =1.82 m, and the soil lateral subgradecoefficient, (N/mJ) .For structuralsteel sectionsand for hollowor concrete-filled steel pipe piles,the plastic hingelengthdepends onthe. sectionshape andthe slope of themoment diagram inthevicinity of the plastic hinge, and should be calibrated byintegration of the section moment-curvature curve. For plastic hinges forminginsteel piles at the deck/pile interface, and where the hinge formsin thesteel sec-tionrather than inaspecial connectiondetail (such as areinforcedconcretedowel connection), allowance should be made for strain penetration into the pilecap. In the absence of experimental data, the increase inplastic hinge length dueto strain penetration may be taken as 0.25 Dp, where D; is the pile diameter (witha circular cross section)or piledepth(witha non ciruc1ar crosssection)in thedirection of the applied shearforce.(4) In-ground hingelocationThe location of the in-ground plastic hingefor a pile may be found directly froman analyses where thepile is modelled as a series of inelastic beam elements, and2,00...q) 1.80ell'6.s1.60

1.40

1.20z 1.00ella:H=20p0.8070 60 20 30 40 50\000 x / OBaff10o.60o(H=height of contraflexure point above ground, Dp=pile diameter)Fig. T5.9. In-ground plastichinge length.242 PlANethe soil is modelled by inelastic Winkler springs. When the pile/soil interaction ismodelled by equivalent-depth-to-fixity piles, thelocation of the in-ground hingeis significantly higher than the depth to effective fixity, as illustrated in Fig. T5.10bythe differencebetweenpoints F, at the effective fixitylocation, andB, thelocation of maximummoment. Notethat whensignificant- inelastic rotationisexpected at the in-ground hinge, the location of B tends to migrate upwards to apoint somewhat higherthanpredicted byapurelyelastic analysis. It is thusimportantthat itslocation, whichis typicallyabout 1-2pilediametersbelowground surface(such asdike surfacebeneath wharf), shouldbe determinedbyinelastic analysis.(5) Pile force-displacement responseThe informationprovided in the subsections(1) through (4) enables an inelasticforce displacement response to be developed individually for each pile. This maybedirectly carried out ona full 2Dsection through the wharf, involving manypiles, as part of a pushover analysis, or it may be on a pile-by-pile basis, with thepushover analysis assembled from thecombined response of the individual piles.Withrespect tothe equivalent-depth-to-fixitymodel ofFig. T5.l0, thepileisinitially represented by an elastic member,lengthL, with stiffness Elctf given byEqn. (T5.13) or (T5.l4), as appropriate, and thedeck stiffness represented by a.spring kd, as shownin thesame figure. Often, it willbe sufficiently accuratetoassume the deck to be flexurally rigid, particularlywith longer piles.Thedeflectionand forcecorresponding todevelopment of nominal strengthMNatthe pile/deck hingecan thenbecalculated. Note that, forelasticdefor-mationcalculations, the interface between the deckand pile shouldnot beIl===::::;:;:=====*=*AR.F(a)B(b)1.1..Q)::o1.1..3Displacement 1!.(c)Fig. T5.1O. Force-displacement response of an isolatedpile.(a) Pile equivalent-depth to fixitymodel. .(b) Displacement profiles.(c) Force-displacement response.(T5.18)Structural Design Aspects of Pile-Deck Systems 2.+3consideredrigid. Theeffectivetopof thepileshouldbe locatedat adistanceO.022!rdh intothe decktoaccount for strain penetration. This is particularlyimportant forshort piles.This additional lengthappliesonlyto displacements.Maximum moment should still be considered to developat the soffit of the deck.The elastic calculations above result in a pile displacement profilemarkedI inFig. T5.10(b), andthecorrespondingpoint ontheforce displacementcurveofFig. T5.10(c). For the next stepin thepile pushover analysis, an additional springkpl mustbeaddedatA, the deck/pileinterface(i.e.thedeck soffit), torepresentthe inelasticstiffnessof the topplastichinge. Thisstiffness canbedeterminedfrom Fig. T5.7 ask =(Mu-MN )pI (tfu -tPy)4Essentially, this springis in serieswiththedeck spring. Additional forcecan beapplied tothe modified structure until the incremental moment at Bissufficientto developthe nominalmoment capacityat thein-ground hinge. Thedeflectionprofilesand force-displacement pointsmarked 2 in Figs. T5.10(b)and (c) referto thestatusatthe endofthis increment. Finally, themodifiedstructure, withplastic hinges at the deck/pile interface and B, is subjected to additionaldisplacement until the limit state curvature tPLS is developed at the critical hinge,which will normally be the deck/pile hinge. Note that the inelastic springstiffnesses at the two hinge locations will normally be