th

The14 Wor ldConfer enceonEar thquakeEngineer ing October 1217,2008,Beijing,China

LIQUEFIABILITYOFSILTSANDTHEVULNERABILITYMAPOF ADAPAZARI
E.Bol ,A.Onalp andA.Ozocak
1

AssistantProfessor,Dept.of Civil Engineering , SakaryaUniversity, Sakarya, Turkey 2 Professor,Dept.of CivilEngineering, IstanbulKulturUniversity, Istanbul,Turkey email: ebol@sakarya.edu.tr, a.onalp@iku.edu.tr,aozocak@sakarya.edu.tr

ABSTRACT: Thecityof Adapazar,TurkeyisnotoriousfortheMw7earthquakesthatstrikes almosteach decade.Theevent of1999wasprobablytheworstinmemory,becauseamajorportionofthecitywasdestroyed.Theground failuresintheformofcyclicmobility,liquefaction andexcessivesettlements of buildingswerewidespread.A majorinvestigationprogramwasinitiatedatSakaryaUniversitytodiagnosetheliquefactionpotentialofsilts andclaysintendedtolead theinvestigatorstowardspreparationofliquefactionand bearingcapacitymapsfor thecity.Theprogramincludeddrillingof700boreholeswithstandardpenetration andlaboratory testing,300 conepenetrationtests(CPTU,SCPTU).AsetofAdapazarCriteriawereproposedasasubstituteforthe ChineseCriteria.Thecriteriawerecomparedtoandaidedbythesurface observations maderight afterthe quake.AdatabaseforthecityofAdapazarbasedonsurfaceobservationsaswellasfieldandlaboratorytest resultsusingGISmethodsisfinallyestablished.Itisnowpossibletoquerythegroundwaterlevel,soiltype, organiccontent,SPNN60,undrainedshearstrength,allowablebearingcapacityandliquefactionpotentialfor thetop15mdepth. TheresultsindicatethatthesoilconditionsintheCityarequitedifferentfromwhathasbeen publishedtodate.Thisdiscrepancyisduetothefactthattheverticalandhorizontalvariabilityofthesoilsis extremelyhighbecauseoftheformationprocessesinvolved. KEYWORDS: 1.INTRODUCTION ThecityofAdapazarisoneoftherarecitiesintheregionthatcarriesextraordinarilyhighseismicrisk.Itis locatedalmostontheNorthAnatolianFault(NAF)whichis knowntobecomeactive(Mw7)almostevery decade(Figure 1). It nevertheless continues to expand, its current population estimated at 300,000. In addition, almost whole of the city is founded on Recent alluvial deposits of the River Sakarya flowing northwards towardtheBlackSea.Arecent borehole drilledat thecenterofthecityfailedtoreachrockat 200m (Bol, 2003) and it was claimed that the depth of the alluvium here may be as deep as 1000m (Komazavaetal.,2001). Numerousgeotechnicalinvestigationscarriedoutfollowingthe1999eventconcludedthatamajorpartofthe soilshadliquefiedthroughoutthecityandbuildingdamagehadbeencausedbythosepoorsoils.Thesurveys andresearchcarriedoutbySakaryaUniversitysincethenhaveprovedthatthis wasahastydiagnosis.Itis 2 now confirmed that liquefaction had materialized in about 30% of the total area of 25.6 km . It was subsequently discoveredthat nonplasticsilts withsandlayers were dominant at sites ofliquefaction,from whichtheAdapazar Criteria(Bol,etal.,2005)werederivedandaliquefactionmapwasdrafted. 2.ADAPAZARIGEOTECHNICALDATABASE About 700 boreholes were drilled and 300 cone penetration tests (CPTU) were performed by Sakarya Universityduringtheinvestigation.Ofthese,133pairs were drilled sidebysidetoenable correlations. Liquefaction,Silt,Finegrained soil,GIS, Mapping,Adapazar criteria

th

The14 Wor ldConfer enceonEar thquakeEngineer ing October 1217,2008,Beijing,China

Figure1GeologicalformationsandlocationofAdapazar(KTa:UpperCretaceousRocks)

A totalof5729SPTand6000standardlaboratorytestswererun.62000linesofoutputfromtheCPTUwere added to the data. Occasional down hole seismic measurements (SCPTU) providing the shear wave velocities werealsoincluded.Allavailableinformationincludingthoseofothercredibleinvestigatorswere ultimatelycollectedintheAdapazarGeotechnicalDatabase whichwillbeavailabletoallresearchers in thefuture.GeographicalInformationSystemwasemployedtoderivethematicmapscomprisingsoilclasses, SPTN,qc,damage,bearingcapacityandliquefaction.

3.GEOMORPHOLOGICALPROCESSESS TheliquefiablesiltsintercalatedbyseamsoffinesandhavebeendepositedbytheRiverSakarya(Figure1) during the past 7000 years. The extensive borehole information procured has indicated that the river has crossedtheplain ofSakaryabymeanderingandtraversing.Figure2illustrates ablock diagramofthesoil profile.ItisobservedthatoutcropsofthebedrockconsistingofUpperCretaceousclaystone,siltstone,marl and limestones are intersected by an inactive fault defining the border of the deep alluvial deposits (Bol, 2003).Thehighlyplasticclays(CH,CI)surroundingthehillsareprobablyweatheringproductsoftherocky slopes,intermixedwithlacustrinedeposits. Figure2depictstwoancient(buried)riverbeds,one4mandtheotherapproximately10mbelowthepresent ground surface that contain gravels rarely encountered elsewhere. These are characterised by notably high penetrationresistances (N60>50 qc>10MPa)whicharenot recorededanywhere else.Itis possiblethat both channelscoexistedwherebyanislandbetweenthemwasformedthatgivestheCityitsname: AdapazarIsland Market,wheremarketsfordifferenttradesandcraftsexistedforseveralcenturies.Themainfeatureofthese ancientriversisthattheyfloodedtheplainofAdapazaralmosteveryyearasthepresentRiverSakaryadid until1965whensedimentationwashaltedbytheconstructionoftwolargedamsupstream. The frequentrapid flooding by the ancient and the present streams often tore their banks at convex sections of the meanders (crevassesplays)and deposited thesedimentsasoverbankdeposits.Therearecertainsectionsofthecitywhere this mechanismis morepronounced,indicatedbyappreciableamountsoffinesediments.Itisbelievedthat thosespeciallocationsarepresentlysitesofliquefaction.Consequently,therearedistrictsofthecitylocatedon thebedrockthathadnotsufferedseriousdamage,regardlessofthequalityofthesuperstructureandground failures were not observed. On the contrary, there were damages of the structures founded on Holocene alluvialdeposits(Bol,2003Bol,etal.2007). Studying the earthquake performance, it is clear that findings from these soils will solve some hitherto unexplained topics of geotechnical engineering like the effect of changing soil properties on deep alluvial basins,variationofbearingcapacitywhichdependonsedimentationconditionsandliquefactionofsiltysoils.

th

The14 Wor ldConfer enceonEar thquakeEngineer ing October 1217,2008,Beijing,China

Figure23DfaciesoftheCity

3.1. TheOriginofAdapazar Silt

AlthoughithaswidelybeensuggestedthatAdapazar isacityofsilt,ithasbeendiscoveredthatfrequently appearinglowplasticitysiltsareinfactoflimitedthickness.Figure3depictsthelocationsofnonplasticsilts foundinthealluvialsedimentsofthecity.Thetop10mthickness,wheremostoftheliquefactionoccurred was dividedintosixzones.The mapshows that nozones werefoundtoconsistofMLinallofsixlayers. Furthermore,zoneswith5or4layersofsiltarealsoextremelyrare.Anappreciablenumberofsiteswith3or 2siltlayersweredetected.Mostsignificantly,anoverwhelmingpartofthecitycontainedsoilprofileswitha singleornolayerofMLsilt. There is a wealth of publications on the liquefaction of Adapazar silt (Sancio et al., 2002 Sancio, 2003 Bray et al., 2004 Bol, 2003 Bol et al., 2005 and 2007). Indeed, the observations and subsequent investigationsbySakaryaUniversityhasshownthatthesectionboundedbythebrokenlineinFigure3was the obvious liquefaction zone. It is noteworthy that no dense sands were found in the soil profiles in this section butfrequentlensesandlayersofsiltwereencountered,apossibleproductofcrevassesplaysfromthe convexpartofthemeanderingriver.Theenvisageddirectionofcrevassesplaysisshownbyathickarrowin thefigure. 4.LIQUEFACTIONOFFINEGRAINEDSOILS Because of high SPTN values and cone tip resistances of sands along the buried channels, and their conspicuous absence in other parts of the alluvium, the liquefaction potential of fine grained soils were evaluated. The most popular plasticity based criteria are the Chinese Criteria which was developed followingtheearthquakesinChina between1966and1976(Wang,1979).

th

The14 Wor ldConfer enceonEar thquakeEngineer ing October 1217,2008,Beijing,China

Figure3 ThedistributionofsiltylayersinAdapazar (total6) Figure 4 illustrates the Modified Chinese Criteria (Wang (1979), and Seed and Idriss (1982)), which represents the criteria widely used for defining potentially liquefiable soils over the past two decades. Accordingtothesecriteria,fine(cohesive)soilsthatplotabovetheAlineareconsideredtobeofpotentially liquefiabletypeandcharacterif:(1)therearelessthan15%clayfines(basedontheChinesedefinitionof claysizesaslessthan0.005mm),(2)Theliquidlimit wLis 35%,and(3)acurrent insitu watercontent greaterthanorequalto90%oftheliquid limit.

Figure4ModifiedChineseCriteria(AfterWang(1979)andSeedandIdriss(1982))

th

The14 Wor ldConfer enceonEar thquakeEngineer ing October 1217,2008,Beijing,China

TokimatsuandYoshimi(1983)wereoftheopinionthatsoilswithclaysizeparticlesgreaterthan20%arenot susceptibletoliquefaction.AccordingtoIshiharaet al.(1981)whostudied minetailings composedofsilty sandsandsandysilts,cyclicstrengthisindependentofvoidratioandgrainsizeandincreaseswithincreasing plasticityindex.IshiharaandKoseki(1989)andYasudaetal.,(1994)claimthataspercentageoffines,clay percentage and plasticity index increase, cyclic strength increases slightly (as reported by Polito, 1999). Polito(1999)impliedtoobtainmeaningfulresultsthatplasticityinvestigationsshouldnotbedoneonlyfor fine particles of a soil sample, but whole sample should betestedevaluate liquefaction potential.Andrews andMartin(2000)reevaluatedtheliquefactionfieldcasehistoriesfromthedatabaseofWang(1979),aswell as a number of subsequent earthquakes, and have transposed the Modified Chinese Criteria to U.S. conventions(withclaysizesdefinedasthoselessthanabout0.002mm).AndrewsandMartinrecommended: (1)thatsoilswithlessthanabout10%clayfines(<0.002mm),andaliquid limit(wL)intheminus#40sieve fractionoflessthan32%,beconsideredpotentiallyliquefiable,(2)thatsoilswithmorethanabout10%clay finesandwL>32%areunlikelytobesusceptibletoclassiccyclicallyinducedinitialliquefaction,and(3)that soilsintermediatebetweenthesecriteriashouldbesampledandtestedinlaboratorytoassesswhetherornot theyarepotentiallyliquefiable. The socalled Chinese criteria worked much better and it is proposed that the liquefaction potential of Adapazarnonsensitive/nonplasticsiltscanbeidentifiedbyhavingsatisfiedallofthethefollowingcriteria (nalpandArel,2002).AsiltisidentifiedasliquefiableifitclassifiesasMLandtheliquidlimitislessthan 30,claycontent(<2mm)lessthan15%,andliquidityindexhigherthan1.Subsequentinvestigatorsdeveloped similarapproachestojudgesubmergedsiltsduringearthquakesofMw>7byrequiring (nalpetal.,2006). a) Liquidlimit<33% b) LiquidityindexIL 0.9 c) Claycontentlessthan10% d) AverageparticlesizeD50 smallerthan0.06mm Inaddition,adescriptionofthegrayzoneswhereadvancedtestingwouldberequiredisgiveninTable1 (nalpet.al.,2008) Table1EvaluationofAdapazarLiquefactionCriteria
Criteria Liquefaction Testzone No Liquefaction wL wL33 33<wL35 wL>35 Clay(%C) %C10 10<%C15 %C>15 D50(mm) D50>0.06 0.02<D500.06 D500.02 IL orwn/wL IL orwn/wL0.9 0.75 IL orwn/wL<0.9 IL orwn/wL<0.75 Value 1.0 0.5 0.0

4.1LiquefactionMap ofAdapazar
InpreparingtheliquefactionmapofAdapazarbytheuseofGIS,allsoilswereinitiallyevaluatedusingthe cyclic stress ratio method CRR/CSR (Seed and Idriss, 1971 and 1981). However, since fine grained soils werealsoencountered,themapwasrevisedbyexcludingthesiteswheresandswerenotpredominant.This providedthemapforcoarsegraineddeposits.TheAdapazarCriteriawerethenappliedtopointswheresilts werepresentbymakinguseofTable1.Accordingly,alayerofsoilwasgraded1ifthecriteriaweresatisfied and zeroifitdidnotliquefy.0.5wasusedforcasesthatrequiredfurthertesting.Thisenabledtheinvestigator toappointfourvaluestoeachlayerusingdifferentpropertiesinTable1. Thetotalsintheendvariedbetweenoneandfour,where4representedsoilsthatareliquefiableand0shows soils satisfying none of the criteria employed. Since the values may attain values between 0 and 4, thus requiringfurthertesting,aclassificationshowninTable2wasadopted. Table2Liquefactionevaluation forfinegrainedsoils
Liquefaction Yes Test No Values 3.5 S Value 4.0 2.5 S Value<3.5 S Value<2.5

th

The14 Wor ldConfer enceonEar thquakeEngineer ing October 1217,2008,Beijing,China

Thedatabasewasthenqueriedforclassofsoil,liquidlimit,averagegrainsizeandliquidityindex(orwn/wL) foreachofthesix1.5mthicklayers.Thisresultedinsixmapsforeachlayer,providing36maps.Eachmap was then evaluated fortheparameterit represented.This resulted in maps withregions representing0,0.5 and 1. The maps were ultimately superposed to obtain the sums using Table 2 for evaluation. Figure 5 presents the maps obtained. Black regions indicate liquefiable coarse grained soils found using the CRR/CSR.Itshouldberememberedthatanotherevaluationwouldbenecessaryhere,totaketheinfluenceof theoverburden onliquefaction(Ishihara,1985).

Figure5LiquefactionVulnerabilityMapsofAdapazar

th

The14 Wor ldConfer enceonEar thquakeEngineer ing October 1217,2008,Beijing,China

Red areas represent soils which satisfy the Adapazar criteria, blues the nonliquefiable, whereas yellow shadesindicateregionswhereliquefactionispossiblebutmaynotbeprobable,thusrequiringfurthertesting. Whiteshades indicateareas ofcoarsegrainedsoils wherethe factorofsafetyagainst liquefactionis above unity and areasoffinegrainedsoilswhereadoptedtotalcriteriavalueequalszero. 5.CONCLUSIONS Adetailedinvestigationfollowingthe1999Marmaraearthquakeconcerningtheliquefactionoffinegrained soils of Adapazar through in situ liquefaction observations, borings, soundings and laboratory testing has enabledtheresearcherstoestablishtheAdapazar GeotechnicalDatabase. TheAdapazar Criteria,similartotheChineseCriteria, havebeendevelopedasaresultofthisresearch.The Adapazar CriteriastipulatethatasoilmustclassifyasML,havealiquidlimit 33,containlessthan10%2 mm sizeclay,haveanaverageparticlesizeD50 of 0.06mm,andaninsitu(natural)watercontent higherthan 90%oftheliquidlimit(wn/wL>0.9)inordertoliquefyduringearthquakesofMw>7. Silt contentdepthliquefaction maps extracted from the database were prepared for the top 10 m where liquefactionismostlikelytooccur.Theseismicbehaviouroffineandcoarsegraineddepositswereevaluated. TheCRR/CSRapproachwasusedtoforthesands,whereas Adapazar criteriaandobservationsonrecordwere usedforfinegrainedsoils.Altoughapictureindicatingsignificantareasofliquefactionemerged,itwouldbe necessarytoconsidertheinfluenceoftheoverburdenontheabilitytoliquefybeforepassingajudgementon vulnerability. Researchisongoingtoincludethisfeature tofinalisetheAdapazar liquefactionmap. REFERENCES
th Andrews,D.C.A.andMartin,G.R.,(2000).Criteriaforliquefactionofsiltysoils.Proc.,12 WorldConf.on EarthquakeEngineering,NewZealand,PaperNo.0312.

Bol,E.,(2003). Geotechnical PropertiesofAdapazar Soils,PhD.Thesis,SakaryaUniversity.Adapazar, p. 195, Turkey (inTurkish). Bol, E., zocak, A. and Arel, E., (2005). Liquefaction of Silty Soils and Adapazar Liquefaction Map. InternationalConferenceEarthquakeEngineeringin21Century,EE21C,Skopje/Ohrid. Bol, E., Arel, E. and nalp, A. (2007). The effects of local site conditions on earthquake damage: the Adapazarcase.SixthNationalConferenceonEarthquakeEngineering,1620October2007,Istanbul,Turkey. Bray,J.D.,Sancio,R.B.,Durgunolu,T., nalp,A.,Youd,T.L.,Stewart,J.P.,Seed,R.B.,etin,.K.,Bol,E., Baturay,M.B.,Christensen,C.,KaradayilarT.,(2004).SubsurfaceCharacterizationatGroundFailureSitesin Adapazar,Turkey.JournalofGeotechnicalandGeoenvironmentalEngineering(ASCE)30:7:673685. Ishihara,K.,Yasuda,S.,Yokota,K.,(1981).Cyclicstrengthofundisturbedminetailings.InProceedingsofthe InternationalConference onRecent Advances inGeotechnicalEarthquakeEngineeringandSoilDynamics, UniversityofMissouriRolla,St.Louis,Mo.1:5358.
th Ishihara, K., (1985).Stability of Natural Deposits During Earthquakes. Proc. of the 11 Int. Conf. on Soil Mech.andFoundationEng1:321376,SanFrancisco.

Ishihara, K., and Koseki, J., (1989). Discussion on the cyclic shear strength of finescontaining sands.

th

The14 Wor ldConfer enceonEar thquakeEngineer ing October 1217,2008,Beijing,China

Earthquakes Geotechnical Engineering, Proceedings of the Eleventh International Conference on Soil MechanicsandFoundationEngineering,101106,RiodeJaneiro. Komazawa,M.,Morikawa,H.,Nakamura,K.,Akamatsu,J.,Nishimura,K.,Sawada,S.,Erken,A.,nalp., A., (2001). Bedrock Structure in Adapazar, Turkey A Possible Cause of Severe Damage by the 1999 KocaeliEarthquake,SoilDynamicsandEarthquakeEngineering,Philadelphia.
th nalp, A. and Arel. E., (2002). Liquefaction potential of silty soils: The Adapazar Criteria, 9 Turkish CongressSoilMech.andFound.Eng.,1:363372,AnadoluUniversity,Eskiehir(inTurkish).

nalp, A., Arel. E. and Bol, E., (2001). A general assesment of the effects of 1999 earthquake on the soilstructureinteractioninAdapazar.JubileePapersinHonourofProf.Dr.ErgnTorol,stanbul,Trkiye.
h nalp,A.,Bol,E.,andUral,N.,(2006).Liquefactionofsilts:ImprovingtheAdapazarcriteria.11 Turkish CongressSoilMech.andFound.Eng.,Trabzon,Turkey(inTurkish).

nalp, A., Arel, E., Bol, E., zocak, A. and Sert S.,(2008). Application of Cone Penetration Disssipation Method for Determination of Liquefaction Potential, Research Report , The Scientific and Technological ResearchCouncilofTurkey,ProjectNo:104M387,pp:250,Ankara,Turkey(inTurkish). Polito,C.,P.(1999).TheEffectsof NonPlasticandPlasticFinesonTheLiquefactionofSandySoils.Ph.D. Thesis,VirginiaPolytechnicInstituteandStateUniversity,USA. Sancio,R.B.,Bray,J.D.,Stewart,J.P.,Youd,T.L., Durgunolu,H.T.,nalp,A.,Seed,R.B., Christensen, C., Baturay, M.B., Karadayilar, T. (2002). Correlation Between Ground Failure and Subsurface Soil ConditionsinDowntown Adapazar,Turkey,.SoilDynamicsandEarthquakeEngineering22:10931102. Sancio, R.B., (2003). Ground failure and building performance in Adapazar, Turkey. PhD Thesis, UniversityofCalifornia,Berkeley,p.485. Seed,H.B.andIdriss,I.M.,(1971).Simplifiedprocedureforevaluationsoilliquefactionpotential.Journalof theSoilMechanicsandFoundationsDivision,ASCE97:9:12491273. Seed,H.B.,Idriss,I.M.,(1981).Evaluationofliquefactionpotentialofsanddepositsbasedonobservationof performance in previous earthquakes. Preprint No. 81544, n Situ Testing to Evaluate Liquefaction Susceptibility,ASCEAnnualConvention,St.Louis. Seed,H.B.andIdriss,I.M.(1982).Groundmotionandsoilliquefactionduring earthquakes,Monograph, Earthquake EngineeringResearchInstitute,Oakland,Ca. Seed,H.B.,Idriss,I.M.,andArango,I.,(1983).Evaluationofliquefactionpotentialusingfieldperformance data. JournalofGeotechnicalEngineering,ASCE109:3:458482. Tokimatsu,K.,andYoshimi,Y.,(1983). EmpiricalcorrelationofsoilliquefactionbasedonSPTNvalueand finescontent.SoilsandFoundations,JSSMFE 23:4:5674. Wang, W., (1979). Some findings in soil liquefaction. Water Conservancy and Hydroelectric Power ScientificResearch nstitue.Beijing,China. Yasuda,S.,Wakamatsu,K.,Nagase,H.,(1994). Liquefactionofartificiallyfilledsiltysands.GroundFailures UnderSeismicConditions.GeotechnicalSpecialPublicationNo44,ASCE91104.