abstract volume 6 swiss geoscience meetinggeoscience-meeting.scnatweb.ch/...symposium_09.pdf · it...
TRANSCRIPT
Abstract Volume6th Swiss Geoscience MeetingLugano, 21st – 23rd November 2008
Apply!Geosciences
6th Swiss Geoscience Meeting 2008 - Lugano6th Swiss Geoscience Meeting 2008 - Lugano
Institute of Earth Sciences
2�0Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s �. Natural Hazards and RisksGiovanni Crosta, Michel Jaboyedoff
Institut de géomatique et d’analyse du risque, Université de Lausanne Dipartimento Scienze Geologiche e Geotechnologiche, Università di Milano-Bicocca
9.1 AkyüzH.Serdar,TaylanSançar,CengizZabcx,PxnarGutsuz,VolkanKarabacak,ErhanAltunel,ZiyadinÇakxr,ÇaglarYalçxner:PaleoseismologicalstudiesonYedisuseismicgap,easternpartofNorthAnatolianFault,Turkey
9.2 AlizadehBahram,BagheriSoheila,HoseiniSeyedHossein:Biomarkersaseffectiveandbeneficialtoolsinpetroleumcausednaturalhazards
9.3 AmbrosiC.,PeraS.:Interdisciplinaryapproachestorecognition,analysisandmodellinginsackungsystemandlargelandslidesinsouthernSwissAlps
9.4 AmbrosiC.,StrozziT.:SarInterferometricPointTargetanalysisandinterpretationofaerialphotographsforlands-lidesinvestigationsinsouthernSwitzerland
9.5 BaruffiniM.,BaruffiniM.,ThüringM.:AGIS-toolforriskassessmentduetonaturalhazardsinmountainregions
9.6 FischerL.,AmannF.,HuggelC. : Multidisciplinaryinvestigationsandback-analysisofaperiglacialrockfallevent:Tschiervarockfall
9.7 ForootanEhsan,SharifiMohammadAli,NikkhooMehdi,DodgeSomayeh: Applyingaltimetryand in-situdata tocomputepoint-wiseMSLforinlandwaters,casestudy:CaspianSea
9.8 FossatiD.,KosA.: TheDeepSeatedGravitationalSlopeDeformationof Landarenca (Graubünden,Switzerland):AGeological-GeotechnicalAnalysis
9.9 HuggelC., Eugster S., Ramírez J.M.,WorniR.: Recent experiences fromSwissprojects in risk reduction in SouthAmerica
9.10 JaboyedoffM.,PedrazziniA.:Theoreticalbasisforshadowanglevariabilityandimplications
9.11 KanevskiM.,PozdnoukhovA.,TimoninV.:Machinelearningalgorithmsforspatialdata.Casestudies:environmentalpollution,naturalhazards,renewableresources
9.12 KünzlerM.,HuggelC.,RamírezJ.M.:Amethodforriskanalysisrelatedtolaharsandfloods–acasestudyatNevadodelTolimavolcano,Colombia
9.13 MattiB.:Geologicalheterogeneityinlandslides:Characterizationandflowmodeling
9.14 MautzR.:Simulationforavolcanomonitoringnetwork
9.15 MeierA.,WilliC.:SystematicrecordingandanalysisofnaturalhazardsalongrailwaylinesusingGIS
9.16 MohamadiMahin: Glass andmagnetitic Spherules associatedwith the solid impact andmass extinction inD/CboundaryinCentralAlborzmountainnorthofIran
9.17 OppikoferT.,BöhmeM.,BlikraL.H.,DerronM.-H.,JaboyedoffM.,SaintotA.:GeologicalandstructuralmodeloftheÅknesrockslide(Norway)
9.18 OstermannM.,SandersD.:Anewapproachtodatingcarbonate-lithicrockslides
9.19 PedrazziniA.,JaboyedoffM.,FroeseC.,HumairF.,LangenbergW.,FranciscoM.:Theroleofregionalfaultandfold-relatedfracturesinthedevelopmentofrockslopefailure
9.20 RossetP.,BonjourC.,CuaG.,KaestliP.,TrendafiloskiG.,WiemerS.,WyssM.: TheearthquakelossestimatingtoolQLARM:Applicationsinreal-timeandforpredictions
9.21 SchweizerJ.:Onthepredictabilityofsnowavalanches
2�1
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks
9.22 StrozziT.,WernerC.,WiesmannA.,WegmüllerU.:Aportableradarinterferometerforthemeasurementofsurfacedeformation
9.23 ThüringM.:Rockfallmodellingappliedtorockfallprotectiondesign
9.24 ThüringM.,CannataM.,HammerJ.:Hazardandriskassessmentoflandslidesonaccumulationreservoirs–afieldapplicablescheme
9.25 TroisiC.:LandslideinvestigationbymeansofPSiNSARTMradarinterferometry:thePiemonteexperience
9.26 VouillamozN.,MosarJ.:MicrozonagesismiqueduCantondeFribourg:CartedeSoldefondations
9.27 ZechnerE.,LewinI.,KonzM.:Effectsoftectonicstructures,groundwaterpumping,andminingactivityonevaporitesubrosionandresultinglandsubsidence
�.1
Paleoseismological studies on Yedisu seismic gap, eastern part of North Anatolian Fault, Turkey
H.SerdarAkyüz*,TaylanSançar**,CengizZabcı*,PınarGutsuz**,VolkanKarabacak***,ErhanAltunel***,ZiyadinÇakır*,ÇağlarYalçıner***
*Istanbul Technical University, Faculty of Mines, Department of Geology, 34469, Maslak, Istanbul ([email protected])** Istanbul Technical University, Eurasia Institute of Earth Sciences, 34469, Maslak, Istanbul ***Eskisehir Osmangazi University, Engineering Faculty, Department of Geology, 26480, Eskisehir
NorthAnatolianFault(NAF)isoneofthemostactivemajorfaultsinEurope.IthasageneralE-Wtrendwithabout1500kmlengththroughoutnorthernTurkeyand22±3mm/yearsliprate.During20thcenturyitisalmosttotallybrokenexcepttwoimportantseismicgaps;MarmaraSeismicSegmentinthewestandYedisuSeismicSegmentintheeast(Figure1).ThispaperpresentspaleoseimologicalstudiesonYedisuSeismicSegment(YSS).
YSSlocatesoneasternpartofNorthAnatolianFaultbetweenErzincancityandYedisutownofBingölcity(Figure2).WesternpartoftheYSSwasbrokenin1992withaMs=6.8earthquakewhileeasternpartwasbrokenin1949withMs=6.7earthquake(Figure1).A80km-longfaultisremainedunbrokenbetweeneasternpartofErzincanBasinandYedisutown.AccordingtoAmbraseys&Finkel(1995)thispartofthefaultwasbrokenlastlyin1784,i.e.224yearsago.TheaimofthisstudywhichissupportedbyTUBITAK(TheScientific&TechnologicalResearchCouncilofTurkey,Projectno:106Y174)istorevealseismicriskofYSSwithpaleoseismologicaldata.
Thefaultgeometry,segmentationandoffsetstructureswereanalysedfirstlyonaerialphotographsandtopographicalmaps,thenobservedanddrowninthefieldona1/25000topographicalmap.Threetrenchsiteswerealsodefinedinconvenientplaces.FirstsitewaschoseninthewesternpartoftheYSS,Sarikayasite(Figure2).Trenchwasopenedperpendiculartothefaulttraceandtrenchwallwaslogged.Finegrainedslopedepositsandfaultbranchesindicatetwoearthquakeevidences.SecondandthirdtrenchsitesareclosetoeasternendofYSS(Figure2).KarapolatTrenchwasopenedperpendiculartofaultondistalpartofalargerecentfandeposits.Trenchstratigraphyevidencedtwopastearthquakesaswell.Thelasttrench,namedasTokmanik,wasdiggeddeeperduetoconvenientundergroundwaterlevel.Atleast5pastearthquakeeventsweredeterminedonthetrenchwalls.CharcoalsamplescollectedincriticallevelinalltrenchesweresenttoArizonaUniversityRadiocarbonLaboratoryfordating.DatingresultstogetherwithhistoricalrecordswillgiveusearthquakerecurrenceperiodofYedisuSeismicSegment.Ifthedateofpastearthquakeshaveperiodicalbehaviour,thiswillhelptopredictthedateoffutureearthquakeonYedisuFault.
REFERENCESAmbraseys,N.N.& Finkel,C.F. 1995: The Seismicity of Turkey and adjacent areas: ahistorical review, 1500-1800. Eren,
Istanbul,p240.Barka,A.,Akyüz,H.S.,&18others,2002:ThesurfaceruptureandslipdistributionoftheAugust17,1999²zmitearthquake,
M=7.4,NorthAnatolianFault.Bull.Seism.Soc.Amer.92,43-60.
2�2Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s
Figure1.20thcenturydestructiveearthquakes(Barkaetal.2002)andseismicgapsonNorthAnatolianFault.
Figure2.ThegeometryofYedisuSeismicSegmentandlocationoftrenchsites.
�.2
Biomarkers as effective and beneficial tools in petroleum caused natural hazards
AlizadehBahram*,BagheriSoheila*&HoseiniSeyedHossein*
*Department of Geology, Faculty of Science, S. Chamran University, Ahvaz, Iran. ([email protected])
IntheZagrosoilrichbasincomprisingmorethan40hugeandtengiantoilfields,unavoideblanaturalhazardssometimesevenforcepeopletoevacuatetheirhomes.Theaimofthisstudywastodeterminethegeneticoriginofoilseepagesoccur-ringinMasjid-e-Soleiman(MIS)oilfieldandinDalakirivernearNargesioilfield.InordertoinjectgasintheNargesiOilfield(SouthDezfulEmbayment),andatthesametimepreventnaturaldisasteritisnecessarytoestablishtheoriginofoilseepspollutedwithH
2SinDalakiarea.
NumerousoilseeppollutedwithH2ShavebeenfoundinMasjid-e-Soleimanoilfield(firstoilfieldintheMiddleEast,North
2�3
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ksDezfulEmbayment),whichpollutedsomeareas,forcinglocalpeopletoleavetheirhouses.
AsmarioilsamplesfromMasjid-e-SoleimanoilfieldcorrelatedwithoilseepsfromSeeberenjandDare-KhersanandAsmari/Jahrum reservoir oil from Naregsi oilfield correlated with Dalaki’s oil seepages by Gas Chromatography and GasChromatography-MassSpectrometry.Dalaki’soilseepsexhibitevidencesofbiodegradationandmixkerogens,thisexplainsthelowsaturatefraction(29-38%),thearomatic-asphalticnatureoftheoilseepsandanimportantdepletioninthehomohopaneseries.Theoilseepsarecha-racterizedbyhighpredominanceC
29toC
30hopaneratios(1.25-1.8),lowratiosofC
34overC
35(1.42-1.62),andthelowdiasterane
abundance (C27-C
29Dia/Reg Steranes: 0.33-0.44). These characteristics suggest that the oil seeps originate from carbonate
marinesourcerocks(Sañchez,C.&Permanyer,A.,2006).OntheotherhandthereservoiroilsamplesofNargesioilfieldarecharacterizedbyhighcontentsofsaturates,Pr/Phratiosbelow2,thedominanceofC
30hopanesovertheC
29hopane(C
29/
C30<1),lowratiosofC
34/C
35(1.02-1.4)homohopanesanddiasteranesaremoreabundant.Thesepropertiessuggestthatthe
oilsweregeneratedmainlyfrommarineshalefacies.TheC29/C
30hopaneandDia/Regsteraneratiosandarelativelyhigh
abundanceofoleananeintheoilseepagesgivesthebestevidenceofdifferentsourcerocksforoilandoilseepsamples.Thisisalsosupportedwiththermalmaturityparameterssuchαα20/(20S+20R)steraneandT
s/T
mratios.
ThediscrepancyinthebiomarkerindicatorsgivestheassuranceofdiverseoriginofoilsfromreservoirandfromSeepages(Fig.1a).Finallytheconstantseepagedischargeevenaftergasinjectionintothereservoirprovedtheoilcorrelationtobeveryreliableandtrustworthy.InMISoilfield,geochemicalparameters,suchashighratioofC
27overC
29sterane(1.01-1.2),thelowratioofC
29toC
30hopanes
(0.63-0.71), aswell as the stardiagramof tricyclic terpane showed thebest similaritybetweenoil andoil seeps. For theAsmarioilandoilseepages,highratiosofsteranestohopanesareverytypical(0.60foroiland0.48-0.63foroilseeps)andlowratioofC
34overC
35homohopanes(0.84-1.15)thatarecharacteristicsforalgaeorganicmatterdepositedwithinanoxic
environment(Alimietal.,2007).ThisisalsosupportedbylowamountsofC30moretanerelativetoC
30hopane(C
30βα/C
30αβ:
0.09–0.11),indicativeofstrongmarineinputtothesourcerock.BothoilandoilseepspresentlowC26overC
25(0.66-0.95)and
C24/C
23(0.45-0.68)tricyclicterpaneratios(lessthan1),lowabundanceofC
29/C
30hopane(0.63-0.71)andtheoccurrenceofdia-
steranes,suggestcarbonate-marlsourcefaciesforoilandoilseeps.Theratiosof20S/(20S+20R)forαααC29steranes(0.48-0.51)
andββ/(αα+ββ)for5α-C29steranes(0.5-0.54)havebeenevaluatedtogetherwiththeratiosof22S/(22S+22R)forC
32homoho-
panes(0.52-0.54).Basedontheseratios,allsamplesarewellwithintheoilwindow(Rogersetal.,1999).Age-specificbiologi-calmarkerincluding,ExtendedTricyclicRatio (0.5-0.52),C
28/C
29 (0.79-0.89)steraneratios,andOleananeIndex(<0.2)show
thatoilseepshavesimilarsourceageasUpperJurassic-Cretaceous.
a
Figure1.CrossplotofC29/C
30hopaneandC
34/C
35homohopane,a,clearlyindicatingtwodifferentorganicfacies.Thetrianglediagramof
diasteranedistributioninMISoilfield,b,demonstratingsimilardepositionalenvironmentforbothreservoiroilsandoilsepages.
Thegeochemicalparametersshowedthattheoilsandoilseepswereformedinsimilardepositionalenvironmentandhaveidenticalthermalmaturitylevel(earlyoilwindow).Intrianglediagram,(Fig.1b),theC
27-C
28-C
29diasteranedistributionfor
Asmarioilandoilseepsarelocatedinthesamevicinityindicatingthesourceofinitialorganicmatterbeingoneandthesame.FinallyitcanbeconcludedthatthesourceofpollutionintheareaisAsmarireservoiroilandduetoitspressuredrop,thepreviouslycondensedH
2Spoisonousgasisnowfreetoescapethroughcaprockfractures,forcingpeopletoevacuate
theirhouses.Allthesewelldemonstratetheproficiencyofgeochemicalparametersinpredictingandmanagingnaturalhazards.
REFERENCESAlimi,H.,Alizadeh,B., Jarvie,D.M.,Adabi,M.H.,Tezheh,F., Jarvei,B. (2007).GeochemicalEvaluationofCrudeOils from
AsmariandBangestanReservoirs inMarunOilfield,SWIRAN.23rd InternationalmeetingonOrganicGeochemistry,IMOG,Torquay,UnitedKingdomSep.2007.
Sa´nchez,C.,Permanyer,A.,2006.OriginandalterationofoilsandoilseepsfromtheSinu´-SanJacintoBasin,Colombia.OrganicGeochemistry,v.37,p.1831-1845.
Rogers,K.M.,Collen,J.D.,Johnston,J.H.,Elgar,N.E.,1999.AGeochemicalappraisalofoilseepsfromtheEastCoastBasin,NewZealand,OrganicGeochemistry,v.30p.593-605.
2��Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s �.3
Sar Interferometric Point Target analysis and interpretation of aerial photographs for landslides investigations in southern switzerland
AmbrosiChristian*,StrozziTazio**
* SUPSI, Istituto Scienze della Terra CP 72, CH 6952 Canobbio ([email protected])** GAMMA Remote Sensing, Worbstrasse 225, CH 3073 Gümligen([email protected])
InformationonlandslidedisplacementfromSARInterferometricPointTargetAnalysis(IPTA)andsketchmapsfromaerialphotographyinterpretationarecombinedforthestudyoflandslidesinTicino,SouthernSwitzerland.ForcurrentITPAinve-stigations,ENVISATandRADARSATSARacquisitionsovertheSwissterritoryareused.Numerousunstablephenomenaareconsideredinthismountainousregion,withanelevationrangefromapproximately200ma.s.l.tomorethan3000ma.s.l.TheresultsachievedwithIPTAareattractivetocomplementaerialphotographsinterpretationfortheevaluationofthestateofactivityoflandslidesovervillagesandinsparselyvegetatedareaswithnumerousexposedrocks.Ontheotherhand,overvegetatedareas(forestsandmeadows)IPTAfailedtoretrievedisplacementinformation.BecausedisplacementfromInSARisrecordedalongthesatelliteline-of-sightdirection,IPTAcannotbedirectlyuseforthedeterminationoftheintensityoflandslidesinhazardmapping.Ingeneral,theactualdisplacementrateislargerthanthatrecordedwithInSAR.Overalpineareascharacterizedbysparsevegetation,wheresnowcover limits theavailabilityofa largenumberofSARacquisitions,conventionalInSARwassuccessfullyappliedtoestimatethemotionofrockglaciersandotherperiglacialphenomena.ForvegetatedareasandrelativelyrapidlandslidesL-bandInSAR(JERS-1SARandALOSPALSAR)hasbeenfoundtobeanefficientsolution.
�.�
Interdisciplinary approaches to recognition, analysis and modelling in sackung system and large landslides in southern Swiss Alps
AmbrosiChristian*,PeraSebastian*
* SUPSI, Istituto Scienze della Terra CP 72, CH 6952 Canobbio ([email protected])
AverylargeDeep-SeatedGravitationalSlopeDeformationsandlargelandslides,affectingtheslopesofSwissSouthernAlps,havebeenrecognisedandmapped.Severalactiverotationalandtranslationaldeepseatedlandslidesoccurinthelowerpartoftheslopes,ofteninassociationwithrockslidesatdifferentscale.Thesephenomenashowadifferentstageofevolutionalongtheflanksoftheridges.Someoftheselandslides(Pontirone,Osco,Faido)hasbeenrecognisedthroughdifferenttechniques.Inparticular,ithasbeencharacterisedbyaerialphotointerpretation,fieldsurveys,tracertestsandanalysisofdetailedairborneLidarDEM.Rockmasses, includingpoly-deformedorthogneisses, paragneisses anddolostones, are affected by systemsof impressivegravitationalmorpho-structuresincludingENEtrendingopentrenches,scarpsandcounterscarpsforminggraben-likestruc-tures.Geodeticmeasurements(opticaltargetsandGPSpoints)andsatelliteradarinterferometry(DinSAR)datademonstratetheactivityofthegravitationalfeaturesaffectingboththeupperandlowerpartoftheslopes.Theanalysisofgeological,geomorphological,hydrogeological, in situ stress,geodeticandupliftdata,havebeenused todemonstratethepossiblerelationshipsamonggeological,structural,topographicandgravitationalfeatures.Numerical2Dmodellinghavebeenusedtoevaluatefailureinpresenceofelastic,elasto-plasticmaterialsandgroundwaterflow.Thisallowtoverify“model”sensitivitytosomefactorsasslopegeometry,structuralfeatures,groundwaterconditionsandtodetermi-natethetriggersfactorsoftheseinstabilities.Onthisbasiswediscussthepossibilitythatthegeological,tectonicandhydrogeologicalframeworkcanplayanactive/pas-siveroleintheonsetanddevelopmentofthesackungsystemandrelatedlandslides.
2��
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks�.�
A GIS-tool for risk assessment due to natural hazards in mountain regions
BaruffiniMirko*,BaruffiniMoreno*&ThüringManfred*
*Istituto Scienze della Terra IST-SUPSI, C.P. 72, CH-6952 Canobbio ([email protected])
Duringthelastdecadesland-useincreasedsignificantlyintheSwiss(andEuropean)mountainregions.Duetothescarcenessofareassuitablefordevelopment,anthropicactivitieswereextendedintoareaspronetonaturalhazardssuchasavalanches,debrisflowsandrockfalls(Smith2001).Consequently,anincreaseinlossesduetohazardscanbeobserved.Tomitigatethe-seassociatedlosses,bothtraditionalprotectivemeasuresandland-useplanningpoliciesaretobedevelopedandimplemen-tedtooptimizefutureinvestments.Efficientprotectionalternativescanbeobtainedconsideringtheconceptofintegralriskmanagement.
Riskanalysis,asthecentralpartofriskmanagement,hasbecomegraduallyagenerallyacceptedapproachfortheassess-mentofcurrentandfuturescenarios(Loat&Zimmermann2004).Theprocedureaimsatriskreductionwhichcanberea-chedbyconventionalmitigationononehandandtheimplementationofland-useplanningontheotherhand:acombina-tionofactiveandpassivemitigationmeasuresisappliedtopreventdamagetobuildings,peopleandinfrastructures.
Consideringdifferenthazardprocessesandtheirimpactonthebuiltenvironment,multiplesolutionsfortheprotectionofnewbuildingsandinfrastructuresandtheupgradeofexistinginventoryexist.Consequently,theconceptoflocalprotectionshouldbeembeddedwithintheframeworkofintegralriskmanagementstrategies.Plannedearly,expendituresfortheim-plementationoflocalstructuralmeasuresarecomparativelylowrelatedtothetotalcostoftheplannedconstruction.Recentstudies suggested a considerable decrease in vulnerability, if local structural protection is implemented (Holub& Fuchs2008).
AspartoftheSwissNationalScienceFoundationProject54“Evaluationoftheoptimalresilienceforvulnerableinfrastruc-turenetworks-Aninterdisciplinarypilotstudyonthetransalpinetransportationcorridors”westudythevulnerabilityofinfrastructuresduetonaturalhazards.
TheSwisssystemforriskanalysisofgravitationalnaturalhazards(BUWAL1999)offersacompleteframeworkfortheanaly-sisandassessmentofrisksduetonaturalhazards,rangingfromhazardassessmentforgravitationalnaturalhazards,suchaslandslides,collapse,rockfall,flooding,debrisflowsandavalanches,tovulnerabilityassessmentandriskanalysis,andtheintegrationintolanduseplanningatthecantonalandmunicipalitylevel.Theschemeislimitedtothedirectconsequencesofnaturalhazards.
Weconductaresearchreferredtotheconceptoftheevaluationoftheresiliencewithintheframeworkofintegralriskma-nagementwiththeaimtodevelopasystemwhichintegratestheproceduresforacompleteriskanalysisinaGeographicInformationSystem(GIS)toolbox,inordertobeappliedtoourtestbed,theAlps-crossingcorridorofSt.Gotthard.
Asimulationenvironment,RiskBox, isdevelopedwithintheopen-sourceGISenvironmentGRASS (GeographicResourcesAnalysisSupportSystem)andadatabase(PostgreSQL)inordertomanageainfrastructuredatacatalog.Thetargetedsimula-tionenvironmentincludestheelementsthatidentifytheconsecutivestepsofriskanalysis:hazard–vulnerability–risk.
Modulehazardintegratesapplicationstosimulatenaturalhazardprocessesinordertoassesstheirdegreeofhazard.Modulevulnerability integratesvulnerableobjectsandattributes themwiththenecessarymeta-data.Modulerisk integrates thenecessaryanalysisapproachesinordertoconducttheriskanalysisbasedontheformertwomodules.
ThefinalgoalofRiskBoxisaversatiletoolforriskanalysiswhichcanbeappliedtoothersituations.Therearespecificneedsforanimprovementofthelevelofinformationforaffectedpeople,legalregulationsandrisktransfermechanisms(ARMONIAProject2007).Theseneedswouldnotonlyresultinanincreasedriskawarenessofpeopleconcerned,butalsoinanenhancedenforceabilityofnecessarylegalregulations,suchasland-useplanningrulesandbuildingcodes.So,theindividualrespon-sibilitycouldbestrengthenedandthesocietywillbeenabledtoanalternativelyuseoftheresourcesinamorecost-efficientway.
TheinitialresultsoftheexperimentalcasestudyshowshowusefulaGIS-basedsystemcanbeforeffectiveandefficientdi-sasterresponsemanagement.WepresenttheconceptandcurrentstateofdevelopmentofRiskBoxanditsapplicationtothetestbed,theAlps-crossingcorridorofSt.Gotthard.
2�6Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s REFERENCESARMONIAProject2007:LanduseplansinRiskyareasfroUnwisetoWisePractices–Materials2ndconference.Politecnicodi
Milano.BUWAL 1999: Risikoanalyse bei gravitativenNaturgefahren -Methode, Fallbeispiele undDaten (Risk analyses for
gravitationalnaturalhazards).Bundesamt fürUmwelt,WaldundLandschaft (BUWAL).Umwelt-MaterialenNr. 107,1-244.
Holub,M.&FuchsS. 2008:Benefitsof local structuralprotection tomitigate torrent-relatedhazards. In:Brebbia,C.A.&Beritatos,E.(eds)2008:RiskAnalysisVI:SimulationandHazardMitigation.InstituteofTechnology,UKandUniversityofThessaly,Greece,401-411.
Loat,R.&Zimmermann,M.2004:LagestiondesrisquesenSuisse(RiskManagementinSwitzerland).In:Veyret,Y.,Garry,G.,MeschinetdeRichemont,N.&ArmandColin (eds) 2002:ColloqueArchede laDéfense22–24octobre2002,dansRisquesnaturelsetaménagementenEurope,108-120.
Smith,K.2001:Environmentalhazards.Assessingtheriskandreducingdisaster.Thirdedition.London
�.6
Multidisciplinary investigations and back-analysis of a periglacial rock fall event: Tschierva rock fall
FischerLuzia*,AmannFlorian**,&HuggelChristian*
* Glaciology, Geomorphodynamics & Geochronology, Department of Geography, University of Zurich, Switzerland ([email protected])** Group of Engineering Geology, Department of Earth Science, ETH Zurich, Switzerland
Slopestabilityofsteeprockwallsinglacierisedandpermafrost-affectedhigh-mountainregionsiscontrolledbymanydiffe-rentfactorssuchasgeologicalandgeomechanicalcharacteristics,topography,hydrologyandalsoglaciationandpermafrostoccurrence.Changesinoneormoreofthesefactorsmayreducetheslopestabilityandeventuallyleadtoarockfallevent.Theatmosphericwarmingduringthe20thcenturyhascausedpronouncedeffectsintheglacialandperiglacialbeltsofhighmountainareas.Thechangesaremadestrikinglyevidentby,forexample,theretreatofAlpineglaciersandlessimmediate-lyvisiblebutalsoverysignificantarechangesinmountainpermafrostdistributionandtemperature.Therefore,cryosphericfactorsaremostpronetoongoingclimatechanges.Inthisstudytheinfluenceofdifferentfactorsandmechanismsdetermi-ningslopestabilityisinvestigated,basedonacasestudyoftheTschiervarockfallevent.
The Tschierva rock fall occurred on October 19, 1988 from thewestern flank of PizMorteratsch (3751m asl, Engadin,Switzerland)onTschiervaglacierwithanestimatedvolumeofapproximately0.3x106m3.Therockmassdetachedatabout3200ma.s.l.andfellontheTschiervaglacierprobablyincorporatingadditionalrock-debrisfromtheslopebelowthedetach-mentzoneandstoppedontheglacieratanelevationofabout2700ma.s.l.
Theprimaryobjectiveofthepresentedstudyistheback-analysisoftheTschiervarockfallandtheinvestigationofpossiblegeological,geomechanicalandclimate-relatedglaciologicaldispositionfactors.Therefore,thedetachmentzoneisinvestiga-ted in amultidisciplinary approach based on in-situ geological field work, associated geotechnical investigations,mor-phometric analyses, permafrostmodelling, glaciation history and subsequent numerical stabilitymodellingwithUDEC(UniversalDistinctElementCodebyItasca).Basedontheseanalyses,theinfluenceandsensitivityofthedifferentfactorsandprocessesontheslopefailuresisassessed.
Numericalslopestabilitymodelingwasperformedtoexaminedifferentscenariosofpossiblefailuremechanisms.Modellingoftheunloadingofthepleistoceneglacialoverburdenshowedthatsubsequentredistributionofstressandstrainfieldsintheflankhaveastronglycontrollinginfluenceonthegeometryofthedetachmentzonebytheopeningofunloadingjoints.Asensitivityanalysisofgeotechnicalparametersadditionallyshowedthatthecohesionofthediscontinuitieswasafunda-mentalparameter.Coupledhydro-mechanicalmodelingdemonstratedthatslopestabilitywasverysensitivetochangesinwaterpressure.Theexistingfaultzonecrossingtherockslopeinducedanelevatedwaterinflowduetothehigherpermea-bilityandmightthereforebe,togetherwiththelong-lastingeffectsoficeunloading,amainfactorfortheslopeinstabili-ty.
Inconclusion,ouranalysiscouldidentifysomeessential factorsofslopestability inrelationwiththegeologicalsetting,glacierretreatandpermafrostdegradation.However,ithasalsoshownthattheunderstandingofthephysicalprocessesandtheadequateintegrationinnumericalslopestabilitymodelsneedsmoreresearch,inparticularinviewofthehazardsthataretobefaced.
2��
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks�.�
Applying altimetry and in-situ data to compute point-wise MSL for inland waters, case study: Caspian Sea
ForootanEhsan*,SharifiMohammadAli*,NikkhooMehdi**,&DodgeSomayeh***
* Surveying and Geomatics Engineering Department, College of Engineering, University of Tehran. Northern Amirabad St, Tehran, Iran, Tel:00982161114251, Fax:00982188008837 ({eforootan | sharifi }@ut.ac.ir)** Surveying and Geomatics Engineering Department, KNT University of Technology, Iran. [email protected]***University of Zurich, Winterthurerstr. 190, CH-8057 Zurich
TheCaspianSeaisthelargestenclosedbodyofwaterontheEarthbyarea,variouslyclassedastheworld'slargestlakeorafull-fledgedsea.TheseaisfedbynumerousriversincludingtheVolga,Ural,Terek,andKurarivers.Inthelast25years,thetotalsurfaceareahasvariedfrom360000to400000km2duetohighwaterlevelvariations.Thesevariationshaveshownoscillationsbetween-26and-29meters(withrespecttothezerooceanlevel)duringthelasthundredsyears.ThelargestuncertaintyfortheCaspianSeaisconcerningwiththeaveragewaterlevel,whichbyitselfisnotsubjecttoastatisticaldistribution,butrathermoretoatrend.Hence,weappliedtwomajordatasetstocomputethewaterfluctuations:1)in-situmeasurements;inthiscaseprovidedbytheIranianCaspianenvironmentalstudycenter.2)derivedwaterlevelsfromsatellitealtimetrymission;weusedT/Paltimeterdata,whichareperformed,attheJetPropulsionLaboratory,CaliforniaandcoverrepeatedT/Pmissioncycles11to400,spanningmorethan10years.Theresultsoftheseobservationsareconsi-deredasasetofvirtualtidegauges,withasamplerateofevery9.915625days.Besidethat,weappliedin-situobservationstorevealthetrendandevaluatethealtimetryresults.AnalysisofT/PobservationsshowthatthereisaphasedifferencesbetweenvariouslocationsintheSea.Infact,80%oftheinputsideofthewaterbalanceinfluxcomesfromtheVolgaRiver.ItseemsthatthecauseofthesedifferencesisrelatedtothedistancebetweenlocationstotheRiverofVolga.ChangeindischargeofVolgadoesnotexpandallovertheseaimmedi-atelybutthelocationsthatarenearertotheoutfalloftheriverareinfluencedsoonerthantheotherlocations.Therefore,itisnecessarytoinvestigatethevariationsofeachpointintheCaspianSeaseparatelyinordertodiscovertheenvironmen-talvariablesusingtimeseriesanalysisandthencomputingthepoint-wiseMSL.InshorttheresultsofouranalysisprovedthatbetweenJanuary1993andMarch1995theCaspianmeansealevelroseatanaveragerateof15mm/monththenfollowedbyafallatarateof-5.5mm/monthuntilJanuary2003.Besidethat,apoint-wiseMSLiscomputedfortheCaspianSeaduring1993to2002.Theresultswillbediscussedinmoredetailinthefullpaper.
REFERENCESAvakian,A.B.andV.M.Shirokov.1994.RationalUseandProtectionofWaterResources.Ekaterinburg:Publ.House"Victor"
(inRussian).Benada, J. R., 1997, TOPEX/POSEIDONUser’sHandbook ,Generation B (MGDR-B) Version 2.0,D-11007, Jet Propulsion
Laboratory,CaliforniaInstituteofTechnologyundercontractwiththeNationalAeronauticsandSpaceAdministration.Birkett,C.M., 1995a:The contributionofTOPEX/POSEIDON to theglobalmonitoringof climatically sensitive lakes, JGR-
Oceans,Vol.100,C12,pp.25,179-25,204CaspianTDA,reportvolume2,2002.,TransboundarydiagnosticanalysisfortheCaspiansea.TheCaspianseaenvironment
programme.,Baku,Azerbaijan.www.caspianenvironment.org
2�8Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s
Figure1.MeanlevelvariationsofSSHalong6passesinCaspianSea.
�.8
The Deep Seated Gravitational Slope Deformation of Landarenca (Graubünden, Switzerland): A Geological-Geotechnical Analysis
FossatiDavid*,KosAndrew*
*Group of Engineering Geology, Geological Institute, Wolfgang-Pauli-Str. 15, 8093 Zürich ([email protected])
Acomplexdeep seatedgravitational slopedeformation (DSGSD) located in the southwesternpart of theCalanca valley(Graubünden,Switzerland),intheareaofLandarencawillbedescribedusinganmulti-disciplinaryapproach,wherefieldsurveys,remotesensingandgeodeticmeasurementsareappliedtothesiteforthefirsttime.TheDSGSDlieswithinisocli-nallyfoldedgneissesandmicaschistsoftheAdulaandSimanonappes.Geomorphologicfeaturessuchasterraces,steeprockcliffs,tensioncracks,scarps,counterscarpsandcombinedscarp-tensioncrackstructurescharacterizethearea.TheareaoftheDSGSDinvolvesatleast1squarekilometrewithanapproximatevolumeof185millioncubicmeters.Anintegrationofresults indicatetheDSGSDtobeslightlyactivewithvelocitiesofupto8mmperyear.Movements involvediscontinuouscreeping,i.e.acombinationofcreepingandslidingandtheDSGSDcanbedefinedasaslightlyactive,deepseatedkinkbandslumping,inadevelopedstage.
REFERENCESAmann, F. 2006:GrosshangbewegungCuolmda Vi (Graubünden, Schweiz) – geologisch-geotechnische Befundeund
numerischeUntersuchungenzurKlärungdesPhänomens.Diss.Univ.Erlangen-Nürnberg.Agliardi,F.,Crosta,G.&Zanchi,A.2001:Structuralconstraintsondeep-seatedslopedeformationkinematics.Eng.Geol.59
(1-2),83-102.Kieffer,D. S. 1998: Rock slumping: A compound failuremode of jointedhard rock slopes. PhDThesis.University of
California,Berkeley.
2��
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks�.�
Recent experiences from Swiss projects in risk reduction in South America
ChristianHuggel*,SebastianEugster**,JuanManuelRamírez***,RaphaelWorni*,****
*Glaziologie, Geomorphodynamik & Geochronologie, Geographisches Institut, Universität Zürich, Winterthurerstrasse 190, 8057 Zürich ([email protected])**Swiss Agency for Development and Cooperation, Lima, Peru***Swiss Agency for Development and Cooperation, Bogotá, Colombia****Department of Environmental Sciences, ETH Zürich
ExtensivepartsofSouthAmericaarecharacterizedbymountaintopographyandseasonally intenserainfall.Accordinglymany regions suffer from repeated landslides, debris flows, floods. Thesehazards are partly linked, and exacerbated byvolcanicactivityalongtheAndeanmountainchain.Additionally,insomeregionsoftheAndes,disastersrelatedtoglaciershaveclaimed thousandsof victims.Suchextremeeventsoccurred in the1970s inPeruwithan ice-rockavalanche fromHuascaranandinthe1980sinColombiawithlaharsfromNevadodelRuiz.Sincethetimeofthesedisastersimportantad-vancesindisasterriskreductionhavebeenmadeonthenationallevels.InColombia,forinstance,thefailuresrelatedtotheRuizdisastertriggeredthecreationofaNationalSystemforPreventionandAttentionofDisasters(SNPAD).However, itcontinuestobeimportantto improvethecurrentsituationstoavoidanysignificanttypeofdisaster.Criticalimprovementscanbemadeonaninstitutionalandcommunity-basedlevel,combinedwiththeimplementationofadvancedtechnologicalandscientificknow-how.Effectivedisasterriskreductiontakesplaceonalocallevelbutisembeddedintheregionalandnationalinstitutionalcontext.HerewereportonrecentexperiencesfromprojectsoftheSwissGovernment(SwissAgencyforDevelopmentandCooperation,SDC) in collaborationwith the University of Zurich and several national institutions in Colombia and Peru. In Peru, aPeruvian–Swissprogrammeonadaptationtoclimatechange (PACC)hasrecentlybeeninitiatedandaimsat identifyingclimatechangeimpactsintheAndeanCuzcoandApurimacregionsinPeru,andtoimplementasetofadaptationmeasurestoreduceadverseeffectsofclimatechange.ThePACCfocusesonthreemajorareas:(i)disasterriskreduction;(ii)waterre-sourcemanagement;and(iii)foodsecurity.Inaclimatechangecontextitisparticularlyimportanttoapproachriskreduc-tioninanintegratedperspectiveconsideringtherelatedareasandcombinedeffects.InColombia,activitieshavefocusedonvolcanoes,glaciersandrainfalltriggeredlandslides.The2007crisisanderuptionofHuilaVolcanohasmadenecessaryaclosecollaborationofscienceandcivilauthoritiestoavoiddisastersthatcouldhavebeen caused bymajor debris flows originating from volcano-ice interaction.With respect to landslides important expe-rienceshavebeengainedwiththeimplementationofanearlywarningsystem.Aneffectivelandslideearlywarningsystemisamajorchallengeinpractice.Difficultiesarefoundonatechnical-scientificlevel (e.g. applicationof rainfall-landslide thresholds) but also on an institutional and community-based level. The earlywarningsystemisonlysuccessfuliftheinter-institutionalcoordinationworkssmoothlyinanemergencyandifitisalsobornebythelocalpopulation.Theperceptionoftheriskbythelocalpeopleintheirlivelihoodcontextexertsasignificantcontrolontheresponsetoriskreductionefforts.Theirviewsareoftendivergingfromthoseofpublicauthoritiesandex-perts,whichhastobetakenintoaccountformoreeffectivedisasterreduction.
�.10
Theoretical basis for shadow angle variability and implications
JaboyedoffMichel*,PedrazziniAndrea*
* Institute of Geomatics and Analysis of Risk, Faculté des géosciences et de l'environnement, University of Lausanne, CH-1015 Lausanne - Switzerland ([email protected])
SinceHeim (1932) the shadow angle or Farböschung has beenwidely studied to estimate runout distance of landslides(Corominas,1996).Thedistanceusedtodeterminetheshadowangleisbasedeitheronthemaximumofrunoutdistanceoronathresholddistance(EvansandHungr,1993;Toppe,1987).Thisdiscrepancyforempiricalapproachhastobeexplained.
Inthepresentpaperweinspectedtheuncertaintyontheparametersofthesimplemodeloftheenergyline.Theuncertain-tythatisrelevantcomesessentiallyfromthefrictionparameter.Asaconsequencethefrictioncoefficientcanbeassumedasarandomvariablealongthelandslidepath.Asitisasumofrandomvariablesitmustfollowanormaldistributionowingtothecentrallimittheorem.
2�0Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s
Thishypothesiscanbeverifiedonseverallandslidedatasetssuchasrockfalls,shallowlandslides,snowavalanches,etc.Inany case, thispermits tounify all thedifferent approaches taking into account thedifferencebetween energy line andFarböschung.
REFERENCESCorominasJ.1996:Theangleofreachasmobilityindexforsmallandlargelandslides.CanGeotechJ33:260–271.HeimA.1932:BergsturzundMenschenleben-FretzundWasmuth,Zurich,218pp.Evans,S.,Hungr,O.,1993:Theassessmentofrockfallhazardatthebaseoftalusslopes.CanadianGeotechnicalJournal,30,
620-636.Toppe,R.1987:Terrainmodels:atoolfornaturalhazardmapping.In:Avalancheformation,movementandeffects.Edited
byB.Salm&H.Gubler.InternationalAssociationofHydrologicalSciences.Wallingford,UK.Publication162,pp.629-638.297,269-281.
�.11
Machine learning algorithms for spatial data. Case studies: environmental pollution, natural hazards, renewable resources
KanevskiMikhail*,PozdnoukhovAlexei*,TimoninVadim*
*Institute of Geomatics and Analysis of Risk (IGAR), University of Lausanne, CH-1015 ([email protected])
This paper presents general problems and approaches for the spatial data analysis usingmachine learning algorithms.Machinelearningisaverypowerfulapproachtoadaptivedataanalysis,modellingandvisualisation.Thekeyfeatureofthemachinelearningalgorithmsisthattheylearnfromempiricaldataandcanbeusedincaseswhenthemodelledenviron-mentalphenomenaarehidden,nonlinear,noisyandhighlyvariableinspaceandintime.Mostofthemachineslearningalgorithmsareuniversalandadaptivemodellingtoolsdevelopedtosolvebasicproblemsoflearningfromdata:classifica-tion/patternrecognition,regression/mappingandprobabilitydensitymodelling.
In thepresent report someof thewidelyusedmachine learningalgorithms,namelyartificialneuralnetworks (ANN)ofdifferentarchitecturesandSupportVectorMachines(SVM),areadaptedtotheproblemsoftheanalysisandmodellingofgeo-spatialdata.Machinelearningalgorithmshaveanimportantadvantageovertraditionalmodelsofspatialstatisticswhenproblemsareconsideredinahighdimensionalgeo-featurespaces,whenthedimensionofspaceexceeds5.Suchfeaturesareusuallygenerated,forexample,fromdigitalelevationmodels,remotesensingimages,etc.Animportantextensionofmodelsconcernsconsideringofrealspaceconstrainslikegeomorphology,networks,andothernaturalstructures.Recentdevelop-mentsinsemi-supervisedlearningcanimprovemodellingofenvironmentalphenomenatakingintoaccountongeo-mani-folds.Animportantpartofthestudydealswiththeanalysisofrelevantvariablesandmodels’inputs.Thisproblemisap-proachedbyusingdifferentfeatureselection/featureextractionnonlineartools.
Todemonstratetheapplicationofmachinelearningalgorithmsseveralinterestingcasestudiesareconsidered:digitalsoilmapping using SVM, automatic mapping of soil and water system pollution using ANN; natural hazards risk analysis(avalanches,landslides),assessmentsofrenewableresources(windfields)withSVMandANNmodels,etc.Thedimensiona-lityofspacesconsideredvariesfrom2tomorethan30.
Figures1,2,3demonstratesomeresultsofthestudiesandtheiroutputs.
Finally,theresultsofenvironmentalmappingarediscussedandcomparedwithtraditionalmodelsofgeostatistics.
Acknowledgements.ThestudywaspartiallysupportedbytheSwissNationalScienceFoundationprojectsGeoKernels(pro-jectNo200021-113944)andClusterville(projectNo100012-113506).
REFERENCESKanevski,M.(Editor),2008.AdvancedMappingofEnvironmentalData.ISTELtd.,JohnWiley&SonsInc,313pp.
2�1
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks
Figure1.Mapofpotentialareaswherelandslidescanoccur(regionofthestudy-northwestChina).Darkercolourspresentregionswith
higherprobabilityofdanger.Trainingareas(outlinedbypolygons)arealsopresented.
Figure2.MapofsoilpollutionautomaticallygeneratedbyANN.
2�2Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s
Figure3.MapofwindfieldsinSwitzerland.
�.12
A method for risk analysis related to lahars and floods – a case study at Nevado del Tolima volcano, Colombia
KünzlerMatthias*,HuggelChristian*,RamírezJuanManuel**
*Department of Geography, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich ([email protected], [email protected])**Swiss Agency for Development and Cooperation, Bogotá, Colombia
Theglacier-coveredNevadodelTolimalocatedintheColombianCordilleraCentral(Figure1)isanactivevolcanowithpo-tentiallaharhazardssimilartothoseonNevadodelRuiz(Thouretetal.,1995).Foreffectivedisasterpreventionariskanaly-sisisanimportanttool.Wepresenthereamethodologythatallowsforarelativelyrapidassessmentoftheriskbasedonafirst-orderanalysisoflaharandrainfallrelatedfloodhazards,andvulnerability.Themethodologyisperformedforfivevil-lagesintheCombeimaValleyandtheregionalcapitalIbagué(≈450,000inhabitants).
Firstly,themodelsLAHARZforlaharsandHEC-RASforfloodswereappliedtogeneratehazardmaps.Laharscenariosarebasedonmeltingof0.5,1,5,and15mofglaciericeduetovolcanicactivity,resultinginlaharvolumesof0.5,1,5,and15millionm3.Forfloodmodelling,designfloodswithareturnperiodof10and100yearswerecalculated.Asecondstepinvol-vestheanalysisofdifferentvulnerabilities.Physicalvulnerabilityisoperationalisedbymarketvaluesofdwellingparcelsandpopulationdensity,whereassocialvulnerabilityisexpressedbypopulationageandpoverty.Thirdly,hazardHandvulnerablyVvaluesarebothtransformedintoascalefrom0to1andsubsequentlymultipliedfollowingtheriskequationR = H * V(Varnes,1984).Theresultsarequalitativelyriskvaluesperparcellevelandquantitativedamageestimates.Figure2schema-ticallyillustratestheriskanalysisconcept.Theleftsidereferstothevulnerabilityandtherightsidetothehazard,respec-tively.GreyparallelogramsareGISlayers.
Whereasfloodhazardhaslimitedeffectsonpopulationandinfrastructure,theimpactoflaharsismoreserious.Anassumedlaharvolumeof15millionm3mayleadtolaharheightsof19mandamaximumhorizontalcrosssectionofca.450m.Over20,000peoplemaybeaffected.
2�3
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ksWhileuncertaintiesyetprevailwithrespecttothetopographicbasisandhazardmappingingeneral,theoverallmethodo-
logyhasproventobeasuitabletooltoprovideafirstoverviewofspatialdistributionofrisk,helpfulfordecision-makers.Weshouldconsider,however,thattheweightingofthedifferentvulnerabilitieshasanimportantcontrolonresultingrisksandshouldthereforebeperformedincoordinationwiththeresponsibleauthorities.
Figure1.CordilleraCentralwithNevadodelTolimaandNevadodelRuiz(fromHuggeletal.,2007).
Figure2.FlowchartoftheriskanalysisconceptbasedonaR=V*Happroach.
REFERENCESHuggel,C.,Ceballos,J.L.,Pulgarín,B.,Ramírez,J.&Touret,J.-C.2007:Reviewandreassessmentofhazardsowingtovolcano-
glacierinteractionsinColombia.AnnalsofGlaciology,45,128-136.Thouret, J.-C.,Cantagrel, J.-M.,Robin,C.,Murcia,A., Salinas,R.&Cepeda,H. 1995:Quaternaryhistory andhazard-zone
modelatNevadodelTolimaandCerroMachinVolcanoes,Colombia.JournalofVolcanologyandGeothermalResearch,66,397-426.
Varnes,D.J.1984:Landslidehazardzonation:areviewofprinciplesandpractice.UnitedNationsEducational,ScientificandCulturalOrganisation,Paris.63pp.
2��Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s �.13
Geological heterogeneity in landslides: Characterization and flow modelling
BorisMatti
SUPSI, Istituto Scienze della Terra CP 72, CH 6952 Canobbio ([email protected])
Significantprogresshasbeenmadetheselastdecadesinthedevelopmentofhydrogeologicalnumericalflowmodellingfordescribingthehydrodynamicbehaviouroflandslides.However,thesenewsophisticatedmethodsarestillveryseldomusedintheproblemsofslopeinstabilityinparticularbecauseofthehydrogeologicalcomplexitywhichcharacterizesthem;thinaquifers,discontinuousmedia,successionofsaturatedandunsaturatedzones,lowpermeabilities,highhydraulicgradients,lithologicalheterogeneity,strongcontrastsofpermeabilitiesandheterogeneousinfiltration.
Predictivemodelsofflowinthesubsurface,whichareoftenbasedonhomogeneousporousmediatypesofrepresentation,arebadlyadaptedtonaturalsystemsthatarecharacterizedbyhighlyheterogeneousmediasuchaslandslides.Thesemodelsaregoodandreliableonalandslidescale(regionalscale),buttheirqualitymaybeaffectedonalocalscalebystronggeolo-gicalheterogeneities.Geologicalheterogeneitiesofthesubsurfacetakepartindeterminingthehydrodynamicalandgeome-chanicalbehaviouroflandslides.However,theirspatialdistributionispartiallyunknown.
Thus,theprincipalobjectivesofthisworkare:(i)TocarryoutanintegratedmultidisciplinarycharacterizationstudyontheinternalstructureoflandslidesinflyschandQuaternaryenvironments,inordertoclarifytheorganisationofthegeologicalheterogeneitiesandtoidentifythehydrodynamicimplications.(ii)Toproposeaconceptualmodelrepresentingthegeologi-calarchitectureandthehydrogeologicalfunctioning.(iii)Toexaminetheeffectsofheterogeneityandanisotropyonflowsystems.(iv)Tobetterunderstandtheinfluenceofgeologicalheterogeneitiesonthemechanicalbehaviouroflargelands-lidesbyperformingnumericalsensitivityanalyses,bymeansofdifferentheterogeneityscenariosonthefieldparameters.(v)Finally,totesttheincidencesonslopestabilizationtechniques;evaluationoftheefficiencyofadrainagegallerywork.Themaintestsiteofla Frasse landslide(VD,Switzerland)waschosen,andcompletedwithadditionallandslidecases.
The main results are the following:
Inmostofthecasestudies,thelandslidemassiscomposedofanoldprehistoricstabilizedmass,pinchedbetweentheactiveslidingmassandthebedrock,andplayinganimportanthydrologicrole.Thestabilizedmassandthebedrockformthesub-stratumofthelandslide.
Landslidesoccurringinthesetypesofmediaaredefinedbyanorganizedheterogeneousenvironmentwith“fracture”flowsand discontinuity porosity. The overall hydraulic conductivity is low, and locally high permeable zones exist. Regionalgroundwatercirculationsarelimitedandformlocalinterconnectedaquicludesorganisedinthinaquifers,andpresentingsaturatedandunsaturatedzones.Thehydrogeologicalanalysesshowedthatthesystempresentsabimodalpermeability;(i)Lowhydraulicconductivitiescha-racterizing theglobalmatrixanddefining thecapacitive fraction,and (ii)highpermeable features,withhighhydraulicconductivitiesdefiningtheconductivefraction,andfavouringstrongchannellingeffects.Besides,theobservationshowsthattheaquifersystemisgenerallyveryreactivewithimportantmagnitudes.Often,thereisastraightcorrelationbetweenwaterlevelvariationandclimaticconditions(rainyevents).Landslidesarecharacterizedby two important inflowsnamelyeffective infiltration fromthe surfaceand lateral inflowsfromtheneighbouringunits.Watertransferbetweenthestabilizedmassandtheactivemassmaybeimportantandthushavetobeconsidered.Theexistenceofwatertransferbetweenthebedrockandthelandslidemass(stabilizedandactive)isnotwellestablished.Thebedrockandthelandslidemasspresentahydrologicalbehaviouralindependence.
Theoreticaltwo-andthree-dimensionalflowmodelsareusedtoinvestigatetheeffectsofthespatialvariabilityofthehydrau-licconductivityontheundergroundflows.Theroleoftheconnectivityingeneratingflowchannellingisexaminedthankstotheobservationofcloserelationsbetweenthepermeabilityandthehydraulicpressures.Thesensitivityanalysisshowsclearlythattherelationbetweenlocalpermeabilityandhydraulicpressuresisnotstraight,andthattheorganizationoftheflowsdependsontheheterogeneityofthehydraulicpropertiesandtheirspatialcorrelation.Strongchannellingeffectsareobservedinhighlyheterogeneousporousmedia.Thedevelopmentofflowchannellingasafunctionofthevarianceofthenaturallogpermeabilityvaluesandthecorrelationlengthsisdemonstrated.
Theintegratedmulti-disciplinarygeologicalcharacterizationattheLa Frassetestsitecombinedwiththehydrogeologicalandlithologicaldataofseveraladditionalcasestudiesledtotheproposalofaglobalconceptualmodel.Thefollowingassump-tionsareconsideredtoenableasubsequentquantificationofflowcomponents:
2��
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks
• Theflowoccursunderconfinedtoleakyconditions,withleakagevaryinginspace;• Theflowframeworkiscontrolledbyacomplexmulti-layersystem,isolatedlensesorperchedaquifer;• Theaquifersystemisdividedintointerconnectedhydrologicalzonespresentingvariousdegreesofsaturation;• Eachhydrologicalzonemayfunctionindividuallyfromtheothers;• Horizontallyandvertically,theflowdirectionintheporousmatrixisaffectedbyprevailingstructuralpatternsgenera-
tingchannelingeffects;• Theflowismultidirectional,freeandchannelized,andisaffectedbytemporalandspatialchanges;• Theaquiferisunderanunsteadyflowregimeduetoseasonalvariationofnaturalgradients.
�.1�
Simulation for a volcano monitoring network
MautzRainer*
* Swiss Federal Institute of Technology (ETH), Institute of Geodesy and Photogrammetry, Wolfgang-Pauli-Str. 15, CH-8093 Zurich ([email protected])
ThispaperinvestigatesthecapabilityofGNSSaidedsmartsensornetworkpositioningbasedonWirelessLocalAreaNetwork(WLAN)signalsincorporatingaccesspoints,tomonitor3Ddeformationassociatedwithvolcanicactivityandothercompa-rablehazardousevents.Manyoftheworld'svolcanoesthaterupt,experiencesignificantpre-eruptionsurfacedeformation.Internalmagmapressuremakesthesurfacebulgeupwardsandoutwards.Thus,precisemonitoringofsurfacedeformationhasthepotentialtocontributesignificantlytotherealisationofapredictivecapabilityofvolcaniceruption.Inparticular,eruptionsourcedepthandevolutiontimecanbeestimatedfromsurfacedeformation.Thescaleofthisdeformationistypi-callycentimetrictodecimetricovertensofsquarekilometresandoverperiodsofweekstoyears.Horizontaldisplacementsshowaradialpatternofmovementofupto10cmwiththedisplacementoftheverticalcomponentstypicallyintherangeof4to6cmperyear.
Inadditiontotheuseofprecisepositioninginformationtofacilitatedeformationmonitoring,thepositioningfunctionisvitalforspatio-temporalreferencingoftherelevantmultipleandcomplementarydatatypesforvolcanomonitoring(e.g.,seismicity,groundsurfacedeformation,geothermal,gravity,andgeomagnetic).
Inarchitecturaltermsthemonitoringnetworkshouldconsistofanarrayofdistributedintelligentnodes (sensormotes),consistingoflow-cost,commerciallyavailable,andoff-the-shelfcomponents(asfaraspossible)withbuilt-inlocalmemoryandintelligence,withself-configuration,communication,interactionandcooperativenetworkingcapabilities.Thenodesshouldbeabletoidentifythetype,intensity,andlocationoftheparametersbeingmeasured,andcollaborateinaninter-nodalmannerwitheachothertoperformdistributedsensingforeventconfirmationandsignificance.
Becauseoftherequirementforhighaccuracypositioningandtheneedtokeepcostsdown(bothintermsoftechnicalcom-plexityandpowerconsumption),buildingcarrierphaseGNSSchipsintoallWLANshouldbeavoided.Acompromisescenarioistohavebothtypesofnodes,someequippedwithWLANaswellascarrierphasechipsthatareusedforabsolutecoordina-tereferencingbutwiththemajorityofnodeswithonlyWLANcommunicationandrangingcapabilities.ThelimitedGNSSaidingproposedshouldenableWLANpositioningtodelivercentimetrelevelpositioning.ThesensorsequippedwithGNSSchipscalculatetheirpositionsinahigherreferenceframewithhighaccuracy,andserveasanchor(=controlorreference)pointsforthemonitoringnetwork.Thecommunicationfunctionofthenetworkshouldenabletheexchangeofthedatarequiredforpositioningwithinthemonitoringnetwork.ThisshouldenabletheWLANnodestopositionthemselvesexploi-tinginter-nodedistancemeasurements.
Suchamonitoring systemrequiresmultiplekey features includingconstructionof thehardware that fulfil the require-mentsintermsofsize,batterylifeandrobustness,theextractionofranges(distances)betweensensornodes,appropriatesupportingnetworkcommunications,protocoldevelopment,optimalroutingandpositioning.Thispaperaddressesspecifi-callythepositionfunctionandcharacterisestheperformanceofanovelhighpositioningalgorithmusingsimulatedrangemeasurements.The3Dpositioningalgorithmusestherangeobservationsformultilateration,clusterisationandgeodeticnetworkadjustment.
Thenovelalgorithmisusedtoinvestigatevarioussimulatedpositioningscenarios.Thechallengesassociatedwiththeuseofwirelesssensornetworksarethatideally,thesensors(nodes)shouldhavereliablepositioningdata,eveninthepresenceofmeasurementnoise,lowinter-nodeconnectivityandbadlyconstrainedgeometry.Thispaperpresentsastrategytoenable
2�6Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s highintegritypositioningandassessesitsperformancebasedonvariousparameterssuchasthenodedensity,maximalsi-gnalrange,requiredfractionofanchornodes(whichhaveGNSSpositioningcapability),rangemeasurementerrors,andlo-cationsofthenodes.
Presentedareresultsfromlargesimulatednetworks(i.e.400nodes)andtheoptimalnetworkparametersarequantified.Therequirementtohavedirectlineofsightsbetweenstationscanbesolvedbylocatingthenodesforamaximumnumberofdirectssights.Thenumberofrequirednodesdependsonthetransmissionrange.TherequiredfractionofGNSSenabledreferencenodeswillbearound10%,dependingonthenetworkdensity.
Figure1.Optimisedpositionsof400sensornodesatvolcanoSakurajima.
REFERENCESMautz, R.,Ochieng,W.Y., Brodin,G.&Kemp,A. 2007:. 3DWirelessNetwork Localization from InconsistentDistance
Observations,AdHoc&SensorWirelessNetworks,Vol.3,No.2–3,pp.141–170.
�.1�
Systematic recording and analysis of natural hazards along railway lines using GIS
AndreasMeier,ChristinaWilli
SBB Infrastruktur, Umwelt; Naturrisiken; Schanzenstrasse 5, 3000 Bern 65
DuetothetopographicalconditionsinSwitzerlandtheSBBrailwaylinesfrequentlyareexposedtonaturalhazardsasrock-fall,debrisflow,landslide,windandothers.Thesafeandhighdisposabletrainserviceisverysensitivetochangesintheterrain.Thisasksforahighlevelofsurveillanceandpreservationintheslopesalongtherailwaylines.Naturalhazarddataoverthepast100yearsanddataaboutthecurrentlyprevailingrisksexistinavarietyofanalogueanddigitalarchivesordatabases.Thereforeitisnecessarytosurveythedataintimeandspace.WiththeGeographicalInformationSystemforNaturalRisks(GISNR)consistingofthetoolsDERINRandWebGIStheSBBestablishedanimportantsystemformanagingnaturalrisks.
2��
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ksData recording using DERI NR
DERINR(Dérangementsàl’Infrastructure„NaturalRisks“)isaninstrumenttoreport,recordanddo-cumentnaturalhazards.Eventsarecontinuouslyrecorded inacentraldatabase (basedon@enter-prise technology; see figure 1) by the people inchargeofmaintainingsafety.In addition, during the next years historic datawill be added. Thereby DERI NR becomes a con-stantly updated database of natural risks alongtherailwaylines.
DERINRisaworkflowinstrumentwithatwo-steprecordingprocess.Traindrivers,routeinspectorsandvegetationkeepersregisterthekeydataofaneventsuchasnaturalhazardtype,routenumberandrouteposition.Thelownumberofobligatorydatafieldshelpstoreportasmanyeventsaspossi-ble.Photographsandtextdocumentscanbeatta-ched.
Dependingontheoperatingpointandthetypeofhazard,thesystemforwardsthereporttothespe-cialist for natural hazards. As a second step theevent is fully documented. The recording endswith the evaluation weather more protectionagainstnaturalhazardsisneeded.Therecordingprocesscanbereactivatedatanytime,e.g.toadddatafromexternalriskcalculations.
Figure1.GraphicaluserinterfaceofDERINR
Visualisation in WebGIS
WebGIS is an application of Intergraph (BM3)andisopentoallemployeesofSBB.Itvisualisesthe data fromDERI NR onmaps or orthopho-tos.Asaresultofthesecondrecordingstepthere-cord appears inWebGIS. The user can select aspecificareatooverviewprevioushazardsintheregion.NaturalhazardscanbesearcheddirectlyinWebGISusingDERINR.Vice-versa, adoubleclick on the visualised events opens the corre-spondingdocumentationinDERINR.The constantly updated risk map (figure 2)shows all sites that require more protectionagainstnaturalhazards.
Figure2.Mapofcurrentrisksites
2�8Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s Further extensionsToday GIS NR mainly centralises the documentation of natural hazards.Additionally the system offers queries about ice falls and avalanches at theGotthardlineandprovidesnaturalhazardmapsfromSILVAPROTECT.
InthecomingyearsGISNRwillbeadatabasecontainingallinformationnee-dedfortheevaluationofrisksites.HenceforthGISNRprovidestheinventoryandtheconditionofprotectivesy-stemssuchasdams,warningsystems,monitors,protectionforestsandothersaswellascalculatedriskdata,hazardmapsandprotectiontargets.GISNRhelpsinspectorsandthetechnicalmanagementtodecideaboutprotec-tionmeasures.Inadditiontothevisualisation,toolsforspatialdataanalysiswillbeavailable.
REFERENCESMeier, A. 2008: Naturgefahren: Herausforderung für die SBB. GeomatikSchweiz,5,242-243.
Figure3.NaturalhazardsinWebGIS
�.16
Glass and magnetitic Spherules associated whit The bolid impact and mass extinction in D/C boundary in Central Alborz mountain north of Iran
MahinMohamadi
Geology Department, Payame Noor University, PNU.Iran .P.O.Box 19395-4697,Tehran, Iran. [email protected]
ManyglassspherulesandtearlikeandalsomagnetiticandmetallicspherulesanddumbbellhaveincomeinupperpartofGeirud Formation in central Alborzmountain. The greymarly limestone bed in upper part of geirud formation (upperDevonian)inGeirudvalley,northofTehranconsistofmanytearandspherulelikeyellowglassandalsoplentifulnumbersofmetallicandmagnetiticspherulesanddumbbellswhithhighdensity.LowerCarboniferousMobarakFormationcoverthisbedinarea.AlloftheseparticlesiscollectedandhavebeenphotographedbySEM.TheformofTheseparticlesaresimilar,(butindiffe-rentstratigraphiclevel)tothosepreviouslydescribedfromF/FboundaryinmanypartoftheworldandfromKellewasermassextinction,Claeys,P.,1994-1996,Sandberg,C.andall,1988andMcLaren,D.andall,1982.Thebiostratigraphyofthisbedbasedonconodontisupperfamennian(youngerthanexpanasazone).(mohammadiandall,inprep.)ThesespherulesarelikelytobeproducedbybolidimpacteventandmaycausedbigmassextinctionatthattimewhichismarkedbyHangenbergeevent.ThisisthefirstreportofthepresenceoftektiteandmagnetiticspherulesinD/CboundarylayersofIranandmoreinspec-tionisneededfromotherequallayerfromotherpartofcountryandalsochemicalandXRDanalysiswillbedoneassoonaspossibleforcampaignwhittheworldsamples.
REFRENCES:Claeys,P.andall,1996:GeochemistryoftheFrasnian_FamennianBoundaryInBelgium:Massextinction,anoxicoceanand
microtektitelayer,butnotmuchiridium?GeologicalSocietyofAmerica,Specialpaper307.pp,491-505Claeys,P.andall,1994:Microtektite-like impact glass associatedwhit the Frasnian-FamennianBoundarymass extinction,
EarthandPlanetaryScienceLettres,V.22,Issues3-4,pp,303-315McLaren,D.andall,1982: Frasnian–Famennianextinction.inSilver,L., T., andSchultz,P.H., eds.,Geological implicationof
Impactsoflargeasteroidsandcometsontheearth,GeologicalSocietyofAmerica,Specialpaper190,pp.477-483Sandberg,C.A.and all,1988: Late Frasnianmass extinction: Conodont event stratigraphy global changes andpossible
causes:CourierForschungsinstitutSenkenberg,v.102,PP267-307.
2��
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks�.1�
Geological and structural model of the Åknes rockslide (Norway)
OppikoferThierry*,MartinaBöhme**,LarsHaraldBlikra**,Marc-HenriDerron**,MichelJaboyedoff*,SaintotAline**
*Institute of Geomatics and Risk Analysis, University of Lausanne, Amphipôle, CH-1015 Lausanne ([email protected])**Geological Survey of Norway, Leif Eiriksons vei 39, NO-7040 Trondheim
ÅknesisacomplexrockslidesituatedontheflankofafamousfjordinWesternNorway(Fig.1a).Therockslidehasavolumeof>35millionm3(Derronetal.,2005)andisinvestigatedandmonitoredwithintheÅknes/TafjordProject,sinceitsfailuremightcauseacatastrophictsunamiinthefjord.Alargeseriesofgeological,structural,geophysicalandboreholeinvestiga-tionsandseveralslopemovementmonitoringtechniquesmaketheÅknesrockslideoneofthemostintensivelystudiedsitesintheworld(Ganerødetal.,inpress).
ThisstudyfocusesontheconceptualmodeloftheÅknesrockslidebasedondisplacementmeasurements,digitalelevationmodelanalysis,terrestriallaserscanning,andfieldinvestigations.Theunderstandingofthemechanismiscrucialtoimple-mentsuitablemonitoringandearly-warningsystems.
TowardstheSWandtheNE,therockslidebodyisdelineatedbytworegionalNNW-SSEfaults(Fig.1b).Sub-verticalNNE-SSWtrendingfaultslaterallydelimitrockslidecompartmentsandactastransfersurfaces.ThemainfoldshaveagentlyplungingESE-trendingaxiscrossingobliquelythelandslidebody.Verticalgneissfoliation(S1)nearthehingezonescreatesweaknessesandfavourablyorientatedplanesthatleadtotheformationofextensionfailuresactingasback-cracks.Suchafractureledtothecreationofafast-movingridgeanda30mwidetrenchintheupperrockslidepartandseveralrockslidescarsontheslopeareexactlylocatedatthefoldshingeswithsub-verticalS1.
Theobserveddisplacementspermittodividetherockslideintoseveralpartsthatmovewithdifferentvelocitiesand/ordi-rections(Fig.1c).SincemostoftheslidingsurfacesreactivateS1,variationsintheorientationofS1–duetofoldsandundu-lations–createchangesintheslidingdirectionbetweenSEandSSW.Theslidingsurfaceisnotcontinuous,butsteppedbysub-verticalfracturesperpendiculartothemainslidingdirection,thathavebeenclearlyidentifiedonthetopographyinthevicinityofÅkneslandslide(Oppikofer&Jaboyedoff,2007).Amodelforthisupper-mostpartoftherockslideimpliesacom-bination of planar sliding along S1, subsidence due to stepped failure surface and toppling towards the opened grabenstructure(Fig.1d)(Oppikoferetal.,2008).
Thisinterpretationgivesacoherentframeworkofthemovementsobtainedbyvariousmethodsandleadstoareinterpreta-tionofthemorphology,indicatingthatseveralrockslidesoccurredinthepast.Thesefindingsenabletoestablishthefailuresurfacetopographyandtodefineblocksmostsusceptibletofailure.
Figure1.a)PictureoftheÅknesrockslide(Derronetal.(2005));b)Rockslidemorphologydisplayingtheback-scars,lateraltransferzonesandtheregional faults;c)Annualdisplacementvectors (data fromGanerødetal. (inpress));d) Instabilitymodelforthefast-movingridge(fromOppikoferetal.(2008)).
REFERENCESDerron,M.; Blikra, L.H.& Jaboyedoff,M. 2005:High resolutiondigital elevationmodel analysis for landslidehazard
assessment (Åkerneset,Norway). In: Landslides andAvalanches: ICFL2005Norway, 101-106. Taylor& FrancisGroup,London.
Ganerød,G.V.,Grøneng,G.,Rønning,J.S.,Dalsegg,E.,Elvebakk,H.,Tønnesen,J.F.,Kveldsvik,V.,Eiken,T.,Blikra,L.H.,&Braathen,A.inpress:GeologicalModeloftheÅknesRockslide,westernNorway.Eng.Geol.
Oppikofer,T.,&Jaboyedoff,M.2007:Åknes/Tafjordproject:DEManalysisoftheRundefjellet/Tårnetarea.Report,UniversityofLausanne,Switzerland,p.44.
Oppikofer,T., Jaboyedoff,M.,Blikra,L.H.,&Derron,M.-H.2008:Characterizationandmonitoringof theÅknes landslideusing terrestrial laser scanning. Proceedings of the 4thCanadianConference onGeohazards, 211–218. Presse del'UniversitéLaval,Québec,Canada.
260Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s �.18
A new approach to dating carbonate-lithic rockslides
OstermannMarc*&SandersDiethard*,
*Geologisch-Paläontologisches Institut, Innrain 52, A-6020 Innsbruck([email protected])
Large-scalerockslidesexceeding106m3involumenotonlyareamajorprocessofmountainerosionandorogenicmassba-lancebut,indenselypopulatedregionssuchastheAlps,alsorepresentamajorthreattohumansandfacilities.Establishingthedistributionofrockslidesintimeisaprerequisiteofhazardassessmentforfutureeventsandforabetterunderstandingofpotentialtriggers,suchasclimaticchangeorphasesofenhancedearthquakefrequencyandpost-glacialstressrelaxation(e.g.Erismann&Abele,2001).Inthelastdecades,rockslidedepositsweredatedby14Cagedeterminationofwoodfragmentsthatarepreserved(a)inlacu-strineorfluvialsuccessionsunderneaththerockslidemass,(b)withintherockslidemass,or(c)innewly-formedlakesontopofthesturzstrom(e.g.Wassmeretal.,2004).Ineachcase,the14Cageprovidesadifferentconstraintfortheageoftherockslideevent,thatis,incase(a)the14Cagerepresentsamaximumagelimitfortheevent,incase(b),whichisveryrare,the14Cageisagoodproxyageoftheevent,andincase(c)the14Cageprovidesaminimumagelimit.Unfortunately,the14Capproachtoage-datingoftencannotbeappliedbecauseofabsenceofsuiteddepositsorexposuresthereof,lackoforganicremnantsorofremnantssuitedforage-dating,and/orbecausetheresulting14Cageisfraughtwithmarkedimprecision(e.g.Prageretal.,2008).Inthepastdecade,anincreasingnumberofsturzstromshavebeendatedby36Clexposuredatingofdetachmentscarsand/orofsurfacesofboulderswithinthesturzstromdeposit(Ivy-Ochsetal.,1998).Atthepresentstate,thus,anincreasingnumberofthenumerousAlpinerockslidesthathithertocouldnotbedeterminedbythe14Cmethodwillforeseeablybedatedbyexposuredating,butacross-checkwithanothermethodofagedeterminationisdesirableineachcase.Ourpreliminaryinvestigationsofmajorcarbonate-lithicrockslidesoftheAlpsrevealedthatindeednearlyallofthemcon-tainpockets,thickercrustsandpatcheswhereintherockslidematerialunderwentcementationintoabreccia.Thesebrecciacementscanprovideaproxyageofthesturzstromeventbydatingthecementwiththe234U/230Thdisequilibriummethod(Ostermannetal.,2007).Tothisend,carefulpetrographicanalysisofsamplesisnecessarytodistinguishdifferentgenera-tionsandtypesofcement.Withintheresearchproject:CatastrophicRockslidesintheAlps,fundedbytheAustrianScienceFund,agedeterminationof17selectedrockslides(Fig.1)shallbedonebybothU/Thdatingofcementsandbysurfaceexpositiondatingwithcosmo-genicradionuclides.Yetexpositiondatinghastheundisputedadvantagethatitistheonlymethodthatveersforthe‘real’ageofasturzstromevent.CombiningtheprecisionofU/Thageswiththecorrectnessof(oftenmoreblurred)expositionageswas,therefore,theidealapproachtodeterminemostpreciseagesforselectedAlpinelandslides.
REFERENCESErismann,H.T.&Abele,G.2001:DynamicsofRockslidesandRockfalls.Springer,Berlin,Heidelberg,NewYork,316pp.Ivy-Ochs, S.,Heuberger,H.,Kubik,P.W.,Kerschner,H.,Bonani,G., Frank,M.&Schlüchter,C. 1998:Theageof theKoefels
event -relative,14CandcosmogenicisotopedatingofanearlyHolocenelandslideintheCentralalps(Tyrol,Austria),Zeitschriftf.Gletscherkundeu.Glazialgeologie34,57-70.
Prager,C.,Ivy-Ochs,S.,Ostermann,M.,Synal,H.-A.&Patzelt,G.2008:Geologyandradiometric14C-,36Cl-andTh-/U-datingoftheFernpassrockslide(Tyrol,Austria).Geomorphology,inpress.
Ostermann,M.,Sanders,D.,Prager,C.&Kramers,J.2007:Aragoniteandcalcitecementationin'boulder-controlled'meteoricenvironments on the Fern Pass rockslide (Austria): implications for radiometric age-dating of catastrophicmassmovements.Facies53,189-208.
Wassmer,P.,Schneider,J.L.,Pollet,N.&Schmitter-Voirin,C.2004:Effectsoftheinternalstructureofarock-avalanchedamonthedrainagemechanismofitsimpoundment,FlimsSturzstromandIlanzpaleo-lake,SwissAlps.Geomorphology61,3-17.
261
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks
Figure1.OverviewmapoftheEastern,SouthernandmostoftheWesternAlpswithprominentcarbonatelithicrockslidedeposits.
Rockslidedepositschosenforfurtherinvestigationarehighlightedbywhiteframedpentagons.
262Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s �.1�
The role of regional fault and fold-related fractures in the development of rock slope failure
PedrazziniAndrea*,JaboyedoffMichel*,FroeseCorey**,HumairFlorian*,LangenbergWillem**,&FranciscoMoreno**
*Institute of Geomatics and Risk Analysis, University of Lausanne, Amphipôle ([email protected])**Alberta Geological Survey, Edmonton, Alberta, Canada
Largerockslopefailuresinfracturatedrockareoftencontrolledbythecombinationofpre-existingtectonicfractureandbrittlefracturepropagationintheintactrockmassduringpre–failurephase(Brideauetal,inpress).Tectonicfeaturessuchfoldandlargefaulthaveinimportantinfluenceonthefrequencyandthespatialdistributionoflargerockslopefailure(AmbrosiandCrosta,2006).Inthispaper,wefocusontheinfluenceofthetectonicfeaturesinthekinematicreleasebutalsoonthecontrolofthelocalreductionofrockmassproprietiesinducedbytectonicdamage.
WepresentherethecasestudyofTurtleMountain,south-westernAlberta,Canada.Thisareaischaracterizedbythepresen-ceofthefamousFrankslide,occurredin1903andinvolving30mioofm3ofmassiflimestone.FromastructuralpointofviewtheareaischaracterizedbethepresenceoftheTurtleMountainanticlineandtheTurtleMountainthrust.TheTurtleMountainfoldcouldbedescribedasamodifiedfault-propagationfold.Adetailedfieldmappingandthemorpho-structuralanalysis(Jaboyedoffetal.,inpress)thehighresolutiondigitalelevationmodel(HRDEM)allowsidentifyingatleast5discontinuitysets.Basedonthelocationtheirorientationthreejointsets(J2,J3,andJ4)couldberelatedtothefoldingphase:
1)Apersistentextensionaljoint,sub-paralleltothefoldaxisandmainlypersistentinthehingearea.2)Twostrike-slipconjugatefracturesmanlypersistentinthewesternfoldlimb.
Theothertwojointsets(J1andJ5)havewithadominantfeatureandcuttingtheentireTurtleMountainanticline.Thelargepersistenceandtheiroccurrencesuggestthatthesediscontinuitiesarepost-foldingandrelatedtotheNE-SWtranspressionduringEocene(Priceetal.,1986).InordertoquantifytherockmassqualityinthedifferentportionsoftheTurtleMountainanticlinetheGeologicalStrengthIndex(GSI)wasestimated(HoekandBrown,1997).InthefoldlimbstheGSIvaluerangebetween35to55.theGSIdecreaseprogressivelyinthehingeareaandrangesbetween20to35.ThevariationoftheGSIvaluesalongthefoldcouldbyexplainanincreasingoffracturation(especiallydiscontinuitysetsJ2)nearthehingezoneduetothestrainconcentrationduringthefoldingphase.Atthesametimeanincreasingof fracturationinthisarea increasethesusceptibilityoftherockmasstoweatheringorfreezeandthawcycle.Fieldmappingshowsthathingeareacorrespondalsotheareaweremostofthepastandpotentialinstabilityareconcentra-ted(Figure1).Inthiscontexttwomainpotentialfailuremechanismshavebeendetected:
1) AplanarslidingonbeddingplanesorpersistentJ1discontinuitysetwiththerearreleasefollowingextensionalorthewedgeformedbystrike-slipeconjugatefaults.
2) Atoppling-slidingmechanismdevelopedontheextensionalfracture.
AllthisobservationshavebeentakingintoaccountinordertoassesthestabilityofthepotentialunstableareasinTurtleMountainandtoreexaminatetheFrankslidemechanism.
REFERENCESBrideau,M-A.,MingY.,Stead,D.(inpress).Theroleoftectonicdamageandbrittlerockfractureinthedevelopmentoflarge
rockslopefailure.Geomorphology(2008),doi:101016/j.geomorph.2008.04.010AmbrosiC. andCrostaG.B. (2006): Large sackungalongmajor tectonic features in theCentral ItalianAlps. Engineering
Geology,83183-200.Jaboyedoff,M.,Couture,R.andLocat,P. (inpress):StructuralanalysisofTurtleMountain(Alberta)usingdigitalelevation
model:towardaprogressivefailure.Geomorphology.(2008).doi:10.1016/j.geomorph.2008.04.012PriceR.A.1994.GeologicalhistoryofthePeaceRiverArch[accessedJune2004];InGeologicalAtlasoftheWesternCanada
SedimentaryBasin,MossopG.D., Shetson I. (comp.),CanadianSocietyof PetroleumGeologists andAlbertaResearchCouncil,Calgary,Alberta,URLhttp://www.ags.gov.ab.ca/publications/ATLAS_WWW/ATLAS.shtml.
Hoek,E.,Brown,E.T.,1997.Practicalestimatesofrockmassstrength.InternationalJournalofRockMechanicsandMiningSciences34,1165–1186.
263
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks
Figure1:(left)FailuremechanismsdrivenbyfoldrelatedfracturesaffectingthehingezoneofTurtleMountainanticline.Thenumberson
theprofilecorrespondtotheevolutionoftheGSIvalue.(right)Picturesshowfieldevidenceofpastinstabilitiesrelatedtothefoldloca-
tion.
�.20
The earthquake loss estimating tool QLARM: Applications in real-time and for predictions
RossetPhilippe*,BonjourCyrill**,CuaGeorgia**,KaestliPhilipp**,TrendafiloskiGoran*,WiemerStefan**&WyssMax*
*World Agency for Planetary Monitoring and Earthquake Risk Reduction (WAPMERR), 2 rue de Jargonnant, CH-1207 Genève ([email protected])**Swiss Seismological Service, ETH Hönggerberg CH-8093 Zürich
QLARMisaloss-estimatingtoolbeingconstructedjointlybyWAPMERRandtheSwissSeismologicalService.Thiscomputerprogramanditsstandardworlddatasetwillbeavailableforusebyanyprofessional.Itspurposeistocalculatemeandama-getothebuiltenvironmentandhumanlossesincaseofearthquakesanywhereintheworld.Theworlddatasetswearecurrentlyupdatinginclude:Population,buildingstock,attenuationofseismicwaves,localsoilconditions,criticalfacilities,andregionalearthquakesourceproperties.
Thepopulationisgivenbynumberofinhabitantsdowntothesmallestsettlementsavailableforeachcountry.QLARMisthusabletoranksettlementsaccordingtotheseverityoflosses,whichisusefulforrescueteamsinearthquakedisasters.
ThebuildingstockisdistributedinvulnerabilityclassesoftheEMS-98scale.Thisdistributionisafunctionoftheregionandofthesettlementsize.TheinformationonbuildingqualitycomesfromtheWorldHousingEncyclopedia,fromthelitera-ture,satelliteimages,groundsurveys,andexpertopinion.
Attenuationfunctionsaretakenfromtheliteratureandbycalibration.Althoughitisclearthatregionaldifferencesexist,theyarenotwellquantified.Insomeareaswemayneedtomodifytheattenuationrelationshipsuchthatintensitiesobser-vedinpastearthquakesarematched.
26�Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s Soilpropertiesaredefinedinvaryingdetailforimportantcitiesworldwideintheliterature.Wederiveamplificationfactors(forintensityoracceleration)aswellaswecanfrompublisheddataonsoilproperties,microzonationmaps,geologicalmaps,andtopography.
Themostprobabledepthforshallowearthquakesisaparameterwearemappingbyregionbecausetheneedforthispara-meterarisesinreal-timelossestimates. Wecalculatelossesforlargeearthquakesworldwidewithinlessthananhouroftheiroccurrencewhenthereisonlyateleseismicestimateofthedepthavailable,withitslargeerror.Thisposesaproblembecauselossesdependcruciallyonthehypocentraldepth.
AnexampleofalossestimateinrealtimeistheSichuanearthquakeofthe12May2008.AtfirstthemagnitudewasgivenasM7.5.Basedonthis,wedistributedanemail28minutesafterthequake,giving3,000+/-1,000asthemostprobablenumb-eroffatalities.AssoonasinformationreachedusthatthemagnitudemaybeM7.9,werevisedourestimateto40,000to100,000fatalities(50minutesaftertheevent).Ouralertsbyemailaredistributedtoanyonewhorequeststhem.Figure1Ashows themap ofmean damage to settlements thatwe placed on ourwebsite for public viewing 52minutes after theSichuanearthquake(www.wapmerr.org).
PredictionsofseveralearthquakesofclassM8+arecurrentlyineffectworldwidebasedontheMScalgorithm(Kossobokov,personalcommunication).Wecalculatedlossesthatareexpectedtoresultintwoofthesecases,iftherespectivepredictionsshouldcometrue.ThedamagestateresultingfromapossiblefuturelargeearthquakeincentralChileisshowninFigure1B.WeexpectthatamajorearthquakedisasterislikelytooccurincentralChileinthefuturewithmorethan1,000fatali-ties.
Tohavesomeconfidenceintheseresults,wecalibratedQLARMbyverifyingthattheselectedattenuationfunctionandthebuildingstockpropertiesyield the intensitiesandhuman lossesobserved inpastearthquakes in theregion inquestion.CalibratingQLARMfordifferentpartsoftheglobeisanimportantongoingactivity.Theparametersweconsidermodifyingregionallyincludevulnerabilitycurves,attenuationfunctions,distributionofbuildingsintoclasses,andthecasualtymatrixonwhichthenumberoffatalitiesandinjuredisderivedforagivendamagestate.
(A) (B)
Figure1.Mapsofmeandamagestateofbuildingsinsettlements,(A)inthecaseoftheSichuanearthquakeM7.9,May122008,and(B)ina
hypotheticalscenariothatmayresultinthefuturefromalikelyM8.5earthquakeoffthecoastofcentralChilenearalocationforwhicha
predictionisineffect.
Thecolourofthedotscorrespondstothemeandamageinsettlementsfromblack(majordestruction)toblue(minordamage).Thesizeof
thedotsisproportionaltothepopulation,theepicenterismarkedbyaringandromannumeralsindicateintensities.
26�
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks�.21
On the predictability of snow avalanches
SchweizerJürg
WSL Institute for Snow and Avalanche Research SLF, Flüelastrasse 11, CH-7260 Davos Dorf, email: [email protected]
Snowavalanchearerareeventsandlargeavalanchesthatcausedamageorharmtopeoplecanbecalledextremeevents.Inthegeneralcontextofpredictionandpredictablyofextremeevents,wewillexaminewhethersnowavalanchesarepredic-tablebasedonafewexamplesofforecastsatdifferentscales.Clearly,avalanchepredictabilitydependsonscale.WhentheregionalavalanchedangerisHighorVeryHigh,anavalancheislikelysomewhereintheregion.However,evenatsuchhighdangerlevels,thereleaseprobabilityinasingleavalanchepathiswellbelow50%(typicallyontheorderof1-10%).Thismeansthatasingleavalancheisarareevent,whichisnotpredictableevenwhenhigherdangerlevelsprevail.Atthelowerdangerlevels–relevantforbackcountryrecreation–thereleaseprobabilityissignificantlylower.Thelowreleaseprobabilitydoesnotmeanthattheriskislow.Evenwithlowoccurrenceprobabilitytheriskmightbetoohightobeacceptablesothatcom-prehensivepreventivemeasures(e.g.temporaryroadclosures)arerequired.Thoughthequalityofsnowavalancheforecastsistypicallyratherpoor,i.e.theskillscoreinastatisticalsenseislow,lowprobabilityforecastscanbeusefuliflargevaluesareatstake.
�.22
A portable radar interferometer for the measurement of surface deformation
StrozziTazio,WernerCharles,WiesmannAndreas&WegmüllerUrs
Gamma Remote Sensing, Worbstrasse 225, CH-3073 Gümligen ({strozzi,cw,wiesmann,wegmuller}@gamma-rs.ch; www.gamma-rs.ch)
Satelliteradar interferometryhasbeenusedextensively forground-motionmonitoringwithgoodsuccess. Inthecaseoflandslides,glaciersandrock-glaciers,forexample,space-borneradarinterferometryhasagoodpotentialtogetanoverviewonsurfacedeformation.Theroleofspace-borneradarinterferometryasanelementinawarningsystemishowevercons-trainedbythespecificspace-borneradarimaginggeometry,thetypicalmultiple-weekrepeat-interval,anduncertaintiesinthedataavailability.Mostoftheselimitationscanbeovercomewithanin-situradarimagingsystem[1,2].
GammaRemoteSensinghasdevelopedaportableradarinterferometerthatutilizesreal-apertureantennas,each2metersinlength,toobtainhighazimuthresolution[3,4].Imagesareacquiredlinebylinewhilerotatingthetransmittingandre-ceivingantennasaboutaverticalaxis(Figure1).Theinstallationeffortisrelativelysmallandtheinstrumentisportableandcanbebatteryoperated.Individualmeasurementscanbetakeninlessthan15minutesandtheacquisitiontimeislimitedprimarilybythespeedoftherotationalscanner.Eachimagelineisacquiredinapproximately2mshencethereislittleornomovementofthescenetointroducetemporaldecorrelation.
Phasedifferencesbetweensuccessiveimagesacquiredfromthesamelocationareusedtodetermineline-of-sightdisplace-mentsδfromthedifferentialphaseφviatherelationshipδ=-λ⋅φ/4πwhereλisthewavelength.Theinstrumentoperatesat17.2GHz(Ku-Band,λ=17.4mm)withadisplacementmeasurementsensitivitybetterthan1mm.Therangeresolutionoftheradarδr=c/2*Bisdeterminedbythe200MHzbandwidthBandisequaltoapproximately75cm.Becausethisisareal-aper-tureimagingsystem,theazimuthresolutionisdeterminedbytheantennabeamwidthandslantrangeR,δaz=Rsinθ.Inthecaseofourterrestrialradar,theazimuthbeamwidthis0.4degreeyieldinganazimuthresolutionofabout7mataslantrangeof1km.
Theinstrumentusestworeceivingantennaswithashortbaselinetoformaninterferometer.Phasedifferencesbetweensi-multaneousacquisitionsbytheseantennasareusedtocalculatethepreciselookanglerelativetothebaseline,permittingderivationofthesurfacetopography.Expectedstatisticalnoiseintheheightmeasurementsisontheorderof1meter.
Inthiscontributionthedesign,measurementprinciplesandcharacteristicsoftheportableradarinterferometerarepresen-ted. Results obtained in a series of experiments started in September 2007 over glaciers (e.g. Rhonegletscher andGornergletscher)andlandslides(e.g.Tessina,PianSanGiacomo)arealsodiscussed.
266Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s
Figure1.TheportableradarinterferometerdeployedovertheGornergletschershowingtherotationalscanner,antennasupportstructure,
antennas,andmicrowaveassembly.
REFERENCES[1]AntonelloG.,N.Casagli, P. Farina,D. Leva,G.Nico,A. J. Sieber andD. Tarchi,Ground-basedSAR interferometry for
monitoringmassmovements,Landslides,1(1):21-28,March2004.[2] PieracciniM., L. Noferini, D.Mecatti, C. Atzeni, G. Teza, A. Galgaro andNicola Zaltron, Integration of Radar
InterferometryandLaserScanningforRemoteMonitoringofanUrbanSiteBuiltonaSlidingSlope,IEEETrans.Geosci.RemoteSensing,44(9):2335-2342,2006
[3] WiesmannA.,C.Werner,T.StrozziandU.Wegmüller,Measuringdeformationandtopographywithaportableradarinterferometer,13th FIG International Symposium onDeformationMeasurements and Analysis and 4th IAGSymposiumonGeodesyforGeotechnicalandStructuralEngineering,Lisbon,Portugal,May12-152008.
[4] WernerC.,A.Wiesmann,T.StrozziandU.Wegmüller,Gamma’sportableradarinterferometer,13thFIGInternationalSymposium on DeformationMeasurements and Analysis and 4th IAG Symposium on Geodesy for Geotechnical andStructuralEngineering,Lisbon,Portugal,May12-152008.
�.23
Rockfall modelling applied to rockfall protection design
ThüringManfred
Istituto Scienze della Terra, Scuola Universitaria Professionale della Svizzera Italiana, CP 72, CH-6952 Canobbio ([email protected])
Rockfallmodelingbycomputersimulationscanbeusedtoplananddesignrockfallprotectionmeasures.RockSim3Disarecentlydevelopedsoftwaretosimulatethethree-dimensionalprocessofrocksfallingdownahillslope.Thefallingrocksaremodeledasindividual,dimensionless,sphericalparticles.Theinteractionwiththedigitalrepresentationoftheterrainoccursatthecenteroftheparticles.Thecomputerprogramisastand-aloneMicrosoftWindowsapplication,readsandwritesESRIrasterandshapefilesasinputandoutputandneedsaGIS(geographicinformationsystem)forpre-andpost-processingofinputandoutput.ThesoftwareisdevelopedinMicrosoftVisualStudioandusestheopen-sourcelibraryGDALforreadingandwritingGISvectorandrasterfiles.Themaininputofthesoftwareisarasterfilewhichdefinesthedigitalterrainandthesurfaceproperties,andashapefile
26�
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
kscontainingthestartinglocationsofrockelements.Afewmodelsareavailabletosimulatetheimpactphaseofrockelements
with theunderlying terrain. Themainoutputs are shapefiles containing the three-dimensional rockfall trajectories andparticlelocations,attributedbyinformationsuchaselapsedtime,velocity,kineticenergyandelevationaboveterrain.Recentlythefunctionalitytoconsiderrockfallprotectionmeasures,suchasfencesandbarrierswasintroducedinthesoft-ware.Theprotectivemeasuresarereadfromashapefileandarerepresentedastwo-dimensionalstructuresinthreedimen-sions.Theprotectionshavecharacteristics,suchasheightandmaximummechanicalresistanceagainstpenetration.Duringthedown-slopemovementofarockelement,impactswiththeprotectivemeasuresaresearchedforandrecognized.Threecasescanoccur,whenarockelementapproachesaprotection: (i) theprotection isnothighenoughandtherockelementjumpsovertheprotectionandcontinuesitsdownslopetrajectory(ii)animpactoccursbetweentherockelementandtheprotectionandthekineticenergyoftherockelementisbelowthemechanicalresistanceoftheprotection,andtherockelementisstopped,(iii)thekineticenergyoftherockelementexceedstheresistanceoftheprotectionandtherockelementpenetratestheprotection,continuingitsdownhilltrajectorywitharesidualvelocity.Asavariant,low-angleimpactsbetweenrockelementsandprotectionscanleadtoareflectionoftherockelement.Foreachprotectivemeasureimportantinformation,suchasnumberofimpacts,ormaximumencounteredimpactenergyissavedandcanbeusedforlateranalysis.Theintroductionofprotectionfencesandbarriersenablesthemodelertoplacetheseelementsonthedigitalterrainandassesstheirefficiencybyconductingrockfallsimulations.Inaniterativeprocess,boththerockelements(i.e.theirstartingposition,numberandmass),andtheprotectivemeasures(position,extension,heightandmaximummechanicalresistance)canbevariedtodeterminetheiroptimalpositionanddesign.Acasestudyispresentedwherethesoftwareisusedtooptimizethedesignandpositionofrockfallprotectionfences.
�.2�
Hazard and risk assessment of landslides on accumulation reservoirs – a field applicable scheme
ThüringManfred*,CannataMassimiliano*,HammerJürg**
*Istituto Scienze della Terra, Scuola Universitaria Professionale della Svizzera Italiana, CP 72, CH-6952 Canobbio ([email protected])**DRM Switzerland SA, CH-6952 Canobbio ([email protected])
Slopeinstabilitiesareathreattoaccumulationreservoirsandcorrespondingdown-streamareasduetopossibleslopecol-lapse,generationoffloodwaves,damovertoppingandfloodingofdownstreamareas.Asimpleandfield-applicableassess-mentscheme,basedonexistingapproaches,ispresentedtoevaluatehazardsandrisksduetolandslidecollapse,inordertoobtainafirstandpreliminaryoverview.Theoutcomesareusedtosupportplanningoffurtherinvestigations.
Thehazardsthatarisefromthepresenceoflandslidesontheslopesofaccumulationreservoirsare(Figure1):(i)Theslopeinstabilitydamagesthedamitselfandbuildingsandinfrastructure.(ii)Themassesofapartialorcompleteslopecollapsereachthereservoirlakeandcauseapulsewave,whichtravelstoanddamagesthedamandbuildingsandinfrastructures.(iii)Thepulsewaveovertopsthedamandcausesfloodingintheareasbelowthedam.Thesethreehazardsituationsarediscussedbrieflyandfield-applicableassessmentschemesarepresentedtopreliminarilyevaluatethehazardsandestimatetheconnectedrisksintermsofworstcasescenarios.
Inafirststeptheprimaryhazardneedstobeassessed,estimatingthecharacteristicsofthelandslidemassanditspathdownslope.Thisismainlydonebyfieldevidence.Keyinformationneededistypeofmaterialoflandslide,volume,elevationabo-ve lake level, inclination of trajectory andwidth of landslide. Themaximum run-out of the hazard is estimated by theFahrböschung-approachorismodeled,evaluatingiftheslopeinstabilityreachesthereservoirlake(Eisbacher&Clague1984,Hungr1995).Withthepreviouslyobtainedinformationthecharacteristicsofthepulsewavewhichmaybegeneratedduetoaslopecollapseisestimated,usingsimplecalculations(Huber1997,Huber&Hager1997).Forthispurposeamapviewofthereservoirandacrosssectionofthedamisneeded,wheredamheightandgeometry,waterdepths,freeboard,anddi-stancescanbereadfrom.Fromthecalculationsitispossibletoestimateifdamovertoppingoccurs.
Ifthecalculatedwaveislikelytoovertopthedam,theconsequencesofdamcollapseandfloodingareestimatedusingme-thodologiesthatpredictfloodheightsandvelocitiesindownstreamareas(Beffa2000,BWG2002).Dataobtainedfrommapsorbyfieldinspectionsisneeded,suchasvalleygeometry,roughnessandinclination.Thevulnerabilityassessmentofthefloodedareasisdonebasedonmapsorfieldinspections,suchasnumberandtypeoffloodedobjects.Theassessmentgivesanestimateofvaluesandfatalitiesatrisk.
268Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s Theschemeisfield-applicablewithdatausuallyavailableorthatcanbeproducedon-siteanddeliversapreliminaryassess-mentofhazardsandrisksconnectedtothepresenceoflandslidesontheslopesofaccumulationreservoirs.WepresentanddiscusstheschemeandshowitsapplicationinaprojectinRomania,whereapreliminaryscreeningonanumberofreser-voirswasconductedtoidentifykeycharacteristicsintermsofhazardandrisk.
Figure1.Hazardandrisksituationsduetothepresenceofalandslideonanaccumulationreservoir.
REFERENCESBeffa C. (2000): Ein Parameterverfahren zur Bestimmungder flächigenAusbreitung vonBreschenabflüssen,Wasser,
Energie,Luft,No.93,Heft3/4.BWG(2002):SicherheitderStauanlagen,BasisdokumentzudenUnterstellungskriterien.BerichtedesBWG,SerieWasser.EisbacherG.H.,Clague J.J. (1984):DestructiveMassMovements inHighMountains:HazardandManagement.Geological
SurveyofCanada,Paper84-16.HuberA.(1997):Quantifyingimpulsewaveeffectsinreservoirs.19ICOLDCongressFlorenceQ.74(R.35),pp.563-581.HuberA.,HagerW.H.(1997):Forecastingimpulsewavesinreservoirs.19ICOLDCongressFlorenceC(31),pp.993-1005.HungrO.(1995):Amodelfortherunoutanalysisofrapidflowslides,debrisflowsandavalanches.CanadianGeotechnical
Journal,32(4):610-623.
�.2�
Landslide investigation by means of PSiNSARTM radar interferometry: the Piemonte experience
CarloTroisi
ARPA Piemonte, Centro Regionale per le Ricerche Territoriali e Geologiche, via Pio VII 9, 10135 Torino, [email protected] , [email protected]
PSiNSARTMtechnique(hereinafterPStechnique)isaregisteredpatentofPolitecnicodiMilano,Italy,anditisoneofthePersistentScatterersMethods,i.e.agroupofapplicationswhichallowtheuseofsatellite-bornradarSAR(SyntheticApertureRadar)interferometrytodetectandmeasuregrounddeformations.ThePSTechniqueidentifiessinglecoherentbenchmarks,referredtoasPermanentScatterersorPS,andreconstructstheirdisplacementhistory.PS’sarenatural“radartargets”thatarelocatedacrosstheearth’ssurfaceandcanbemonitoredbysatellites.WhenPSremaincoherentwithinamulti-temporalradardata-set,itispossibletodetectandmeasuremillimetervariationsinthesensor-targetdistanceovertime.PS’stypicallycorrespondtoobjectsonman-madestructuressuchasbuil-dings,bridges,dams,water-pipelines,antennae,aswellastostablenaturalreflectors(typicallybareexposedrocksfaces).StartingwiththeidentificationofsuchPS’s,thetechniqueanalyzesthousandsofsquarekilometersofterritory,withinanextremelyshorttime.Indeed,thePS’scompriseasortof“naturalgeodeticnetwork”foraccuratelymonitoringsurfacede-formationphenomena,aswellasthestabilityofindividualstructures.
26�
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ksWhereverPS’sarepresent,itispossibleto:
• Establishtheirgeographiccoordinates.• EstimatetheirdisplacementratealongtheLineofSight(LOS)connectingthesatelliteandtheradargroundtarget,with
aprecisionwhichcanbeasgoodas0.1mm/a,dependingontheamountofavailabledataandthePSdensity.• Reconstruct the displacement history of an individual PSwhere the precision can be as good as 1mm on a single
measurements.
AllPSmeasurementscanbeeasilydownloadedontoGeographicalInformationSystems(GIS)andcomparedwithotherin-formationlayers,forbetterinterpretationofresults.PStechniqueprovedtobeexcellentforlandslideinvestigation;sincethemaximumvelocitywhichmayberecordedbyPS’sisintheorderof10cm/a,thetechniqueisusefulbasicallyforslow-movinglandslides.Themainadvantagesare:
• Satelliteradarimagesareavailablesince1992,sothatitispossibletoobtaindisplacementdatasincethatyear.• Alargenumberofslow-movinglandslidecanbecheaplyidentifiedandmonitoredoverawidearea.• Thereisnoneedforanyfielddevice,benchmark,monumentetc.,neitherthemonitoredareaneedstobeaccessible.• DataareeasilyimportedinGIS.
Therearealso,however,someproblems:
• Themethodrecordsonlyonecomponentofthedisplacement,alongtheline-of-sight(LOS)betweenthesatelliteandthePS.Thedeterminationoftotal3Ddisplacementsisnotstraightforwardandmaybetroublesome.
• Themethodisnotsuitabletodetectlandslideswithdisplacementratesexceedingabout10cm/aalongtheLOS.• Thetechniqueworksaslongasgoodradarreflectorsarepresent(buildings,barerocksetc.);inwoodedorgrass-covered
areasitisnotapplicable.Moreover,unfavourableslopeattitude,withrespecttosatelliteacquisitiongeometries,leadstoradar"shadowing"ofwideareas.Asaresult,largeportionsofagiventerritorymaybeunsuitedforPStechnique.InthePiemonteexperience(seebelow),aboutonethirdoftheoverallsurfacewastotallyblindtoPS.
• SincesatelliteorbitsareNSoriented,displacementsalongEWorientedslopesaredifficulttodetect.• Sinceradarsatellitespassoverthesameareaonceevery35days(average),real-timelandslidemonitoringisnotpossi-
ble.• PSanalysisisnotrivialmatter.Fewcompaniesintheworldprovidethistypeofanalysis• IfthetargetisaffectedbyLOSdisplacementvaluesapproaching14mmbetweentwosuccessiveacquisitions,measure-
mentscanbe“aliased”andtheestimationofthedisplacementmaynolongerbecorrect.
Intheperiod2005-2007ArpamadeaPSsurveyovertheentirePiemonte,about25000km2,thefirstcaseintheworldofsuchawideareasurveyedwiththistechnique.ThesurveywasmadeusingdatafromEuropeanradarsatellitesERS1and2,cove-ring the timespan1992-2001.Since themainaimof theArpa survey is landslidedetection, this timespan isextremelymeaningful,forthisperiodwasaffectedbyatleastsevenperiodsofveryheavyorprolongedrainscausingbothfloodsandlandslides.Thesurveyidentifiedabout2.2millionsPS,outofwhichabout10%showsomedisplacement.Topresentthedatafordiffusionbyaweb-gisserviceonArpasite(www.arpa.piemonte.it)wedevisedamethodbasedonwhatwedefinedareaanomala(anomalousarea).Anareaanomalaisdefinedasapolygoncontainingatleast3PS’sshowingadisplacementrateexceeding±2mm/aandlikelytorefertoasingledefinitephenomenon(landslide,subsidence,structuralfailureetc.).Uptonowwedefinedabout2200areeanomale.Foreachareaitispossibletodownloadaformdescribingalltherelevantfeatures.
ThePSsurveyisprovingextremelyusefulfortheintegrationofthePiemontelandslideinventory,whichincludes,uptonow,about35000landslides,mappedatthescale1:10k(http://marcopolo.arpa.piemonte.it/website/geo_dissesto/arpa_ib_iffi/viewer.htm).
UptonowwemainlyusedPS’sfor:
• Identificationanddefinitionofformerlyunknownlandslides.• Betterdefinitionoflandslidelimits.• Definitionofthestateofactivity.• Zoningoflargelandslides.
Last,ArpaPiemonteisalsoexperimentingtheuseofPS’sforidentificationanddefinitionofrecenttectonicdisplacements.Thefirstresultsareextremelypromising.
2�0Sy
mp
osi
um
9:
Na
tura
l H
aza
rds
an
d R
isk
s �.26
Microzonage sismique du Canton de Fribourg: Carte de Sol de fondations
VouillamozNaomi*&MosarJon*
* Département de Géosciences Université de Fribourg, Chemin du Musée 6, CH-1700 Fribourg ([email protected])
Lacoordinationdescartesdesolsdefondationàl’échelle1:25'000ducantondeFribourgvientterminerenjuillet2008uneétudedemicrozonagesismiquelancéeparl’Établissementcantonald’assurancedesbâtiments(ECAB)encollaborationaveclasectionRisquesgéologiquesdel’Officefédéraldel’environnement(OFEV).LanormeSIA261défini6classesdesoldefondation(deAàF)surlabasedescaractéristiques(lithologie,épaisseuretvites-sedesondesS)des30premiersmètresdescouchessuperficielles.Lacartographiedesdifférentesclassesdesolsdefondationpermetainsid’établirlesdonnéesd’aléasismiquelocal.Cesdonnéessontcalculéesàpartirdupotentield’amplificationdesondessismiquesdechaqueclassepourles3zonesd’aléasismiquerégionalquecomportelecantondeFribourg(zones1,2et3a,approximativementduNordauSud).Cettedémarchepermetd’évaluerlerisquesismiquedesagglomérationsducan-tond’unepartetdedéfinirledimensionnementdesbâtimentsquidevraits’ensuivred’autrepart.Lesmandatairessuivantsontparticipéàl’élaborationdescartesdesolsdefondationpourlesdifférentespartiesduterritoirecantonalfribourgeoisdélimitéesparlesfeuillesdel’Atlastopographiquesuisseau1:25'000(Figure1):
• lebureauABA-GEOLSA(Fribourg)pourlespartsfribourgeoisesdesfeuilles1164Neuchâtel,1165Morat,1184Payerne,1204Romont,1244Châtel-St.-Denis,1245Château-d’Oex,1264Montreuxet1265LesMosses;
• leDépartementdeGéosciences–SciencesdelaTerredel’UniversitédeFribourgpourlesfeuilles1185Fribourg,1205Rossens et 1224Moudon ainsi que pour les klippes ou débordements nonmandatés 1145 Bielersee, 1166 Bern, 1183Granson,1203Yverdon-les-Bainset1246Zweisimmen;
• lebureauInstitutGéotechniqueSA(Berne)pourlespartsfribourgeoisesdesfeuilles1186Schwarzenburg,1206Guggisberget1226Boltigen;
• lebureauGéovalIngénieurs-GéologuesSA(Sion)pourlafeuille1225Gruyères.
L’élaborationdescartessefaitparl’interprétationentermedeclassesdesolsdefondationdedifférentstypesdedonnées.Lescartesgéologiques,lesorthophotos,lesmodèlesnumériquesdeterrain(MNT1m)etombrages,serventd’informationdebaseencequiconcernelanatureduterrain.Lesdonnéesdeforages,lacartedesinstabilitésetglissementsdeterrainsducanton ainsi que des études géophysiques ponctuelles (géothermie, géoélectricité, géoradar, «petite sismique», …) don-nentdesindicationsquandàlaprofondeurdusubstratrocheux.LescartesdesolsdefondationducantondeFribourgontétéélaboréessousformedeSIGàl’aidedelasuitedelogicielsArcGIS©.Uneréactualisationcontinueestainsipossiblelorsquedenouvellesdonnéessontàdisposition.Lorsdelasynchro-nisation,unecertainesystématiqueaétéappliquéedanslebutd’uniformiserlesfichierssourcesd’unepartetdecoordon-nerlesborduresdescartesselonlesmêmescritèresd’autrepart.
Figure1.Cartedelarépartitiondesfeuilles1:25000couvrantlescantonfeFribourg.
2�1
Sym
po
siu
m 9
: N
atu
ral
Ha
zard
s a
nd
Ris
ks�.2�
Effects of tectonic structures, groundwater pumping, and mining activity on evaporite subrosion and resulting land subsidence
EricZechner*,InaLewin**,MarkusKonz*
*Geologisch-Paläontologisches Institut, Universität Basel, Bernoullistrasse 32, CH-4056 Basel ([email protected])** now at Institut für Angewandte Geowissenschaften, Technische Universität Darmstadt, Schnittspahnstr. 9, D-64287 Darmstadt (change of name from Spottke to Lewin 4/2008)
Evaporitesofgypsumorrocksaltarewidelyseenasthemostsolublecommonrockformation.Percolatingundersaturatedgroundwater leads to subsurfacedissolution, or subrosionof evaporites and, consequently, to thedevelopmentof karst.Dependingonthehydrogeologicalsettingandtheanthropogeniccircumstances,thesubrosionmaycausewidespreadlandsubsidence.Evencomparablysmallsubsidenceratescansignificantlyaffectsensitiveurbaninfrastructure,suchaslargerbuildings,dams,powerplants,ortherailwaytracksforhigh-speedtrainswhichcrossthepresentedMuttenz-PrattelnsiteinNorthwesternSwitzerland.
Fortheobservedwidespreadlandsubsidencesseveralcauseswereconsidered:(1)naturaldissolutionoftheevaporitesoftheMiddleMuschelkalk(anhydriteandhalite),whichisinducedbythetectonicsettingwithasetofHorstandGrabenstruc-tures,(2)saltsolutionmining,whichhasbeenpursuedatdifferentlocationsoverthelast150years,(3)large-scaleextractionofgroundwaterintheUpperMuschelkalkaquiferwithsuccessivehydrostaticconnectionalongthenormalfaults.Thetec-tonicsettingofthestudyareaischaracterizedbyhorstandgrabenstructuresdelimitedbynormalfaults.Theconstructionofa3Dgeologicalmodelincluded47faultsand4faultedhorizonsofthemainaquifers-aquitardsboundariesandprovidedabasis fornumericalgroundwatermodelingandthecomparisonwithgeodeticsurveydata (Spottkeetal.2005).The3Dnumericalgroundwatermodelwasusedtodelineateareaswithincreasedhydrostaticgradients.Intermediatescalelabora-toryexperimentswereconductedtounderstanddensity-coupledflowmechanismsinporousmedia.Thesetupofthedensi-tyflowexperimentwaschosentosimulatethegeologicalconditionsfoundatthesite(Konzetal.2008).Recentlandsubsi-denceshavebeensurveyedat6separatelocationswithintheMuttenz-PrattelnareaEastofBasel,Switzerland.Thediametersoftheaffectedsurfaceareasrangefrom100to1500m,andcorrespondingsubsidenceratesreachmorethan100mm/year.ThreesitesshowelongatedshapesofdepressionconesalongaSSW-NNE-orientedaxiscorrespondingtothestrikingoftheHorstandGrabenstructures,and,therefore,indicatingastrongrelationtothetectonicsetting.Thesubsidencehazardareaaboveeachformerorrecentsolutionminingwellwasestimatedbasedonamodelconcept,wherethecollapsedroofofex-cavatedsaltpropagateswitha45degreesangletothesurface.Forthreesites,whicharesaltsolutionminingfields, thepredictedhazardareacorrespondstotheobservedareaofsubsidence.Onesite,wheresolutionmininghasstoppedmorethan100yearsago,showsevidencethattheinitialcauseforsubsidence,i.e.solutionmining,hasbeenreplacedbylarge-scalegroundwaterwithdrawalinitiatingasignificanthydrostaticgradient,whichpropagatesalongthegrabennormalfaultnetworkfromtheUpperMuschelkalkAquifertotheevaporitesoftheMiddleMuschelkalk.Forthetwolastsites,acombina-tionofanaturaldissolutionprocessduetothetectonicsettingandaslightlyincreasedhydrostaticgradientseemsthemostprobablecausefortheongoinglandsubsidence.
REFERENCESKonz,M.,Ackerer,P.,Younes,A.,Gechter,D.,Zechner,E.&Huggenberger,P.2008:Newhomogeneousandheterogeneous
laboratory-scale2-Dbenchmarkexperimentfordensity-coupledflowmodels.AcceptedtoWaterResour.Res.Spottke,I.,Zechner,E.&Huggenberger,P.2005:ThesoutheastborderoftheUpperRhinegraben:A3Dstructuralmodelof
geologyanditsimportanceforgroundwaterflow.Int.Jour.EarthSci.,94,580–593.