PrefaceWelcome to HESSS4, the fourth International Conference on Hydrology delivers Earth SystemScience to Society! This isnotanordinary conferenceorganizedby scientific societiesorunions,but isanexcitingopportunitytoparticipate incross-disciplinarydiscussionsandtobeacquaintedwithresearchersfromdifferentfields,organizedbyasmallvoluntarygroupinanadhocsetting.The first HESSS was held in late February 2007 in Tsukuba, Japan, and motivated the bridgebetween hydrometeorological sciences and social demands, applying the latest scientificknowledgeforsolvingglobalwater issues.Estimationandreductionofuncertainty insimulationswere identifiedaskey leveragepointsalongwithanappropriateuseofobservationaldatasets invalidation and data assimilation. Itwas the dawnof applying social-ecologicalmodeling into theglobal scale hydrological modeling system, which incorporates reservoir operation andanthropogenicwithdrawalsinnaturalhydrologicalcyclemodeling.IncontrasttoHESSS1,whichwasmainlyfocusedonhydrologicalsciences,HESSS2consistsofwidercross-disciplinary sciences ranging from plot-scale ecosystem science to global radiation budgetstudies.HESSS2wasjointlyorganizedbytheGlobalSoilWetnessProject(GSWP)undertheGlobalLand/Atmosphere System Study (GLASS), the AsiaFlux under FLUXNET, the LandFlux-EVAL undertheLandFluxinitiativeoftheGEWEX(GlobalEnergyandWaterCycleExperiment)RadiationPanel,andCEOP(CoordinatedEnergyandWaterCycleObservationsProject)underGEWEX.HESSS3washeldatSeoulNationalUniversity,Seoul,KoreainAugust2013.Itwasjointlyorganizedwith AsiaFlux and Korean Society of Agricultural and Forest Meteorology under the slogan of“CommunicatingScience toSociety:Copingwithclimateextremes for resilientecological-societalsystems”.Threeplenarysessionsand18breakoutsessionswerehosted,andavisionwassetup:acommunity where science and technologyworkmore directly for sustainable ecological-societalsystems.To synthesize and advance knowledge of climate change impacts, land surface processes, andglobal energy and water cycles in a further comprehensive manner, the HESSS4 gatherscontributionsfromcommunitiesacrosswiderdisciplinessuchasGEWEX/WorldClimateResearchProgramm (WCRP), Coupled Model Intercomparision Project (CMIP)/WCRP, Climate andCryosphere(CliC)/WCRP,Inter-SectoralModelIntercomparisonProject(ISIMIP),InternationalLandModel IntercomparisonProject (ILAMB), JetPropulsionLaboratory (JPL)/NASA,EarthObservationResearch Center (EORC)/JAXA, Data Integration and Analysis System (DIAS), and Half a degreeAdditionalwarmingPrognosisandProjectedImpacts(HAPPI).It is expected this conference will provide an agora to strengthen synergies between researchcommunitiesofclimatemonitoring,modeling,andimpactsandtodiscusshowtodeliverscientificachievementstooursociety.Itwillprovideachanceforearlycareerresearchersaswellthosefromdeveloping countries to engage with frontier earth system science approaches to dealing withtheseproblems.Aparticularscientificfocusofthemeetingwillbeonclimateextremesandtheirimpacts,acknowledgingtheimportanceoftheseeventsinachangingclimate.Finally, the meeting organizers would like to acknowledge our supporters from nstitute ofIndustrial Science, theUniversityofTokyo (IISUTokyoSYMPOSIUMNo.90), JapanSociety for thePromotionofScience(JSPS,16H06291),InternationalAssociationofMeteorologyandAtmosphericSciences(IAMAS)/InternationalUnionofGeodesyandGeophysics(IUGG),andGlobalEnergyandWaterCycleExchanges(GEWEX).

Let’senjoytheHESSS4!HyungjunKim,

Co-chairoftheOrganizingCommitteeforHESSS4SpeciallyAppointedAssociateProfessor,IIS,theUniversityofTokyo,Japan

TaikanOki,

Co-chairoftheOrganizingCommitteeforHESSS4Professor,IIS,theUniversityofTokyo,Japan

OnbehalfoftheOrganizingCommitteeforHESSS4May16th,2017

OrganizingCommittee

ScientificCommitteeHyungjunKim

The University of Tokyo, Japan(Chair)JacobSchewe

PotsdamInstituteforClimateImpactResearch,GermanyGabAbramowitz

University of South Wales,AustraliaMichaelEk

NOAA/NCEP/EMC,USAAaronBoone

CNRMMeteo-France/CNRS,FranceSoniaSeneviratne

ETH,SwitzerlandForrestHoffman

OakRidgeNationalLaboratory,USAJamesFamiglietti

NASA Jet Propulsion Laboratory,USAToshiyukiNakaegawa

MeteorologicalResearchInstitute,JapanTaikanOki

The University of Tokyo,JapanLocalOrganizingCommittee

AyakoKurosawa,YukiTsukada,YukihikoOnuma,TomokoNitta,DaiYamazaki,TakaoYoshikane,ToshiyukiNakaegawa,MasashiKiguchi,KeiYoshimura,YukikoHirabayashi,HyungjunKimandTaikanOki

TheUniversityofTokyo,JapanVenueMap

Program

05.16TUE-DAY115:00–16:00 Registration@An401/40216:00– Icebreaker/*Lightningtalks@An301/302

05.17WED-DAY208:00–08:50 Registration@ConventionHall08:50–09:00 Welcomeaddress@ConventionHall09:00–10:30 S1.ChangesofClimateForcingandTerrestrialFeedback

@ConventionHall10:30–11:00 *Postercore-time/Coffeebreak11:00–12:30 S2.ChangingClimateandNatural-HumanSystem

@ConventionHall12:30–14:00 Lunch14:00–15:30 W1@ConventionHall T1@An401/40215:30–16:00 *Postercore-time/Coffeebreak16:00–17:30 W2@ConventionHall O1@An401/402

05.18THU-DAY308:30–09:00 Registration@ConventionHall09:00–10:30 S3.SatelliteRemoteSensingandModelIntegration

@ConventionHall10:30–11:00 *Postercore-time/Coffeebreak11:00–12:30 S4.DealingwithUncertaintiesinModelSimulations

@ConventionHall12:30–14:00 GroupPhoto/Lunch14:00–15:30 W3@ConventionHall T2@An401/40215:30–16:00 *Postercore-time/Coffeebreak16:00–17:30 W4@ConventionHall O2@An401/40217:30– Banquet

05.19FRI-DAY409:00–10:30 S5.DeliveringSciencetoSociety

@SBlockPresentationRoom10:30–11:00 Postercore-time/Coffeebreak11:00–12:30 S6.GeneralDiscussion&ClosingRemarks

@SBlockPresentationRoom12:30–15:30 JointProjectMeeting(invitationonly)@An401/402

*PostershouldbepreparedinportraitlayoutwhichsizesuptoA0.*AllposterpresentersareinvitedtoLightningtalkssessiontointroducetheirpostersin2-min.

SymposiumSessions /*AllInvited*/S1.ChangesofClimateForcingandTerrestrialFeedback<Day2> 9:00-9:20 Assessmentofland-climatefeedbacksandsystematicbiasesofLSMs:

LandSurface,SnowandSoilMoistureModelIntercomparisonProject(LS3MIP)-HyungjunKim(TheUniversityofTokyo,Japan)

9:20-9:40 EarthSystemModel-SnowModelIntercomparisonProject(ESM-SnowMIP)-GerhardKrinner(LGGE,France)

9:40-10:00 AdvancingourunderstandingoftheimpactsofhistoricandprojectedlanduseintheEarthSystem:TheLandUseModelIntercomparisonProject(LUMIP)-DaveLawrence(NCAR,USA)

10:00-10:30 Discussion

S2.ChangingClimateandNatural-HumanSystem<Day2> 11:00-11:20 Inter-SectoralImpactModelIntercomparisonProject(ISIMIP):

Modelingclimateimpactsacrosssectors-JacobSchewe(PIK,Germany)

11:20-11:40 Opportunitiesandconstraintsforimprovedwaterresourcesmanagementusingdifferentlensesandscales-YoshihideWada(IIASA,Austria)

11:40-12:00 Specifyingsourcesofhumans'waterabstractionbyusingtheH08globalhydrologicalmodel-NaotaHanasaki(NIES,Japan)

12:00-12:30 Discussion

S3.SatelliteRemoteSensingandModelIntegration<Day3> 9:00-9:20 DeliveringHydrologicSciencetoSociety:SuccessesandChallengesin

theWesternUnitedStates-JamesFamiglietti(JPL/NASA,USA)

9:20-9:40 SatelliteRemoteSensingandModelIntegrationinJAXA-RikoOki(EORC/JAXA,Japan)

9:40-10:00 DIASContributingtoClimateChangeAnalysisandDisasterRiskReduction-AkiyukiKawasaki(TheUniversityofTokyo,Japan)

10:00-10:30 Discussion

S4.DealingwithUncertaintiesinModelSimulations<Day3> 11:00-11:20 SystematicEvaluationofLandSurfaceModelsUsingtheInternational

LandModelBenchmarking(ILAMB)Package-ForrestHoffman(OakRidgeNationalLaboratory,USA)

11:20-11:40 ProtocolfortheAnalysisofLandSurfacemodels(PALS)andmodelevaluation.org-GabAbramowitz(UniversityofSouthWales,Australia)

11:40-12:00 TargetedexperimentstoassesstheParisAgreementonClimateChange-DanielMitchell(BristolUniversity,UK)

12:00-12:30 Discussion

S5.DeliveringSciencetoSociety<Day4> 9:00-9:20 GlobalHydrologyintheAnthropoceneandSDGs-TaikanOki(The

UniversityofTokyo,Japan)

9:20-9:40 Socio-hydrology:Use-inspiredBasicScienceintheAgeoftheAnthropocene-MurugesuSivapalan(UniversityofIllinoisUrbanaChampaign,USA)

9:40-10:00 SustainableDevelopmentGoals(SDG)Lab:RuralSystemsVisioneering-JoonKim(SeoulNationalUniversity,Korea)

10:00-10:30 Discussion

S6.GeneralDiscussion&ClosingRemarks<Day4> 11:00-12:30 Reports/Discussion/Closingremarks

Rapporteur: Aaron Boone (Meteo-France, Franch), Ahmed Tawfik(NCAR,USA),KurtSolander(LoaAlamosNationalLab.,USA)

WorkshopSessionsW1.ClimateExtremes<Day2> 14:00-14:15 /*Invited*/Climatechangeandadaptationstrategiesforflood

controlplaninHokkaido,Japan-TomohitoYamada(HokkaidoUniversity,Japan)

14:15-14:30 /*Invited*/California’swatercycleextremes:fromdroughttodeluge-Jin-HoYoon(GwangjuInstituteofScienceandTechnology,Korea)

14:30-14:45 Impactsofclimatechangeonhydrologicextremesunder1.5/2.0Cdegreesglobalwarming-SatoshiWatanabe(TheUniversityofTokyo,Japan)

14:45-15:00 ConvectionandLandCoverChange:Interpretingsimulations,gaps,andmovingforward–AhmedTawfik(NCAR,USA)

15:00-15:15 FutureprojectionofextremeprecipitationeventslinkedtotemperatureoverJapanunderdifferentfuturescenarios-SridharaNayak(NationalResearchInstituteforEarthScienceandDisasterResilience,Japan)

15:15-15:30 Recentextremeevents:What'spredictable,what'snot,andwhattodo-SimonS.-Y.Wang(UtahStateUniversity/UtahClimateCenter,USA)

W2.InteractionbetweenNatureandSociety<Day2> 16:00-16:15 /*Invited*/Localandremoteclimateresponsetodeforestationin

MaritimeContinent-Min-HuiLo(NationalTaiwanUniversity,Taiwan)

16:15-16:30 /*Invited*/Historicalsocio-hydrologyinJapan:capturingtheregimeshiftsandphenomenainthemodernera-ShinichiroNakamura(NagoyaUniversity,Japan)

16:30-16:45 Multi-ModelRegionalSimulationofClimateChangeImpactsonAgricultureandEcosystemsintheSouthwesternUnitedStates-SeungHeeKim(ChapmanUniversity,USA)

16:45-17:00 100-yearGlobalWarmingPotentialofRiceCropIntensity-DIMWanndet(TokyoInstituteofTechnology,Japan)

17:00-17:15 EvaluationoftheapplicationoftheISI-MIPbias-correctionmethodoffuturesimulationsofclimateoverIndonesiafortheimplementationofClimateChangeAdaptationPlans-MartinGomezGarcia(NipponKoeiCo.,Ltd.R&D,Japan)

17:15-17:30 ModelingImpactsofIrrigationonLandSurfaceHydrologyandSubseasonalForecast-YaduPokhrel(MichiganStateUniversity,USA)

W3.HydrologicForecastandDataAssimilation<Day3> 14:00-14:15 /*Invited*/Developmentofglobalandregionalfloodforecasting

systemandtheirvalidation-KeiYoshimura(TheUniversityofTokyo,Japan)

14:15-14:30 Sensitivityanalysisofhistoricalweatherdocumentsforreconstructingpastclimate-PandukaNeluwala(TheUniversityofTokyo,Japan)

14:30-14:45 TheComputableCatchment:Interactivemediaformodel-datasharingwithimplicationsforscientist-stakeholderaccessandparticipation-ChristopherDuffy(PennStateUniversity,USA)

14:45-15:00 HydrologicSciencesforFloodDisasterMitigations:IntegrationofLocalInformationandFloodModeling-TakahiroSayama(DPRI,KyotoUniversity,Japan)

15:00-15:15 Towardsecohydrologicaldroughtmonitoringandpredictionusingalanddataassimilationsystem-YoheiSawada(MeteorologicalResearchInstitute,JapanMeteorologicalAgency,Japan)

15:15-15:30 TheperformanceoflandsurfaceandcumulusconvectionschemeinthesimulationofIndianSummerMonsoonusingRegCM4-SumanMaity(IndianInstituteofTechnologyKharagpur,Malaysia)

W4SatelliteRemoteSensingforExtremeMonitoring<Day3> 16:00-16:15 /*Invited*/Progressinunderstandinghydrologicfloodingusing

GRACE-JohnReager(NASAJPL,USA)

16:15-16:30 Distributionofdebristhicknessanditseffectonglaciermeltatlarge

scale-YukikoHirabayashi(TheUniversityofTokyo,Japan)

16:30-16:45 Reconstructingclimate-drivenwaterstoragevariability-VincentHumphrey(ETHZurich)

16:45-17:00 LocalEnsembleTransformKalmanFilter(LETKF)dataassimilationofaltimetrydataforSWOTmission-Jean-FrancoisVuillaume(TheUniversityofTokyo,Japan)

17:00-17:15 Globaldistributionofgroundwater-vegetationspatialcovariation-SujanKoirala(MaxPlanckInstituteforBiogeochemistry)

17:15-17:30 Heavyrainpredictionapplyingsatellite-basedclouddataassimilationoverland-RieSeto(TokyoInstituteofTechnology,Japan)

TutorialSessionsT1.ILAMBModelBenchmarkingPackageI<Day2> 14:00-15:30 Description,Installation,andBasicOperation-NathanCollier(ORNL,

USA)

T2.ILAMBModelBenchmarkingPackageII<Day3> 14:00-15:30 Userissues,PackageDesign,andAdvancedUsage-NathanCollier

(ORNL,USA)

OpenScienceSessionsO1.OpenGLASS/GEWEXSession<Day2> 16:00-16:15 TBD

PetrusJvanOevelen(Int.GEWEXProjectOffice,USA)

16:15-16:30 WCRPandGEWEXscienceofrelevancetoGLASSSoniaSeneviratne(ETH,Switzerland)

16:30-16:45 ThecurrentstatusofactivitiesintheGEWEXGlobalLand/AtmosphereSystemStudy(GLASS)MikeEk(NCEP,USA)

16:45-17:00 GSWP3asaliaisonacrosscommunitiesHyungjunKim(TheUniversityofTokyo,Japan)

17:00-17:30 Discussion

O2.OpenISIMIPSession<Day3> 16:00-16:15 Simulatingtheimpactsofclimateextremes:thecaseofthe2003

Europeanheatwave-JacobSchewe(PotsdamInstituteforClimateImpactResearch,Germany)

16:15-16:30 BenchmarkingcarbonfluxesoftheISIMIP2abiomemodels-JinfengCHANG(LOCEAN-IPSL,France)

16:30-16:45 ImpactsimulationsforISI-MIPwithaprocess-basedbiomemodel,

VISIT-AkihikoIto(NationalInstituteforEnvironmentalStudies,Japan)

16:45-17:00 Humanimpactparameterizationinglobalhydrologicalmodelsimprovesestimatesofmonthlydischargesandhydrologicalextremes:amulti-modelvalidationstudy-TedVeldkamp(VUUniversityAmsterdam,Netherlands)

17:00-17:30 Discussion

JointProjectMeetings<Day4> 12:30-15:30 JointProjectMeetings/*invitationonly*/

PosterSession

01. Regionalclimatedownscalingforriskinformation-EstablishmentofCORDEXAsiaEmpirical-StatisticalDownscaling(ESD)Group-KojiDairaku(NationalResearchInstituteforEarthScienceandDisasterResilience,Japan)

02. DoSoilWaterLimitationsSuppressConvection?-AhmedTawfik(NCAR,USA)

03. EvaluationofhighlatitudeslandclimatesimulatedbyMIROC5AGCMandawetlandscheme-TomokoNitta(TheUniversityofTokyo,Japan)

04. CurrentStatusofGlobalandRegionalCO2Budgets-MasayukiKondo(CenterforEnvironmentalRemoteSensing(CEReS),ChibaUniversity,Japan)

05. Reconciliationoftop-downandbottom-upCO2fluxesinSiberialarchforest-KumikoTakata(NationalInstituteforEnvironmentalStudies,Japan)

06. Hydroclimaticintensityin1.5&2.0°Cwarmerworld-GavinMadakumbura(TheUniversityofTokyo,Japan)

07. AnalysisofnegativeemissionapplicationstatusinSouthKoreaForest.-NahuiKim(KoreaUniversity,Korea)

08. StatisticalEvaluationofSoilWetnessChangesinFutureClimateinCMIP5Multi-ModelEnsemblesinEastAsia-TosiyukiNakaegawa(MeteorologicalResearchInstitute,Japan)

09. DynamicaldownscalingsimulationoverEastAsiausingacoupledatmosphere-oceanregionalclimatemodelRSM-ROMS-XiaojunGUO(TheUniversityofTokyo,Japan)

10. ValuationIrrigationwaterbyyieldcomparisonapproach:globalscale-SobhanAfraz(TheUniversityofTokyo,Japan)

11. TheCommunityWaterModel(CWATM)-Developmentofacommunitydrivenglobalwatermodel-YusukeSatoh(InternationalInstituteforAppliedSystemsAnalysis,Austria)

12. Global-scaleriverfloodvulnerabilityinthelast50years-MasahiroTANOUE(TheUniversityofTokyo/InstituteofIndustrialScience,Japan)

13. Understandingthesoilmoisture-climatecouplinginirrigationregionsoftheIndiathroughsatelliteandassimilationdatasets-ZohaibiMuhammad(SungkyunkwanUniversity,Korea)

14. Long-termprojectionsofglobalwateruseforelectricitygeneration-SayakaYOSHIKAWA(TokyoInstituteofTechnology,Japan)

15. Interactionofwater,landuse,andecosysteminIntegratedTerrestrialModel:abio-

geophysicallandsurfacemodelwithhumancomponents-TokutaYokohata(NationalInstituteforEnvironmentalStudies,Japan)

16. ImpactsofLarge-scaleSolarPhotovoltaicDeploymentonGlobalClimateVariability-MasahitoOmori(TheUniversityofTokyo,Japan)

17. Intercomparisonofregulatedriverdischargeamongmultiplehydrologicalmodels-YoshimitsuMasaki(HirosakiUniversity,Japan)

18. Springtimetrans-PacifictransportofAsianpollutantsrelatedtotheWesternPacificteleconnection-Ja-HoKoo(YonseiUniversity,Korea)

19. UncertaintyOfRainfallAmountAndRiverDischargeInMountainousCatchmentsAssociatedWithSpatialDistributionOfRainfallObservationAndMeteorologicalPrediction-KonosukeShibata(HokkaidoUniversity,Japan)

20. ClassificationandforecastofheavyrainfallinnorthernKyushuduringBaiuseasonusingSelf-OrganizingMaps(SOM)-DzungNguyen-Le(FacultyofEngineering,HokkaidoUniversity,Japan)

21. EstimationofFloodDischargebyAssimilatingWater-LevelandRatingCurveintoQuasi-2DHydraulicModel-TsuyoshiHoshino(Hokkaidouniversity,Japan)

22. DevelopmentofaEulerianglobalriverwatertemperaturemodel-DaisukeTokuda(TheUniversityofTokyo,Japan)

23. EffectofanexponentialdecayofsaturatedhydraulicconductivityonequilibriumsoilmoistureandwatertabledepthinMAT-GW-NatsukiYoshida(TheUniversityofTokyo,Japan)

24. Benchmarkingstrategyformulti-modelensemblesimulationsandoptimizationofensemblemean-RyotaroDoi(TheUniversityofTokyo,Japan)

25. EstimatingEvapotranspirationwithlimitedclimatedatainAndongDamWatershedusingFAO56Penman-Monteithmethodology-SeaJinKim(KoreaUniversity,Korea)

26. Evapotranspirationprocessesareevaluatedinglobalmodels(hydrologicalandLandSurfaceModels)usingtheiLAMBsystemandusinganewmetrictowhichquantifiesdifferenttimescalesofdryingafterarainevent-EleanorBlyth(CentreforEcologyandHydrology,UK)

27. ImpactsofsoilmoistureinitializationonborealsummersubseasonalforecastsintheGloSea5predictionsystem-EunkyoSeo(UNIST,Korea)

28. HotDroughtsandHotterExtremes:InvestigatingChangesinExtremeHydrologyintheColoradoRiverBasin,USA-KurtC.Solander(LosAlamosNationalLab.,USA)

29. InterannualvariabilityofsnowpropertiesinanupperreachbasinofSapporoareabetween2005and2017.-AtsushiNunokawa(GraduateschoolofEngineering,HokkaidoUniversity,Japan)

30. SeaSprayCharacteristicsunderAir-SeaBoundaryLayer-HirokiOkachi(HokkaidoUniversity,Japan)

31. MeteorologicalCharacteristicsOfAHeavySnowfallEventBasedOnMulti-DopplerRadarsOverSapporoAreaIn2015-YutaOhya(HokkaidoUniversity,Japan)

32. CorrectionoftherainrateestimatesofgroundradarusingGPM/DPRdata-TatsuyaShimozuma(NagasakiUniversity,Japan)

33. Highresolution,highaccuracyglobaltopographymapforlandhydrologymodeling-DaiYamazaki(TheUniversityofTokyo,Japan)

34. ImpactanalysisofmeteorologicalfactorsandSWIonmonthlyforestNDVIvalues-DongfanPiao(KoreaUniversity,Korea)

S1

AbstractAssessmentof land-climate feedbacks and systematicbiasesof LSMs: LandSurface,SnowandSoilMoistureModelIntercomparisonProject(LS3MIP)HyungjunKimTheUniversityofTokyoThesolidandliquidwaterstoredatthelandsurfacehasalargeinfluenceontheregionalclimate,itsvariabilityanditspredictability,includingeffectsontheenergyandcarboncycles.Notably,snowand soilmoisture affect surface radiation and flux partitioning properties,moisture storage andlandsurfacememory.Recently,theLandSurface,SnowandSoil-moistureModelIntercomparisonProject(LS3MIP)wasinitiatedasanintercommunityeffortbetweenGlobalEnergyandWaterCycleExchangesProject (GEWEX) andClimate andCryosphere (CliC) to contribute to the6thphaseofCoupledModelIntercomparisonProject(CMIP).TheexperimentstructureoftheLS3MIPwasdesignedtoprovideacomprehensiveassessmentoflandsurface,snow,andsoilmoisturefeedbacksonclimatevariabilityandclimatechange,andtodiagnosesystematicbiasesinthelandmodulesofcurrentAtmospheric-OceanGeneralCirculationModelsandEarthSystemModelswiththefollowingobjectives:

• evaluate the current state of land processes including surface fluxes, snow cover and soilmoisturerepresentationinCMIP6DECKruns;• estimate multi-model long-term terrestrial energy/water/carbon cycles, using the surfacemodules of CMIP6 models under observation constrained historical (land reanalysis) andprojectedfuture(impactassessment)conditionsconsideringlanduse/landcoverchanges;•assesstheroleofsnowandsoilmoisturefeedbacksintheregionalresponsetoalteredclimateforcings,focusingoncontrolsofclimateextremes,wateravailabilityandhigh-latitudeclimateinhistoricalandfuturescenarioruns;•assess the contributionof land surfaceprocesses to the currentand futurepredictabilityofregionaltemperature/precipitationpatterns.

The outcomes of the LS3MIP will eventually contribute to the improvement of climate changeprojections by reducing the systematic biases and representing better feedback mechanisms incoupledmodels.Further,theimpactsofclimatechangeonhydrologicalregimesandavailablefreshwaterresourcesincluding extreme events, such as floods and droughts, will be assessed based on multi-modelensemble estimates of long-term historical and projected future changes in energy, water, andcarbon cycles over land surfaces. Those achievements will contribute to the next cycle of theIntergovernmentalPanelonClimateChange.Correspondingemail:hjkim@iis.u-tokyo.ac.jp

S1

ESM-SnowMIPGerhardKrinnerIGE/CNRSSnowhasbeenrecognizedasanimportantsourceofclimatefeedbacksforalongtime.Alreadyinthe very first three-dimensional climate simulations, snow caused a substantial drift in thesimulated climate, and even today, snow is a sustantial source of inter-model spread incoordinated climate modeling exercises. The ESM-SnowMIP coordinated modeling exercise is acomplementaryeffort toLS3MIP inwhichwepursueseveralgoals.The firstgoal is toassess thefitness forpurposeof current snowmodules in theEarthSystemModelsparticipating toCMIP6.Thisisprimarilydonethroughsite-scaleandglobal-scaleuncoupledsimulations.Onthesitescale,theESMlandsurfacemodelsarecomparedwith(putatively)moreelaboratespecificsnowmodelsinordertoidentifytheoptimumcomplexityandtheminimumsetofsnow-relatedprocessesthatneed tobe represented in current-generationEarth SystemModels.On larger scales, theofflinesimulations are extensions to LS3MIP simulations that specifically aim at assessing snow-relatedmodelperformance.Asecondgoalistobetterquantifysnow-relatedlandsurfacefeedbacksintheEarthSystem.TherequiredsimulationsaretightlylinkedtorelevantLS3MIPsimulations.Correspondingemail:gerhard.krinner@cnrs.fr

S1

AdvancingourunderstandingoftheimpactsofhistoricandprojectedlanduseintheEarthSystem:TheLandUseModelIntercomparisonProject(LUMIP)DavidLawrenceNationalCenterforAtmosphericResearchHuman land-use activities have resulted in large changes to the Earth surface, with resultingimplicationsforclimate.Inthefuture,land-useactivitiesarelikelytoexpandandintensifyfurtherto meet growing demands for food, fiber, and energy. ÂThe Land Use Model IntercomparisonProject(LUMIP)aimstofurtheradvanceunderstandingoftheimpactsofland-useandland-coverchange(LULCC)onclimate,specificallyaddressingthequestions:(1)WhataretheeffectsofLULCCon climate and biogeochemical cycling (past-future)? (2) What are the impacts of landmanagement on surface fluxes of carbon, water, and energy and are there regional land-management strategies with promise to help mitigate against climate change? ÂIn addressingthese questions, LUMIP will also address a range of more detailed science questions to get atprocess-levelattribution,uncertainty,data requirements,andother related issues inmoredepthandsophisticationthanpossibleinamulti-modelcontexttodate.Therewillbeparticularfocusonthe separation and quantification of the effects on climate from LULCC relative to all forcings,separationofbiogeochemicalfrombiogeophysicaleffectsofland-use,theuniqueimpactsofland-coverchangeversuslandmanagementchange,modulationofland-useimpactonclimatebyland-atmosphere coupling strength, and the extent that impacts of enhanced CO2 concentrations onplantphotosynthesisaremodulatedbypastandfuturelanduse.ÂLUMIPinvolvesthreemajorsetsof scienceactivities: (1)developmentofanupdatedandexpandedhistoricaland future land-usedataset,(2)anexperimentalprotocolforspecificLUMIPexperimentsforCMIP6,and(3)definitionofmetricsanddiagnosticprotocolsthatquantifymodelperformance,andrelatedsensitivities,withrespect to LULCC. In this manuscript, we describe the LUMIP activity (2), i.e., the LUMIPsimulationsthatwillformallybepartofCMIP6.ÂTheseexperimentsareexplicitlydesignedtobecomplementarytosimulationsrequested intheCMIP6DECKandhistoricalsimulationsandotherCMIP6MIPs including ScenarioMIP, C4MIP, LS3MIP, and DAMIP. LUMIP includes a two-phaseexperimental design. Phase one features idealized coupled and land-only model simulationsdesigned to advance process-level understanding of LULCC impacts on climate, as well as toquantify model sensitivity to potential land-cover and land-use change. Phase two experimentsfocus on quantification of the historic impact of land use and the potential for future landmanagement decisions to aid inmitigation of climate change.Wewill present the experimentalprotocol,explaintherationale,outlineplansforanalysis,anddescribeanewsubgridland-usetiledata request for selected variables (reporting model output data separately for primary andsecondaryland,crops,pasture,andurbanland-usetypes).Correspondingemail:dlawren@ucar.edu

S2

ISIMIP:ModelingclimateimpactsacrosssectorsJacobSchewePotsdamInstituteforClimateImpactResearchIn this talk I will present the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP,www.isimip.org). ISIMIP is a community-driven climate-impacts modelling initiative, bringingtogether teams from different disciplines including hydrology, ecology, or environmentaleconomics. ISIMIP offers a consistent framework for cross-sectoral, cross-scalemodelling of theimpacts of climate change. The key goal of ISIMIP is to contribute to the comprehensive (cross-sectoral) understanding of the impacts of politically and scientifically-relevant climate-changescenarios.Iwilloutlinetheprojectstructure,showsomeselectedresults,andgiveanoutlookonthe upcoming, third phase of ISIMIP. Most importantly perhaps, I will explain how people canparticipateinISIMIP,andhowitcanlinktoothercommunityinitiatives.Correspondingemail:jacob.schewe@pik-potsdam.de

S2

Opportunities and constraints for improved water resources management usingdifferentlensesandscalesYoshihideWadaInternationalInstituteforAppliedSystemsAnalysisThequest forwatersecurityhasbeenastrugglethroughouthumanhistory.Only inrecentyearshasthescaleofthisquestmovedbeyondthelocal,tothenationalandregionalscalesandtotheplanet itself. Absent or unreliable water supply, sanitation and irrigation services, unmitigatedfloods and droughts, and degraded water environments severely impact half of the planet’spopulation. Over the past few years, water insecurity has become recognized in the WorldEconomicForumglobalriskstudiesasoneofthegreatestthreatsthatbusinessleadersthemselvesseethattheyfaceinthefuture,bothintermsoflikelihoodandscale.Thescaleandcomplexityofthewaterchallengesfacedbysociety,particularlybutnotonlyintheworld’spoorestregions,arenow recognized, as is the imperative of overcoming these challenges for a stable and equitableworld.Howcanweensure thewell�beingof all people andecosystemswith thewater, human,technological,andfinancialresourcesavailable?IntheframeworkoftheSustainableDevelopmentGoalswaterhastobemanagedmoreeffectivelyandwiselybyunlockingscientific,managerial,andbusinesscapabilities;breakingoutoftechnologicallock-in;andinnovativeandadaptiveportfoliosofsolutionshavetobedevelopedwhileremovingbarrierstoprogressonsoundwatergovernance.IIASA’s Water Futures and Solutions Initiative (WFAS) is an unprecedented inter-disciplinaryscientific initiativeto identifyrobustandadaptiveportfoliosofoptionalsolutionsacrossdifferenteconomicsectors, includingagriculture,energyandindustry,andtotestthesesolution-portfolioswithmulti�model ensembles of hydrologic and sectormodels to obtain a clearer picture of thetrade�offs,risks,andopportunities.TheresultsofWFaSscenariosandmodelswillprovideabasisforlong�termstrategicplanningofwaterresourcedevelopment.Andgiventhecomplexityofthewatersystem,WFaSwilluniquelyprovidepolicymakerswithoptionalsetsofsolutionsthatworktogetherand that canbeeasily adaptedas circumstances change in the future.Water is alsoallaboutrelationships.AsWFaSprogresses,itwillestablishanetworkinvolvinginformationexchange,mutual learning and horizontal cooperation across teams of researchers, public and privatedecisionmakers and practitioners exploring solutions at regional, national and local scales. Theinitiative includes amajor stakeholder consultation component, to informand guide the scienceandtotestandrefinepolicyandbusinessoutcome.Correspondingemail:wada@iiasa.ac.at

S2

Specifyingsourcesofhumans'waterabstractionbyusingtheH08globalhydrologicalmodelNaotaHanasakiNationalInstituteforEnvironmentalStudiesHumansabstractwaterfromvariouswatersourcestosustaintheirlivelihoodandsociety.Someofglobal hydrological models include explicit schemes of humans’ water abstraction, but theirrepresentations and performances are still largely limited. Here we substantially enhanced thewaterabstractionschemesoftheH08globalhydrologicalmodel. Itenabledustoestimatewaterabstraction fromsixmajorwater sources,namely, streamflowregulatedbyglobal reservoirs (i.e.reservoirsregulatingtheflowoftheworldmajorrivers),aqueductwatertransfer,localreservoirs,seawater desalination, renewable groundwater, and non-renewable groundwater. In a standardsetup, the model covers the whole globe at the 0.5°×0.5° of spatial resolution, and thecalculation interval isaday.Waterbalance isalwaysclosedatanycalculationgrid-cellsandtimeintervals.Aglobalhydrologicalsimulationwasconductedtovalidatetheperformanceofmodelforthe period of 1979-2013. The simulated water fluxes on water abstraction were extensivelyvalidatedwiththosereported inearlierpublicationsandshowedgoodandfairagreementat theglobalandcountryscalerespectively.Thesimulatedmonthlyriverdischargeandterrestrialwaterstorage of six heavily human-affected river basins in the world were compared with theobservationat river gauging stations anda satelliteproductof theGravityRetrieval andClimateExperiment mission (GRACE). The results indicated that in circa 2000, of total 3628 km3yr-1freshwaterrequirementglobally,2839km3yr-1wastakenfromsurfacewater,789km3yr-1 fromgroundwater. Streamflow, aqueduct water transfer, local reservoir, and seawater desalinationaccountedfor1786,199,106,1.8km3yr-1ofsurfacewaterrespectively.Forremaining747km3yr-1ofsurfacewaterrequirementwasunmet,orsurfacewaterwasnotavailablewhenandwhereitis needed inour simulation.Renewableandnon-renewable groundwater accounted for607and182km3yr-1ofgroundwaterrespectively.Eachsourcediffers in renewability,economiccosts fordevelopment,andenvironmentalconsequencesforusage.Themodelwouldbeusefultoenhancethe global water resources assessments by adding the aspects of sustainability, economy, andenvironment.Correspondingemail:hanasaki@nies.go.jp

S3

Delivering Hydrologic Science to Society:Successes and Challenges in the WesternUnitedStatesJayFamigliettiCaliforniaInstituteofTechnology,NASAJetPropulsionLaboratoryDelivering hydrologic science to society is a noble cause that our community should engage in.However, while the delivery of hydrologic information is straightforward, its uptake and use bystakeholdersandthegeneralpublic ismorechallenging.OurteamattheUniversityofCalifornia,Irvine,andatCaltech’sNASAJetPropulsionLaboratory,hasbeendeeplyengaged insucheffortsfornearlyadecade.Fromthelocalwaterdistricttofederallevels,andfromtheeverydaycitizen,toregionalwatermanagersandtotheU.S.Congress,wecontinuetotraversethewaterlandscapetoinform,educate,andshareimportantdata,observationsandmethods.InthispresentationIwillshare someofour successand failures,whilehighlighting critical challenges thatour communitymustfaceinitseffortstobetteraddresssocietalneeds. Correspondingemail:James.Famiglietti@jpl.nasa.gov

S3

SatelliteRemoteSensingandModelIntegrationinJAXAR.Oki,T.Kubota,M.Kachi,K.YoshimuraTheJapanAerospaceExplorationAgencyThe Japan Aerospace Exploration Agency (JAXA) Earth Observation Research Center providesvarious kinds of dataset from JAXA Satellite Monitoring for Environment Study (JASMES)[http://kuroshio.eorc.jaxa.jp/JASMES/], Global Satellite Mapping of Precipitation (GSMaP), andHimawari-Monitor [http://www.eorc.jaxa.jp/ptree/].Thesedatasetsarebaseduponobservationsfrom JAXA’s "SHIZUKU" or the Global Change ObservationMission -Water (GCOM-W) satellite,GlobalPrecipitationMeasurement(GPM)mission,theJMA’sHimawari-8,andsoon.DatafromtheSecond Generation Global Imager (SGLI) for the future GCOM – Climate (GCOM-C)mission andEarth Cloud, Aerosol and Radiation Explorer (EarthCARE) will be also used when it becomesavailableaftertheirlaunches.Userscanaccessthosedatabysearchingsensor/satellitenamesandatthesametimetheycanfinddatasets by topic by topic. In recent years,multi sensor/satellite data processing becomesmorecommon. For example, JAXA processes GSMaP, which is a blended Microwave-IR precipitationproductandhasbeendevelopedinJapantowardtheGPMmission.Progress insatelliteremotesensingdataandmodel integrationarea isremarkable.JAXA/EORCisdeveloping both global and local (Japan area) land surface simulation systems, called theYesterday’s Earth at EORC (YEE), in collaboration with the University of Tokyo for water cycleresearch.Satelliteobservationdataisusedasinputstothelandsurfacemodelaswellasobjectiveanalysisorforecastdataofmeteorologicalparametersatthesurface.GSMaPisalsousedasinputforYEE.Inaddition,GSMaPwillbeusedfordataassimilationinNICAM-LETKFdevelopedbyRIKENforshortrangeweatherforecast.These datasets can contribute to global water cycle studies. Furthermore, the JAXA/EORC is developing Joint-Simulator (Joint Simulator for Satellite Sensors) with universities and research institutes, which is helpful for evaluations of climate models with satellite data.

S3

DIASContributingtoClimateChangeAnalysisandDisasterRiskReductionAkiyukiKawasakiTheUniversityofTokyoIn 2015, global attempts were made to reconcile the relationship between development andenvironmental issues. This led to the adoption of key agreements such as the SustainableDevelopment Goals. In this regard, it is important to identify and evaluate under-recognizeddisaster risks thathinder sustainabledevelopment:measures tomitigate climate changeare thesameasthosethatbuildresilienceagainstclimate-relateddisasters.Todothisweneedtoadvancescientificandtechnicalknowledge,builddata infrastructurethatallowsustopredicteventswithgreater accuracy, and develop data archives. For this reason we have developed the DataIntegration and Analysis System (DIAS). DIAS incorporates analysis, data andmodels frommanyfields anddisciplines. It collects and storesdata from satellites, groundobservation stations andnumericalweatherpredictionmodels; integrates thisdatawithgeographicaland socio-economicinformation; then generates results for crisis management of global environmental issues. ThispresentationgivesanoverviewofDIASandsummarizes itsapplicationtoclimatechangeanalysisanddisasterriskreduction.DIASaimstoinitiatecooperationbetweendifferentstakeholders,andcontribute to the creation of scientific knowledge. DIAS provides a model for sharingtransdisciplinaryresearchdatathatisessentialforachievingthegoalofsustainabledevelopment.Inthepresentation,followingtheintroductionofDIAS,theCoupledModelIntercomparisonProject5 (CMIP5) analysis toolwill be introduced for supporting climate changeanalysis. Then, theAsiaWaterCycle Initiative(AWCI)and InternationalFlood Initiative(IFI)and itsrelatedactivatesweresummarizedasdatasharingandnetworkingforsocialproblemsolving.AnalysisonclimatechangeandfloodriskreductioninAsianandAfricancountrieswillbedemonstratedasacasestudy.Correspondingemail:kawasaki@hydra.t.u-tokyo.ac.jp

S4

Systematic Evaluation of Land Surface Models Using the International Land ModelBenchmarking(ILAMB)PackageForrestM.HoffmanOakRidgeNationalLaboratoryAs Earth system models (ESMs) become increasingly complex, there is a growing need forcomprehensive andmulti-faceted evaluation ofmodel predictions. To advance understanding ofbiogeochemical processes and their interactionswith hydrology and climate under conditions ofincreasing atmospheric carbon dioxide, new methods are needed that use observations toconstrainmodelpredictions, informmodeldevelopment,andidentifyneededmeasurementsandfieldexperiments.Improvedprocessparameterizationsareneededtoconstrainenergyandwaterpredictions in land surface models and better representations of biogeochemistry–climatefeedbacks and ecosystem processes in ESMs are essential for reducing uncertainties associatedwith projections of climate change during the remainder of the 21st century. The InternationalLand Model Benchmarking (ILAMB) project seeks to 1) develop internationally acceptedbenchmarks for land model performance, 2) promote use of benchmarks for modelintercomparison projects, 3) strengthen linkages between experimental, remote sensing, andmodeling communities, and 4) support the design and development of an open sourcebenchmarking software system. Leveraging work on past model evaluation studies, we havedevelopedtwogenerationsofsuchbenchmarkingsoftwarepackagesthatassessmodelfidelityon24variables in fourcategories fromabout45data sets;producegraphicalglobal-, regional-,andsite-level diagnostics; and provide a hierarchical scoring system. The ILAMBv2 package, publiclyreleasedinMay2016,hasbecomeanintegralpartofmodelverificationworkflowforrapidmodeldevelopment and calibration cycles for the U.S. Department of Energy’s Accelerated ClimateModeling for Energy (ACME) model and the Community Earth System Model (CESM). We willpresent results from model analysis using the ILAMB packages, discuss techniques for routinemodel evaluation, propose coordinated evaluation of the Sixth Phase of the Coupled ModelIntercomparisonProject (CMIP6) output, anddescribenewmetrics that integrate across carbon,surfaceenergy,hydrology,andlandusedisciplines.Correspondingemail:forrest@climatemodeling.org

S4

ProtocolfortheAnalysisofLandSurfacemodels(PALS)andmodelevaluation.orgGabAbramowitzUNSWSydneyThefirstphaseofthePALSwebapplicationprovidedbenchmarkingandevaluationoflandsurfacemodelsataround40fluxtowersitesusingafixedsetofmetrics.PALSwasusedforthePALSLandsUrfaceModelBenchmarkingEvaluationpRoject(PLUMBER,amodelintercomparisonexperiment),which used out of sample empirical models as benchmarks, defining model performanceexpectationsapriori.Themodelevaluation.orgprojectbuildsonthestrengthsofPALS,focusingonbenchmarkingandworkflow traceability inanonlineenvironment,whileexpanding the scopeofpotentialexperiments.Itisdesignedtoabletoutilisedifferentpackagesasanalysis‘engines’,suchas ILAMB, LVT or ESMValTool, as well it’s own package based around the PALS R package, butexpandedtoregionalandglobalanalyses.Thedistributeddesignoftheapplicationalsomeansthatanalysisenginescanbesetupinvirtualmachinescolocatedwithlargemodeloutputs,withusersstillmaintaininga single interfacewith themainwebapplication. In thisnewsystem, thenatureandscopeofanexperimentiscontrolledbyusers,meaningthatglobalgriddedanalysesfunctioninthesamewayassite-basedfluxtoweranalyses,orindeedanyotherapplicationoutsideclimateorecosystemmodelling.Thisallowsforrelativelyeasyuseofthesystemforbothteam-basedmodeldevelopment and internationalMIPs, in an environment that stores ancillary information, aidingreproducibility and allowing the potential for data mining of ancillary information as part ofanalyses.Thispresentationwilldetailthedesignandworkflowofthemodelevaluation.orgsystem.Correspondingemail:gabriel@unsw.edu.au

S4

TargetedexperimentstoassesstheParisAgreementonClimateChangeDannMitchellUniversityofBristolThe Intergovernmental Panel on Climate Change (IPCC) has accepted the invitation from theUNFCCCtoprovideaspecialreportontheimpactsofglobalwarmingof1.5Cabovepre-industriallevels and related global greenhouse gas emission pathways. Many current experiments in, forexample, the Coupled Model Inter-comparison Project (CMIP), are not specifically designed forinform- ing this report.Here,wedocument thedesignof theHalf a degreeAdditionalwarming,Projections, Prognosis and Impacts (HAPPI) experiment. HAPPI provides a framework for thegenerationofclimatedatadescribinghowtheclimate,and inparticularextremeweather,mightdifferfromthepresentdayinworldsthatare1.5Cand2.0Cwarmerthanpre-industrialconditions.Output from 10 participating climate models includes variables frequently used by a range ofimpactmodels. The key challenge is to separate the impact of an additional approximately halfdegree of warming from uncertainty in climatemodel responses and internal climate variabilitythat dominate CMIP-style experiments. Large ensembles of simulations (>50 members) ofatmosphere-onlymodels forthreetimesliceexperimentsareproposed,eachadecade in length;the first being the most recent observed 10-year period (2006-2015), the second two beingestimatesoftheasimilardecadebutunder1.5and2Cconditionsacenturyinthefuture.WeusetheRepresentativeConcentrationPathways2.6(RCP2.6)toprovidethemodelboundaryconditionsforthe1.5Cscenario,andaweightedcombinationofRCP2.6andRCP4.5forthe2Cscenario.Correspondingemail:d.m.mitchell@bristol.ac.uk

S5

GlobalHydrologyintheAnthropoceneandSDGsTaikanOkiTheUniversityofTokyoDue to the high temporal and spatial variabilities in hydrological cycles, humanbeings has beenrequired to develop facilities to regulate water resources in order to mitigate the seasonallyuneven distribution of available freshwater resources by averaging the temporal variation withmillions of artificial reservoirs, lakes, and ponds in the world, and the geographically unevendistribution by physically transferring water through enormous lengths of canals and channels,sometimes tans-boundarymanner, even thoughonly 10%ofmaximumavailablebluewater and30%ofgreenwaterresourcesareusedgloballyatpresent.Consequentially,realhydrologiccyclesonthecurrentEarthisnotnaturalanymoreandweseealotoftracesofhumaninterventionsonhydrologiccyclesevenontheglobalscale.Further,owingtothewater’scomparatively lowpriceperweightandthecostoftransportation,itisrarelytransportedbyitselfoverlongdistancesmorethan 500 km (with the exception of bottled water), and water is an absolutely local resourcewithout a single international price. On the contrary, the prices of agricultural products aresufficiently high compared to transportation costs, and their trades are feasible. From waterresources management point of view, the international trade in water-intensive commoditiesbetweenwater-abundantandwater-poorcountriescanhelpwater-poorcountriessaveresources,andthisprocessiscalledas"virtualwatertrade".Basedonthelatestestimateofthevirtualwatertrade, it can be illustrated that rich but water-scarce countries tend to reduce local waterconsumptionbyimportingvirtualwater.Inaddition,itwasprevailedfromananalysisthatnationscharacterizedbynetvirtualwaterexportshavehigherwaterresourcesandincomepercapitaandthat no countries fall below a certain thresholdwith respect to both GDP andwater resources.Thesepointssuggestthatthevirtualwatertradeisexplainedbyeconomiccharacteristicsofwaterand that sustainable development depends on promoting the co-development of povertyalleviationandwaterresourcedevelopment.Intheeraofthe2030AgendaoftheUnitedNations,waterisnotonlythe7thGoaloftheSustainableDevelopmentGoals(SDGs)butoneofkeycross-cuttingissuesamongmanySDGs.Thecomparativelylowprice,hugeamountofdailyconsumption,and natural recycling characterize water as a unique natural resources and we should promotedeveloping more scientific and technological capacity on monitoring, understanding, predicting,andmanagingwatertosupportachievingSDGs.Correspondingemail:taikan@iis.u-tokyo.ac.jp

S5

Socio-hydrology:Use-inspiredBasicScienceintheAgeoftheAnthropoceneMurugesuSivapalanUniversityofIllinoisatUrbana-ChampaignWearewellandtrulyintheAnthropocene.Humanscannolongerbeconsideredasmereexternaldrivers or boundary conditions in water and environmental systems. The interactions andfeedbacksbetweenhumanactionsandwatercycledynamicsarethrowinguparangeofemergent‘‘bigproblems.’’Understandingandofferingsustainablesolutionstothese‘‘bigproblems’’requirea broadening of the water sciences to embrace the perspectives of both social and naturalscientists.Thenewscienceofsocio-hydrologywasintroducedwiththisinmindanddealswiththedynamics arising from thebi-directional feedbacksbetweencoupledhuman-water systems.Howthewatersystemrespondstohumanalterationsorinterferencesisgovernedbynaturallawsandhasbeenwellstudiedthroughdecadesofhydrologicresearch.Howthehumansystemrespondstohydrologic variability and change involves human choices and/or tradeoffs governed loosely bysocietal laws. Tradeoffs can be ofmany kinds: between humans and the environment, betweendifferent sectors using water, or between different groups in society. Within the prevailingenvironmentalandresourcemilieu,tradeoffsaremediatedbyhumanvaluesandnorms,whicharein turn shaped by the dynamics of the human-water system and must therefore be treated asendogenoustothesystem.Valuesandnormscanvaryalongupstream-downstream,urban-rural,humid-arid, rich-poor, or technological-green society gradients. In thisway tradeoffs can have aspatialdimension,andoveraperiodoftimecanmanifestinpatternsoflegacyeffectsrecordedinthe land/human-scapes. Bringing together the perspectives of both social and natural scientistsdealingwithwaterisgoodforwatersciences,havingthesalutaryeffectofrevitalizingthemasuse-inspired basic science. It is good formanagement too, in that the broader, holistic perspectivesprovidedby socio-hydrology canhelp recognizepotential ‘‘big’’ problems thatmayotherwisebeunforeseen and, equally, identify potential ‘‘alternative’’ solutions to otherwise intractableproblems.Correspondingemail:sivapala@illinois.edu

S5

SustainableDevelopmentGoals(SDG)Lab:RuralSystemsVisioneeringJoonKimSeoulNationalUniversityTo make breakthroughs on the Sustainable Development Goals (SDG) Labs initiatives, we areestablishing the ‘Rural Systems Visioneering (RSV)’ SDG Lab in the Arusha region in northernTanzania. ‘RSV’ SDG Lab will focus on clean water and sanitation (SDG 6), affordable and cleanenergy (SDG 7), sustainable cities and communities (SDG 11), responsible consumption andproduction(SDG12)andlifeonland(SDG15)by(1)co-creatinginnovativesustainabilityscience&technology necessary for the renewable energy-based electrical, climate-smart agricultural, andresilience-based educational fields with the ‘Tanzania-Korea Innovative Energy Technology (IeT)Center’ through cooperation with the local ‘Nelson-Mandela African Institute of Science andTechnology and (2) co-growing with rural villages in Arusha through visioneering - a triad ofgovernance,management andmonitoring for sustainable rural-urban systems. Tomobilize ruralpeople and villages, the following applications will be supplemented to the already initiatedTanzania-Korea IeT Center: (1) the conceptual framework of ‘self-organizing hierarchical opensystemswithvisioneering’,(2)sustainabilityeducationwithfocusesonnurturingthebasicandkeycompetences for sustainability (e.g., systems thinking, normative, strategic, and anticipatorycompetences), (3) climate-smart agriculture and its quantitative assessment based onbiotic/network/thermodynamicindicatorsbymonitoringandmodelingenergy-matter-informationflows in and out of rural systems using (eventually inexpensive) flux measurement, computermodeling and remote sensing, (4)multiagent-based systems analysis for emergent solutions forbetter productivity and profit for heterogeneous smallholder farmers, and (5) linking the above-mentioned efforts to the rural communities with feedback loops (i.e., guided self-organizationprocess) to create profit, to promote businesses, and to nurture entrepreneurship withstewardshipthatensuressustainability.Themainobjectiveofthispresentationistointroduce‘RSV’SDGLabinitiativetotheHESSScommunityandtosharecriticalinformationandpotentialresourcesforcollaborations.Correspondingemail:joon@snu.ac.kr

W1

ClimatechangeandadaptationstrategiesforfloodcontrolplaninHokkaido,JapanTomohitoYamadaHokkaidoUniversityThemodernhistorywithmodernhydrologicaltechnologiesinJapanwasstartedfromMeijimodernera (1868-1912) with import of Dutch river engineering. After this technological revolution,Japanese society had been rapidly changed from “green society” to “technological society” (DiBaldassarreetal.,2015)withinjustacentury.Duringthischanging,ingeneral,Japanhadalsofacedsomeoftypicalsocio-hydrologicalphenomenawhichobservedinothercountriesandbasins:leveeeffect,pendulumswing,lock-insituationetc.However,thesephenomenahavenotbeenexplainedin the context of socio-hydrology and quantitative analysis with historical dataset. Then, ourresearchgrouphasbeentried todevelop thehistoricaldataset relatedhuman-flooddynamics inthe Japanesemoderneraandcapturehistorical regimeshifts in thesociety. In thispresentation,we show that Japanesemodern era is separated into three specific historical classificationswithsocio-hydrological regimeshifts, anda typicalphenomenon in this rapid changing society, “leveegrowth”,whichisobservedbyourhistoricalspatialanalysiswithusingoldmaps.Correspondingemail:tomohito@eng.hokudai.ac.jp

W1

California’swatercycleextremes:fromdroughttodelugeJin-HoYoonGwangjuInstituteofScienceandTechnologyMoreclimateextremeeventshaveoccurredinrecentyears,includingthecontinualdevelopmentof extreme drought in California, the severe cold winters in the eastern U.S. since 2014, 2015Washingtondrought, andexcessivewildfireeventsoverAlaska in2015. In caseofCaliforniaandtheWesternUnitedStates,thisseveredroughtischangingtoextremefloodconditioninthewinterof2016-17.Thesehavebeencasuallyattributed toglobalwarming.However,aneed for furtherunderstanding ofmechanisms responsible for climate extremes is growing. In this presentation,we’ll use sets of climate model simulation that designed to identify the role of the oceanicfeedbackinincreasingclimateextremesunderglobalwarming.Correspondingemail:yjinho@gist.ac.kr

W1

Impacts of climate change on hydrologic extremes under 1.5/2.0C degrees globalwarmingSatoshiWatanabeTheUniversityofTokyoThechangesofhydrologicalextremesunder1.5/2.0Cdegreesglobalwarmingareestimatedbasedon themulti-model super ensemble experiments. After the Paris agreement, the importance ofprojectionfor1.5/2.0Cdegreesglobalwarmingisemerging.Thisstudyprojectshydrologicalcyclesunder1.5/2.0Cdegreeswarmerclimatesandassesstheassociatedimpacts.Sincetheestimationofuncertainty is indispensable in the assessment of extreme, using a large set of ensemblesimulationsisastraightforwardwaytodeliverappropriateinformation.Inthisstudy,theresultsofHalf a degree Additionalwarming, Prognosis and Projected Impacts (HAPPI) project are used. Intotal, 400 ensemble experiments from four climate models are prepared for each warmingcondition (i.e., historical, 1.5C, and 2.0C warming). To effectively use the super ensembleexperimentsattheaffordablecomputationalcosts,weadoptastrategytodirectlyusetheoriginalrunoff projections by climatemodels in contrast to the typical approach running a hydrologicalmodelforcedbyatmosphericboundaryconditionsbyclimatemodels.Inthisstudy,wesuggesttocorrectbiasof the runoff projectedby climatemodels rather thanmeteorological variables. Theproposedmethodadjuststhebiasbasedontherunoffsimulatedbythepreviousstudywhichaimstoreproducehistoricalhydrology.Hydrologicalprojectionsbasedonthebiasadjustedrunofffromsuper ensemble experiments reveal the impact of climate change on hydrology, especially forextremes,under1.5/2.0Cdegreesglobalwarming.Correspondingemail:stswata@sogo.t.u-tokyo.ac.jp

W1

Exploringthepropagationofuncertaintyinrainfallextremesthrough toarangeofhydrologicaloutcomes:floods,droughts,soilmoisture,evapotranspiration,surfacerunoffandgroundwaterrechargeEleanorBlythCentreforEcologyandHydrologyRainfall data is the most important and the most uncertain of all the climate variables whenconsideringhydrologicalimpacts.Thisistruebothofobservationsandofmodelforecastsofrainfall.This uncertainty drives many initiatives: new satellite products are newly available which mayimprove information about rainfall in data-sparse areas and numerical weather prediction iscontinually improving. But how good is ‘good enough’ in terms of rainfall for predictinghydrological outcomes such as floods, droughts and water resources? Uncertainties relating tototal amounts, spatial and temporal distributions of intensity and specific locations especially inextremerainfalleventsallhaveadifferentimpact.Hereweexploretheissueof‘howgoodisgoodenough’ in terms of precipitation data provided at large scales (25km).We use a suite of landsurface andhydrologymodels to explorehowuncertainty in the rainfall propagates through thelandtoessentialhydrologicaloutputs:floods,droughts,soilmoisture,evapotranspiration,surfacerunoffandgroundwaterrecharge.WewillexaminethisissueoverIberiabystudyingtheoutputsofanensembleofmodelsrunbyasuiteofdistributedrainfallproductsthatsampletheuncertaintyofrainfall amounts in this area. All themodels are runwith the samemeteorological forcing dataincludinga20-memberensembleofrainfallcreatedfromsatellitedataproductswhichsamplestheuncertaintyinthatregion.ThisstudyispartoftheEuropeanUnionFP7EartH2Observeproject.Wewill analyse the convergence or divergence of the model outputs in terms of the hydrologicaloutputs. This analysis will indicate the accuracy of the rainfall amount needed for varioushydrologicalapplications.Wewillexpandtheanalysis to theglobal scalebyanalysing theresultsfromasetofmodelsrunwitharangeofglobalsatellitederivedproducts.Theresultinganalysiswillindicatewhataccuracyisneededintermsofprecipitationforuseindifferenthydrologicalimpacts.The conclusions will inform both the satellite data-retrieval community and atmospheric modelcommunityaswellaslandsurfaceandhydrologymodellers.Correspondingemail:emb@ceh.ac.uk

W1

FutureprojectionofextremeprecipitationeventslinkedtotemperatureoverJapanunderdifferentfuturescenariosSridharaNayakNationalResearchInstituteforEarthScienceandDisasterResilienceAnumberofstudieshavesuggestedthattheextremeprecipitationeventsare increasedinmanyregionsacrosstheworld(IPCC2007,2012)duetoatmosphericwarming.ThisargumentisbasedontheClausius-Clapeyron(CC)relationshipwhichstatesthatatmospherecanholdmoremoistureatarate of 7% per degree rise in temperature (Trensberth, 2003). In our study, we analyzedmulti-GCMs/RCMs ensemble experiment results to provide the possible scenarios of extremeprecipitationevents linked temperature inpast50yearsand their future changes in coming100years under difference future scenarios (A1B, RCP4.5, 4Kwarming).We analyzed 167 ensembleresults [140experimentswith theNHRCM (50experiments for present climate: 1951-2010&90experiments for future climate: 2051-2110, 4K warming) and 27 experiments with the NHRCM,WRF,NRAMS(15experimentsforthecurrent:1981-2000;9forthefuture:2081-2100,RCP4.5and3 experiments with A1B scenario)]. Precipitation intensities of wet days (defined as ≥ 0.05mm/d) are stratified to different bins with 1°C temperature interval and different percentiles(50%,75%,85%,90%,95%,97%,and99%)ofprecipitationarecomputedineachbin.Theresultsofthe current climate are validated against APHRODITE and AMeDAS observations. The resultsindicate that all percentiles of precipitation intensities basically follow the CC relationship andincreasewith temperatureup to a certaindegree (~19-21°C), abovewhich further increasesoftemperaturecausedecreasesofprecipitationintensities.Therangeofuncertaintyvarieswiththepercentileofextremeeventsandgreateruncertaintycorrelatestoahigherpercentiles.However,observations are almost within the range of uncertainty of one standard deviation. Extremeprecipitation intensities are significantly increased by 5-15 mm/d in future climate fortemperatures above ~21°C in under SRES A1B, while the same is increase by 5 mm/d fortemperaturesabove~21°CunderRCP4.5scenarioandby~30mm/dforthetemperaturesabove~24°Cunder4Kwarming.Theratesofchangeofextremeprecipitationintensitiesarenoticedtobeconsistentunderallfuturescenariosandareincreasedby~2%/°Cinfuture.Wefindthattheincreaseofextremeprecipitationintensitiesisassociatedwithstrongupwardmotionofair(whichisfoundtobestrongerinfuture)andsubstantialincreaseofwatervaporunderthefuturescenario.Correspondingemail:sridharanayakiitkgp@gmail.com

W1

Recentextremeevents:What'spredictable,what'snot,andwhattodoSimonS.-Y.WangUtahStateUniversity/UtahClimateCenterIncreasing extreme weather and climate events have resulted in tremendous loss of life andproperty worldwide. Record-breaking extreme heat and heavy precipitation events are steadilygrowing,drivinga significant surgeofacademic interest inprojectionandattributionanalysesofsuchextremes.Emergingnewtheoriesregardingtheincreasedweather/climateextremespresentan exciting new development in science; however, newly proposed mechanisms can contradicteachotherorbeovergeneralized.Controversial theories(suchasrecentdebatesregardingArcticamplification) can confuse the public and concerned stakeholders who are not familiar withmeteorology and climate dynamics. Likewise, overgeneralized concepts do little to advance thepredictionofextremeevents.Thispresentationisdesignedtohighlightthedifferentpathwaysthatclimate mechanisms engender extreme events as well as harness that information to improveextremeeventprediction.Thispresentationwill illustrate recentstudiesof floodevents inNorthAmerica, including the 2017 California flood (that dramatically reversed the unprecedenteddrought),2016Louisianaflood(thatrivalsHurricaneKatrina),2015-2016Texasfloods,2016WestVirginia flood (aneventexceeding1-in-1000-yearprobability), and thegreat2011MissouriRiverflood (a success story!). The goal of this talk is to be inclusive of the diverse theories by (a)integrating the different views and approaches in research of climate extremes and, (b)transitioningthesystematicallyacquiredscientificknowledgetobenefitthepotentialinforecastingfromameteorologicalperspective.Correspondingemail:simon.wang@usu.edu

W2

LocalandremoteclimateresponsetodeforestationinMaritimeContinentMin-HuiLoNationalTaiwanUniversityDeforestation in tropical regions would lead to changes in local energy and moisture budget,resultinginfurtherimpactsonregionalandglobalclimate.Previousstudieshaveindicatedthatthereductionofevapotranspirationdominatestheinfluenceoftropicaldeforestation,whichcausesawarmer and drier climate. Most studies agree that the deforestation leads to an increase intemperature and decline in precipitation over the deforested area. However, unlike Amazon orAfrica,MaritimeContinentconsistsofislandssurroundedbyoceanssothedryingeffectsfoundinAmazonorAfricamaynotbethecaseinMaritimeContinent.Thus,ourobjectiveistoinvestigatethe local and remote climate responses to deforestation in such unique region. We conductdeforestation experiments using NCAR Community Earth System Model (CESM) and throughconvertingthetropicalrainforestintograssland.ThepreliminaryresultsshowthatdeforestationinMaritime Continent leads to an increase in both temperature and precipitation, which is notpredictedbyearlierstudies.Wewillfurtherperformmoisturebudgetanalysistoexplorehowtheprecipitationchangeswiththedeforestationforcing.Correspondingemail:minhuilo@ntu.edu.tw

W2

Historicalsocio-hydrologyinJapan:capturingtheregimeshiftsandphenomenainthemoderneraShinichiroNakamuraNagoyaUniversitySocio-hydrology,whichisanewscienceproposedbySivapalanetal.(2012),aimstounderstandacoupledhuman-watersystemdynamics.Inthisnewfield,historicalanalysishasbeenrecognizedasone of fundamental way to seek the coupled system because analysis of the past at differentlocations helps to discover principles or regularities behind the systems. Therefore, Pande andSivapalan (2016) defined the historic way of socio-hydrological research as “historical socio-hydrology”withtheothertwopathways.Althoughpurposeofthehistoricalsocio-hydrology istocaptureacouplesystemfromitsdistantpastwithusinghistoricaldataset,werecognizedthatthelimitationoflonghistoricdataisabigbarriertoconductthisresearchbecausesuchdatahadoftenbeendocumented inhistoricalnarrativesandnot in format.Thefurtherstudyofhistoricalsocio-hydrologyandcapturingthecoupledsystemswithlonghistoricaldatasetisneeded.ThemodernhistorywithmodernhydrologicaltechnologiesinJapanwasstartedfromMeijimodernera (1868-1912) with import of Dutch river engineering. After this technological revolution,Japanese society had been rapidly changed from “green society” to “technological society” (DiBaldassarreetal.,2015)withinjustacentury.Duringthischanging,ingeneral,Japanhadalsofacedsomeoftypicalsocio-hydrologicalphenomenawhichobservedinothercountriesandbasins:leveeeffect,pendulumswing,lock-insituationetc.However,thesephenomenahavenotbeenexplainedin the context of socio-hydrology and quantitative analysis with historical dataset. Then, ourresearchgrouphasbeentried todevelop thehistoricaldataset relatedhuman-flooddynamics inthe Japanesemoderneraandcapturehistorical regimeshifts in thesociety. In thispresentation,we show that Japanesemodern era is separated into three specific historical classificationswithsocio-hydrological regimeshifts, anda typicalphenomenon in this rapid changing society, “leveegrowth”,whichisobservedbyourhistoricalspatialanalysiswithusingoldmaps.Correspondingemail:shinichiro@civil.nagoya-u.ac.jp

W2

Multi-ModelRegionalSimulationofClimateChangeImpactsonAgricultureandEcosystemsintheSouthwesternUnitedStatesSeungHeeKimChapmanUniversityNatural and agricultural ecosystems play a major role in various processes in the earth systemincludingthecarbon,energyandwatercycles.Tounderstandclimate impacts inthenextseveraldecades,itisessentialtocoupletheregionalprocessestotheoverallglobalclimatesystem.Inthisstudy,weemployedasystemutilizingregionalmodelingtodownscalefromtheglobaltoregionallevelsandtocapturethepotentialimpactsofclimatevariabilityandchangeonagroecosystem.Toassess climate variability and change impacts on agricultural system, we employed AgriculturalProductionSystemssIMulator(APSIMregions)process-basedcropmodel,havingabilityofregionalscaleassessment,toevaluatemaizeyieldpotentialandassessproperadaptationstrategiesinthefuture. The sensitivity of maize yields to the regional climate in the Southwestern U.S. (SWUS)showed that potential crop production responded nonlinearly to variations in Tmax, Tmin, anddownwelling solar radiation at the surface. Future projection based on RCP8.5 scenario showssignificant decrease of maize yield potential over the warmer climate region but the increasedtemperatureappearstobebeneficialincoolerregion.Theresultsalsoindicatethattheyieldsarehighlysensitivetoplantingdatesbothspatiallyandinterannually,dependingon localclimates. Inthe study region,daily temperaturesarepositivelyaffectedbyNorthAmericanOscillation (NAO)duringtheearlywarmseason.TheresultsimplythatNAOisassociatedwithbothoptimalplantingdate andwater resources in the regionbymodulating timing andperiodof snowmelting in theregion. The results also emphasize that interdisciplinary collaborations across different fields areneededtobetterunderstandfutureclimatechangeanditsimpacts.Correspondingemail:sekim@chapman.edu

W2

100-yearGlobalWarmingPotentialofRiceCropIntensityDIMWanndetTokyoInstituteofTechnologyThe complex nexus between the food balance and climate change currently essential to have abetter understanding the food security challenge and production in term of climate changemitigation.Mostrecentlystudyisthelifecycleassessment(LCA)methodswhereprovidesawaytoquantify the climate impacts of a food product by accounting for all greenhouse gas (GHG)emissionsassociatedwithitsproduction,includingupstreamanddownstreamfromthefarm.ThemaincharacterizationfactorreportedbythemidpointimpactLCAisthe100-yearglobalwarmingpotential (GWP), based on carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O); andexpressed as kgCO2-equivalent. Here, themain aspect is by using the existing data from globalcrop-specificcirca2000:i)tospatialthedifferentemissionscompositionofGHGemissionsandii)estimatesGWP100, in rice crop intensity in two specifications as per area and production units.Riceproductionhasreceivedsignificantattentionintheglobalclimatechangediscourseduetoitsuniqueness among cultivated crops for emitting both CH4 andN2O, twoGHGs aremore potentthanCO2 indriving climate change. CH4andN2Oemissionwere converted intoCO2equivalentusingtheIPCC4threportofglobalwarmingpotential(GWP)coefficientof25and298,respectively.CH4hadthehighestcontributiontoglobalwarmingwithmorethana50%ofCO2equivalent.CH4compriseroughly90%oftheGWPoffieldemissionsinfloodedricesystems.Correspondingemail:wanndetdim@gmail.com

W2

Evaluation of the application of the ISI-MIP bias-correction method of futuresimulations of climate over Indonesia for the implementation of Climate ChangeAdaptationPlansGomez Garcia Martin1, Ogawada Daikichi1, Matsumura Akiko1, Takahashi Kiyoshi2, HanasakiNaota2,FukushiKensuke31ResearchandDevelopmentCenter,NipponKoeiCo., Ltd.2National Institute forEnvironmentalStudies(NIES)3IntegratedResearchSystemforSustainabilityScience(IR3S),UniversityofTokyoThe second volume of the Fifth Assessment Report of the IPCC, which focuses on impact,adaptationandvulnerability,highlightsthefactthatitdiffersfrompreviousreportsintheinclusionoftopicsaimedtoevaluatetheinteractionsbetweenclimatechangeandbiophysicalandsocietalstressors. This kind of new analysis has the objective of providing information from scientificresearch that can be used to design adaptation plans that also include human security andlivelihoods.AspartofthenationaldevelopmentframeworkofIndonesia,theNationalActionPlanforClimateChangeAdaptation (RAN-API)was formulated,andhasbeensupportedbyThe JapanInternationalCooperationAgency(JICA).TheRAN-API is integrallycoordinatedbyalltheinvolvedstakeholders, includingthegovernment,andcommunityorganizations fromboththeprivateandpublicsector.Morerecently,theMinistryoftheEnvironment(MOE)ofJapanhaslaunchedanewprojectforsupportingthedevelopmentofprovincialadaptationplansinIndonesiaadheredtotheRAN-API.FourinstitutesinJapan,theIntegratedResearchSystemforSustainabilityScienceoftheUniversity of Tokyo, College of Agriculture in Ibaraki University, the National Institute forEnvironmental Studies and the Nippon Koei Co., Ltd. Research and Development Center, haveformedaconsortiumforexecutingtheMOE-fundedproject.ThisconsortiumhasconductedintheFY2016 studies for theadaptation to climatechangeand theconsequentvulnerabilities inNorthSumatra, East Java, and Bali Provinces. The studies encompass food production (mainly rice),health-related problems like heat stress and water-borne diseases caused by flood, and ocean-related problems like negative impacts on fishery, shrimp farms and rise of sea level. BecauseIndonesia has a fast development of human activity caused by the growth of population, theestimationof the future impactsmaycarrygreatuncertainty.Having reliableestimatesof futureclimate (precipitation, minimum/maximum/mean daily temperature, downward shortwaveradiationandhumidity)becomesamajorconcerninclimate-change-adaptationstudies.Therefore,a process of bias-correction of several climate models was carried out based on the ISI-MIPapproachdevelopedbyHempeletal.(2013).Theobjectiveofthisreportistoshowtheresultsataregional level, but most importantly to share the setbacks found during the application of themethodologytotherequiredmeteorologicalvariables.ThecorrectionwasdonebycomparingdataofasetofpreselectedGCMmodels(CMIP5)toforcingdatasetsbasedonre-analysisdata(WFDEI).The preselectionwas considered necessary since not allmodels are able to reproduce the localannualclimatology,andtheISI-MIPapproachismeanttostatisticallycorrectsystematicdeviations.Thismethodattempts to adjust themean-trendaswell as thedistributionof daily variationsbyanalyzingbothdatasetsinahistoricalperiod.Theproceduremakesadistinctionbetweenvariablesthatshouldbecorrectedadditively(e.g.,temperature)ormultiplicatively(e.g.,precipitation).Thecorrection of specific humidity and downward short-wave radiation is not detailed explicitly,therefore,asimilarcorrectionwasdevelopedfollowingthesameassumptions/considerationsdonefor rainfall and temperature. Even though there is a short explanation for the correction ofmaximum/minimum daily temperature, the results were quite unreal, thus, an alternativeprocedure was implemented to adjust the distributions of the differences of mean dailytemperatureandmaximum/minimumdaily temperatures. Theoutcomehelped realize themeanmeteorologicalchangesthatareexpectedacrosstheIndonesianArchipelago.Theresultswerealsoevaluatedbyobservingthefrequencyandmagnitudeofextremes.TheISI-MIPapproachassumes

thatthedistributionsofdeviationsofbothdatasetsareofthesamekind,consequently,extremevalues are hardly adjusted in many cases. The observation of these results suggests that animprovement of the adjustment-methodology of daily variations is necessary to reduce theuncertainty of meteorological information of future simulations. The differentiated analysisperformedwithin this report revealed statistical conditions of eachmeteorological variable thatmay be useful to design alternative ways of correcting the daily deviations, considering thepossibilitythatthedeviationsfrombothdatasetshavedifferentprobabilitydistributions."Correspondingemail:gomez-mr@n-koei.jp

W2

ModelingImpactsofIrrigationonLandSurfaceHydrologyandSubseasonalForecastYaduPokhrelMichiganStateUniversityIrrigation alters land surface water and energy balances, consequently affecting soil-vegetationprocessesandland-atmosphereinteractionsoverarangeofspatialandtemporalscales.Inrecentyears,therehasbeenincreasedattentioninstudyingirrigationimpactsusingobservationsrecordandmodeling;however,irrigationstillremainscrudelyrepresentedinlarge-scalehydrologicalandclimatemodels. In this study,we use an improved irrigation scheme to examine the impacts ofirrigation on land surface hydrologic water balance, water availability and use, and subseasonalforecastover the continentalUS.Results froma regionalwater cyclemodel andanatmosphericgeneral circulation model (AGCM) indicate that irrigation significantly alters the partitioning ofsensible and latent heat and results in substantial surface cooling over the irrigated areas.Furthermore,resultsofsubseasonalforecastsuggestthat incorporationof irrigationinthemodellargelyimprovesforecastskillsoftheAGCM.Finally,weperformacomparativeanalysisofresultsfrom two sets of simulations, onewith and the other without irrigation, to identify the regionswhereirrigation-inducedsoilmoisturechangeplaysakeyroleinland-atmospherecoupling.Correspondingemail:ypokhrel@egr.msu.edu

W3

DevelopmentofglobalandregionalfloodforecastingsystemandtheirvalidationKeiYoshimura*,YutaIshitsuka,KenshiHibino,MisakoHatono,HyungjunKim,MisakoKachi,TakujiKubota,RikoOki,TaikanOki*InstituteofIndustrialScience,TheUniversityofTokyoEvennow,wearesufferingfromfloodsallovertheworld.JapaneseKinugawafloodin2015duetoa torrential rain associated with two typhoons was a typical example. Whereas it is highlyanticipatedthatcurrentwarmingclimatewouldincreasetheriskoffloods,earlyfloodwarning,inaddition to resilient infrastructural flood control, is needed to avoid flood damage and casualty.Here,wehavedevelopedaprototypeofglobalandregionalfloodwarningsystemusingour landsurfacemodel,MATSIRO,andhydrologicalmodel,CaMa-Flood.Fortheregionalsystem,wetesteda hindcast experiment for the Kinugawa flood using atmospheric forcing data from JMA’smesoscale forecast data (MSM-GPV) and ECMWF’s global ensemble forecast precipitation data.Thesystemwithhorizontalresolutionof5km-meshcoveringJapanesedomainwellreproducedtheflood event with slight earlier peak of the river discharge. Moreover, it was revealed that thesystemhadaskillofensembleforecast,i.e.,17,23,and43membersoutof51memberspredictedtheriverwaterheightwouldexceedthecriteriaof“highlydangerouslevelofwaterlevel”in35,23,and 11 hours before the actual flood event started (2015/9/10 2AM JST), respectively.We alsotested a 1km-mesh regional system for east side of Japan using newly developed atmosphericforcing data, i.e., assimilated regional forecasting product, provided from JAXA/MRI. The globalsystem with horizontally 10-kmmesh was also validated and compared with GloFAS system byECMWF.Correspondingemail:kei@iis.u-tokyo.ac.jp

W3

SensitivityanalysisofhistoricalweatherdocumentsforreconstructingpastclimatePandukaNeluwalaTheUniversityofTokyoReconstructionofhistorical climate is anessential issue forabetterunderstandingof the futureclimate.Lackofhistoricalobservationisachallengeforthistask.Thus,severalresearchhavebeencarriedouttoreconstructhistoricalclimateusingclimateproxies(e.g.,tree-ring,andisotopedata)asanalternativedatasource.Asaresult,reconstructinglong-termextremeeventssuchasthelittleiceagecouldbeachieved.However,thesestudiesarelimitedtoannualordecadalscaletemporalresolutionduetothepropertiesoftheproxies.Therefore,thepersonaldiaryrecordsthathaveacoursedetailsofdailyweathersuchas ‘sunny’, ‘cloudy’or ‘rainy’canbeemployed.Manydiariesduring 16th - 19th century have been collected over japan to extract weather information andconvertedintoadigitizedfrom.Theultimategoalofourresearchistoincorporatethesedataandovercome uncertainty associated with these data to investigate the impact of the climate onsocietyinpast.Thecoursedetailsmentionedinthediariescanbeconvertedtousablefieldssuchascloudcover,radiation,andprecipitation.ArecentstudywithidealizeexperimentsusingmoderntotalcloudcoverobservationshasshownthepossibilityofassimilatingthehistoricalcoursedetailsusingaDataAssimilationscheme.Inthisstudy,wefollowasimilarapproachtofurtherinvestigatethesensitivityofassimilatingotherfieldswhichcanberetrievedfromdiaryrecords.TheJapanesedaily data is randomly distributed over japan with an average number of 17 spots. In order tomimic the uncertainty of diary information, a total of 17 observation stations from JapanMeteorological Agency (JMA) over Japan were selected. A random error to the JMA data wasaddedtoconsideruncertainty.GlobalSpectralmodel(GSM)ofNationalCentersforEnvironmentalPrediction(NCEP)wasusedastheclimatemodelwiththeLocalEnsembleKalmanfilter(LETKF)asthe data assimilation scheme. Daily average values of observations from the selected fields areassimilated daily at 3 pm. The sensitivity of these observation fields will be shown in thispresentation.Correspondingemail:neluwala@gmail.com

W3

TheComputableCatchment:Interactivemediaformodel-datasharingwithimplicationsforscientist-stakeholderaccessandparticipationChristopherDuffyPennStateUniversityThis paper demonstrates the development of an interactive document designed to integrategeospatialdata,watershedmodeling,dataanalyticsandscenariovisualizationthatisapproachableby experts and non-experts alike. The objective of the research is a sharable, interactive open-source document that summarizes and documents the underlying modeling concepts, provideslinkagetodescribingscientificsoftwaremetadata(http://www.ontosoft.org),andimplementstheGeoscienceStandardNamesOntologyschemafordescribingcomputationalmodelsanddatasets(http://www.geoscienceontology.org). The notebook provides simple access to the EssentialTerrestrial Variable geospatial data (ETV) from HydroTerre (www.hydroterre.psu.edu), a dataservice thatprovides fastaccess tohigh resolutionnationaldata sources for soils,hydrogeology,climate,landcoverandtopographyattheHUC-12scale(30mresolution)forCONUS(LeonardandDuffy, 2015). In addition to the ETV’s thenotebookmakes accessibleUS reanalysis climatic data(NLDAS-2,hourly1979-present)forinteractivemodel-datavisualizationandanalysis.Anexecutabledocument for each HUC-12 is stored in the cloud with automatic provisioning and a uniqueidentifierallowingcollaborativemodelanddataenhancementsforhistoricalreconstructionand/orfuture landuse or climate change scenario development. The executable document is based onWolframCDForComputableDocumentFormatwithaninteractiveopen-sourcereaderaccessiblebyanymoderncomputingplatform.TheCDFfileandcontentsareavailableonGitHubandsharedas a normal pdf-style document except that it maintains all interactive features and links tonationalmodels anddata. TheComputableCatchment concept representsone applicationof anextensibledocumentthatcombinestheory,models,dataandanalysisthatcanbedigitallyshared,documentedandreusedamongresearchcollaborators,students,educatorsanddecisionmakers.Applicationsarepresented forcatchments subject toa rangeof landuse, climaticandecosystemsettings.Correspondingemail:cxd11@psu.edu

W3

HydrologicSciencesforFloodDisasterMitigations:IntegrationofLocalInformationandFloodModelingTakahiroSayamaDPRI,KyotoUniversityFlood plains provide tremendous benefits for human settlements. Since olden days people havelivedwithfloodsandattemptedtocontrol themifnecessary.Modernengineeringworkssuchasbuilding embankment have enabled people to live even in flood prone areas, and over timepopulation and economic assets have concentrated in these areas. In developing countries also,rapid land use change alters exposure and vulnerability to floods and consequently increasesdisaster risk. Floodhazardmapping is anessential step for any countermeasures. It has variousobjectives including raising awareness of residents, finding effective evacuation routes andestimatingpotentialdamages through flood riskmapping.Dependingon theobjectivesanddataavailability,therearealsomanypossibleapproachesforhazardmappingincludingsimulationbasis,community basis and remote sensing basis. In addition to traditional paper-based hazardmaps,recent advancement in Information and Communication Technology (ICT) promotes moreinteractivehazardmappingsuchasreal-timehazardmappingforeffectivedisasterresponsesandsafe evacuations. This study presents recent advancement of flood hazard modeling using aRainfall-Runoff-InundationModel and its application to Japanese case studies aswell as inAsiancountries.Itintroducesalsotheintegrationoflocalinformationfromfloodaffectedareasandthesimulation for real-time hazard mapping. The advancement and deliver of hydrologic science isexpectedtocontributeinthemitigationflooddisasters.Correspondingemail:sayama.takahiro.3u@kyoto-u.ac.jp

W3

TowardsecohydrologicaldroughtmonitoringandpredictionusingalanddataassimilationsystemYoheiSawadaMeteorologicalResearchInstitute,JapanMeteorologicalAgencyDespite the importanceof theecologicalandagriculturalaspectsof severedroughts,nodroughtmonitoringandprediction frameworkbasedona landdata assimilation system (LDAS)hasbeendevelopedtomonitorandpredictvegetationdynamicsinthemiddleofdroughts.Inthisstudy,weapplied a LDAS that can simulate surface soil moisture, root-zone soil moisture, and vegetationdynamics to the Horn of Africa drought in 2010-2011 caused by the precipitation deficit in twoconsecutive rainy seasons.We successfully simulated the ecohydrological drought quantified bythemodel-estimatedsoilmoisturesandleafareaindex(LAI).Theroot-zonesoilmoistureandLAIare good indicators of prolonged droughts because they reflect the long-term effects of pastprecipitation deficit. The precipitation deficit in 2010 significantly affected the land surfaceconditionofthenextrainyseason in2011,which indicatedthe importanceofobtainingaccurateinitial soil moisture and LAI values for prediction of multi-seasonal droughts. In addition, thegeneral circulation model (GCM)-based seasonal meteorological prediction showed goodperformanceinpredictinglandsurfaceconditionsoftheHornofAfricadrought.Correspondingemail:ysawada@mri-jma.go.jp

W3

TheperformanceoflandsurfaceandcumulusconvectionschemeinthesimulationofIndianSummerMonsoonusingRegCM4SumanMaityIndianInstituteofTechnologyKharagpurAnattempthasbeenmadetoinvestigatethesensitivityofIndianSummerMonsoon(ISM)tolandsurface model (LSM) and cumulus convection scheme (CCS) using RegCM4. Seasonal (May-September) simulations of ISM are conducted using with two LSM viz., Biosphere AtmosphereTransferScheme,CommunityLandModel-CLM3.5(NCAR)andfiveCCSviz.MIT,KUO,GRELL,andtwomixedscheme(GO_MLandGL_MO)forthreeconsecutiveyears:2007,2008and2009.SomeoftheimportantsynopticfeaturesofISMviz.,Heatlow,Somalyjet,Tropicaleasterlyjet(TEJ),Sub-tropical westerly jet (STWJ), Tibetan anticyclone etc. including monsoon rainfall are validatedsystematically using CRU, NCEP and TRMM datasets. It is observed that seasonal surfacetemperature is reasonablywell simulated bymodelwithMIT, GO_ML andGL_MO schemewithslightwarmbiaswhilesignificantcoldbiasisnoticedwithKUOandGRELLschemeusingboththeLSM. Although the model is able to simulate the Somaly jet, TEJ, STWJ using each of thecombinationsbutlocationandstrengthofthejetsvariescombination-wiseandisbettersimulatedusingMITschemewithslightweaker(stronger)Somalyjet(TEJandSTWJ).Seasonalrainfallisnotwell simulated by the model with none of the combinations. Significant underestimation isobservedoverCentralandNorthWestIndiawitheachcombination.SpatialdistributionofseasonalrainfallisbettersimulatedbythemodelwithMITfollowedbyGO_MLschemeincombinationwithCLM although it overestimates over heavy precipitation zones. Monthly rainfall over fivehomogeneous zones is not represented with none of the combinations. However, in terms ofspatialdistributionandaccuracyinpredictingmagnitudeoftherainfall,performanceofthemodelvariesregion-wise.Consideringoverallperformances,RegCM4showsbetterskillinsimulatingISMwithMITschemeusingCLM.Correspondingemail:suman.buie@gmail.com

W4

ProgressinunderstandinghydrologicfloodingusingGRACEJohnReagerNASAJPLGRACEdatahavebeenshowntocapturetheextremehydrologicconditionsassociatedwithfloods—asthelandsurfacebecomessaturated,thecapacityofsoilstoholdmorewaterdecreases—animportanthydrologicmechanisminmanyinlandfloodevents.Recentresearchandprogressonthistopicwillbepresented, includingresultsandanalysisfromthreeprojects:(1)evaluationofamodel-datafusionapproachforhigh-resolutionGRACEdownscaling;(2)acoupledanalysisoflandwaterstorageandhigh-resolutionriverstoragetoestimatetherunoffgenerationprocessandthecontributions of baseflow to streamflow; and (3) a global assessment of the historic floodoccurrence data base, to determine the frequency of flood intensity based on storage. Theseresults demonstrate the extent to which knowledge of storage conditions can help in floodprediction, and explores the range of conditions, scales and intensities forwhich storage-drivenfloodprocessesarerelevantandobservablebysatellitegravity.Correspondingemail:john.reager@jpl.nasa.gov

W4

DistributionofdebristhicknessanditseffectonglaciermeltatlargescaleYukikoHirabayashiIIS,TheUniversityofTokyoSupraglacialdebrisaffectstheresponseofglacierstoclimatechangebyalteringthereflectivityofsolar radiationandconductiveheat flux.Toaccuratelyassess thecontributionofglaciermelts tosea level rise,water resources and natural hazards, it is important to account for the effects ofdebris.However, due to thepractical difficultiesof global-scale fieldmeasurements, informationregarding the spatial distributionof the thickness and thermal propertiesof debris on glaciers islimited; hence, the effects of debris on glacier melting are not explicitly taken into account incurrentglobal glaciermodels. In this study,wedevelopedadatasetof the thermal resistanceofdebris on glaciers at 90-m resolution derived frommulti-temporal satellite images and satellite-derivedradiationdataattheglobalscale,excludingGreenland,Antarctica,andsomeoftheArctic.Wefoundthatsupraglacialdebriswiththermalresistance>0.10m2KW^-1covered16.8%oftheentireglacialareaanalyzed.ThehighestdebriscoverpercentageoccurredinNewZealand,andthelowestwasinIceland.Thederiveddebristhicknessinformationwasthenappliedtoenergybalanceglaciermassmodelatgloballydistributed51glacierswhere long-termobservationwasavailable.The results indicated that the insulation effect of thick debris was larger than the accelerationeffect of thin debris at most analyzed glaciers and the insulation effect was large in glaciers inAlaskaandCentralEurope.Correspondingemail:hyukiko@rainbow.iis.u-tokyo.ac.jp

W4

Reconstructingclimate-drivenwaterstoragevariabilityVincentHumphreyETHZurichSince 2002, the Gravity Recovery and Climate Experiment (GRACE) mission has providedunprecedentedobservationsofglobal terrestrialwaterstorage (TWS)changes.However, there isstillalimitednumberofindependentsourcesforthepre-2002period.BeforeGRACEobservationswere available, traditional approaches to retrieve past TWS relied on either global hydrologicalmodelsorbasin-scalewaterbalanceestimates.Whiletheseapproachescanprovidereliablefiguresfor certain regions, they do not take advantage of the full information content brought by thecurrent GRACE record. Here, we reconstruct past TWS changes using first-order processapproximationsandBayesianinferencetoidentifyrelationshipsbetweenGRACEobservationsandthe main atmospheric drivers (e.g. daily precipitation and temperature). The resulting griddedestimatesofmonthlysub-decadalclimate-drivenTWSchangesrangefrom1985to2015andtheirrobustness is assessed through a comparison with a set of global hydrological and land surfacemodels*. At midpoint between GRACE observations and hydrological models, this data-drivenapproach represents a new opportunity for studying climate-driven TWS variability as well aspotentiallybridgingdatagapsbetweenGRACEand theGRACEFollow-Onmission.Asanongoingdevelopment, we also present a 1901-2010 gridded reconstruction of climate-driven TWS usingGSWP3forcing.Wediscussthepotential(andlimitations)ofusingsuchstatisticalreconstructionsfor documenting rare extreme events (e.g. megadroughts) or benchmarking hydrological andclimate models. *Humphrey, V., L. Gudmundsson, and S.I. Seneviratne, 2017: A globalreconstructionofclimate-drivensubdecadalwaterstoragevariability,GeophysicalResearchLetters,44,2300-2309.Correspondingemail:vincent.humphrey@env.ethz.ch

W4

LocalEnsembleTransformKalmanFilter(LETKF)dataassimilationofaltimetrydataforSWOTmissionJean-FrancoisVuillaumeIIS,TheUniversityofTokyo"Jean-FrancoisVuillaume1,DaiYamazaki1,DaikiIkeshima2,ShinjiroKanae21IIS,TheUniversityofTokyo,4-6-1Komaba,Meguro-ku,Tokyo153-8505,Japan2DepartmentofCivilandEnvironmentalEngineering,TokyoInstituteofTechnology2-12-1-M1-6O-okayama,Meguro-ku,Tokyo152-8552,JapanTheSurfaceWaterandOceanTopography(SWOT)missionprojectedtobeaunchedin2021isexpected toprovidehigh spatial resolutionaltimetrydata (from50 to100m) for river surface.Theprime advantageof thismission is the use of an interferometric radar knownas KaRIN (Ka-bandInterferometer)whichwillprovidea120kmswath.Severalstudieshavebeenconductedtoevaluate thepotentialof suchobservationonhydromodelquality improvement in theUS,Ohioriver (Andreadis., 2007), Niger (Pedinotti., 2014, Munier 2015), Amazon (Paiva., 2013) andTennesseeRiverBasin(Yoon.,2013)butwithdifferentassimilationdata,methodsandresultsevenifallofthemshowdischargeforecastmeasurementimprovement.Inthisstudy,weusetheglobalfloodmodelCaMa-Flood(Yamazakietal.,2014)andtheglobaldataassimilationmodeldevelopedbyIkeshima(2017).ThemodelusedaLocalEnsembleTransformKalmanFilter(LETKF)schemewith20ensemblemembers.Inourexperiments,weassimilatetheanomalyofriverelevationtoremovethebiaserrorofthetopographyandchannelbathymetryofriverelevation.Weevaluatetheglobalperformance of the model and in particular the Amazon river basin to improve dischargecomputation. The framework of the study is based on a twin experiment to simulate (1) the""true"" system state and then (2) an ensemble of corrupted model states. The virtual SWOTobservations of river equivalent to the CaMa-Flood global hydrological model level high wereassimilatedintothemodelwitharepeatcycleof21daystoassimilatetheelevationanomalyofthepseudoSWOTobservationsovertheAmazonbasin.Finally,theperformanceoftheassimilationisdiscussed intermofseasonality,upstream-downstreamlocalizationandobservationassimilation.Inaddition,wecompare(1)non-assimilated,assimilatedandthe""true""CaMaFloodforecast.(2)theperformanceoftheassimilationofboththeabsoluteandtheanomaliesofwaterelevation."Correspondingemail:vuillaume@rainbow.iis.u-tokyo.ac.jp

W4

Globaldistributionofgroundwater-vegetationspatialcovariationSujanKoiralaMaxPlanckInstituteforBiogeochemistryGroundwaterisanintegralcomponentofthewatercycle,anditalsoinfluencesthecarboncyclebysupplying moisture to ecosystems. However, the extent and determinants of groundwater-vegetation interactions are poorly understood at the global scale. Using several high-resolutiondata products, we show that the spatial patterns of ecosystem gross primary productivity andgroundwatertabledeptharecorrelatedduringatleastoneseasoninmorethantwo-thirdsoftheglobalvegetatedarea.Positiverelationships,i.e.,largerproductivityundershallowergroundwatertable, predominate in moisture-limited dry to mesic conditions with herbaceous and shrubvegetation. Negative relationships, i.e., larger productivity under deeper groundwater,predominateinhumidclimateswithforests,possibly,indicatingadrawdownofgroundwatertabledue to substantial ecosystem water use. Interestingly, these opposite groundwater-vegetationinteractionsareprimarilyassociatedwithdifferences invegetationthanwithclimateandsurfacecharacteristics.Thesefindingsputforththefirstevidence,andaneedforbetterrepresentation,ofanextensiveandnon-negligiblegroundwater-vegetationinteractionsattheglobalscale.Correspondingemail:skoirala@bgc-jena.mpg.de

W4

Heavyrainpredictionapplyingsatellite-basedclouddataassimilationoverlandRieSetoTokyoInstituteofTechnologyFor accurate flood prediction, warning systems, and optimized dam control, information ofpositional relationship between rain areas and river basins is crucial. This requires very fineprecision inthepredictionofrainfallareas.Assimilationofsatellite-basedmicrowaveobservationof cloud has great potential to improve precipitation areas because it can directly obtaininformation on rainfall locations as well as amount of cloud. However, it is difficult to observeclouds over land using satellite microwave remote sensing, because land emissivity is muchstronger and more heterogeneous than that of cloud. To overcome this challenge, appropriaterepresentation of heterogeneous land emissivity is needed. Thus, We developed a CoupledAtmosphereandLandDataAssimilationSystemwiththeWeatherResearchandForecastingmodel(CALDAS-WRF),whichcanassimilatesoilmoisture,vertically integratedcloudwatercontentoverland, and atmospheric variables such as heat and moisture within clouds simultaneously. TheCALDAS-WRFassimilatessoilmoisture,usingpassivemicrowavebrightness temperatureat lowerfrequency,whichhasahighsensitivitytosoilmoisture,andthenassimilatescloudandatmospherewithin clouds, using higher-frequency brightness temperature and optimized emissivity of landsurfaceasabackgroundinformation.TheassimilationalgorithmsintheCALDAS-WRFareEnsembleKalmanFilter (EnKF) for landandone-dimensional variationaldataassimilation (1dVar) for cloudandatmosphericvariables.WeappliedthissystemtoheavyraineventsinJapan.Resultsshowthatthesystemeffectivelyassimilatedcloudsignalsandproducedveryaccuratecloudandprecipitationdistributions with appropriate intensity. Also, the local atmospheric fields are modifiedappropriatelyaroundtheareaofassimilatedclouds.Furthermore,byusingoperationallyanalyzeddynamicalandmoisturefieldsas initialandboundaryconditions,thesystemimprovedpredictionofprecipitationduration.Theresultsdemonstratethemethod’spromiseindramaticallyimprovingpredictionsofheavyrainandconsequentflooding.Correspondingemail:seto.r.ac@m.titech.ac.jp

O2

Simulatingtheimpactsofclimateextremes:thecaseofthe2003EuropeanheatwaveJacobSchewePotsdamInstituteforClimateImpactResearchExtremeclimateorweatherevents,suchasdroughtsorfloods,canaffectsocietythroughmultiplechannels; e.g. by impacting crop yields, infrastructure, ecosystem services, the spreading ofdiseases, etc. Sophisticated numerical models exist of some of the related processes, but asystematic evaluation of their performance for extreme conditions is missing, as is anunderstanding of the extent to which the aggregate impact of a given extreme event canpotentiallybeanticipatedbycombiningthesemodels.Hereweuseawell-documentedinstanceofwidespreadextremeweatherconditions - the2003Europeanheatwaveanddrought (EHW) - totest, for the first time, a suite of impact models covering several major sectors of the humanenvironment and economy: Agriculture, freshwater, terrestrial ecosystems, electric energygeneration,marineecosystemsand fisheries,andhealth. Ineachsector, severaldifferent impactmodels(onlyonemodelforhealth)areforcedwithobservations-basedhistoricalclimatedata.Thesimulatedimpactsarecomparedwithmajorobservedimpactsofthe2003EHW.Sinceeventslikethe2003EHWarewidelyseenasharbingersofmoreextremefutureclimateconditions,ourresultshaveimportantimplicationsforourunderstandingoffutureclimatechangeimpactsingeneral,andspecificallyourabilitytoquantifyexpectedimpactsrelatedtoextremeweatherregimes.Moreover,ourstudycouldprovideusefulinsightsforthefurtherdevelopmentofimpactmodels.Correspondingemail:jacob.schewe@pik-potsdam.de

O2

BenchmarkingcarbonfluxesoftheISIMIP2abiomemodelsJinfengChangLOCEAN-IPSLThe purpose of this study is to evaluate the eight ISIMIP2a biomemodels against independentestimates of long-term net carbon fluxes (i.e., Net Biome Productivity, NBP) over terrestrialecosystemsfortherecentfourdecades(1971-2010).WeevaluatemodeledglobalNBPagainst1)theupdatedglobalresiduallandsink(RLS)pluslanduseemissions(ELUC)fromtheGlobalCarbonProject (GCP),presentedasR+L in this studyby LeQuéréetal. (2015), and2) the landCO2fluxes from two atmospheric inversion systems: Jena CarboScope s81_v3.8 and CAMS v15r2,referredtoasFJenaandFCAMSrespectively.Themodelensemble-meanNBP(thatincludessevenmodelswithland-usechange)ishigherthanbutwithintheuncertaintyofR+L,whilethesimulatedpositiveNBPtrendoverthelast30years is lowerthanthatfromR+Landfromthetwoinversionsystems. ISIMIP2a biomemodels well capture the interannual variation of global net terrestrialecosystemcarbon fluxes.TropicalNBP represents31±17%ofglobal totalNBPduring thepastdecades, and the year-to-year variation of tropical NBP contributes most of the interannualvariationofglobalNBP.Accordingtothemodels,increasingNetPrimaryProductivity(NPP)wasthemaincauseforthegenerallyincreasingNBP.SignificantglobalNBPanomaliesfromthelong-termmeanbetweenthetwophasesofElNiñoSouthernOscillation(ENSO)eventsaresimulatedbyallmodels (p<0.05),which isconsistentwith theR+Lestimate (p=0.06),alsomainlyattributedtoNPPanomalies,ratherthantochangesinheterotrophicrespiration(Rh).TheglobalNPPandNBPanomaliesduringENSOeventsaredominatedby theiranomalies in tropical regions impactedbytropical climate variability. Multiple regressions between R+L, FJena and FCAMS interannualvariations and tropical climate variations reveal a significant negative response of global netterrestrial ecosystem carbon fluxes to tropical mean annual temperature variation, and a non-significant response to tropical annual precipitation variation. According to themodels, tropicalprecipitationisamoreimportantdriver,suggestingthatsomemodelsdonotcapturetherolesofprecipitationandtemperaturechangesadequately.Correspondingemail:jinfeng.chang@locean-ipsl.upmc.fr

O2

ImpactsimulationsforISI-MIPwithaprocess-basedbiomemodel,VISITAkihikoItoNationalInstituteforEnvironmentalStudiesAs a contribution to ISI-MIP, we have applied a process-based terrestrial biome model, VISIT(Vegetation Integrative SImulator for Trace gases), to climate change impact simulations. In theFastTrackphase,wesubmittedoursimulationresults fordifferentRCPsandclimateprojections,contributing to uncertainty assessments. In addition to global analyses, we conducted severalregionalanalyses,focusingonimpactsonecosystemfunctionsinMonsoonAsiaandintheArctic.In thebenchmarking (ISIMIP2a) phase,we submitted simulation results on thebasis of differenthistorical climate data and conducted benchmarking of gross primary production simulated bybiomemodels. In this phase, impacts of extreme events such as large volcanic eruption and ElNiño were discussed. Now, in the ISIMIP2b phase, we are conducting simulations for theassessmenton2-degreeor1.5-degreetargetoftheParisAgreement.ThesetargetsarearoundtheRCP2.6 projections, and therefore climatic impactsmay not be so serious. However, we shouldfocusonremainingimpactsinpotentiallyvulnerableregionssuchashigh-latitudesandalpineareas.Throughtheparticipationin ISIMIP,weobtaineduseful implicationsonmodelcharacteristicsandsourcesofuncertainty,encouraging furthermodeldevelopmentand in-depthanalyses for futureclimaterisks,resilience,andadaptation.Correspondingemail:itoh@nies.go.jp

O2

Humanimpactparameterizationinglobalhydrologicalmodelsimprovesestimatesofmonthlydischargesandhydrologicalextremes:amulti-modelvalidationstudyTedVeldkampVUUniversityAmsterdamHuman impacts on freshwater resources andhydrological features form the coreof present-daywater relatedhazards, like flooding,droughts,waterscarcity,andwaterquality issues.Drivenbythe societal and scientific needs to correctlymodel suchwater related hazards a fair amount ofresources has been invested over the past decades to represent human activities and theirinteractionswiththehydrologicalcycleinglobalhydrologicalmodels(GHMs).UseoftheseGHMs-includingthehumandimension-iswidespread,especiallyinwaterresourcesresearch.Evaluationor comparative assessments of the ability of such GHMs to represent real-world hydrologicalconditionsare,unfortunately,howeveroftenlimitedto(near-)naturalriverbasins.Suchstudiesare,therefore,notabletotestthemodelrepresentationofhumanactivitiesanditsassociatedimpacton estimates of freshwater resources or assessments of hydrological extremes. Studies that didperform a validation exercise - including the human dimension and looking into managedcatchments-eitherfocusedonlyononehydrologicalmodel,and/orincorporatedonlyafewdatapoints (i.e. river basins) for validation. To date, a comprehensive comparative analysis thatevaluates whether and where incorporating the human dimension actually improves theperformance of different GHMs with respect to their representation of real-world hydrologicalconditionsandextremesismissing.TheabsenceofsuchstudylimitsthepotentialbenchmarkingofGHMs and their outcomes in hydrological hazard and risk assessments significantly, potentiallyhamperingincorporationofGHMsandtheirmodellingresultsinactualpolicymakinganddecisionsupportwith respect towater resourcesmanagement. Toaddress this issue,weevaluate in thisstudytheperformanceoffivestate-of-the-artGHMsthat includeanthropogenicactivities intheirmodelling scheme, with respect to their representation of monthly discharges and hydrologicalextremes.Tothisend,wecomparedtheirmonthlydischargesimulationsunderanaturalizedandatime-dependent human impact simulation, with monthly GRDC river discharge observations of2,412 stations over the period 1971-2010. Evaluation metrics that were used to assess theperformance of the GHMs included themodified Kling-Gupta Efficiency index, and its individualparametersdescribingthelinearcorrelationcoefficient,thebiasratio,andthevariabilityratio,aswell as indicators for hydrological extremes (Q90, Q10). Our results show that inclusion ofanthropogenicactivitiesinthemodellingframeworkgenerallyenhancestheoverallperformanceoftheGHMsstudied,mainlydrivenbybias-improvements,andtoa lesserextentduetochanges inmodelled hydrological variability. Whilst the inclusion of anthropogenic activities takes mainlyeffect in the managed catchments, a significant share of the (near-)natural catchments isinfluencedaswell.Togetestimatesofhydrologicalextremesright,especiallywhenlookingatlow-flows,inclusionofhumanactivitiesisparamount.Whilsthigh-flowestimatesaremainlydecreased,impactofhumanactivitieson low-flows is ambiguous, i.e. due to the relative importanceof thetimingof return flowsand reservoiroperations.Evenwith inclusionof thehumandimensionwefind,nevertheless, apersistentoverestimationofhydrological extremesacross allmodels,whichshouldbeaccountedforinfutureassessments.Correspondingemail:ted.veldkamp@vu.nl

P01

Regional climate downscaling for risk information -Establishment of CORDEX AsiaEmpirical-StatisticalDownscaling(ESD)Group-KojiDairakuNationalResearchInstituteforEarthScienceandDisasterResilienceClimate informationandservicesfor Impacts,AdaptationandVulnerability(IAV)Assessmentsareofgreatconcern.Tomeetwith theneedsof stakeholderssuchas localgovernments inAsia, theCORDEX Asia Empirical-Statistical Downscaling (ESD) Group was established in November 2016.　　Scientific vision of the WCRP CORDEX is to advance and coordinate the science andapplicationof regional climatedownscaling throughglobalpartnerships. InAsia, severalCORDEXactivitiesforDynamicalDownscaling(DDS)suchasCORDEX-EastAsia,SoutheastAsia,SouthAsia,andCentralAsiaareon-going.TheDDSneedsrelativelylargecomputerresourcestoresolvesmallscalephenomena.Ontheotherhand, theESDdoesnotneedmuchcomputer resourcesandnotnecessarilywellrepresentlocallyforcednonlinearphenomena,extremeeventssuchasheavyrain,heavy snow, etc. TheDDS and ESD can complement each other. The CORDEXAsia ESD group ismainlyforenhancingandintegratingthescienceandapplicationofdownscalingactivitiesinAsia.OurESDoutputsbasedonastandardexperimentalset-upandacommonframeworkwillprovideabenchmarkforDynamicalDownscalingandalsomakeitpossibletofurtherintegratedanalysesofthe added values of statistical/dynamical downscalingmethods. Themain activities and goals oftheCORDEXAsiaESDgroupare;1.toproduceregionalclimatestatisticaldownscalinginformationbasedonacommonprotocols foracommonbenchmarkfor investigatinguncertaintyofregionalclimatescenarios.2.casestudiesinsmalldomainfordevelopingandimprovingthemethodswheresufficientobservationisavailable(e.g.,city,coastalareas,agriculturallands,etc.)andforobtainingbest/goodpracticesofco-production/coordinationwith IAVcommunity.3. to improvereferencedata for improving ESD and DDS skills by collecting better observational data and its update. 4.training workshops to share and exchange knowledge and techniques. 　　Wewould like toenhance well-coordinated downscaling activities in Asia to provide regional climate informationand service which is required for risk assessment on sub-national scale for climate changeadaptationandIPCCAR6.

Correspondingemail:dairaku@bosai.go.jp

P02

DoSoilWaterLimitationsSuppressConvection?AhmedTawfikNCARIsolating the influence of the land surface on convection and precipitation has larger beenquantifiedusingeitherEarthSystemModels(ESMs)orsomestatisticaltechniques.Unfortunatelyeach of thesemethods have their shortcomings. In the case of ESMs, it is unclearwhether thefeedbacks between soil moisture and precipitation are a consequence of a particular model'sparameterizationsor reflecta"true" feedbackprocess. Statisticalapproachesontheotherhandreturnanoverallsoilmoisture-precipitationsignalbuttheactualprocesschaingenerallyremainsunclear,andthereforetheutilityforhowto improvemodelsremainsunclear.Herewepresentanew process-based methodology that clearly isolates the role of surface energy fluxes onconvective initiation.Using theHeated Condensation Framework (HCF) a surface flux-convectiveinitiation(termedtheflux-CIequation)equationisderivedthatallowsforthesurfacesensibleandlatent heat flux contributions to be separated. Specifically, the flux-CI equation has separatesensibleandlatentheatfluxtermsinadditiontootherfactorsthatcontributedirectlytoCI,suchasmoistureconvergence.Weusetheflux-CIequationtoexplorewhetherincreasinglatentheatflux,through removing soil moisture supply limitations, tends to increase the number of convectiveinitiation events. Essentially the latent heat flux is replaced with the equivalent potentialevaporationestimatestherebyremovingsoilwaterlimitations.TheanalysisisperformedusingtheNorthAmericanRegionalReanalysisfrom1979-2015andfocusesonMay-September.Wefindthatconvective initiationoccurs10-20%moreoftenacrossthePlainswhenevaporation isnot limitedbysoilwater. TheGulfcoastoftheUSalsoshows increases inCI inJulyandAugustof lessthan10%.Thisprovidesthefirstobservedquantificationofhowsoilwaterlimitationsmaybeinhibitingconvective initiation. Future work will extend this analysis to the joint effects of increasingevaporationandreducingsensibleheatflux.Correspondingemail:abtawfik@ucar.edu

P03

EvaluationofhighlatitudeslandclimatesimulatedbyMIROC5AGCMandawetlandschemeTomokoNittaTheUniversityofTokyoWithin the global climate system, land areas exhibit complex processes in the cryosphere,hydrosphere, and the land surface itself, which all interact with the climate system throughcomplexfeedback.Oneoftheissuesinherentinclimatemodelingismodelbias.Thewarmbiasinthesurfaceairtemperatureoverlandduringtheborealsummeriswellknownfrommanyclimatemodeling experiments. In this study, we evaluated high latitudes land climate simulated by theModelforInterdisciplinaryResearchonClimate5(MIROC5)atmosphericgeneralcirculationmodel(AGCM)usingmultiplerecentreferencedatasets.IntheMinimalAdvancedTreatmentsofSurfaceInteractionandRunOff(MATSIRO)landsurfacemodelinMIROC5,aschemerepresentingasnow-fedwetland, inwhich snowmelt canbe storedwith considerationof sub-grid terrain complexity,wasimplemented.Weconductedtwokindsofsimulationswithandwithoutthewetlandscheme.Theexperiment showednotonly abetter surfacehydrologybut also aweaker land-atmospherecoupling strength and larger (smaller) latent (sensible) heat flux due to the delayed snowmeltrunoff.Themeanabsoluteerror(MAE)of2-mairtemperatureandprecipitationoverlandat45–90 °N in summer decreased by 19% and 4%, respectively. Thus, through the representation ofwetlandsinaclimatemodel,itwasrevealedthatthesummerwarmanddrybiascouldbepartiallyimprovedoversnowyandflatareas,particularlyovermuchofwesternEurasiaandNorthAmerica,withoutanapparentdeteriorationofsimulatedsurfacehydrologyandclimateovertherestoftheland in the other seasons. The next step of model development will involve implementing anexplicitrepresentationofsub-gridscalesurfacewaterfractionandrelatedprocesses.(e.g.wetlandfraction,waterandenergybudget).Correspondingemail:t-nitta@iis.u-tokyo.ac.jp

P04

CurrentStatusofGlobalandRegionalCO2BudgetsMasayukiKondoCenterforEnvironmentalRemoteSensing(CEReS),ChibaUniversityTerrestrialecosystemsplayacriticalroleinformationofafeedbackloopofcarbondioxide(CO2)inatmospherebyinteractingwithatmosphericreservoirandclimate,andthusdirectingacourseofthe futureprojectionof climate change. The research communityhas spent significantefforts tounderstand behaviors of terrestrial ecosystems under a steady rise in atmospheric CO2concentration and temperature during the recent decades and deepen knowledge about theregional and global patterns of terrestrial CO2 sinks and sources. Currently, two principalapproachesareusedtoestimateterrestrialCO2exchange:top-downandbottom-upapproaches.The top-downapproachestimates the terrestrial CO2exchange that is optimally consistentwithatmospheric CO2 concentration measurements, via atmospheric inversion models. This methoduses a forward-runningatmospheric transportmodel andapriori initial estimatesof the surfacefluxes to predict observed concentrations and to estimate the terrestrial CO2 exchange, whileseeking consistency between simulated and observed CO2 concentrations. The bottom-upapproach estimates the terrestrial CO2 exchange using ecosystem models, which simulate theecosystem-scale carbon cycle by considering the internal biogeochemicalmechanisms of carbonflows for eachprescribed vegetation type and soil.However, the current estimates of terrestrialCO2exchangeby thebottom-upand top-downapproaches remain inconsistent.As illustrated intherecentIPCCAssessmentReport(AR5),thetop-downapproachtendstoindicatestrongerCO2sinks in temperateandboreal regions than thebottom-upapproachdoes.Furthermore, the twoapproachesexhibitedcontrastingCO2sink-sourcepatternsinthetropics;thebottom-upapproachindicated CO2 sinks and the top-down approach CO2 sources. As illustrated by theseinconsistencies,aconsensusonthegeographicdistributionoftheterrestrialCO2exchangehasyettobeestablishedamong the research community. In this study,weelaborate the current statusand issues of terrestrial CO2 flux estimations by the top-down and bottom-up approaches.Specifically,wecomparethebottom-upestimatefromdynamicglobalvegetationmodelsthatareforcedbyinterannualvariationsofCO2concentration,climateandlandusechanges,withthetop-down estimate from atmospheric CO2 inversions. We show an improved level of agreementbetweenthetwoestimatesinrelationtoregionalandglobalbudgets,sincetheIPCCAR5.Wealsodiscusstheremainingissuescausinginconsistencybetweenthetwoestimates.Correspondingemail:redmk92@gmail.com

P05

Reconciliationoftop-downandbottom-upCO2fluxesinSiberialarchforestKumikoTakataNationalInstituteforEnvironmentalStudiesCarbondioxide(CO2)fluxesbydifferentmethodsvary largelyatglobal, regionaland localscales.ThenetCO2fluxesbythreebottom-upmethods(towerobservation,biogeochemicalmodels,anda data-driven model), and an ensemble of atmospheric inversions (top-down method) arecompared in Yakutsk, Siberia for 2004-2013. The region is characterized by highly homogeneouslarchforestonaflatterrain.TheecosystemaroundYakutskshowsanetsinkofCO2(0.0598kg-Cm-2 mon-1 by tower observation). The monthly-mean seasonal cycles agree among the fourmethodswithintherangeofinter-modelvariations.Thepeak-to-troughamplitudeoftheseasonalcycle is greater for the inverse models than bottom-up methods. The ensemble mean of theinterannualvariabilityestimatedbytheinversemodelsismoresimilartothetowerobservation,aswell as the reduction in CO2 uptake after 2008 due to unusualwaterlogging, than those by thebiogeochemicalmodelsandthedata-drivenmodel.Theresultsofthedata-drivenmodelatvariousscalesshowthelocalinfluenceofexceedinglymatureforestaroundthesite,indicatingapotentialtofillthescalegapsbetweentop-downandbottom-upestimates.Correspondingemail:takata.kumiko@nies.go.jp

P06

Hydroclimaticintensityin1.5&2.0°CwarmerworldGavinMadakumburaTheUniversityofTokyoThemomentousParisagreementof2015wasadoptedatthe21stconferenceofpartiestopursueeffortstolimittheglobalaveragetemperatureincreasetowellbelow2°Cpotentiallyeven1.5°Cabove pre-industrial levels. Many previous climate model experiments such as Coupled Modelinter-comparisonProject(CMIP)arenotdesignedtocateracomparativeevaluationofimpactsof1.5°Cand2°Cwarming.Toovercomethisdrawback,halfadegreeadditionalwarming,prognosisandtheprojectedimpacts(HAPPI)projectwasconductedtoquantifytherelativerisksassociatedwith 1.5 °C and 2 °C ofwarming. Large ensemble simulations of atmosphere-only global climatemodels for two-timeslices (2006-2015,2106-2115)where twoclimateprojectionscorrespondingto1.5°Cand2°Cwarmingwereassociatedwiththefuturetimeperiod.Inthisstudy,weutilizedthe HAPPI multi-model (i.e., MIROC, NorESM, CanAM4 and CAM4) large ensemble experiments(100ensemblesperexperimentpermodel)toinvestigatethehydroloclimaticintensitychangesindifferentregionsoftheworldwiththewarmingclimate.Hydroclimaticintensity(HY-INT;Giorgietal. 2011), as the multiplication of dry spell length (DSL) and precipitation intensity (PINT), wasfoundsignificantlydifferentbetween1.5°Cand2°Cwarmerclimatesfordifferentregionsoftheworld. PINT indicated an increase globally under 2 °Cwarmer climatewhereas DSL showed anincrease in regions such as South America and Europewhile a decreasewas observed in NorthAfrica andEastRussia. Europe showed increasedHY-INTowing to the increases inbothDSLandPINT(exceptforScandinavia).Theprobabilitydistributionofvariablesineachregionindicatedthatdespite the differences inmagnitudes, formost of the regions, DSL, PINT andHY-INT showed asimilarchangefromhistoricalto1.5°Cto2°Cclimates.Furthermore,thelinearityofthesechangesandthepossiblereasonswerealsoinvestigatedduringthestudy.Correspondingemail:gavindayanga09@gmail.com

P07

AnalysisofnegativeemissionapplicationstatusinSouthKoreaForest.NahuiKimKoreaUniversityIn 2015, United Nations Framework Convention on Climate Change (UNFCCC) agreed onestablishingpost-2020climate change regime system to limit global average temperature rise tobelow2°C.TheamountofGreenHouseGas(GHG)reductionsgoalwasvoluntarilyestablishedbyeachcountryandNationalDeterminedContributions(NDCs)hadrespectedthegoal.SouthKoreahastoreducetheGHGby37%fromBusinessAsUsual(BAU)levels.Koreangovernmentidentified3methodstoachievetheambitiousgoalbuttheyrequiremuchefforts.Negativeemission(NE)isatechnologythatdirectlyisolatescarbonfromtheatmosphereanditisrecognizedasanewmethodto achieve the NDCs. Thus, the quantity and quality improvement of forest can result in theexpansion of the carbon sink which increase the NE rate. This study analyzed the recent NEapplicationstatusinSouthKoreainregardtothepost-2020climatechangeregimeandsuggestedthepossibleapplicationofNEinSouthKorea.Correspondingemail:nahee0o0@naver.com

P08

StatisticalEvaluationofSoilWetnessChangesinFutureClimateinCMIP5Multi-ModelEnsemblesinEastAsiaTosiyukiNakaegawaMeteorologicalResearchInstituteThepresentstudyexploresasimpleunderstandingofchangesinsurfacesoilmoistureinthelate21stcenturyunderaglobalwarmingbasedon thestatistical significanceaside fromcomplicatedmechanisms.Significantincreaseinsurfacesoilmoistureinthemulti-modelanalysisisconfinedattwoeasternedgeareas,whilesignificantdecreaseinsurfacesoilmoistureisseeninaninlandareaofsouthernChinaandanorthernarea.Significantchangesinevaporationamongthe3waterfluxvariables:precipitation(P),evaporation(E),andtotalrunoff(R):explainsonly10%ofthetotalareawithsignificantchangeinsurfacesoilmoisture.SignificantchangesinP-Eamongthe32-combinedwaterfluxvariables:P-E,E+R,andP-Risidentifiedasadominantvariablebutnooverlappedareaisfound between significant changes in P-E and the significant change in surface soil moisture.Respectiveanalysisrevealsthatsignificantincreasesinevaporation,P-R,andE+Rexplain26%,13%and9%oftheareawiththesignificantdecreaseinsurfacesoilmoistureasMMEmeanrespectively.This indicates that ensemble mean of MME may hinder the geophysical understanding ofmechanismbetweensurfacesoilmoistureandwaterfluxvariables.Correspondingemail:tnakaega@mri-jma.go.jp

P09

DynamicaldownscalingsimulationoverEastAsiausingacoupledatmosphere-oceanregionalclimatemodelRSM-ROMS

XiaojunGUO

*InstituteofIndustrialScience,TheUniversityofTokyo

The climate over East Asia in 2006 is evaluated using a coupled regional climate model RSM-ROMS at a resolution of 25 km. The result shows that the coupled model can better capture the spatial patterns of rainfall and effectively reduce the heavy precipitation over some areas of the ocean compared to the uncoupled one. In addition, it can better reproduce the amount of released latent heat flux from ocean than the uncoupled one. The major components contributing to the negative net heat flux bias are shortwave and latent heat flux. Compared to the observation, the coupled model exhibits prominent systematic cold bias in sea surface temperature (SST) over Indian Ocean and Northwest Pacific Ocean. As for the subsurface ocean state simulation, it is also evident that the coupled model tends to display poor simulation of the ocean variables in the mixed layer. The ocean temperature in the upper 100m of the ocean is underestimated while the salinity and density are overestimated. The negative (positive) SST bias is well consistent with deeper (shallower) mixed layer depth over Northwest Pacific Ocean. Further analysis indicates that more upward latent heat flux caused by the stronger near-surface wind speed, less net heat flux and stronger wind-driven equatorial and coastal upwelling motion are mainly responsible for cold SST bias over Indian Ocean in summer.

Correspondingemail:guoxjblicy@gmail.com

P10

ValuationIrrigationwaterbyyieldcomparisonapproach:globalscaleSobhanAfrazTheUniversityofTokyoWater scarcity is amajor global issue. During the last few decades, it has become evident thatbecauseofa steadily increasingdemand, freshwater scarcity isbecominga threat to sustainabledevelopmentofhumansociety. In itsmostrecentannualriskreport, theWorldEconomicForumlistswatercrisesasthelargestglobalriskintermsofpotentialimpact.CleanWaterandSanitationandclimateactionareSDGsgoals.Soincreasingefficiencyofirrigationwateruseisveryimportant,inhydro-economicmodels,waterallocationbetweencompetitiveuses isbasedontheeconomicbenefits they generate (Harou et al., 2009). Taking into account the economic value of waterenablesnotonlytoallocatewatertothemostvaluableuses,butalsotoestimatethedirectcostsofwaterscarcity,intermsofunrealizedeconomicbenefits.Todoso,afirststepistomeasurethebenefitsofwaterinitsdifferentuses,i.e.todeterminetheeconomicvalueofwater(Young,2005).Thenassessingthemissingquantityofwatergivesthecorrespondingunrealizedeconomicbenefits.Knowing water value in its different uses, it is possible tomanage the available water so as tominimizetheeconomiccostsofscarcity.Theaimthisstudyishowmuchbenefitwecangainfromusingofunitamountofwaterinirrigationsector.Maximizetheeconomicbenefitsoftheallocatedwatercanhelpreducethecostsoffuturewaterscarcityunderglobalchangesintheworld.Inthisstudy irrigation and rain-fed yield and its revenue comparisonwill be applied for some crops todefinewhichcropswithwhichmethodhasthebesteconomicrevenue.Correspondingemail:sobhan.afraz@gmail.com

P11

TheCommunityWaterModel(CWATM)-DevelopmentofacommunitydrivenglobalwatermodelYusukeSatohInternationalInstituteforAppliedSystemsAnalysisWith a growing population and economic development, it is expected that water demands willincrease significantly in the future, especially in developing regions. At the same time, climatechangeisexpectedtoalterspatialpatternsofhydrologicalcycleandwillhaveglobal,regionalandlocalimpactsonwateravailability.Thus,itisimportanttoassesswatersupply,waterdemandandenvironmentalneedsovertimetoidentifythepopulationsandlocationsthatwillbemostaffectedby these changes linked to water scarcity, droughts and floods. The Community Water Model(CWATM)willbedesigned for thispurpose in that it includesanaccountingofhowfuturewaterdemandswillevolveinresponsetosocioeconomicchangeandhowwateravailabilitywillchangeinresponsetoclimate.CWATMrepresentsoneofthenewkeyelementsofIIASA’sWaterprogram.Ithasbeendevelopedtoworkflexiblyatbothglobalandregionallevelatdifferentspatialresolutions.Themodel isopensourceandcommunity-driventopromoteourworkamongst thewiderwatercommunity worldwide and is flexible enough linking to further planned developments such aswaterqualityandhydro-economicmodules.CWATMwillbeabasistodevelopanext-generationglobal hydro-economic modeling framework that represents the economic trade-offs amongdifferent water management options over a basin looking at water supply infrastructure anddemandmanagements.The integratedmodeling frameworkwill considerwaterdemandfromallsectors, investment needs to alleviate future water scarcity, and will provide a portfolio ofeconomically optimal solutions for achieving future water management options under theSustainableDevelopmentGoals (SDG)forexample. Inaddition, itwillbeabletotracktheenergyrequirementsassociatedwiththewatersupplysysteme.g.,pumping,desalinationand interbasintransfer to realize the linkagewith thewater-energyeconomy. Inabigger frameworkofnexus -water,energy,food,ecosystem-CWATMwillbecoupledtotheexistingIIASAmodelsincludingtheIntegrated AssessmentModelMESSAGE and the global land and ecosystemmodel GLOBIOM inorder to realize an improvedassessmentsofwater-energy-food-ecosystemnexusandassociatedfeedback. Our vision for the short to medium term work is to introduce water quality (e.g.,salinizationindeltasandeutrophicationassociatedwithmegacities)intoCWATMandtoconsiderqualitativeandquantitativemeasuresoftransboundaryriverandgroundwatergovernanceintotheintegratedmodellingframework.Correspondingemail:satoh@iiasa.ac.at

P12

Global-scaleriverfloodvulnerabilityinthelast50yearsMasahiroTanoueTheUniversityofTokyo/InstituteofIndustrialScienceThe impacts of flooding are expected to rise due to population increases, economic growth andclimate change. Hence, understanding the physical and spatiotemporal characteristics of riskdrivers (hazard, exposure and vulnerability) is required to develop effective flood mitigationmeasures.Here,thelong-termtrendinfloodvulnerabilitywasanalysedgloballybetween1960and2013, calculated from the ratio of the reported flood loss or damage derived from a disasterdatabase to themodelled flood exposure using a global river and inundationmodel. A previousstudy showeddecreasing global flood vulnerability over a shorter period using different disasterdata.Inaccordancewiththestudy,floodvulnerabilitytomortalitiesandeconomiclossesshowedadecreasing trend, and inverse relationships were found between flood vulnerability and grossdomesticproductper capita, indicatingan improvement in floodvulnerabilityat theglobal scalesince 1960 associated with economic growth. However, although a significant negative trend inglobalfloodvulnerabilitytomortalitieswasseenacrossthewholeanalysisperiod,thevulnerabilityfor themost recentperioddidnot showastatistically significant trend.This resulthighlights theimportanceoftheanalysisperiod,aswellasuncertainties,whencalculatingfloodvulnerability.Thelong-term analysis demonstrated for the first time that flood vulnerability to economic losses inupper-middle, lower-middleand low-incomecountries showsan invertedU-shape, as a resultofthe balance between economic growth and various historical socioeconomic efforts to reducedamage, leading to non-significant upward or downward trends. We also show that the flood-exposed population is affected by historical changes in population distribution, with changes infloodvulnerabilityofupto48.9%.Bothincreasinganddecreasingtrendsinfloodvulnerabilitywereobserved in different countries, implying that population growth scenarios considering spatialdistributionchangescouldaffectfloodriskprojections.Correspondingemail:masatano@rainbow.iis.u-tokyo.ac.jp

P13

Understandingthesoilmoisture-climatecouplinginirrigationregionsoftheIndiathroughsatelliteandassimilationdatasetsZohaibiMuhammadSungkyunkwanUniversity"Theunderstandingofcouplingbetweenthelandsurfaceandtheatmosphereisessentialinorderto predict the weather and climate patterns. The linkage between the land surface and theatmosphere takes place throughwater, energy, and carbon fluxes. These fluxes are significantlygovernedbythesoilmoisture(SM),knownasoneofthekeyvariablesintheearth’sclimatesystem.Thenatureof landprocessesand the feedback loop is complexover theglobedue tosignificantheterogeneity of the natural land cover and its temporal variability. Moreover, human-inducedmodifications also influence the nature of land-atmosphere coupling by changing the surfaceconditions.Especially, irrigation,definedastheartificialapplicationofwatertolandorcrops,isawell-known anthropogenic activity to alter the natural land surface processes, thereby largelyinfluencing the regional andglobalwaterbalance. In this study, the land-atmosphere coupling isanalyzedintheirrigationregionsoftheIndiabythecorrelationanalysisoftheSMwiththesurfacetemperature(ST)andevapotranspiration(ET).Therelationshipbetweenthesevariablesarewidelyusedtoidentifythedirectionofland-atmospherecoupling.Generally,negativeSM-ETandpositiveST-ET relationships are found in the humid regions under energy-limited conditions, whereaspositive SM-ET and negative ST-ET relationships are prominent in the arid regions underwater-limitedconditions.Thefindingsofthisstudywillhelptounderstandtheland-atmospherecouplingintheirrigationregions.Moreover,theresultswillalsogiveaninsightintotheincorporationoftheirrigationactivitiesinthelandsurfaceandclimatemodels.Keywords: Irrigation, Land-Atmosphere coupling, Soil Moisture, Evapotranspiration, SurfaceTemperature"Correspondingemail:zohaib557@skku.edu

P14

Long-termprojectionsofglobalwateruseforelectricitygenerationSayakaYoshikawaTokyoInstituteofTechnologyElectricity generationmaybecomea key factor that accelerateswater scarcity. In this study,weestimatedthefutureglobalwateruseforelectricitygenerationfrom2005to2100in17globalsub-regions. Twenty-two future global change scenarios were examined, consisting of feasiblecombinationsoffivesocioeconomicscenariosoftheSharedSocioeconomicPathways(SSPs)andsixclimatemitigationscenariosbasedonfourforcinglevelsofrepresentativeconcentrationpathways(RCPs) and two additional forcing levels, to assess the impacts of socioeconomic and climatemitigation changes on water withdrawal and consumption for electricity generation. Climatepoliciessuchas targetsofgreenhousegas (GHG)emissionsaredeterminedbyclimatemitigationscenarios.Bothwaterwithdrawalandconsumptionwerecalculatedbymultiplying theelectricitygenerationof each energy source (e.g., coal, nuclear, biomass, and solar power) and the energysource-specificwateruseintensity.ThefutureelectricitygenerationdatasetwasderivedfromtheAsia-Pacific Integrated/Computable General Equilibrium (AIM/CGE) model. Estimated waterwithdrawal and consumption varied significantly among the SSPs. In contrast, waterwithdrawaland consumption differed little among the climate mitigation scenarios even though GHGemissions depend on them. There are two explanations for these outcomes. First, electricitygeneration forenergysources requiringconsiderableamountsofwatervariedwidelyamong theSSPs, while it did not differ substantially among the climate mitigation scenarios. Second, theintroduction of more carbon capture and storage strategies increased water withdrawal andconsumption under stronger mitigation scenarios, while the introduction of more renewableenergy decreased water withdrawal and consumption. Therefore, the socioeconomic changesrepresentedbytheSSPshadalargerimpactonwaterwithdrawalandconsumptionforelectricitygeneration,comparedwith theclimatemitigationchanges representedby theclimatemitigationscenarios. The same trendswere observed on a regional scale, even though the composition ofenergysourcesdifferedcompletelyfromthatonaglobalscale.Correspondingemail:yoshikawa.s.ad@m.titech.ac.jp

P15

Interactionofwater,landuse,andecosysteminIntegratedTerrestrialModel:abio-geophysicallandsurfacemodelwithhumancomponentsTokutaYokohataNationalInstituteforEnvironmentalStudiesFuture climate changes possibly affect eco-system services, water resources, food production,energy supply, etc. It is important to understand the interaction between the changes in thesecomplicated factors. In the present study, we develop an integrated terrestrial model whichdescribes the natural biogeophysical environment as well as human activities. In the integratedmodel,aglobalvegetationmodelVISIT(Itoetal.2012),waterresourcemodelH08(Hanasakietal.2008,Pokhreletal.2012),cropgrowthmodelPRYSBI2 (Sakuraietal.2015),and landusemodelTeLMO(Kinoshitaetal.,inpreparation)arecoupledtoalandsurfacemodelMATSIRO(Takataetal.2003,Nitta et al. 2014),which is a component of global climatemodelMIROC (Watanabe et al.2010). Output variables of each sub-model are passed to other sub-models during the timeintegration. The time intervals of variable exchange are from hourly to monthly or yearly. Forexample,thecropyields[ton/ha]calculatedbyPRYSBI2isusedinTEMOwhichcalculatethelandusechange(cropornaturalvegetationarea)ofnextyear.Theprojectedland-usemapisusedinallother sub-models. The water resource model H08 considers the irrigation process (waterwithdrawalfromrivers)aswellasdamoperationsinlargerivers,whichaffectsthestateofthesoilmoisture and the river flows in the land surfacemodel.Wewill present the state of themodeldevelopment,andresultsfromthehistoricalandfuturesimulation.Inthehistoricalsimulation,wevalidated the model output such as river flow, irrigated water, crop yield, and ecosystemproductionsbycomparingtotheobservedorreanalysisdata.Basedonthefuturesimulation,wealso assessed the risk of future climate change by investigating the relationship of possiblecroplandareaexpansionandcropproductionsandsoon.Correspondingemail:yokohata@nies.go.jp

P16

ImpactsofLarge-scaleSolarPhotovoltaicDeploymentonGlobalClimateVariabilityMasahitoOmoriTheUniversityofTokyoConsidering the impact of fossil fuel combustion CO2 emission on the global warming and theprojecteddepletion,itisexpectedthatrenewableenergywillplayanimportantroleinthefuture.However,mostofthepreviousresearcheshavedemonstratedclimateconsequenceofrenewableenergydeploymentwithinregionalscales.Inthisresearch,itisaimedtoestimateimpactsofsolarphotovoltaics deployment on climate variability at the global scale. A general circulationmodel(GCM),ModelforInterdisciplinaryResearchonClimate(MIROC)5.0isusedinanatmosphericGCMmode for 10-year experiment span (1998-2007). As expected, temperature decreased in regionswheresolarphotovoltaicswereinstalled,butatthesametime,remoteimpactswerealsofoundinsome regions. Especially, temperature increase in western Russia and temperature decrease inwesternCanadawereobserved,whichisspeculatedasbecauseofanomalousshiftofRossbyWave.This indicatesthatsomeareasaresubjecttoashiftofmeanstatechange intemperature,whichurgestotaketheimpactsonglobalclimateintoconsiderationbeforelarge-scalesolarphotovoltaicdeployment.Correspondingemail:omori@rainbow.iis.u-tokyo.ac.jp

P17

IntercomparisonofregulatedriverdischargeamongmultiplehydrologicalmodelsYoshimitsuMasakiHirosakiUniversityWe performed an intercomparison of river discharge regulated by dams among multiplehydrological models by simulation under the framework of the Inter-Sectoral Impact ModelIntercomparison Project (ISIMIP) 2a. This is the firstmultimodel intercomparison study on dam-regulatedriverflow.Weexaminedriverdischargeonalongitudinalsectionoftwocase-studyriverchannelstoinvestigatetheeffectsofdamsonsimulateddischarge,especiallyattheseasonaltimescale. We found that the magnitude of dam regulation differed considerably among thehydrologicalmodels.Thedifferencewasattributablenotonlytodamoperationschemesbutalsotothemagnitudeofsimulatedriverdischargeflowingintodams.Intermodeldiscrepanciestendedtodecreasetowardthelowerreachesoftheseriverbasins,whichmeansmodeldependenceislesssignificant toward lower reaches.Thesecase-study results imply that, intermodelcomparisonsofriverdischargeshouldbemadeatdifferentlocationsalongtheriver'scoursetocriticallyexaminetheperformanceofhydrologicalmodelsbecausetheperformancecanvarywiththelocations.Correspondingemail:ymasaki@hirosaki-u.ac.jp

P18

Springtime trans-Pacific transport ofAsianpollutants related to theWesternPacificteleconnectionJa-HoKooYonseiUniversitySpringtime trans-Pacific transport of Asian pollutants has been examined to understand itsproperties. Basedon the satellite pollutantmeasurements and reanalysismeteorologydata, thisstudy focus on how the springtime trans-Pacific transport relates to the Western Pacific (WP)pattern, one of northern-hemispheric teleconnections. Positive WP patterns well indicate thestrenghened dipole structure between the northern Aleutian Low and the southern Pacific Highover theNorthPacific.We find that theTOMS/OMIAerosol Index (AI)andMOPITTCOshowtheenhancementoftrans-PacificpollutanttransportduringperiodsofpositiveWPpattern,especially40-50Nlatitudes.ThispatternisalsoconfirmedbyhighcorrelationsofWPindexwithAIandCO.Toevaluate the effect of the WP pattern, we investigate several cases of trans-Pacific transportreported in previous studies. As a result, we reveal that most trans-Pacific transport cases areassociatedwiththepositiveWPpattern.FortheperiodofnegativeWPpattern,however,cyclonicwavebreaking is consistently foundover thewesternNorthPacific,which inhibits thehorizontalair-masstransport.Somecasesshowthetrans-PacifictransportofCOintheperiodofnegativeWPpattern, implyingthattheWPpattern ismore influentialonthetransportofparticles(e.g.,Asiandust)mostlyemittednear~40N.ThisstudyshowsthepossibilitytoutilizeWPpatternasaproxytopredicttheextentofspringtimeAsianparticletransportacrossthePacific.Correspondingemail:talc45@hotmail.com

P19

UncertaintyofRainfallAmountandRiverDischargeinMountainousCatchmentsAssociated with Spatial Distribution of Rainfall Observation and MeteorologicalPredictionKonosukeShibataHokkaidoUniversity　Four sequential typhoons,which recordedprecipitationequal toannualprecipitation, causedserious disasters such as severe floods overmany river basins,mudslides, bridges collapse, andinundationofagriculturallandsinHokkaidoduringthemiddletolatterhalfofAugustin2016.Thisstudy discusses uncertainty of rainfall amount and river discharge associated with spatialdistributions of rainfall observation in the basin of Kanayamadam, located in the upper Sorachiriverbasin,whereafloodoccurredbythecollapseoftheembankmentbyheavyrainfallduetothetyphoonnumber10.Theriverimprovementplansandthemeasurestodealwithnaturaldisastersaremainlybasedon the ground rainfall observations. Rainfall is observedby tippingbucket raingaugeswith20cmdiameter.Thereisanaverageofonegaugeper114㎢inthebasinofthefirst-class river in Hokkaido. However, uncertainties associatedwith the basin representativeness forspatialdistributionofthegaugeexist,becauseofthefew-kilometersspatialscaleofcumulonimbusandthecharacteristicofhighdeviationoftimeandspaceinrain-fallintensityduetotopographyinmountainousareas.The rainfalldatausedon this study is fromC-BandradarsofMLITTransport(MinistryofLand, InfrastructureandTransport),C-BandradarsoftheMeteorologicalAgency,theground rainfall gauges (by AMeDAS (Automated Meteorological Data Acquisition System) andMLIT)andanalyzedrainfalldata.Analyzedrainfalldataisrainfalldatarevisingtheplaneregiondataobserved on radar comparedwith the ground rainfallwith good confidence in quantity.We ranrainfalltorun-offmodelbyusingobservationsbythegroundrainfallgaugesandC-BandradarsoftheMeteorological Agency as input conditions and conducted run-off calculation for three sub-basinsintheIkutorabasin.Andalsowerantherainfalltorun-offmodelataupperreachofSorachiriver basin using the results of the ensemble forecast of typhoon number 10 with the domainclimaticmodelastheinputdata.Asimilaranalysisregardingtheuncertaintyover235catchmentsalloverJapanwasinvestigated.Theresultssuggestthattherecognitionoftotalrainfallamountforthetop10rainfalleventsduringthelast11years(January2006-August2016)highlydependsonthespatialdensityofrainfallobservationsbyAMeDASandMLIT.Furthermore,uncertaintyinriverdischarge associatedwith the spatial distribution of rainfall observation is larger than its of theensembleforecastfrom2daysinadvancebyweatherforecastmodel.Correspondingemail:calib.nosu-0213@i.softbank.jp

P20

ClassificationandforecastofheavyrainfallinnorthernKyushuduringBaiuseasonusingSelf-OrganizingMaps(SOM)DzungNguyen-LeFacultyofEngineering,HokkaidoUniversityInthisstudy,theSelf-OrganizingMapsincombinationwithK-meansclusteringtechniqueareusedfor classification of synoptic weather patterns inducing heavy rainfall exceeding 100 mm/dayduring the Baiu season (June-July) of 1979-2010 over northern Kyushu, southwestern Japan. Itsuggests that these localextreme rainfall eventsareattributed to four clusteredpatterns,whichareprimarily relatedtotheBaiu frontandtheextratropical/tropicalcyclone/depressionactivitiesand represented by the intrusion of warm and moist air accompanied by the low-level jet orcycloniccirculation.Theclassificationresultsarethenimplementedwiththeanaloguemethodtopredicttheoccurrence(yes/no)oflocalheavyrainfalldaysinJune-Julyof2011-2016byusingtheprognosticsynopticfieldsfromtheoperationalJapanMeteorologicalAgency(JMA)GlobalSpectralModel (GSM). In general, the predictability of our approach evaluated by the Equitable ThreatScore up to 7-day lead times is significantly improved than that from the conventional methodusingonlythepredictedrainfallintensityfromGSM.AlthoughtheFalseAlarmRatioisstillhigh,itisexpectedthatthenewmethodwillprovideausefulguidance,particularlyforrangeslongerthan2days, for decision-making and preparation by weather forecasters or end-users engaging indisaster-proofingandwatermanagementactivities.Correspondingemail:dzungnl@gmail.com

P21

EstimationofFloodDischargebyAssimilatingWater-LevelandRatingCurveintoQuasi-2DHydraulicModelTsuyoshiHoshinoHokkaidoUniversityRiverdischarge is important informationtograspwatermovement inariverbasin,becauseriverdischarge has a close relation with time space distribution of precipitation and runoffcharacteristics. However, flood discharge information has low reliability because of difficulty ofdischarge observation. Observation of river discharge has some difficulties. First, observation atmultipoint isgenerallydifficultsinceobservationofriverdischargerequiresexpensiveequipmentand human labor as comparedwith observation of water level. Second, rainfall outflow time isshort Ina steepriverand it isdifficult tocapture the riseofdischarge.Third, safeobservation isdifficult when a rise in water level approaches the embankment height. For these reasons, arelationalexpressionbetweenwaterlevelanddischarge(ratingcurve)arebuilt,andriverdischargeis often estimated from observation results of water level. However, water level and dischargedon’tnecessarilyhavea fixed relationshipbecauseofnon-stationarityofa flood.Moreover, It isexpectedthattheapplicabilityoftheratingcurveislow,especiallyinthecaseofaninexperiencedbigflood.Inordertosolvesuchproblems,authorsdevelopedadischargeestimationmethodwhichis a hydraulic model with data assimilation. The developedmethod is consist of quasi-2D floodcomputationasahydraulicmodelandparticlefilterasadataassimilationmethod.Theestimationtarget by data assimilation is conveyance, because conveyance is a basis of a rating curve andreliable discharge can be obtained through reliable conveyance. Conveyance is constituted by acrosssectionalareaofariverchannelandaroughnesscoefficientwhichhaveuncertaintybecauseitisdifficulttomeasurethemduringaflood.Observationdatawhichusedtoassimilationarewaterlevelanddischargederivedfromaratingcurve.Inordertoexpressuncertaintyofratingcurve, itwasassumedthatthereliabilityofaratingcurvedecreaseswiththeincreaseofdischarge.Authorsapplied thedischargeestimationmethodto the floodof2011 in theShinanoriver.Comparedtothecasewithoutusingdataassimilation,theestimateddischargewashighinagreementwithsomeobserveddischargedataduring the floodperiod.Also, the trajectoryof the timevariationof thewaterlevelanddischargewasreasonable.Fromtheabove,thedischargeestimationmethodusedin this study is an effectivemethod, and it can contribute to grasping time spacedistributionofprecipitationandtherunoffcharacteristicsofa riverbasin.Authorswillapply themethodtothefloodsof2016inHokkaidoregion.Correspondingemail:hoshino@eng.hokudai.ac.jp

P22

DevelopmentofaEulerianglobalriverwatertemperaturemodelDaisukeTokudaTheUniversityofTokyoRiverwater temperature is a physical quantitywhich affects not only ecology andwater qualitysuch as disolved oxygen but also water resources such as cooling water of industrial activitieselectricity generation. It is known that there are regional differences in response of river watertemperatureduetoclimatechangeandhumanactivitiessuchasreservoirconstruction,regulationandwaterintake.Therehavebeenfeweffortstoestimatethesephenomenaonaglobalandlong-time scale. In this presentation,we introduce a global riverwater temperaturemodel based onEulerian description. The difference from the existing semi-Lagrangian global model is that it issuperiorinheatconservation,anditiseasytoapplytocomplexboundaryconditions(topographysuch as confluence, lakes and so on) and the situation where it flows down at different flowvelocities insteadofhaving restrictionson the stabilityof calculation. Thismodel alsodealswithphysicalprocessessuchaswaterphasechange(i.e.riverice)andheatbalanceinfloodplains.Thephase change amount is calculated to satisfy themass and heat conservation law, and the heatbalance in the flood plain is expressed by calculating the average absorption rate of shortwaveradiation on the floodplain.We validated thismodel using global observation data collected byGEMS/Water and compared reproducibility of river water temperature with and without floodplainconsiderationintheMekongriverbasin.Thisresultsindicatesthatnotonlytheflowrateandheatcapacitybutalsothehydrologicalprocessoffloodingaffectstheriverwatertemperature.Andfromtheheatbalanceanalysis,wediscusstheinfluenceofheatconductionfromriverbedonriverwatertemperature.Fromtheseresults,itisconsideredthatthehydrologicalprocessnotonlyflowrate,heatcapacitybutalsofloodingisaffectingtheriverwatertemperature.Thismodelseemstohaveapotential in extrapolating simulationas climate changeand impact assessmentofhumanactivity.Correspondingemail:daisuke@rainbow.iis.u-tokyo.ac.jp

P23

EffectofanexponentialdecayofsaturatedhydraulicconductivityonequilibriumsoilmoistureandwatertabledepthinMAT-GWNatsukiYoshidaSchoolofengineering,theuniversityofTokyo,ResearchFellowofJapanSocietyforthePromotionofScienceDCAppropriate initial soil moisture and water table depth reduces uncertainty in hydrologicalsimulation.Theroleofgroundwateronlandsurfaceprocesshavebeendiscussedrecentlyandsoilmoisture have a connectionwith groundwater. There is a problem that initialization needs longtimesimulationataridregioninlandsurfacemodelwithrepresentationofwatertabledynamics,and this relates tohydrologicalquestion thathowterrestrialequilibriumsoilmoistureandwatertabledepthisdeterminedbyclimateconditions.Findingoneofanswertothisquestioncontributesto hydrological knowledge and also leads to solve the above problem. To answer the question,obtainingappropriate simulatedsoilmoistureandwater tabledepthatglobal scale isneededatfirst.Thisstudyquantifytheaccuracyofannualsoilmoistureandwatertabledepthinlandsurfacemodel with representation of water table dynamics (MAT-GW) at first. Second, an exponentialdecayofsaturatedhydraulicconductivitywasadoptedintoMAT-GWandeffectofanexponentialdecayof saturatedhydraulic conductivityon soilmoistureandwater tabledepthwasexamined.SoilprocessinMAT-GWwascalculatedbyClappandHornbergerequation.Simulatedsoilmoistureonsurfacesoillayer(0-0.05m)atAfricaandChinashowshighspatialcorrelationcoefficientwithinsituobservation inboth simulation,while simulatedwater tabledepth shows lowconsistencyatespeciallydeepareaincontrolsimulation.Anexponentialdecayofsaturatedhydraulicconductivityon soil moisture improved simulated water table depth at deep area, while similar water tabledepthweresimulatedinbothsimulationatshallowareabecauseexponentialprofileofsaturatedhydraulic conductivitywas adoptedbelow1.5m.This result shows that theexponential profileofsaturatedhydraulicconductivity isreasonabletosimulatewatertabledepthatdeeparea in landsurfacemodelwith representationofwater tabledynamics.The impactofexponentialprofileofsaturatedhydraulicconductivityonrunoffwillbeexaminedandshownatconference.Correspondingemail:yoshida@rainbow.iis.u-tokyo.ac.jp

P24

Benchmarking strategy for multi-model ensemble simulations and optimization ofensemblemeanRyotaroDoiTheUniversityofTokyoThereisagrowinginterestinstandardizationofmodelevaluationforincreasingnumberofmulti-model based research initiatives such as Couple Model Intercomparison Project (CMIP). Theperformanceofamodel,however, canbeevaluatedwithdifferent scopesdependingonvariousskillscores(e.g.,Taylor’sS-score(Taylor,2001)). Inordertocharacterizethosemetrics,15modeloutputs on evapotranspiration from The Inter-Sectoral ImpactModel Intercomparison Project 2(ISIMIP2)arevalidatedagainstmultiplereferenceobservationsandtheskill scoresarecalculatedusing the International Land Model Benchmarking (ILAMB) package. Model performances areevaluateddifferentlydependingonmetrics.Forexample,RMSEmetricfromILAMBpackagegivesahighscoretoamodelincludingoutliers,whileSpatialDistributionmetricfromILAMBpackagegivesa low score to the same model. To study the potential of skill scores for ensemble averaging,ensembleaveragesarecalculatedusingeachskillscoresasweights. It isfoundthattheweightedensembleaveragetendstomarkhigherperformancecomparedtoasinglememberandalsothesimplearithmeticmeanof theensemble.Taking intoaccountmorenumbersofvariablessuchasterrestrialwaterstorageanomalyanddischarge,thedifferenceofmodelperformancesdependingonvariablesneedstobestudied.Correspondingemail:doi@rainbow.iis.u-tokyo.ac.jp

P25

Estimating Evapotranspirationwith limited climate data in AndongDamWatershedusingFAO56Penman-MonteithmethodologySeaJinKimKoreaUniversityThisstudy isconductedtoestimatepotentialevapotranspiration inAndongDamWatershedwithFAO56Penman-Monteith (FAO56PM)methodologyusingtheclimatedata from2013to2014of10 weather stations. Also, the potential evapotranspiration were estimated when there was nosolar radiation, humidity or wind speed data by FAO56 PM and the results were evaluated todiscuss whether the methodology is applicable when climate dataset is not available. As tocompare thepotentialevapotranspirationestimated fromthecompleteclimatedatasetand thatestimated from limited dataset, statistical analysis was performed using the RootMean SquareError (RMSE), and the coefficientofdetermination (R2). From the result, evenwhen the climatedata is limited, FAO56 PM showed a relatively high accuracy in calculating potentialevapotranspirationbyestimatingtheclimatedata.Correspondingemail:bluegulcy@gmail.com

P26

Evapotranspirationprocessesareevaluated inglobalmodels (hydrological andLandSurfaceModels)usingtheiLAMBsystemandusinganewmetrictowhichquantifiesdifferenttimescalesofdryingafteraraineventEleanorBlythCentreforEcologyandHydrologyOneofthekeypropertiesofthelandishowitdriesoutafterarainfallevent.Itaffectshowwarmthe landbecomesbetween the rain storms,howmuchwater is available for growing food,howmuchwaterisreceivedbytherivers.Thisdryingisduetomanyfactors:thetypeoflandcover,theefficiencyofthedryingduetowindandsunshine,theresponseofthesoils.Thisprocessofdryingis referred to as Evapotranspiration and many models and dataset aspire to quantify it for allclimatetypesandacrossallecosystems.AspartoftheEUFP7projectEartH2Observe,asuiteof10modelswereusedtosimulateglobalhydrologyincludingevapotranspiration.Allthemodelsweregiventhesamedrivingdataandallofthemodelsreproducedtheirestimatesofevapotranspiration(as well as soil moisture, surface runoff and drainage: the key water balance components of amodel grid). All of the models treat the processes leading to the eventual evapotranspirationslightlydifferentlyand the resultsweresurprisinglyvariedaswillbepresentedhere.Thesecondstageof the study is toevaluate themodels. This isnot straightforwardas therearenogloballyavailable observed datasets of evapotranspiration. Instead there are options to evaluate themodels against proxies of evaporation such as soil moisture, or an independentmodel, or at arangeofsiteswheredirectobservationshavebeenmade.Inthispresentation,wedemonstratethemanymethodsofevaluatingthemodels.Firstly,weusetheiLAMBbenchmarkingsystem.Initiallyusing the data sets already in the system (GRACE for the total soilwater, GLEAM for a satellitedrivenmodelofevaporation), thesystemwas thenexpandedto includeSoilMoisture (using theESA-CCI soil moisture product). New metrics were also developed to evaluate a specificevaporationprocesses: thespeedofdry-downafterrainfall.Forthis,weusedasetof fluxtowersites that provides half-hourly evapotranspiration data and represents different land covers andclimatesaroundtheworld.Weanalysedthedry-downattwokeytimescales:theinterceptiontimescale(hours)andthetranspirationtimescale(days).Thedry-downanalysischaracterizesdifferentmodels on theirmedium time-scale response towater limited conditions and their interceptionfraction. We show that this analysis with the flux data provides an important new potentialconstraintonthemodels.Correspondingemail:emb@ceh.ac.uk

P27

ImpactsofsoilmoistureinitializationonborealsummersubseasonalforecastsintheGloSea5predictionsystemEunkyoSeoUNISTThis studyexamines the contributionof soilmoisture initializationwithGlobal Seasonal ForecastSystem version 5 (GloSea5;MacLachlan et al. 2015) to the improving prediction skill of surfacetemperaturefocusedonborealsummerseason.TwosetsofAtmosphericModelInter-comparisonProject (AMIP) style hindcast experiments perform for 15 years (1996-2010) either with non-realisticsoilmoistureconditionfromGloSea5long-termAMIPintegrationorrealisticonewhichisnot reanalysis land condition but JULES (Best et al. 2011) offline Land Surface Model (LSM)simulation driven by data assimilated atmospheric forcing dataset (Sheffield et al. 2006). For agiven15-yearperiod,thetwosetofhindcastsarecommonlyperformedwith7-memberensemblesimulations.We put the variation of the initial soil moisture condition from the offline LSM onAMIPstatisticsinordertoavoidclimatedriftbycouplingwithatmosphericcomponent(Dirmeyer2000;KosterandSuarez2003;Kosteretal.2004).Thisstudyadaptedasoilmoistureinitializationmethodology,StandardNormalDeviateScaled (SNDS;Kosteretal.2004),where thevariationofthesimulatedsoilmoistureby JULESofflineexperiment is rescaledontoGloSea5AMIPstatistics.Resultsrevealstatisticallysignificantincreaseinforecastskillofsurfacetemperatureparticularlyinthe mid-latitude land area where background soil moisture condition is the transitional regionbetween relatively wet and arid regimes. The 10-15 % grid points among total become skillincreasing, in which the squared forecast skill difference between soil moisture initializedexperimentandwithout initialization isover0.1.The initializationhighlycontributestomodulatethe heat wave developing processes for 2003 European and 2010 Russian extreme heat waveevents. In contrast, although the realistic soil moisture is initialized in the GloSea5 modelexperiment, there is limitation tomodulate the realisticdegreeof surfaceanomaly conditions inextremeevents.Theproblemisassociatedwiththedifficultyresolvingrealisticprecipitationbythedynamical model even though the anomalous incoming radiation strongly amplify the extremeclimate.Correspondingemail:ekseo90@gmail.com

P28

AssessinguncertaintyofenvironmentalmodelstopredictstreamwaterqualityShirinKarimiUniversityofTehranIn recentdecades, the importanceofwaterquality inhuman lifedevelopment,especially inaridand semi-arid countries facing lack of freshwater resources such as Iran, is more perceived.Modelingtherelationshipbetweenwaterqualityparametersandphysicalattributesofcatchmentplay a significant role for the integrated watershed management. Moreover, considering theuncertaintyanalysisofmodeloutputstopredicttheireffectsondecision-makingisessential.Thisstudy focusedon theprobabilisticuncertaintyanalysisusingMonteCarlo simulation. Inorder toobtain therelevanceofphysicalattributesof50catchments located insouthwestof theCaspianSea and in-stream water quality parameters, stepwise Multiple Linear Regression was applied.According to the regression analysis and the coefficient of determination, seven water qualityparameters,withR squaredcoefficientshigher thanorequal to0.5,wereselected,whichareasfollows:SAR(r2=0.68,pCorrespondingemail:Shirin.karimi@ut.ac.ir

P29

Interannual variability of snow properties in an upper reach basin of Sapporo areabetween2005and2017.AtsushiNunokawaGraduateschoolofEngineering,HokkaidoUniversityHokkaido, thenorthernmostprefectureof Japan,hasoneof theheaviest snowfallsof theworld.Snowoftencausessevereeffectssuchasavalancheandblizzards.WaterresourcesinSapporocityhighlydependsonsnowamount in itsupperreachbasin.Continuoushydrologicalobservation intheuppercatchmentofSapporoincludingsomesnowpropertieshasbeenconductedbytheCivilEngineeringResearchInstituteforColdRegion(CERI)from2005topresent.Inadditiontotheseobservations, the authors started to measure the volumetric water content, the volumetric icecontent, snowdensity and snowwater equivalent at three vertical levels by a sensor associatedwith dielectric constant at the sameobservation site. In this presentation,wewill discuss aboutinterannualvariabilityofsnowpropertiesatthissite.Furthermore,interseasonalcharacteristicsofsnowpropertieswillbeanalyzedassociatedwiththecross-sectionalobservationofsnow.Correspondingemail:peachtennenyyy@gmail.com

P30

SeaSprayCharacteristicsunderAir-SeaBoundaryLayerHirokiOkachiHokkaidoUniversitySea spray is a carrier which exchanges momentum and enthalpy between air-sea interaction.Influenceofseaspraytotheintensificationoftyphoonhasbeendiscussedinpreviousstudies.Theseasurfaceiscompletelycoveredbythetwo-phaseenvironmentduringwindspeedsexceeding25-30m/s.The10-mdragcoefficienthasamaximumvalueforwindspeedsbetween30to40m/sanddecreaseswith strongerwind speed. This studyhas twoapproaches, these are fieldobservationand wind-tunnel experiment, which are used to estimate the sea spray size distribution. Inperformingthefieldobservation,weinstalledopticaldisdrometerontheobservationtower.Thisisusually used for rain observation. The observation site is located in Tanabe bay, Wakayama-prefecture,where typhoonoftenhits.Wegotdatawhen itwasnot rainyandblewstrongwind.Measuringthedropletremainselusivebecauseoftworeasons.Firstly,whendropletsfallthroughtheedgesofthelaser,thesizeofdropletisunderestimated.Secondly,dropletsdisperseasithitsthedisdrometer (KatjaandStephanie2013). In thisstudy,wediscussedhowtoestimatetheseaspray size distribution and rain size distribution. By performing thewind-tunnel experiment, weinvestigatethedroplets’verticaldistributionanditsdependenceonwindspeedandheight.Inthisposter, I will present the results of the field observation and the wind-tunnel experiment,respectively.Correspondingemail:hokachiroki@gmail.com

P31

Meteorological Characteristics of a Heavy Snowfall Event Based on Multi-DopplerRadarsoverSapporoAreain2015YutaOhyaHokkaidoUniversity　It is possible to discuss three-dimensional fields of atmosphere by using multi-doppler andvariational principle. This study adopted two X-band multi-parametric radars installed andmanagedbyMinistryofLand,Infrastructure,TransportandTourism,andanalyzedmeteorologicalfields for a heavy snowfall event which was associated with a small low pressure system overIshikariBayresulting in localheavysnowfallwhich isdifficult tobepredictedfromweathermap.　Thevariationalmethodused inthisstudyhasbeendevelopedforairborne-typeandadjustedtobeappliedforground-basedradars.Thismethodbasedoncalculatingwiththreetermstogethighqualityprecisionofbase linebetween radars.These termsare improved for lowconvectivecloud inwinter fromoriginal system. First term is fit termwhich interpolatesobservation valuefrom polar coordinates to rectangular coordinates. Least squaresmethod is used by this study.Secondtermismassconservationwhichusingmassflux.Whenthereisdeepconvectivecloudovera flat plane,we can use continuous system. Third term is filter termwhich smoothwind vectorusing differential operator. The differential operator showed the second and third differentialcalculus,andexcludedthethirddifferentialcalculusbythis research.　Numericalmodelof theMeteorologicalAgencywhichassimilatedfromobservationvaluearoundJapanshowssmallvortexontheIshikariBay.Resultsofradaranalysisshowedthatwindvectorconversantlyandhorizontalwindconvergencewasdetectedin lowertroposphereoveracoastalzoneof IshikariBaywhich isnorthwesterly located from Sapporo area. In addition, stronger reflectivity propagates tosoutheasterlydirectionincludingSapporoareaandshowedhorizontaladvectionofsnowfall.Windvectors obtained in the variationalmethodwas validatedwith those in radiosonde observation,respectively.Correspondingemail:saving_energy@eis.hokudai.ac.jp

P32

CorrectionoftherainrateestimatesofgroundradarusingGPM/DPRdataTatsuyaShimozumaNagasakiUniversityCorrection of the rain rate estimates of ground radar using GPM/DPR data 1. Introduction andobjective of this study Accurate and high-resolution distributed rainfall dataset is required tocalculatethereturnperiodofpreviousheavyrainfalleventsandtosimulatepossiblerainfalleventsunder the climate change. Japan Meteorological Agency (JMA)’s precipitation analysis dataset“Radar AMeDASâ€� was used in a previous study. By using ground radar datawith superiorresolution,itispossibletocreatehigh-resolutionspatialrainfalldata.Inthisstudy,JapanMinistryof Land, Infrastructure, Transport and Tourism (MLIT)’s X-band multi-polarimetric RadarInformation Network (XRAIN) is used to produce high-resolution spatial rainfall data. XRAIN hashigher spatial and temporal resolutionsof250mand1minute,but thereare someproblems:1)XRAIN ismore subject to rain attenuation than C-band radars. 2) XRAIN sometimes shows falseheavy rainestimatesat specific area. To reduce theerror inXRAIN,Dual-frequencyPrecipitationRadar (DPR) on the Global PrecipitationMeasurement (GPM)mission’s core satellite is used. 2.Decide thecorrectionobjectgrid (1)MethodFirst, the rain rateestimatesofXRAINandDPRarecomparedtodecide thecorrectionobjectgrids.Here, thecorrectionobjectgridsareareawherethedifferenceoftheXRAINandDPR’srainrateislarge.ThestudyisfocusedonXRAINradarstobeoperatedinKantoregions.AsDPRandXRAINmakedifferentobservationarea,theirmatch-updatathat satisfy the several conditions were extracted: 1) The observation area is overlap. 2) Somedegree of the rain rate is observed. 100 matchup cases of DPR and XRAIN are found betweenMarch2014andJune2016.5kmgridsarecreatedthroughouttheKantoregion,andcomparetheXRAIN andDPR's rain rate in each grid by interpolation. TheDPR’s rain rates at the clutter freebottomwhichisstoredasthevariablenameof[precipRateNearSurface]isused.AndtheproductversionofDPRisV04A.(2)ResultAsaresultofcomparingtherainratebetweenXRAINandDPR,XRAIN>DPR tendency is confirmed inmountainous areas ofMt. Fuji, Gunma prefecture, Niigataprefecture, Tochigi prefecture, Fukushima prefecture and the plain of Kanto. The tendency ofXRAIN>DPR in the mountainous areas is considered to be caused by ground clutter. However,ground clutter is unlikely to cause XRAIN>DPR tendency in the plain of Kanto. Therefore, thecorrection object grids are decided by following condition, altitude is over 500m, and grid withXRAIN/DPR>2.0atover50%ofallraincases.Asaresult,correctionobjectgridconcentrateinthemountainousarea.3.CorrectionoftheXRAIN’srainrate(1)MethodXRAIN'srainrateiscorrectedfromthecorrectionobjectgridsandexaminethedifferenceoftherainratebetweenbeforeandafter correction. In this study,using thedataofXRAIN in2015, the sumandmaximumrain ratebefore and after correction are calculated. In the correction method, XRAIN's rain rate in thecorrectionobjectgridisdividedbythecorrectioncoefficient.However,iftherainrateiscorrecteduniformlyinthecasewheretherainiswidelydistributed,therainratesuddenlychangeatthegridboundary.Therefore,the0.5kmgridisdefinedtocovertheobservationarea,andtheaveragerainrateofthecorrectionobjectgrid(R1)andfoursurroundinggrids(R2)arecalculated,andcorrectionis performed only when the ratio of both are over 2.0 (R1/R2>2.0). The smaller value of"XRAIN/DPRratiooftherainrate"and"ratiooftherainratebetweencorrectionobjectgridandsurroundinggrids"isusedasthecorrectioncoefficient.(2)ResultAsaresultofcomparingtherainrate before and after correction, the rain rate estimates are decreased about 10% in themountainous area. Particularly, areaswhere the ratioof after/before correction is small, aroundthebukumahighlandofFukushimaprefecture,Mt.FujiandNikkoofTochigiprefecture.Eachareasarelocatedbetweenradarsiteandthehighaltitudemountain,andit isconsideredthattheeacharea is subject to the ground clutter. The ratio of after/before correction of each area is about

0.42:AbukumahighlandofFukushimaprefecture,about0.45:Mt.Fujiandabout0.60:NikkoofTochigi prefecture. From the above result, it is suggested that the correction can improve theXRAIN's overestimation by the ground clutter. 4. Conclusions and future work MLIT’s XRAIN iscorrectedbytheGPM/DPRtoproducehigh-resolutionspatialrainfalldata.Inthisstudy,tomitigatetheoverestimationof theXRAIN’s rain rate, the correctionobject grids aredecidedby followingcondition,altitude isover500m,andareawherethedifferenceof therainrate is largebetweenXRAINandDPR. XRAIN's rain rateof 2015 is correctedbasedon the correctionobject grid.As aresult, it is suggested that thecorrectioncan improve theXRAIN'soverestimationby thegroundclutter.Thefutureworkistoincreasethematch-upcasesbetweenXRAINandDPRandtoimprovethecorrectionmethod.Correspondingemail:shimozuma@nagasaki-u.ac.jp

P33

Highresolution,highaccuracyglobaltopographymapforlandhydrologymodelingDaiYamazakiTheUniversityofTokyoGlobal-scaleDigitalElevationModels (DEMs)areessential formanystudies, suchas landsurfacehydrologymodelling,floodinundationmodelling,andterrainanalysis.WhilepreciseairborneDEMsareavailableindevelopedregions,mostpartsoftheworldrelyonspaceborneDEMswhichincludenon-negligible height errors. Here we show the most accurate global DEM to date at ~90mresolutionbyeliminatingmajorerrorcomponentsfromtheSRTMandAW3DDEMs.Usingmultiplesatellitedata andmultiple filtering techniques,weaddressedabsolutebias, stripenoise, specklenoise and tree height bias from spaceborne DEMs. After the error removal, significantimprovements were found in flat regions where height errors were larger than topographyvariability,andlandscapesfeaturessuchasrivernetworksandhill-valleystructuresbecameclearlyrepresented.WefoundthetopographyslopeofthepreviousDEMswaslargelydistortedinmostofworldmajorfloodplains(e.g.Ganges,Nile,Niger,Mekong)andswampforests(e.g.Amazon,Congo,Vasyugan).ThedevelopedDEMwilllargelyreducetheuncertaintyinbothglobalandregionallandhydrologymodeling.Correspondingemail:yamadai@rainbow.iis.u-tokyo.ac.jp

P34

ImpactanalysisofmeteorologicalfactorsandSWIonmonthlyforestNDVIvaluesDongfanPiaoKoreaUniversityClimatechangeisnowanissueinmanyareas,suchasforestecosystems,marineecosystemsandwetland ecosystems, and the impacts have become serious. In order to identify the impacts ofclimate change on forests, predicting NDVI by analyzing the correlation between NDVI value offorests and environmental factors in Northeast Asia was conducted in this study. Seasonalcorrelation between monthly NDVI values of Northeast Asian forests and the meteorologicalfactors,andseasonalcorrelationbetweenmonthlyNDVIvaluesandtheSoilWetnessIndex(SWI)were analyzed. As a result, factors eliminating solar radiation and precipitation had highsignificancewithNDVIvalues(PCorrespondingemail:pdf0920@gmail.com