DOC

Simulating the Impacts of Global Warming on Wheat in China Using a Large Area Crop Model

By Alex Hill,2014-02-18 20:39
8 views 0
Simulating the Impacts of Global Warming on Wheat in China Using a Large Area Crop Modelof,on,in,the,Wheat,China,Using,Large,are,wheat

    Simulating the Impacts of Global Warming

    on Wheat in China Using a Large Area

    Crop Model

    VOL.24NO.1ACTAMETE0R0L0GICASINICA2010

    SimulatingtheImpactsofGlobalWarmingonWheatinChina

    UsingaLargeAreaCropModel

    LISanai1,2(李三爱),TimWheeler,AndrewChallinor.,LINErda(林而达),

    XUYinlong(许吟隆),andJUHui(居辉)

    1DepartmentofMeteorology}UniversityofReading,ReadingRG66BB,UnitedKingdom 2DepartmentofAgriculture,UniversityofReading,ReadingRG66AR,UnitedKingdom 3InstituteforAtmosphericScience,SchoolofEarthandEnvironment,

    UniversityofLeeds,LeedsLS29JUnitedKingdom

    4AgroEnvironmentandSustainableDevelopmentInstitute}ChineseAcademyo{ AgriculturalScience,Beijing100081,China

    (ReceivedMarchl9,2009;revisedNovember12,2009)

    ABSTRACT

    Temperatureisoneofthemostprominentenvironmentalfactorsthatdetermineplantgrowth.devel

    opment.andyield.Coolandmoistconditionsaremostfavorableforwheat.Wheatislikelytobehighly

    vulnerabletofurtherwarmingbecausecurrentlythetemperatureisalreadyclosetooraboveoptimum.In

    thisstudy.theimpactsofwarmingandextremehightemperaturestressonwheatyieldoverChinawere

    investigatedbyusingthegenerallargeareamodel(GLAM)forannualcrops.Theresultsshowedthateach

    1.CriseindailymeantemperaturewouldreducetheaveragewheatyieldinChinabyabout4.6%5.7%

    mainlyduetotheshortergrowthduration.exceptforasmallincreaseinyieldatsomegridcells.

    Whenthe

    maximumtemperatureexceeded30.5.C,thesimulatedgrain

    setfractiondeclinedfrom1at30.5.Ctoclose

    to0atabout36.C.Whenthetotalgrainsetwaslowerthanthecriticalfractionalgrain

    setf0.5750.6),

    harvestindexandpotentialgrainyieldwerereduced.Inordertoreducethenegativeimpactsof

    warming.

    itiscrucialtotakeseriousactionstoadapttotheclimatechange,forexample,byshiftingsowin

    gdate,

    adjustingcropdistributionandstructure,breedingheat

    resistantvarieties,andimprovingthemonitoring,

    forecasting,andearlywarningofextremeclimateevents.

    Keywords:climatechange,warming.wheatyield.cropmodel Citation:LiSanai,TimWheeler,AndrewChallinor,eta1.,2010:Simulatingtheimpactsofglo

    balwarming

    onwheatinChinausingalargeareacropmode1.AetaMeteor.Siniea,24f11123135.

    1.Introduction

    Anaverageincreaseof0.13.Cperdecadeinglobal

    surfacetemperaturehasbeenfoundoverthepast50

    yearswiththelast12years(1995-2006)beingthe12

    warmestyears(IPCC,2007).Temperatureisoneof

    themostprominentenvironmentalfactorsthatdeter

    mineplantgrowth,development,andyield(White,

    2003).Astemperaturerises,ittendstoincreasethe

    probabilityoftheexposureofacroptotemperatures

    abovethethresholdforheatstress,consequentlyaf-

    fectingcropyield.Insomepartsoftheworld,the

globalwarminghasexertedameasurablenegativeim

    pactoncropyield(e.g.,LobellandAsner,2003;Peng eta1.,2004;Taoeta1.,2006).Furtherwarmingcould imposeadramaticnegativeimpactoncropyield. Coolandmoistconditionsaremostfavorablefor wheat,andtheoptimumtemperaturewasapproxi

    mately15.Cforgraingrowth(Paulsen,1994),Weight pergrainofwheatwasreducedby3%foreach1.Crise

    anthesismeantemperature oftemperaturewhenpost

    wasabove18.C/13.Cday/night(Wardlaweta1., 1989).Attheglobalscale,wheatyieldwas

    reducedbyabout3.2%8.4%foral.Crisein

    SupportedbyDorothyHodgsonPostgraduatc~NERC

    HutchinsonWhampoaPh.D.ScholarshipandKeyProjectsinthe NationalScience&TechnologyPillarPrograminthe11thFive-YearPlanPeriod:Demo

    nstrationofAdaptationtoClimate

    ChangeinVulnerableRegionofChina(2007BAC03A06). tCorrespondingauthor:lisanai@163.com. 124ACTAMETE0R0L0GICASINICA

    seasonalmeantemperaturefrom1961to2000(Lobell andField,2007).InChina,wheatyieldsatTianshui Stationhasbeenreducedby10.2%foreach1.Cin

    creaseinthegrowing-seasonfrom1981to2000(Tao eta1..2006).Wheatislikelytobehighlyvulnerable tofurtherwarmingbecausecurrentlythetemperature isalreadyclosetooraboveoptimum.Wheatisthe secondmostimportantfoodcropinChinaintermsof bothplantingareaandproduction(ZGTJNJ,2001). Therefore,itiscriticaltoinvestigatetheimpactsof globalwarmingonwheatyieldacrossChina.

    Weatherextremessuchashightemperaturestress tendtohaveanegativeimpactoncropproductivity. Itwasreportedthatamaximumtemperatureabove 32.2.Ccanhaveastressfuleffectoncornproductiv

    ity(Thompson,1975).Forwheat,duringflowering, ashortduration,highmaximumtemperaturetreat

    ment(>31.C)candramaticallyreducegrainnum

    bers,harvestindex,andgrainyield(Wheelereta1., 1996a,b;Ferriseta1.,1998).So,wheatissuscepti

    bletoahighextremetemperatureatanthesis.The responseofgroundnuttomaximumtemperatureat, floweringhasbeenparameterizedthroughreducing theharvestindexandthencropyieldbyChallinor eta1.f2005).Eachoftheflowers/spikeletdistribution patternsandthetemperatureeffectsonpod-set/grain--

    setofacropcanbesimulatedbyamode1.Therefore, inthisstudytheeffectofextremehightemperature stressonwheatduringfloweringwillbeparameterized basedonmodelingstudiesongroundnut(Challinor eta1.,2005)andpreviousexperimentalresultswith wheat.

    Thegenerallargeareamodel(GLAM)foran

    nualcropshasbeendesignedtooperateatthespatial scaleofglobalandregionalclimatemodels(Challinor eta1.,2004).Originally,theGLAMmodelwasde

    signedtosimulategroundnutyieldoverlargeareasin India(Challinoreta1.,2004).Basedontheexisting GLAMgroundnutmode1.aGLAMWheatversionof

    themodelwasdevelopedbydefiningthewheatgrowth anddevelopmentparametersetsfromtheliterature

andmodifyingsomemodelprocessessuchascropphe

    nologyandleafareadevelopment(Li,2008).Ithas V0L.24

    beenevaluatedthatthemodelcansimulatewheat yieldwellinthelargerareasinChina(Lieta1.,2007; Li,2008).Inthisstudy,theGLAMWheatmode1was

    usedtosimulatetheresponseofwheatgrowth,devel

    opment,andyieldtotemperature.

    InChina,theimpactoftemperatureoncropde

    velopmentandyieldinsomespecificsites(e.g.,Tao eta1.,2006)andthecriticalthresholdsoftemperature forcropsyieldinChina(Xiongeta1.,2007)havebeen investigated.However,limitedstudieshaveevaluated theimpactsofextremetemperaturewarmingonwheat yieldacrossChina.Theobjectiveofthisstudyisto investigatetheimpactsofwarmingonwheatyieldover Chinaandanalyzehowwarmingaffectswheatdiffer

    entlyamongregions.Also,theeffectofextremehigh temperaturestressonwheatisassessed.

    2.Methodsandmaterial

    Figure1showsaflowchartoftheGLAMmodel

    structureandtheprocessesofyieldandbiomassfor

    mation.InGLAM,theabovegroundbiomass(W)

    isaccumulatedthroughactualtranspiration(),the leastfavourabletranspirationefficiency(ET/V),and maximumtranspirationefficiency(ETN,m),asshown below,

    =

    (,,哳…(1)

whereWfkghaday)isthedailycropabove

    groundbiomassproduction,ET(Pa)isthenormalized transpirationefficiency.V(kPa)isthedailymeanva

    porpressuredeficit(V=e8at(T)e,whereeistheva-

    porpressure),andETN,max(gkg)isthemaximum transpirationefficiency.Tw(cmday)isthedaily actualtranspiration,whichisaffectedbythetemper

    ature,leafareaindex(LAI),rootgrowth,andavailable soilwater.

    Cropyieldissimulateddailyastheproductof accumulatedbiomassandtheharvestindex(theratio ofgrainyieldtobiomass)

    Y=HI×

    N0.1LISanai,TimWheeler,AndrewChallinor,etal125 Fig.1.Schematicdiagramoftherelationshipsamongthevariablesinfluencinggrowthandyie

    ldinGLAM.Rectangles

    representstatevariables;ovalsrepresentauxiliaryvariables.Solidlinesrepresentflowsofma

    terial;dashedlinesrepresent

    flowsofinformation.Radiation,temperature,rainfall,andRH(relativehumidity)aredailym

    eanweatherinputs.HI

    istheharvestindex.

    Timefromanthesis(day)

    O2040

    Fig.2.RelationshipbetweentheharvestindexandtimefromanthesisforwheatgrownataCO2

    concentrationof380

    #moltool-1(Wheelereta1.,1996b).

    whereYisyield(kgha-1),Histheharvestin

    dex,andWistheabovegroundbiomass.Before

    podfilling(i<2),theyieldcomponentandOHI/Ot

arezero.Fromthebeginningofpodfilling(i?2),

    OHI/Ot=prescribedconstant(%day-i),asshown inFig.2.Thedescriptionsforcropdevelopmentand growthcanbefoundinChallinoreta1.f20041and Lif20081.ThemainparametersusedintheGLAM

    WheatmodelwereshowninTablesIAand2Aofthe Appendix.

    ThedailyweatherinputsusedbytheGLAM modelare:themaximumandminimumtempera- tures,precipitation,andsolarradiation.Thebase

    linePRECISfHadleyCentreRegionalClimateMod

    clingsystemProvidingRegionalClimateforImpacts Studies)outputsdrivenbyHadCM3(HadleyCen

    treCoupledModel,version3)GCMareavailable forentireChinafrom1961to1990(Xu,2004).So, baseline(196119901PRECISweatherdatadrivenby HadCM3wereusedastheinputtoGLAMoverChina. Onlysixyears(19851990)ofobservedwheatyield dataoverlappedwiththetimeseriesofthePRE- CISweatherdataf196119901.Thesowingdateand

    yieldgapparameteraretwoimportantparameters oftheGLAMmode1.Sowingdate8aredetermined bythelocalcalendarandconditionofsoilmoisture. Theyieldgapparameter(YGP),G,varieswithin arangeof0.051.Theoptimalvalueischosenby

    minimi'zi.ngtherootmeansquareerror(RMSE)be- tweenobservedandsimulatedyields.Theoptimal valueoftheYGPineachgridcellwaschosenbvmin- imizingtheRMSEbetweenobservedandsimulated yieldsforthewholeperiodfrom1986to1990.Wheat

    yieldforChinaata0.5.scalewassimulatedusing acalibratedoptimalYGPvalueforthewholetime period.

    一一q?

    126ACTAMETEoRoLoGICASINICA

    3.Parameterizationofhightemperaturestress Oilwheatgrain-set

    InGLAM,theimpactofhightemperatureon

    groundnutflowershasbeenparameterizedandas

    sessedbyChallinoreta1.(2005).Thissub-routinewas modifiedtosimulatethee?_ectofhightemperatestress

    onwheatduringfloweringbasedonstudiesofGLAM

    Groundnutandpreviousexperimentalwheatstudies. Duringtheflowering,thegrain-setofwheatisreduced whenthedailymaximumtemperature()exceeds acritica1value(Inin).Thereductioninthetotal grain-setisdeterminedbythefractionofgrain-setand thenumberofopenflowers(spikelets)oneachofthe days.Notal1spikeletsonawheatspikereachanthesis atthesametime,sothedistributionofspikeletflow

    eringneedstobequantified.Therateofchangeof harvestindexisreducedwhenthegrainsetislower

    thanacriticalvalue.Thisleadstoareductioninthe harvestindexandhencethepotentialgrainyield.The relevantequationdescribingtheeffectofhightemper

    atureatfloweringonwheatisgiveninChallinoreta1. f2005).Thedetailedprocessisdefinedasfollows: (t,d)=min[Tmrin,37.8+1.8t3d],0<t?8,

    i(,d)=.+0.75t1.5d,0<t?8,(3)

    wheretisthetimeofthehightemperatureepisode

    indaysrelativetothedaysofanthesisanddisthe durationoftheepisodeindays.randimarethe criticaltemperature(.C1andthetemperatureatzero grain.set(.C),respectively..istheinterceptofthe postanthesisimparameterization(.C)andis theminimumvalueofr(.C).Thevaluesof"and nare30.5and36.C,respectively(Wheelereta1., 1996a).ThegrainsetfractionisgivenbyEq.(4)f0r atemperaturebetweenthecriticaltemperatureand thetemperatureatzerograinset.

    P(i,:1Tmax()

    i

    'Tm>r(4)

    whereiisthetimeindaysrelativetothestartofthe grainfillingperiod,andP(i)isthefractionofseeds fromdayi,whichsets

    i=Np

    Pt.t=?P(i)Ff(i),:1

    (5)

    VoL.24

    t0tisthefractionofthetotalyield-determininggrain- set,whichisdeterminedbythesumoftheimpactof thatepisodeoneachofthedaysduringtheflowering developmentalstage.Ff()isthefloweringdistribu- tion,itisexpressedasthepercentageofflowersopen- ingondayi.Ff()isgivenbythegammadistribution, whichisobtainedbyfittingtotheobservedvaluesof deVries(1973)(seeFig.9).

    ()i~-le-irf1

    whereoL(=4)istheshapeparameterand(:1.3)is

thescaleparameter.

    OHI

    =

    (OHI)Ot/0(\/\1PcrPt01,R.<Pc,/(7)

    whereOHI/Otistherateofchangeoftheharvest index.BelowacriticalfractionalgrainsetPcr,theto-

    talfractionalgrainsetPt0tisusedtoreduce(OHI/Ot fromitsnonstressedvalue.ThevalueofPcrisgiven bythesuggestedrangeof0.5750.6fromWheeleret

    a1.(1996b).

    4.Modelevaluation

    AtthereprehensivesitesGuyuan,Guyang, Huma,andZhengzhou,thepredictedgrowthduration fromGLAMdrivenbystationweatherdatashoweda goodagreementwithobservations(Table1).Formost oftheyears,theGLAMsimulationscancapturethe variabilityofobservedgrowthduration. The0.5.wheatyieldacrossChinawansimulated usingGLAMwiththecalibratedoptimalyieldgappa- rameters,andthemodelwasdrivenby0.5.gridPRB CISweatherdata.Thespatialdistributionofspring wheatandwinterwheatplantingregionsinChinawas determinedaccordingtoFig.3.Theobservedaverage wheatyieldatthe0.5.scalevariedfrom419to5950 kghal,andthesimulatedmeanyieldwasfrom222 Table1.Correlationcoefficientsbetweenobserved andsimulatedgrowthdurationsatfieldleve1.GLAM N0.1LISanai,TimWheeler,AndrewChallinor,etal127 Fig.3.ThespatialdistributionofspringwheatandwinterwheatplantingregionsinChina.Red

    pointsaredistribution

Report this document

For any questions or suggestions please email
cust-service@docsford.com