Performance of the SUBSTOR-potato model across contrasting ...pure.iiasa.ac.at/12914/1/Performance...

PerformanceoftheSUBSTOR-potatomodelacrosscontrastinggrowingconditions

RubíRaymundoa,⁎

[email protected]

SentholdAssenga

RishiPrassada

UlrichKleinwechterb

JuanConchac

BrunoCondoric

WalterBowenj

JoostWolfd

JørgenE.Olesene

QiaoxueDongf

LincolnZotarellia

ManuelGasteloc

AshokAlvag

MariaTravassoh

RobertoQuirozc

VijayArorai

WendyGrahama

CherylPortera

aAgricultureandBiologicalloridaEngineeringDepartment,UniversityofFlorida,Gainesville,FL,USA

bInternationalInstituteforAppliedSystemsAnalysis(IIASA),Laxenburg,Austria

cInternationalPotatoCenter(CIP),Lima,Peru

dPlantProductionSystemsGroup,WageningenUniversity,TheNetherlands

eDepartmentofAgroecology,AarhusUniversity,Tjele,Denmark

fChinaAgriculturalUniversity,Beijing,China

gUnitedStatesDepartmentofAgriculture-AgriculturalResearchService(USDA-ARS),Sidney,MT,USA

1IntroductionPotatoisthemostimportantnon-graincropworldwidewithaproductionof330milliontonnesgloballyin2010(FAO,2010).Potatoproductionhasincreaseddramaticallyduringthelastdecadeinthedevelopingworld,surpassingtheproduction

levelsofthedevelopedworld(FAO,2010).Potatoconstitutesthemainsourceoffoodsecurityandincomeinthedevelopingworld(LutaladioandCastaidi,2009),andwillbecomeincreasinglyimportantasthepopulationisgrowingmorerapidlyinthe

developingworldthandevelopedregions(LutzandKC,2010LutzandKC,2010).Agrowingpopulation,alongwithclimatechangeandincreasingclimatevariability,willputadditionalpressureonpotatofoodsystems.Assessingtheimplicationsofthese

trendsrequiresintegratingcropmodelswhenevaluatingtheimpactofnewtechnologiesandstrategiesforadaptingtoclimatechange.

Atmosphericconcentrationofcarbondioxide(CO2)isexpectedtoincreasefrom400ppmin2010to–550ppmby2050(IPCC,2013).Potato,aC3crop,willrespondwithhigherphotosynthesisrate(Finnanetal.,2008)andwateruseefficiency

underelevatedatmosphericCO2concentrations(DeTemmermanetal.,2002b;Fleisheretal.,2013).ButhighlevelsofatmosphericCO2arethemaindriverofclimatechangeandwillincreaseglobaltemperatureandhigherrainfallvariability,leadingto

heatwavesandmoredroughtsinsomeregions(IPCC,2013).StudiesincontrolledexperimentchamberssuggestthatelevatedatmosphericCO2concentrationscanmitigatestressesduetowatershortage,buthightemperaturescanalsonegatethe

positiveeffectsofincreasedatmosphericCO2concentrationsoncropproduction(Kaminskietal.,2014).

Cropmodelsarepowerfultoolsthatdescribecropdevelopmentandgrowthasafunctionofcropmanagement,weather,andsoilconditions(HaverkortandTop,2011).Morethan30cropmodelshavebeendevelopedforpotato,andmanyof

themhavebeenusedtostudytheimpactsofclimatechangeonpotatoproduction(Raymundoetal.,2014).Overall,thesestudieshighlightthatdespitethepositiveeffectofatmosphericCO2concentrations,potatoproductionwilldeclineacrossmany

regionsintheworldby2100(Raymundoetal.,2014).However,Stockleetal.(2010) indicatedthat, taking intoaccount theeffectofCO2,adaptationstrategiesoncropproductionmightguaranteethecurrentproduction levelsunder futureclimate

changeconditions in thestateofWashington in theUnitedStates.Othershaveusedpotatocropmodels toassess the impactofclimatechangeonregional (Tubielloetal.,2002;Supitetal.,2012)andglobalpotatoproduction (Hijmans, 2003).

Nevertheless,modelshavebeendevelopedforspecificcultivarsandgeographicdomains(Griffinetal.,1993;;MacKerron,2004).Globalsimulationsrequiretakingintoaccountthecropvariabilityacrosstheglobeandtestingthemodelfunctionality

withastandardcultivaracrosslatitudes.Inmostoftheclimatechangestudies,potatomodelswereusedwithcultivarsandspeciesfromthedevelopedworld(Tubielloetal.,2002;Hijmans,2003;Supitetal.,2012),neglectingthecultivardiversityof

othercultivatedspecies,aswellastraditionalandmoderncultivars.CultivarsofthespeciesSolanumtuberosumaremostwidelygrown,whereassevencultivatedpotatospecies,includingSolanumandigenum(flourypotato),andSolanumjuzepzukii(bitter

potato),coexistinthetropicalAndes(HuamanandSpooner,2002).Also,severalhybridsofvariousspeciesaregrowninthedevelopingworld(Thieleetal.,2007),wheretheuseofpotatomodelsislimited.

hInstitutodeClimayAgua(INTA),LosReserosyLasCabañas,Castelar,Argentina

iDepartmentofSoilScience,PunjabAgriculturalUniversity,Ludhiana,India

jInstituteofFoodandAgriculturalSciences(IFAS),UniversityofFlorida,loridaGainesville,FL,USA

⁎Correspondingauthor.

Abstract

Cropmodelsareessentialtoolsinclimatechangeimpactassessments,buttheyoftenlackcomprehensivefieldtesting.Inthisstudy,wetestedtheSUBSTOR-potatomodelwith87fieldexperiments,including204treatmentsfrom19

countries.Thefieldexperimentsvariedinpotatospeciesandcultivars,Nfertilizerapplication,watersupply,sowingdates,soiltypes,temperatureenvironments,andatmosphericCO2concentrations,andincludedopentopchamberand

Free-Air-CO2-Enrichment(FACE)experiments.TuberyieldsweregenerallywellsimulatedwiththeSUBSTOR-potatomodelacrossawiderangeofcurrentgrowingconditionsandfordiversepotatospeciesandcultivars,includingSolanum

tuberosum,Solanumandigenum,Solanumjuzepczukiispecies,aswellasmodern, traditional,early,medium,and latematurity-typecultivars,witha relativeRMSEof37.2%for tuberdryweightand21.4%for tuber freshweight.Cultivars

‘Desiree’and‘Atlantic’weregrowninexperimentsacrosstheglobeandwellsimulatedusingconsistentcultivarparameters.However,themodelunderestimatedtheimpactofelevatedatmosphericCO2concentrationsandpoorlysimulated

hightemperatureeffectsoncropgrowth.Othersimulatedcropvariables,includingleafarea,stemweight,cropN,andsoilwater,differedfrequentlyfrommeasurements;someofthesevariableshadsignificantlargemeasurementerrors.

TheSUBSTOR-potatomodelwasshowntobesuitabletosimulatetubergrowthandyieldsoverawiderangeofcurrentgrowingconditionsandcropmanagementpracticesacrossmanygeographicregions.However,beforethemodelcan

beusedeffectivelyinclimatechangeimpactassessments,itrequiresimprovedmodelroutinestocapturetheimpactsofelevatedatmosphericCO2andhightemperaturesoncropgrowth.

Keywords:SUBSTOR-potato;Potato;Cropmodeling;Modelperformance;CO2;Hightemperature

Mostpublishedpotatocropmodelshadlimitedexposuretofieldmeasurementsfortesting,andnoneofthemhaveeverbeentestedwithobserveddataunderhightemperatureanddroughtconditions(Raymundoetal.,2014).Somepotato

cropmodelsstillignoretheeffectofincreasingatmosphericCO2concentrationsoncropgrowth(Hijmans,2003;Gobin,2010;SaueandKadaja,2011).MostmodelsincludeatheoreticalC3cropresponsetoelevatedatmosphericCO2(Raymundoetal.,

2014),butonlytwopotatomodels,LOPTCOandAQUACROP,weretestedwithexperimentaldataofyieldresponsetoelevatedlevelsofCO2concentrations(WolfandVanOijen,2003;Vanuytrechtetal.,2011).TheSUBSTOR-potatoandtheLINTUL-

potatomodelsare themostwidelyusedmodels for climatechangestudies (Frankeetal., 2013;Haverkortetal., 2013;Raymundoetal., 2014); however, bothmodels lackmodel testingwithexperimental dataunderelevatedatmosphericCO

concentrationexpectedinthefuture.Currently,publicationsofmodelapplicationsoutnumberpublicationsofmodelperformancetesting(Raymundoetal.,2014).Therefore, fieldtestingwithcurrentandpossiblefuturescenarios isrequiredtobuild

confidenceinanycropmodelapplication.ThemostextensivefieldpotatoexperimentaldatasetfromaroundtheworldhasbeenassembledtoevaluatetheperformanceoftheSUBSTOR-potatomodeltoguidemodelimprovementneedsandsupport

futuremodelapplications.

2Materialandmethods2.1Themodel

TheSUBSTOR-potatomodelbelongstoafamilyofcropmodelsintheDSSAT-CSM(DecisionSupportSystemsforAgro-technologyTransfer−—CropSimulationModel)software(Jonesetal.,2003;Hoogenboometal.,2012).Themodelinputsaredailyweather

data,soilprofileparameters,cultivarparameters,andcropmanagementinformation.TheSUBSTOR-potatomodelsimulatesthedailydynamicsofphenology,biomass,andyieldaccumulation.Themodelaccountsforsoilwaterdeficitfactorsthatreducephotosynthesis

(SWFAC)andgrowth(TURFAC)(Ritchie,1998Ritchieetal.,1995).Similarly, themodelusesanitrogendeficiencyfactor(NFAC)computedbytheactual leafnitrogencontent, thecritical leafnitrogencontentandminimumleafnitrogencontent toreducephotosynthesis

(NSTRES)andgrowth(AGEFAC)(GodwinandSingh,1998).Underwaterornitrogenstress,SWFACandNSTREShastentuberinitiationandincreasethecarbondemandoftubers.ThemodelhasbeenextensivelydescribedbyGriffinetal.(1993),Ritchieetal.(1995)

andSinghetal.(1998).Followingisabriefsummaryofthemodel.

TheSUBSTOR-potatomodelsimulatesfivephenologicalstages,including(1)pre-planting,(2)plantingtosproutelongation,(3)sproutelongationtoemergence,(4)emergencetotuberinitiation,and(5)tuberinitiationtoharvest.Fivecultivar-specificparameters

controlcropdevelopmentandgrowth.Theparameterstuberinitiationsensitivitytophotoperiod(P2,dimensionless)anduppercriticaltemperaturefortuberinitiation(TC,°C)affectphenology;andleafareaexpansionrate(G2,cm2m−2day−1),potentialtubergrowthrate

(G3,gm−2day−1),andanindexthatsuppressestubergrowth(PD,dimensionless)affectbiomassaccumulation(Griffinetal.,1993).

TheSUBSTOR-potatomodelhasdifferent trapezoidal temperature impact functions,whichsimulate theeffectof temperatureon leafgrowth (RTFVINE), rootand tubergrowth (RTFSOIL),photosynthesis (PRFT),and tuber initiation (RTFTI).Eachof these

functionshasarangefromzerotoone.ForRTFVINE,dailymeantemperatureisoptimalbetween18°Cand24°Candpotentialleafexpansionstopsat<2°Cand>35°C.ForRTFSOIL,soiltemperature(computedinthemodelfromdailymeantemperature)isoptimal

between15°Cand23°C,androotandtubergrowthstopsat<2°Cand>35°C.ForPRFT,meandailytemperatureisoptimalbetween15°Cand30°C,andphotosynthesisstopsat<3°Cand>42°C.ForRTFTI,aweightedaveragetemperatureisused(meanof0.75

timestheminimumtemperatureplus0.25timesthemaximumtemperature)andisoptimalbetween10°CandtheuppercriticaltemperaturesetwiththecultivarparameterTC.Tuberinitiationstopsat<4°Cand>TC+8°C(Griffinetal.,1993).

2.1.1TuberinitiationParametersTCandP2playakeyroleat tuber initiation. If temperature isaboveTC, the tuber initiationandtuberbulking is reducedor inhibited.Thus, theuppervalueofTCcanbe interpretedasrepresentinghigh temperature tolerance.P2describes the

sensitivitytodaylengthandhasadimensionlessvaluebetween0and1.ThecloserP2isto0,thelesssensitiveacultivaristolongphotoperiods.Bothparameters,TCandP2,areembeddedinfunctionsthatdeterminethetuberinitiationandinfluencetuberbulking.

Therelativetemperaturefunctionfortuberinitiation(RTFTI)isdescribedasfollows:

PleasealignformulatoleftRTFTI=1−(1/64)*(TEMP-TC)2;(TEMP>TC&TEMP<=TC+8)

Here,RTFTIisafunctionofweightedaveragetemperature(TEMP=tmin*0.75+tmax*0.25)andcriticaltemperature(TC).

Therelativedaylengthfunctionfortuberinitiation(RDLFTI)isdescribedasfollows:

RTFTI=0;(TEMP<=4)

RTFTI=1−(1/36)*(10-TEMP)2;(TEMP>4&TEMP<=10)

RTFTI=1;(TEMP>10&TEMP<=TC)

RDLFTI=(1-P2)+0.00694*P2*(24-PHPER)2

Here,RDLFTIisafunctionofdaylengthinhours(PHPER)andsensitivitytodaylength(P2).RDLFTIequals1whenphotoperiodislessthan12hours.

2.1.2BiomassaccumulationaftertuberinitiationandpartitioningIntheSUBSTOR-potatomodel,CO2concentrationsaffectthedailypotentialcarbonfixationandpotentialtubergrowth.Thepotentialcarbonfixationrateisdescribedasfollows:

Here,thePCARBisafunctionofradiationuseefficiency(RUE,gMJ−1),photosyntheticallyactiveradiation(PAR,MJm−2),andleafareaindex(LAI,dimensionless),whereplantdensity(PLANTS,plantsm−2)isusedtoexpressthepotentialcarbonfixationperunit

area.RUEis3.5gMJ−1fromemergencetotuberinitiationand4.0gMJ−1fromtuberinitiationtoharvest.PCARBismodifiedwithincreasedatmosphericCO2byapplyingarelativeCO2responsefactor(PCO2)forC3crops(Curryetal.,1990).Thisfactoris1atatmospheric

CO2concentrationof330ppmandincreasesasymptoticallyupto1.43ataCO2concentrationof990ppm.Theactualcarbonfixationrateiscalculatedbymultiplyingthepotentialcarbonfixationratewiththeminimumreductionfactorsforwatershortage(SWFAC),nitrogen

stress(NSTRES),ortemperaturefactorthataffectsphotosynthesis(PRFT)(Griffinetal.,1993).

Biomassaccumulationatthisstageisinfluencedbythethreeremainingcultivar-specificparameters,PD,G3,andG2.Aftertuberinitiation,themodelcomputestubergrowthintwosteps.First,itestimatesthepriorityformaximumtubergrowth(TIND)usingthe

sinkstrength(DTII)andthecarbondemandoftubersaftertuberinitiation(DEVEFF):

DTIIavgisathree-daymovingaverageofdailyvaluesofthesinkstrength(DTII).DTIIisestimatedasarelativefunctionoftemperature(RTFTI)andstressconditions.DEVEFFrepresentsthecarbondemandoftubersaftertuberinitiation,whereXSTAGEindicates

theprogressionthrougheachphenologicalstageasafunctionofthecumulativeleafthermaltime(CUMRTFVINE);theparameterPDrangesbetween0and1anddetermineshowfasttubersgetfullpriorityoverleafgrowth.Theconstant10multiplythefactorstomaintain

theresultbetween0and1.

Second,themodelestimatesthepotentialtubergrowth(PTUBGR,gplant−1day−1)asafunctionofpotentialtubergrowthrate(G3),relativetemperaturefactorforrootgrowth(RTFSOIL),andplantdensity(PLANTS,plantsm−2):

Actualtubergrowth(GROTUB,gplant−1day−1)isafunctionofpotentialtubergrowthaffectedbyTIND,andwaterandnitrogenshortages:

Actualleafexpansion(PLAG)isafunctionofpotentialleafexpansion(G2),limitedbytemperature(RTFVINE),water(TURFAC)andnitrogen(AGEFAC)shortages:

Leaf(GROLF),stem(GROSTM)androot(GRORT)growtharecomputedasfollows:

wheretheleafareatoleafweightratio(LALWR,270cmg−1)isaconstantthroughthecropdevelopment.

PCARB=RUE*PAR/PLANTS*(1-EXP(-0.55*LAI))*PCO2

TIND=DTIIavg*(1/NFAC)*DEVEFF;NFAC>1

TIND=DTIIavg*DEVEFF;NFAC<1

DTII=RTFTI;ifnostress

DTII=RTFTI+0.5*(1-min(SWFAC,NSTRES,1))

DEVEFF=min((XSTAGE-2)*10*PD,1)

XSTAGE=2.0+(CUMRTFVINE)/100

PTUBGR=G3*PCO22*RTFSOIL/PLANTS

GROTUB=PTUBGR*min(TURFAC,AGEFAC,1)*TIND

PLAG=G2*RTFVINE/PLANTS*min(TURFAC,AGEFAC,1)

GROLF=PLAG/LALWR

GROSTM=GROLF*0.75

GRORT=(GROLF+GROSTM)*0.2

Themodelconvertstuberdryweighttotuberfreshweightassumingadrymattercontentof20%.Otherwise,themodelestimatesonlydryweightforleaves,stems,androots.

Finally,inDSSAT-CSM,includingtheSUBSTOR-potatomodel,atmosphericCO2concentrationof550ppmincreasesthestomatalresistanceby37%(Allen,1990).Thiseffect,whichisassociatedwiththeboundarylayerandcanopyresistance,causesanincrease

intranspirationefficiency.

3ExperimentaldataFor thisstudy,weuseddata fromexperimentsconducted inpotatoproduction regionsacross theworld (Fig.1a).Theexperimentswereclassified into temperate,subtropics,and tropic regions (CIP,1992).Theaveragephotoperiods for

experimentsintemperate,subtropics,andtropicregionswere14.90,11.21,and12.06hours,respectively.Inthesubtropics,thephotoperiodwasshorttomoderatelylongdependingontheseason,whereasinthetropicsthephotoperiodwasconstant

throughouttheyear(Fig.1b).Theexperimentsrepresentawiderangeofgrowingenvironments,includingdifferentelevationsandsoiltypes.Thus,theexperimentscovereddifferenttemperatureregimes(Fig.1c)andalsoawiderangeofcultivartypes,

cultivarspecies,andcropmanagementpractices,acrossatotalof87experimentsconductedfrom1970to2013in19countries,consistingof12soiltypes,32cultivars,and204treatments.Theexperimentshavebeencarriedouttostudydrymatter

allocation;yieldresponsetovarioustreatments,includingnitrogen(N)fertilizerapplications,watersupplyrates,radiationdeficit,impactofhightemperature;theadaptabilityofcultivarsacrosslocationsandyears;andtheeffectofincreasedatmospheric

CO2concentrationsoncropdevelopmentinOpenTopChambers(OTC)andFree-Air-CO2-Enrichment(FACE)facilities(Table1a).Experimentsselectedforthisstudywerefullycontrolledtoprevent,pests,diseases,andweeds.

Fig.1(a)Globalpotatoproducingareas(gray)withmodeltestsites(Monfredaetal.,2008),(b)photo-period,and(c)weightedaveragetemperatures(=0.75×minimumtemperature+0.25×maximumtemperature)duringcropgrowthperiodsatmodeltestsites.

alt-text:Fig.1

Table1(a)Experimental−sitesandmeasuredvariablesusedinthesimulation–observationcomparisons–sortedalphabeticallybycountryname.(b)Experimentalsitesandmanagementtreatmentsusedinthemodeltesting—sortedalphabetically

bycountryname.

alt-text:Table1

(A)Location Year lata lonb altc Objectived tre cultf repg In-seasonsampling Measuredvariablesh SoilTexturei Reference

Argentina,Balcarce 1991 −37.8 −58.3 97 Nrates 4 1 5 tu CL Travassoetal.(1996)

Australia 1970 −35.0 149.0 Radiationdeficit 3 1 10 tuf,LAI SC Hoogenboometal.(2012)

Belgium,Tervuren 1998–1999 50.8 4.5 97 CO2OTC 1 1 3–6 2 tu,le,st,to,LAI SiL DeTemmermanetal.(2002a)

Bolivia,Belen 1997 −16.0 −68.7 3640 Cultivaradaptation 1 3 3 3 tu,le CL Condorietal.(2010)

Bolivia,Chinoli 1997 −19.6 −65.3 3450 Cultivaradaptation 1 1 3 4 tu,le,st,ro,to SL Condorietal.(2010)

Bolivia,Koari 1997 −17.4 −65.6 3500 Cultivaradaptation 1 3 3 3 tu,le,st CL Condorietal.(2010)

Bolivia,Patacamaya 1997 −17.2 −68.0 3780 Cultivaradaptation 1 2 3–4 tu,le,st,ro SiCL Condorietal.(2010)

Bolivia,Patacamaya 1998(2) −17.2 −68.0 3780 Cultivaradaptation 1 2 3 3–4 tu,le,st,ro SCL Condorietal.(2010)

Bolivia,Toralapa 1993 −17.5 −65.7 3430 Cultivaradaptation 2 3 4 7 tu,le,st,ro, CL Condorietal.(2010)

China,Huhhot 1996 40.5 111.4 1065 Partitioning 1 1 6 tu,LAI SL Gaoetal.(2003)

China,Huhhot 1998 40.5 111.4 1065 Partitioning 1 2 6 tu,le,st,to SL Liuetal.(2003b)

China,Jining 1999 41.0 113.0 Partitioning 1 2 6 tu,le,st,to SL Gaoetal.(2004)

China,Zhalan 1997 48.0 123.0 Various 1 2 7 tu,le,st,to SL Liuetal.(2003a)

China,Zhalan 1998 48.0 123.0 Partitioning 1 3 7 tu,le,st,to SL Gaoetal.(2004)

Colombia,Cundinamarca 1999 4.4 −74.1 Modelvalidation 1 1 9 tu,le,st,to,LAI SiL ForeroHernandezandGarzonMontaño(2000)

Denmark,Jyndevad 1982–1983 54.9 9.1 10 Various 3 4 14–15 tu,to S Jørgensen(1984)

Denmark,Jyndevad 1990–1993 54.9 9.1 10 Nrates 4 1 6–7 tu,ro,to,tuN,roN,toN S Edlefsen(1991)

Denmark,Jyndevad 1984–1986 54.9 9.1 10 Nrates 2 1 5–7 tu,to,tuN S JørgensenandEdlefsen(1987)

Denmark,Tylstrup 1981–1983 57.2 10.0 10 Various 4 4 13–14 tu,to S BachandNielsen(1985)

Ecuador,SanGabriel 1985 0.6 −77.8 Nrates 3 2 4 tu SL ClavijoPonce(1999)

Finland,Jokioinen 1998–1999 60.8 23.5 84 CO2OTC 1 1 3 2 tu,le,st SL DeTemmermanetal.(2002a)

Germany,Giessen 1998–1999 50.6 8.7 68 CO2OTC 1 1 1–6 2 tu,le,st,to,LAI SL DeTemmermanetal.(2002a)

Germany,Giessen 1998–1999 50.6 8.7 68 CO2FACE 1 1 2–6 2 tu,le,st,to,LAI SL DeTemmermanetal.(2002a)

India,Ludhiana 2008–2011 30.9 75.8 244 Nratesandirrigation 4 1 1 tuf SL Aroraetal.(2013)

India,Modipuran 2002–2009 28.7 77.2 228 Various 1 1 1 tuf L,SiCL

Ireland,Carlow 1998–1999 52.9 −6.9 57 CO2OTC 1 1 3 2 tu,le,st,to SiC DeTemmermanetal.(2002a)

Italy,Rapolano 1998–1999 42.7 11.9 38 CO2FACE 1 1 3–4 2 tu,le,st,to,LAI SL DeTemmermanetal.(2002a)

Peru,LaMolina 2003–2010 −12.1 −77.0 244 Cultivaradaptation 1 1 1 tuf SL

Peru,LaMolina 2013 −12.1 −77.0 244 Cultivaradaptation 1 3 3 5 tu,le,st SL

Peru,LaMolina 1985 −12.1 −77.0 244 Cultivaradaptation 2 3 5 tu,le,st,to,LAI SL,SCL,SL TrebejoandMidmore(1990)

Peru,SanRamon 1984 −11.1 −75.3 800 Hightemperaturetolerance 1 3 4 tu,le,st,to SL Nelson(1987)

Peru,SanRamon 2013 −11.1 −75.3 800 Hightemperaturetolerance 1 3 3 5 tu,le,st,to SL

Scotland,Dundee 1984–1985 56.5 −3.1 40 Nrates 2 1 8 tu,le,st,to,LAI,de,tuN,leN,stN,toN SL MarshallandVanDenBroek(1995)

Scotland,Dundee 1986–1987 56.5 −3.1 40 Irrigation 4 1 8 tu,le,st,to,LAI,SWC SL MarshallandVanDenBroek(1995)

Sweden,Goteborg 1998–1999 57.9 12.4 58 CO2OTC 1 1 4–6 2 tu,le,st,to SL DeTemmermanetal.(2002a)

Uganda,Kalengyere 2001–2009 −1.2 29.8 2400 Various 1 1 1 tuf C

UnitedKingdom,Sutton 1998–1999 52.8 −1.3 87 CO2OTC 1 1 3 2 tu,le,st,to,LAI SL

UnitedStates,Benton 2003 45.9 −119.5 Nrates 2 1 5–6 tu,le,st,to,LAI,tuN,leN,stN,toN S Alvaetal.(2010)

UnitedStates,Hastings 2011–2012 29.7 −81.5 2 Nrates 2 3 3 1 tu,le,st,tuN,leN,stN,toN,soN S Zotarellietal.(2014)

UnitedStates,NewYork 1980 42.4 −76.5 Various 1 1 5 tu,le,st,to SL Hoogenboometal.(2012)

UnitedStates,Idaho 1988 45.8 −119.3 Various 1 9–15 tu,le,st,to,LAI SL Hoogenboometal.(2012)

UnitedStates,Suwanee 2010–2013 30.1 −83.1 13.7 Nmassbalance 1 1 12 1 tuf S Prasadetal.(2015)

UnitedStates,Suwanee 2001 30.1 −83.1 13.7 Nmassbalance 1 1 4 3–5 tu,le,st,soN,SWC S Albert(2002)

UnitedStates,Suwanee 2003 30.1 −83.1 13.7 Nmassbalance 1 1 8 3–5 tu,le,st,soN,SWC S Warren(2003)

(B)Location Year Ntr* Sowingdate

DOYEmergence

dateDOY

HarvestdateDOY Napplication(kg ha−1)** Irrigation(mm) Rainfall(mm) Typeofirrigation CO2(ppm) Cultivar

Argentina,Balcarce 1991 1 298 n.a. 64 0 113 540 Mixed Default Spunta

2 298 n.a. 64 60(1) 113 540 Mixed Default Spunta



Australia 1970 1 222 273 356 425(2) n.a. 219 Full Default Sebago

2 222 273 356 425(2) n.a. 219 Full Default Sebago

3 222 273 356 425(2) n.a. 219 Full Default Sebago

Belgium,Tervuren 1998+ 1N 127 135 257 205(2) 373 423 Mixed 380 Bintje

2N 127 135 257 205(2) 373 423 Mixed 386 Bintje

3N 127 135 257 205(2) 373 423 Mixed 676 Bintje

1999+ 1N 131 144 250 220(2) 182 204 Full 365 Bintje

2N 131 144 250 220(2) 181 204 Full 370 Bintje

3N 131 144 250 220(2) 183 204 Mixed 664 Bintje

Bolivia,Belen 1997 1 288 316 84 110(1) n.a. 264 Full Default Waycha

2 288 316 84 110(1) n.a. 264 Full Default Lucky

Bolivia,Chinoli 1997 1 301 340 62 124(1) n.a. 275 Full Default Desiree

Bolivia,Koari 1997 1 281 329 111 100(1) – 540 Rainfed Default Waychacal

2 281 329 111 100(1) – 540 Rainfed Default Alphacal

3 281 329 111 100(1) – 540 Rainfed Default Luckycal

Bolivia,Patacamaya 1997 1N 300 351 112 110(1) n.a. 341 Full Default Waycha

2N 300 351 112 110(1) n.a. 341 Full Default Lucky

Bolivia,Patacamaya 1998–1 1 292 347 110 110(1) n.a. 334 Full Default Waycha

2 292 347 110 110(1) n.a. 334 Full Default Lucky

Bolivia,Patacamaya 1998–2 1N 292 342 110 110(1) n.a. 334 Full Default Waycha


Bolivia,Toralapa 1993 1N 295 351 103 120(1) n.a. 440 Full Default Waycha

2N 295 351 103 120(1) n.a. 440 Full Default Alpha


4N 295 351 103 120(1) – 440 Rainfed Default Waycha

5N 295 351 103 120(1) – 440 Rainfed Default Alpha

6N 295 351 103 120(1) – 440 Rainfed Default Lucky

China,Huhhot 1996 1 118 n.a. 250 150(2) n.a. 252 Full Default Desiree

China,Huhhot 1998 1 115 n.a. 253 150(2) n.a. 470.9 Full Default Desiree

China,Jining 1999 1 119 n.a. 234 74(2) n.a. 156 Full Default Desiree

2 119 n.a. 234 74(2) n.a. 156 Full Default Kexin1cal

3 119 n.a. 234 74(2) n.a. 156 Full Default Jinguancal

China,Zhalan 1997 1 119 n.a. 253 180(2) n.a. 89 Full Default Desiree

China,Zhalan 1998 1 119 n.a. 242 90(2) – 754 Rainfed Default Kexin1

2 119 n.a. 242 90(2) – 754 Rainfed Default Neishu7cal

Colombia,Cundinamarca 1999 1 119 143 262 100(1) – 392 Rainfed Default Capiro

Denmark,Jyndevad 1981 1 119 149 225 155(1) 169 431 Mixed Default Bintje

1982 1 119 148 236 155(1) 222 521 Mixed Default Bintje


Denmark,Jyndevad 1990 1 107 n.a. 267 180(1) 107 499 Mixed Default Bintjecal

2 107 n.a. 267 180(4) 104 499 Mixed Default Bintjecal

3 107 n.a. 267 180(4) 105 499 Mixed Default Bintjecal


2 101 149 273 180(3) 137 390 Mixed Default Bintje








Denmark,Jyndevad 1984 1 117 151 278 150(1) 71 519 Mixed Default Tilvacal

2 117 151 278 200(1) 71 519 Mixed Default Tilva

1985 1 119 149 270 150(1) 27 465 Mixed Default Tilva


1986 1 118 146 293 150(1) 186 488 Mixed Default Tilva


Denmark,Tylstrup 1981 1 120 155 272 140(1) 25 405 Mixed Default Bintje

2 120 152 272 140(1) 25 405 Mixed Default Savacal

3 120 149 272 140(1) 25 405 Mixed Default Posmocal

4 120 150 272 140(1) 25 405 Mixed Default Kaptahcal

5 120 145 272 140(1) 25 405 Mixed Default Dianellacal


2 109 155 270 180(1) 185 607 Mixed Default Sava

3 109 152 270 180(1) 185 607 Mixed Default Posmo

4 109 152 270 180(1) 185 607 Mixed Default Kaptah

5 109 149 298 180(1) 185 607 Mixed Default Dianella


2 109 154 298 160(2) 135 566 Mixed Default Sava

3 109 150 298 160(2) 135 566 Mixed Default Posmo

4 109 154 298 160(2) 135 566 Mixed Default Kaptah

5 109 149 298 160(2) 135 566 Mixed Default Dianella

Ecuador,SanGabriel 1985 1 45 76 241 168(2) – 426 Rainfed Default INIAP-maria

2 45 76 241 117(1) – 426 Rainfed Default INIAP-gabriela

3 45 76 241 168(2) – 426 Rainfed Default INIAP-maria

4 45 76 241 117(1) – 426 Rainfed Default INIAP-gabriela

Finland,Jokioinen 1998+ 1 152 163 270 80(2) – 329 Full 375 Bintje

1999+ 1 149 159 264 80(2) – 122 Full 550 Bintje

Germany,Giessen 1998+ 1N 124 134 250 150(2) 67 365 Mixed 373 Bintje

2N 124 134 250 150(2) 67 365 Mixed 541 Bintje

3N 124 134 250 150(2) 67 365 Mixed 690 Bintje

1999+ 1N 130 147 258 116(2) 152 267 Full 380 Bintje

2N 130 147 258 116(2) 152 267 Full 541 Bintje

3N 130 147 258 116(2) 148 267 Full 708 Bintje

Germany,Giessen 1998 + + 1N 125 135 257 150(2) 186 417 Mixed 401 Bintje

2N 125 135 257 150(2) 186 417 Mixed 429 Bintje

1999 + + 1 130 145 239 156(2) 128 250 Full 374 Bintje

2 130 145 239 156(2) 126 250 Full 491 Bintje

India,Ludhiana 2008 1 290 n.a. 29 0 80 18 Mixed Default KufriBahar

2 290 n.a. 29 136(2) 80 18 Mixed Default KufriBahar



5 290 n.a. 29 0 160 18 Mixed Default KufriBahar








2010 1 285 n.a. 29 0 80 32 Mixed Default KufriBahar












India,Modipuran 2002 1 288 n.a. 3 181(2) n.a. 0 Full Default KufriBahar

2003 1 299 n.a. 18 181(2) n.a. 30 Full Default KufriBahar







Ireland,Carlow 1998N 1 128 138 280 250(2) 61 343 Full 372 Bintje

2 128 138 280 250(2) 31 343 Full 693 Bintje

1999N 1 140 152 249 250(2) 89 392 Full 372 Bintje

2 140 152 249 250(2) 91 392 Full 670 Bintje

Italy,Rapolano 1998 + + 1 141 149 237 240(2) 309 554 Full 366 Bintje

2 141 149 237 240(2) 294 554 Full 552 Bintje

3 141 149 237 240(2) 285 554 Full 367 Bintje

1999 + + 1 126 147 237 250(2) 462 146 Mixed 367 Bintje

2 126 147 237 250(2) 462 146 Mixed 552 Bintje

3 126 147 237 250(2) 462 146 Mixed 367 Bintje

Peru,LaMolina 2003 1 181 n.a. 314 310(2) 450 0 Irrigated Default Amarilis

2004 1 177 n.a. 307 310(2) 350 0 Irrigated Default Amarilis







Peru,LaMolina 2013 1 179 200 288 210(2) 215 14 Full Default Achirana

2 179 198 288 210(2) 215 14 Full Default Atlantic

3 179 198 288 210(2) 215 14 Full Default Sarnavcal

Peru,LaMolina 1985 1 30 60 122 160(2) 449 3 Irrigated Default DTO-33

2 30 60 130 160(2) 480 3 Irrigated Default LT1

3 30 60 144 160(2) 485 3 Irrigated Default Revolucion

4 176 206 285 160(2) 315 1 Irrigated Default DTO-33

5 176 206 291 160(2) 257 1 Irrigated Default LT1

6 176 206 274 160(2) 272 1 Irrigated Default Revolucion

Peru,SanRamon 1984 1 109 122 199 200(2) n.a. 336 Full Default DTO-33

2 109 126 199 200(2) n.a. 336 Full Default Desiree

3 109 128 199 200(2) n.a. 336 Full Default Revolucion

Peru,SanRamon 2013 1 213 228 301 300(2) 449 303 Mixed Default Achiranacal

2 213 228 301 300(2) 449 303 Mixed Default Atlantic

3 213 228 301 300(2) 449 303 Mixed Default Sarnav

Scotland,Dundee 1984 1N 104 150 268 0 187 202 Mixed Default Marispiper

2N 104 150 268 240(1) 187 202 Mixed Default Marispiper

1985 1N 114 154 262 0 15 373 Mixed Default Marispiper

2N 114 154 262 240(1) 15 373 Mixed Default Marispiper

1986 1N 135 166 266 175(1) 91 200 Mixed Default Marispiper

1987 1N 119 155 258 175(1) 39 346 Mixed Default Marispiper

Sweden,Goteborg 1998+ 1N 145 154 252 88(2) 480 357 Mixed 708 Bintje

2N 145 155 252 88(2) 480 357 Mixed 404 Bintje

Uganda,Kalengyere 2004 1 61 n.a. 169 100(1) – 294 Rainfed Default Asante

2005 1 80 n.a. 189 100(1) – 318 Rainfed Default Asante



UnitedKingdom,Sutton 1998+ 1N 126 136 239 110(2) 98 251 Full 379 Bintje

2N 126 136 239 110(2) 97 251 Full 563 Bintje

3N 126 136 239 110(2) 99 251 Full 673 Bintje

1999+ 1N 132 144 249 250(2) 131 247 Full 399 Bintje

2N 132 144 249 250(2) 134 247 Full 543 Bintje

3N 132 144 249 250(2) 132 247 Full 694 Bintje

UnitedStates,Benton 2003 1 87 110 231 324(9) 666 164 Mixed Default RussetBurbank

2 87 110 231 669(5) 666 164 Mixed Default RussetBurbank

UnitedStates,Hastings 2011 1 12 43 108 168(3) n.a. 255 Full Default Atlantic

2 12 43 108 224(3) n.a. 258 Full Default Atlantic





UnitedStates,NewYork 1980 1 143 157 257 275(2) 209 297 Full Default Kathadin

UnitedStates,Idaho 1988 1 105 129 264 367(10) 669 136 Full Default RussetBurbank

2 89 115 174 354(15) 580 123 Full Default RussetBurbank









UnitedStates,Suwanee 2010 1 41 74 140 265(4) 281 758 Mixed Default RedLasoda

2011 1 28 57 118 278(5) 297 537 Mixed Default RedLasoda

2 43 62 140 285(5) 291 558 Mixed Default RedLasoda




UnitedStates,Suwanee 2001 1 46 62 141 313(5) 507 290 Mixed Default RedLasoda





cal:Treatmentsusedforcalibration.N:TreatmentswithestimatedinitialsoilN.

*:Numberortreatments.

**:Numberofnitrogenapplicationsareinparenthesis.

+:OTC.

++:FACE.

n.a.:Notavailable.

−:Notapplicable.

Full:Automaticirrigation.

Default:AtmosphericCO2concentrationcalculatedbyDSSAT-CSM.

a Latitude.

b Longitude.

c Altitude.

d FACE:Free-Air-CO2-Enrichment,OTC:OpenTopChambers.

e nNumberoftreatments.

f nNumberofcultivars.

g sSamplingrepetitions.

h tu:tTuberdryweight(Mgha−1,tubf:Ttuberfreshweight(Mgha−1,le:Lleafdryweight(Mgha−1,st:Sstemdryweight(Mgha−1,to:abovegrounddryweight(Mgha−1,LAI:leafareaindex,de:deadtissuedryweight(Mgha−1;tuN:tuberNuptake

(kgha−1;le:LeafNuptake(kgha−1;stN:StemNuptake(kgha−1,toN:abovegroundnitrogenuptake(kgha−1),SoN:SoilNcontent(ppm),SWC:Soilwatercontent(m3m−3.

i C:clay;S:sand;Si:silt;L:loam.

Measurementsinmanyofthesepotatoexperimentsdiffered.Forexample,65oftheexperimentshadinformationabouttuberdryweight,while22experimentshadinformationabouttuberfreshweight.Mostexperimentshadin-seasontuber

growthmeasurements(oftenwithmore than twomeasurements),and21experimentsonlyhad informationabout the final tuberyield.Table1aandb lists theexperimentsweusedwith themeasuredexperimentalvariablesand themanagement

information.

3.1WeatherdataThe experiments included daily measurements of solar radiation, maximum and minimum temperatures, and precipitation. Missing data were filled in with data from NASA Prediction of Worldwide Energy Resource (http://power.larc.nasa.gov/cgi-

bin/cgiwrap/solar/agro.cgi).

3.2SoilparametersSeveralofthefieldexperimentshadmeasurementsofsoilparameters.Ifthisinformationwasnotavailable,itwascomputedusingtheSbuildsoilparameterestimationtoolavailableintheDSSAT-CSMsuiteofapplicationswhenquantitativeinformationofsoil

textureandorganiccarbonwereavailable.Genericsoilprofiles,availableinDSSAT-CSM,wereassignedonlywhenqualitativesoiltypeinformationwasavailable.

3.3CalibrationofcultivarparametersCultivarparametersP2,TC,G2,G3,andPDwereobtainedfromliteratureandotherwisefromtheDSSAT-CSMdatabase(Table2).ToestimatenewparametersfornewcultivarsweusedtheGeneralizedLikelihoodUncertaintyEstimation(GLUE)toolofDSSAT-

CSM(Jonesetal.,2011),whichrequiresadefaultsetofparametersandobserveddata.Thedefaultsetofparameterswasassignedfromarandomlyselectedcultivarorwasassignedfromaspecificcultivaronlyifanewcultivarsharedthepedigree(Berlooetal.,2007Berloo

etal.,2007)withaknowncultivarinTable2.Table1bshowsthetreatmentsusedforcalibration.WeevaluatedthemodelperformancewiththedefaultsetofparametersandwiththenewsetofparametersestimatedwithGLUE.ParametersTCandP2weremanually

adjustediftuberyieldsimulationsfailedinoneofcontrastinglocations(LaMolinaandSanRamon).Theseresultswerediscussedwithpotatoexperts(breeders)whoprovidedtheexperimentaldatasets.Aftercultivarparametersweredetermined,theywerekeptunchanged

acrossexperimentsandlocations.Table2liststhecultivarparametersusedinsimulationsperformedwiththeSUBSTOR-potatomodel.

Table2CultivarparametersusedintheSUBSTOR-potatomodel.

alt-text:Table2

Cultivar Species Maturitytype Country G2(cm2m−2day−1)G3

(gm−2day−1)PD(−) P2(−) TC(°C) Source

Achirana S.tuberosum Early Peru 2000 21 0.8 0.5 17

Alpha S.tuberosum Late Bolivia 1000 25 0.9 0.4 20

Amarilis S.tuberosum Medium Peru 2000 30 0.9 0.9 20 (U.Kleinwechter,pers.comm.,2014)

Asante S.tuberosum Medium Uganda 2000 26 0.7 0.9 21 (U.Kleinwechter,pers.comm.,2014)

Atlantic S.tuberosum Early Peru,UnitedStates 1000 25 0.9 0.6 17 Hoogenboometal.(2012)

Bintje S.tuberosum Late Europe 1000 30 0.8 0.1 19

Capiro S.tuberosum Late Colombia 1000 22 0.6 0.7 17 Bowenetal.(1999)

Desiree S.tuberosum Early China,Peru,andBolivia 2000 25 0.9 0.6 16 Hoogenboometal.(2012)

Dianella S.tuberosum Late Denmark 1500 24 0.6 0.3 18

DTO-33 S.tuberosum Early Peru 2000 25 0.7 0.6 17 Bowenetal.(1999)

INIAP-Gabriela S.tuberosum Late Ecuador 2000 19 0.9 0.3 19 Bowenetal.(1999)

INIAP-Maria S.tuberosum Late Ecuador 2000 22 0.4 0.5 19 Bowenetal.,(1999)

Jinguan S.tuberosum – China 2000 24 0.9 0.8 17

Kaptah S.tuberosum Late Denmark 1800 24 0.6 0.4 18

Kathadin S.tuberosum Late UnitedStates 2000 25 0.5 0.7 20 Hoogenboometal.(2012)

Kexin1 S.tuberosum – China 2000 25 0.9 0.8 17

KufriBahar S.tuberosum Early India 2000 22 0.9 0.8 23 (U.Kleinwechter,pers.comm.,2014)

LT-1 S.tuberosum Early Peru 2000 25 0.9 0.8 17 Hoogenboometal.(2012)

Luky S.juzepczukii Late Bolivia 2000 21 0.4 0.8 15

MarisPiper S.tuberosum Late Scotland 2000 25 0.8 0.4 17 Hoogenboometal.(2012)

Neishu7 S.tuberosum – China 1600 23 0.7 0.8 18

Posmo S.tuberosum Late Denmark 1500 24 0.6 0.3 18

RangerRusset S.tuberosum Late UnitedStates 1100 26 0.9 0.6 17 Hoogenboometal.(2012)

RedLasoda S.tuberosum Medium UnitedStates 2000 22 0.7 0.4 19 Hoogenboometal.(2012)

Revolucion S.tuberosum Late Peru 2000 30 1 0.6 17 Bowenetal.(1999)

RussetBurbank S.tuberosum Late UnitedStates 1100 26 0.9 0.6 17 Hoogenboometal.(2012)

Sarnav S.tuberosum Late Peru 1000 30 0.2 0.6 18

Sava S.tuberosum Late Denmark 1300 24 0.6 0.4 21

Spunta S.tuberosum Medium Argentina 1800 24 0.5 0.1 19 Travassoetal.(1996)

Tilva S.tuberosum Late Denmark 1500 24 0.9 0.3 18

Waycha S.andigenum Late Bolivia 1200 23 0.6 0.3 15

Zibaihua S.tuberosum – China 2000 25 0.9 0.8 17

G2:Leafexpansionrate,G3:Ttubergrowthrate,PD:Iindexthatsuppresstubergrowthaftertuberinduction,P2:Ssensitivitytophotoperiod,TC:Uuppercriticaltemperaturefortuberinitiation(°C).−:nNotavailable.

3.4SimulationsSimulated experiments were set to non-water-limiting conditions (automatic irrigation), when irrigation was known to be applied but actual rates weremissing and when the involved researchers confirmed non-water-limiting conditions. Additionally, some

experimentswereconfirmedtobenon-nitrogenlimitedbytheresearchers.Initialsoilconditionsweremissinginabout20%oftheexperiments;inthesecases,simulationswithvariousamountsofinitialmineralsoilNwerecarriedoutforonetreatmentandtheamountof

initialmineralNthatfittedthefinalyieldbestwasappliedtoallothertreatments.ThisprocesswasrepeatedforeveryexperimentwithnoinformationofinitialsoilN.InexperimentswithNratetreatments,theplantNuptakeofthelowestNtreatment(usuallyatreatment

withoutNfertilizer)wassetastheinitialmineralsoilN(Ritchieetal.,1995).Inexperimentswithinformationonsoilorganiccarbon,weinitializedthesoilcarbonpoolswiththeCENTURYmodel(Bassoetal.,2011;Porteretal.,2014),providingthestableorganiccarbonfor

eachsoillayer.Weassumed82%and90%ofthetotalorganiccarbonasstableorganiccarbonbetween0 to60cmandbelow60cm,respectively.

WeperformedsimulationswiththeSUBSTOR-potatomodel−–embodiedinDSSAT-CSM−–forallexperiments.EvapotranspirationwascalculatedwiththePriestley-Taylor/Ritchieformula;soilwaterinfiltrationwascomputedwiththecapacityapproachmethod;

soilevaporationwasestimatedwiththeSuleiman-Ritchiemethod;andthedynamicofcarbonandnitrogenwassimulatedwiththeCENTURYmodel(Hoogenboometal.,2012).

Finally,wesimulatedtheexperimentsfromemergencetoharvestdatewithinformationaboutweather,soilandcultivarcharacteristics,andcropmanagementpractices,asgiveninTable1aandb.Simulationsstartedatplantingdateiftheemergencedaywasnot

available.

3.5EvaluationofmodelperformanceCropmodelperformancewasevaluatedbycomparingsimulatedandobservedin-seasonandend-of-seasonvalues.Experimentalcropmeasurementsincludedtuberdryweight(Mgha−1),tuberfreshweight(Mgha−1),abovegrounddryweight(Mgha−1),rootdry

weight(Mgha−1),leafdryweight(Mgha−1),stemdryweight(Mgha−1),deadmaterialdryweight(Mgha−1),leafareaindex(LAI,−),abovegroundN(kgNha−1),tuberinitiation(days),tuberNcontent(kgNha−1),rootNcontent(%),leafNcontent(kgNha−1),andstemN

content(kgNha−1).Experimentalsoilmeasurementsincludedsoilwatercontent(m3m−3),soilNO3

-N(ppm),andsoilNH4

-N (ppm).We evaluated themodel simulations by comparing withmeasured data from experiments, using statistical indices of coefficient of determination (R2), slope of a linear regression (m), root mean square error (RMSE) (Wallach andGoffinet, 1987), and

relativeRMSE(RRMSE).TheregressiontocalculatetheR2wasfor the1:1 lineandforcedthroughtheorigin.ThisR2valuemeasuresthetruedeviationof theestimates fromtheobservations(Yangetal.,2014).Theslopemquantifiesapossibleoverestimationor

underestimationbythemodel.TheRMSEwascomputedtoprovideameasureoftheabsolutemagnitudeoftheerror.AllcalculationsandgraphsweremadeusingtheRstatisticalsoftware(RCoreTeam,2015).

4ResultsFig.2showsacomparisonofsimulatedandobservedyieldsofthreepotatospecies,S.andigenum(cv. ‘Waycha’),S.tuberosum(cv. ‘Alpha’),andS.juzepczukii(cv. ‘Lucky’)atToralapa,Bolivia(1993)at3430m.a.s.l.Thein-seasontuberdry

weightwaswell-simulatedforthesethreespecies.Thesimulatedin-seasonleafdryweightaccumulation(Fig.2b,d,andf)followedthetrendoftheobservations,althoughitwaslessaccuratethanthetuberdryweightsimulations(Fig.2a,c,ande).

–and

Fig.3showstheperformanceofthemodelforcv.‘Desiree’indifferentlocationswithcontrastingtemperatureandphotoperiodenvironments.Thephotoperiodinthesefourlocationsrangedfrom11.4hoursinPeru(SanRamon)to14.8hours in

northernChina(HuhhotandWumeng).InHuhhot(Fig.3a)andWumeng(Fig.3b),thesimulationcapturedthedynamicsoftheobservationswell,buttendedtooverestimatefinalyieldsinHuhhot(Fig.3a).InSanRamonandChinoli,thesimulatedtuber

growthcorrespondedwellwiththeobservedtubergrowthduringtheinitialpartofthegrowthperiod(Fig.3Fig.3candd).However,attheselocationsobservedtubergrowthstoppedearlierthanassumedinthesimulation,resultinginadifferenceinthe

finaltuberyield.

Fig.2Simulated(—)versusobserved(●)tuberdryweight(Mgha−1)andleafdryweight(Mgha−1)forthreepotatospecies:(a,b)Solanumandigenum−cv.‘Waycha’,(c,d)Solanumtuberosum−—cv.‘Alpha’and(e,f)Solanumjuzepczukii−—cv.‘Lucky’inToralapa,Bolivia,1993.Errorbars

indicatestandarderrorofmeasurementswhenavailable.

alt-text:Fig.2

Fig.4showsacomparisonofsimulatedandobservedvaluesoftuberdryweight,leafareaindex,andtuberNuptakefortwoNtreatments(N0andN240,withapplicationratesof0and240kgNha−1,respectively)onasandyloamsoilat

Dundee,Scotlandin1985.ThemodelsimulatedaccuratelythetimecourseoftheobservedtuberdryweightsforthetwoNtreatments(Fig.4a).ThesimulatedresultsforbothLAIandtuberNuptakesometimesdifferedfromthemeasuredvalues(Fig.

4Fig.4bandc).Forexample,themodeloverestimatedthetuberNuptakefortheN0treatmentbutshowedagoodagreementfortheN240treatment.TheseresultsshowthatingeneralthemodelcanpredicttuberproductionwellfordifferentNfertilizer

treatments,althoughsometimesthesimulatedtimecoursesofLAIandtuberNuptakemaydifferfromtheobservedvalues(Fig.4).

Fig.3Simulated(—)versusobserved(●)tuberdryweight(Mgha−1)forcv.‘Desiree’at(a)Huhhot,China,1996,(b)Wumeng,China,1999,(c)SanRamon,Peru,1984,and(d)Chinoli,Bolivia,1997.Dottedlinesshowtheweightedaveragetemperatureforeachlocation.Errorbarsindicate

standarderrorofmeasurementswhenavailable.

alt-text:Fig.3

Fig.5presentsthemodelresultsfor87experimentsundervarioustreatmentsandconditions, includingnon-stress,water limited,hightemperature,andNlimitedconditions,andbothcurrentandelevatedatmosphericCO2concentrations.

Overall,themodelresultscorrespondedwellwiththeobservedtuberdryandfreshweights(Fig.5a,b).NotethatforsomeOTCtreatmentsthemodelunderestimatedthetuberdryweightdespitethenon-limitingconditionsofwaterandNmanagement

(Fig.5a).Forleafdryweight,stemdryweightandLAI(Fig.5c–f)thecorrespondencebetweensimulatedandobservedvaluesappearedtobelimited.Thetuberinitiationperiodwasgenerallysimulatedwellforallregions,althoughsomeunderestimated

valuesweresimulatedforthehighAndes(Patacamaya−3700m.—3700m.a.s.l)andhightemperatureconditions.

Fig.4Simulated(lines)versusobserved(symbols)tuberdryweight(Mgha−1),leafareaindex(LAI),andtuberNuptake(kgNha−1),fortwonitrogentreatments:N0,andN240atDundee,Scotland,1985.TreatmentN0(dottedlines,opensymbols)andtreatmentN240(solidlinesandsolid

symbols).

alt-text:Fig.4

Thesimulated tuberNcontentsand leafNcontentscorrespondedwellwith theobservedvalues(Fig.6aandc),but thesimulatedabovegroundandstemNcontentsshowed largediscrepancies (Fig.6bandd).Fig.6eand f presents a

comparisonofsimulatedandmeasuredsoilmineralNandsoilwatercontentsintheexperimentsatSuwanee,UnitedStates(subtropicalregion)andatDundee,Scotland(temperateregion).TheseresultsshowthatsoilmineralNasNO3

-NandNH4-NandNH4

N(notshowninfigure)werepoorlysimulated;additionally,theobserveddatahadlargemeasurementerrors.ThesoilwatercontentsweremoderatelywellsimulatedforSuwanneebutunderestimatedforDundee.

Fig.5Modelperformancefor(a)tuberdryweight(Mgha−1),(b)tuberfreshweight(Mgha−1),(c)stemdryweight(Mgha−1),(d)leafdryweight(Mgha−1),(e)leafareaindex,and(f)tuberinitiation(dap−—daysafterplanting)intropical(○),subtropical(△),andtemperate(+)regionsandfor

atmosphericCO2concentrations:380ppm(*),400ppm(×),500ppm(♢),550ppm(▽),680ppm(●),700ppm(□),forFACEandOTCexperiments.Errorbarsindicatestandarderrorofmeasurementswhenavailable.

alt-text:Fig.5

Fig.7providesthetuberyieldresponseofcv. ‘Achirana’undernon-high(LaMolina,Peru)andhightemperature(SanRamon)environments.ThiscultivarhasaTCof18°C;therefore,weightedaveragetemperaturesabovethisthreshold

inhibited tuber initiationandreduced tuberbulking. InLaMolinaandSanRamon,weightedaverage temperatures for tuber initiationweresuitable (<17°C)andunsuitable (>20°C), respectively.Undernon-high temperature conditions, themodel

providedsatisfactorysimulations.However,underhightemperatureconditions,themodelreproducedthefinalyieldwell,butoverestimatedthetuberinitiationandunderestimatedthetuberdryweightsduringthegrowthperiod.Thisshowsthatunder

hightemperatureconditionsthroughoutthegrowingseason,SUBSTOR-potatofailedtosimulatetheobservedgrowthdynamicswell.

Fig.6Modelperformanceof(a)tuberNuptake(kgNha−1),(b)abovegroundNuptake(kgNha−1),(c)leafNuptake(kgNha−1),(d)stemNuptake(kgNha−1),(e)soilmineralNasaNO3

N(ppm),and(f)soilwatercontent(m3m−3)forsubtropical(△),andtemperate(+)regions.Errorbarsindicatethestandarddeviationofmeasurementswhenavailable.Forpanel(e)and(f)errorbarswereavailablebutnotshown.

alt-text:Fig.6

Fig.8showsthetuberyieldresponsetoelevatedatmosphericCO2concentrationsinaFACEexperimentin1999atRapolano,Italy.SUBSTOR-potatosimulatedthetubergrowthunderambientCO2concentrationsreasonablywell,butfailedto

simulatethelargepositiveobservedyieldresponsetoelevatedatmosphericCO2.

Table3summarizestheperformanceoftheSUBSTOR-potatomodelversusexperimentalobservationsfor17variables.Simulationswiththemodelwerecomparedto5345actualmeasurements.ValuesofRRMSEwere37.2%fortuberdry

weight,21.0%fortuberfreshweight,22.6%fortuberinitiation,40.4%fortuberNuptake,and52.3%forleafNuptake.RRMSEforstemNcontent,abovegroundNcontent,andsoilNO3andNH4were>80%.

Table3SummaryofmodelperformanceofSUBSTOR-potatomodellingforallexperiments(seeTable1).

alt-text:Table3

Variable Numberofpaireddata Observedrange R2 Slope RMSEa RRMSEb

Tuberdryweight(Mgha−1) 946 0.00c–32.08 0.93 1.01 2.12 37.20

Tuberfreshweight(Mgha−1) 79 0.50–54 0.97 0.96 5.23 21.04

Abovegrounddryweight(Mgha−1) 758 0.03–9.75 0.69 0.64 1.93 85.33

Rootsdryweight(Mgha−1) 213 0.02–1.08 0.63 0.45 0.34 253.07

Fig.7Simulated(lines)versusobserved(symbols)tuberdryweight(Mgha−1)underhightemperatureatSanRamon,Peru,2013andalowtemperatureenvironmentatLaMolina,Peru,2013forcv.‘Achirana’.Hightemperaturesconditionswithdottedlinesandopensymbols,andlow

temperaturesconditionswithsolidlinesandsolidsymbols.Errorbarsindicatestandarderrorofmeasurementswhenavailable.Weightedaveragetemperatureforbothlocationsisembeddedinthegraphic.

alt-text:Fig.7

Fig.8Simulated(lines)versusobserved(symbols)cumulativepotatotuberdryweight(Mgha−1)forcv.‘Bintje’,FACEexperimentwithambient(370ppm)andelevatedatmosphericCO2concentration(560ppm)atRapolano,Italyin1999.TreatmentwithelevatedatmosphericCO2withdottedlines

andopensymbols,andtreatmentwithambientCO2withsolidlinesandsolidsymbols.Errorbarsindicatestandarderrorofmeasurementswhenavailable.

alt-text:Fig.8

Leafdryweight(Mgha−1) 504 0.01–3.67 0.61 0.64 0.97 90.34

Stemdryweight(Mgha−1) 471 0.02–4.32 0.51 0.48 1.32 141.42

Deaddryweight(Mgha−1) 57 000d–2.56 0.26 0.79 0.62 151.42

LAI 378 0.01–8.39 0.70 0.67 2.24 81.95

Tuberinitiation(dap) 116 26.00–113 0.96 0.93 10.54 22.65

TuberN(kgha−1) 184 0.10–318.20 0.92 0.90 44.78 40.40

AbovegroundN(kgha−1) 160 76.48–132.72 0.78 0.65 76.48 86.23

RootsN(kgha−1) 96 0.63–8.19 0.60 0.37 6.95 153.83

LeafN(kgha−1) 64 21.48–52.32 0.86 0.89 21.48 52.31

StemN(kgha−1) 64 0.62–59.25 0.74 0.28 47.18 309.57

Soilwatercontent(m3m−3) 891 0.03–0.49 0.89 0.89 1.55 62.08

NO3 -N(ppm) 240 0.10–58.10 0.42 1.35 9.82 95.45

NH4 -N(ppm) 240 0.60–185.60 0.05 1.66 19.42 140.09

a Rootmeansquareerror(Mgha−1).

b Relativerootmeansquareerror(%).

c 0.00002.

d 0.00002.

5Discussion5.1Cultivarsofthedevelopedworld

Thisstudypresentsacomprehensivemodeltestingforapotatocropmodel,usingmultiplecultivars,locations,andtreatments.Previousstudiesalsotestedvariouspotatocultivarsbutwereoftenlimitedinthenumberofcultivars.Otherstudieshadanalyzedmodel

performanceforasinglecultivarwiththemodelsSIMPOTATOandDANUBIA(Hodgesetal.,1992;Gayleretal.,2002;Lenz-Wiedemannetal.,2010);sixcultivarswiththemodelSUBSTOR-potato(Griffinetal.,1993);sevencultivarswiththemodelDAISY(Heidmannet

al.,2008);and10cultivarswiththemodelINFOCROP(Aggarwaletal.,2006).Onlyonemodel,SOLANUM,wasusedtocomparesimulatedandobservedyieldsforalargenumberofpotatospecies(Condorietal.,2010).Inourstudy,weanalyzedtheperformanceofthe

SUBSTOR-potatomodelforthreepotatospeciesand32potatocultivars.TheresultsindicatedthattheSUBSTOR-potatomodelcaningeneralreproducethetuberandleafdevelopmentfordifferentpotatospeciesandcultivars.

5.2CultivaruseacrossclimaticregionsAnimportanttestforanycropmodelistoperformsimulationsforonecultivargrownacrossawiderangeofclimaticregions.Cultivarparametersinsuchevaluationstudymustbekeptconstantacrossthedifferentenvironmentstobeconsideredtrulycultivar-

specific.WolfandVanOijen(2003)usedcv.‘Bintje’totestthemodelLPOTCOacrosseightlocationsinEurope,whereasHeidmannetal.(2008)usedcv.‘Agria’totestthemodelDAISYatthreelocationsofEurope.However,thelocationsusedintheirstudieswereallpart

ofthesametemperateregion.Inourstudy,acropmodelwastestedforthefirsttimewithacultivaracrossseveralclimaticregionsusingconstantcultivarparameters.Observedtuberyieldswerewellsimulatedforcv.‘Desiree’intemperate(China)andtropical(Peru,

Bolivia)regions.Similarly,simulationswereacceptableforcv.‘Atlantic’inthetropics(Peru)andsubtropics(southernUnitedStates)regions.Notethatconditionsbetweenthetemperateandtropicalregionsweremorecontrasting,whereasconditionsofthesubtropicsand

tropicsweremorealike.Tuberyieldsofcultivars‘Sarnav’and‘Achirana’werewellsimulatedinthetropics.Despitethelackofexperimentalmeasurementsofthesecultivarsintemperateandsubtropicsregions,wepresumethatthesamesetofparameterscanbeusedin

temperateregionsofUzbekistan,Argentina,andTheNetherlands,wherethesecultivarswereusedinvariousstudies(Butzonitchetal.,1994;Inceogluetal.,2010;Carlietal.,2014).TheseresultsindicatethatthecultivarparametersintheSUBSTOR-potatomodeldo

representcultivar-specificcharacteristicsandsupportthegeneralfunctionalityandtransferabilityofthismodel.

5.3NresponseTheSUBSTOR-potatomodelreproducedthetuberyieldresponsetovariousNtreatments.TheresultsvariedwithapplicationofNandthelevelsofsoilorganiccarbon(OC).InsoilswithhighOC,themineralizationofsoilorganicNcontributestothecropNsupply

duringthegrowingseason(Bassoetal.,2011).Therefore,themodelsimulatedhightuberyieldsforN0treatments(7.7Mgha−1ofdryweight)inDundee(OC=2.2%).Also,othermodelshavebeenshowntoreproducepotatoyieldsunderoptimumNrates(higherthan

150kgha−1ofN)(Hodges,1998;Gayleretal.,2002).TheSUBSTOR-potatomodelshowedreasonableresultsforN0treatmentsinasoilwithloworganiccarboncontent(OC=0.3%)(Aroraetal.,2013);however,usingthesameexperimentalsoilparameterizationdidnot

allowustoreproduceobservedplantgrowthdynamicsfortheN0simulationsobtainedbyAroraetal.(2013).ThesimulatedNmassbalancesuggestedthatothersourcesofNmusthavebeenavailable,buttheinformationprovidedbyAroraetal.(2013)didnotallowthese

sourcestobeidentified.EstimatesbyBobbinketal.(2010) indicatedthatatmosphericNdeposition inIndia,calculatedwithatransportanddepositionNmodel,couldrangebetween15 to–30kgyear−1andcouldpotentiallybeasignificantsourceofadditionalN for

croppingsystems,asalsoshownforotherregionsoftheworld(Assengetal.,2000).Similarly,wecouldnotsatisfactorilyreproducetheobservationsforsomeoftheNratetreatmentsforArgentinawhereNavailabilitylimitedcropgrowth(Travassoetal.,1996).

5.4SimulationoftuberyieldandothervariablesTheSUBSTOR-potatomodel simulated tuber dry and tuber freshweights in current growing conditions reasonablywell (Fig.5a and b). However, themodel indicated limitations for simulations under high temperature environments (Fig. 7) and elevated

atmosphericCO2concentrations(Fig.8).CropmodelsforcerealshavebeenshowntosimulateelevatedatmosphericCO2concentrationswell(O’learyetal.,2014),butthesemodelsstillrequireimprovementstosimulatetheimpactofheattemperaturestress(Assenget

al.,2015)andinteractionsofCO2withhightemperatures(Assengetal.,2013).

Othercropvariableswerelesswellsimulatedthantuberyields,includingabovegroundbiomass,LAI,androotdryweight.Thisshowsthatitisdifficulttosimulateleafandstemdryweightsandleafareaindexprecisely,butalsomeasurementerrorsareoftenlarge

forthesevariables(vanOijenandEwert,1999).Thepotentialbiomassisafunctionofphotosyntheticallyactiveradiation(PAR),radiationuseefficiency(RUE),andlightinterception.LightinterceptionsaturatesatLAIhigherthanthree;therefore,anyLAIhigherthanthis

valuewillhavethesameimpactonpotentialbiomass.

Leafareaismodeledbyconvertingtheincrementofleafweightintoleafareausingthespecificleafarea-weightratio(LALWR,270cmg-1).NgandLoomis(1984)reportLALWRfromdifferentpotatoexperimentsrangingfrom202to303cmg−1,howeverthe

variabilityofLALWRinpotatocropduringthecropdevelopmentandphenologyisnotwelldocumented.TheuseofvariousmethodstomeasuretheLAIofamaizecrophaveshownsubstantialdifferencesinasingleplotexperiment(Yangetal.,2012Yangetal.,2014

Consequently,someofthediscrepanciesbetweensimulationsandobservationsareduetofieldexperimentswithrelativelargemeasurementerrorcomprisedfromexperimentsfromdifferentsources.Inaddition,inthepotatomodel,biomassandyieldaccumulationarea

functionofLAI,lightinterceptionandRUE.However,otherfactors,suchnon-optimaltemperatures,waterandnitrogenstresscanlimitactualcarbonfixationinthemodel.Inaddition,understressconditionstubergrowthhaspriorityforcarbonallocationwithlesscarbon

allocationtoleavesandstems.Hence,whileLAIisimportantforlightinterception,itisoftennotthelimitingfactorforyield.

ThispartlyexplainsthediscrepanciesofobservedandsimulatedLAInotaffectinggrowthandyieldinthesameway.Thisphenomenonwasalsoreportedforothercrops(Assengetal.,1998;Assengetal.,2000).Otherpotatomodels,suchastheSPUDSIM

model,alsohaddifficulties insimulatingrootdryweight(Datheetal.,2014).Growthhabits(HuamanandSchmiediche,1999)androoting traits (Wishartetal.,2013)varyacrosscultivars,species,and regions.Forexample,S.tuberosum ismore robustand taller in

temperate regions,andsmaller in the tropics (VanderZaagetal.,1990); thiswould implicatediscrepancies instemweightofasamecultivar incontrasting regions.Traditionalandmoderncultivarsgivehigherharvest indexes thannon-improvedspeciessuchasS.

andigenumandS.juzeppzukii(Condorietal.,2010).Assomeof theexistingcultivarvariability isnot taken intoaccountvia thecurrentcultivar-specificmodelparameters, thiscouldpartlyexplain theratherpoorlysimulatedresults forabovegroundbiomassandroot

weights.Inaddition,thequalityofrootmeasurementsinpotatoescouldbeafactorinsomemodelobservationdiscrepancies(Ahmadietal.,2014).

Someofthemodeldiscrepancieswiththeobservedsoilwatercontent(SWC)andsoilmineralNcanbeattributedtotheone-dimensionalwatermovement(tippingbucket)oftheDSSATmodel(Ritchieetal.,1995)andthetwo-dimensionalstructureofridgesand

valleysinpotatofields.

5.5OverestimationattheendofthegrowingseasonIngeneral,theSUBSTOR-potatomodelresultsweremoreaccurateinexperiments,inwhichNsupplyfromthesoilwasexhaustedtowardtheendofthegrowingseason.However,themodeloftenoverestimatedthefinalyieldinsituationsinwhichthesoilNsupply

was large.Such overestimated yields are attributed to the lack of simulatingmaturity in theSUBSTOR-potato. The simulated crop growth requires constantwater andN supplies and stops under terminalwater andN limitations. The concept of cropmaturity and

senescenceiswidelydiscussedforpotato(MackerronandDavies,1986;Khanetal.,2013),andtheimpactofwaterorNstressoncropsenescenceandthematuritytypeofcultivars(early,medium,orlate)isnotclear.Themodelalsolackstheabilitytoaccelerateleaf

senescenceduetohightemperatures.Increasingtemperaturesinthesubtropicsattheendoftheseason(Rahmanetal.,2014)oftenacceleratecropsenescence(KoomanandHaverkort,1995).Thispartlyexplainswhypotatocultivarsinthesubtropicsusuallyhave

shortergrowingperiods(Santhoshetal.,2014).

5.6SimulationsunderhightemperaturesIntheSUBSTOR-potatomodel,hightemperaturehasnoeffectonabovegrounddevelopment,buthasadirectimpactontuberinitiationandtuberdevelopment.Therefore,underconstantlyhightemperaturesinlowlandtropics,themodelsimulatesadelayedtuber

initiationandunderestimatedthein-seasontubergrowth.Incontrast,experimentalobservationsinSanRamon,Peru,suggestedthattuberinitiationoccurreddespitehightemperatures.Somestudieshaveexploredtheeffectofhightemperaturesontuberinitiation(O’Brien

etal.,1998;LevyandVeilleux,2007).Theirresultsindicatedthathightemperaturesaffectedtheallocationofglucose,stimulatingabovegroundbiomassgrowthandsimultaneouslyreducingtuberaccumulationbutnotnecessarilyinhibitingtuberinitiation(Ewing,1981;

Gawronskaetal.,1992;BasuandMinhas,1999).OurresultsshowedthelimitedcapabilityoftheSUBSTOR-potatomodeltosimulatehightemperatureresponseswhenhightemperaturesarefrequentthroughoutthegrowingseason(e.g.,lowlandtropics).Ontheother

hand,themodelmatchedtheobserveddatawhenhightemperaturesoccurredaftertuberinitiationorattheendofthegrowingseason.Theseconditionswerefrequentinthesubtropics(UnitedStates)andinthetemperate(northernChina)regions.

5.7CO2effectFinnanetal.(2008)showedthatthefertilizationeffectsofincreasedCO2(550ppmto680ppm)ontuberyieldsofpotatoinOTCandFACEfacilitieswerehighlyvariable,rangingfrom−7.3%to+54%.TheFACEexperiments(550ppm)fromItalyandGermany

showedanincreaseontuberyieldof46%and5.75%respectively(Finnanetal.,2008).Jaggardetal.(2010)usedtherelativechangeofFACEexperimentsfromItalytoindicateayieldincreaseof36%by2050.Migliettaetal.(1998)indicatedanincreaseoftuberyieldof

10%forevery100ppmofCO2.TheFACEexperiments in Italypresented in thisstudywerealsoused to test theperformanceof themodelsLPOTCOandAQUACROP(WolfandVanOijen,2003;Vanuytrechtetal.,2011).Theobservedyield responseofpotato to

increasedCO2intheseexperimentswashighbutunderestimatedintheSUBSTOR-potatosimulations.AnunderestimationdidalsooccurbuttoalesserextentinthesimulationresultsfromthemodelLPOTCO(WolfandVanOijen,2003).IntheSUBSTOR-potatomodel,

thepotentialcarbonfixationandpotentialtubergrowthincreasesby17%,whenatmosphericCO2increasesfrom330ppmto550ppm.FortheFACEexperimentsinRapolano,Italy in1999,themodelsimulationsindicatedthatNstressshouldhavelimitedthecarbon

fixation.ThisresultedinalowersimulatedyieldresponsetoincreasedCO2ofonly3%,whereastheobservedyieldresponsewasashighas45%.However,simulationsoftheseexperimentsundersufficientNsupplyresultedinoverestimationoftheobserveddata.The

interactionsbetweenincreasedCO2anddifferentNfertilizerrateshavenotbeenstudiedinpotatofieldexperiments,althoughstudiesinothercropsindicatedthatincreasingNshortagereducestheyieldresponseofmostcropstoincreasedatmosphericCO2(Wolf,1996;

Kimetal.,2003;;Franzaringetal.,2011).Notethatintheexperimentsdoneinthe“CHangingclimateandpotentialImpactonPotatoyieldandquality”(CHIP)project(DeTemmermanetal.,2002b),theyieldresponseofpotatotoincreasedCO2wasmuchlower(i.e.,about

+20%to+30%iftheCO2concentrationincreasedfrom380to550ppm)thanthatobservedintheFACEexperimentsinRapolano(DeTemmermanetal.,2002b;WolfandVanOijen,2003).

6ConclusionTuberyieldsweregenerallywellsimulatedwiththeSUBSTOR-potatomodelfordifferentpotatospeciesandcultivars,acrossawiderangeofmanagementandenvironmentsundercurrentgrowingconditions.Thesimulationresultsforother

cropgrowthvariables(e.g.,leafareaindex,leafandstembiomass)werelessaccurateincomparisontoexperimentaldata,whichwaspartlyduetothelimitedparametersforcultivarcharacterizationintheSUBSTOR-potatomodel.However,someof

thesevariablesalsohadsignificantlargemeasurementerrors.ConsistentunderestimationsoccurredunderhightemperatureandelevatedatmosphericCO2concentrationsandrequireimprovementsbeforethemodelcanbeusedforclimatechange

impactassessments.

Implementingasenescenceroutineaffectedbyhighmaximumtemperaturesandmaturitytypeshouldimprovethemodelsimulationsunderhightemperatureenvironments.Thesenescenceroutineshouldtriggertuberinduction,affecttherate

anddurationof tuberbulking,anddecrease thewaterandnitrogenuptakeat theendof thegrowingseason. In theSUSBTOR-potatomodel, the relative response function toatmosphericCO2 is too lowanddoes requireadjustmentsbasedon

experimentaldata.UncitedreferencesAlbert(2002),BerlooR.v.Huttenetal.(2007),ZZZZ(2016),LutaladioandCastaidi(2009),Lutz(2010).

AcknowledgementsWethanktheInternationalResearchFoodPolicyResearchInstitute(IFPRI)andtheUSAIDlinkagefundproject“CollaborationbetweentheInternationalPotatoCenter(CIP)andtheUniversityofFlorida(UF)tobetterunderstandandprioritize

climatechangeadaptationneedsforfoodsecurityintheAndes”.

ReferencesAggarwalP.K.,KalraN.,ChanderS.andPathakH.,InfoCrop:Aadynamicsimulationmodelfortheassessmentofcropyields,lossesduetopests,andenvironmentalimpactofagro-ecosystemsintropicalenvironmentsI.Modeldescription,Agricultural

Systems.Syst.89,2006,1–25.

AhmadiS.H.,SepaskhahA.R.,AndersenM.N.,PlauborgF.,JensenC.R.andHansenS.,Modelingrootlengthdensityoffieldgrownpotatoesunderdifferentirrigationstrategiesandsoiltexturesusingartificialneuralnetworks,FieldCrops

Research.162,2014,99–107.

AlbertM.,MonitoringandmodelingthefateandtransportofnitrateinthevadosezonebModelingtheFateandTransportofNitrateintheVadoseZoneBeneathaSuwanneeeRiverBasinvegetablefVegetableFarm,2002,DepartmentofAgriculturalandBiological

Engineering;UniversityofFlorida,Gainesville,Florida,pp.1–179.

AlvaA.K.,MarcosJ.,StockleC.,ReddyV.R.andTimlinD.,Acropsimulationmodelforpredictingyieldandfateofnitrogeninirrigatedpotatorotationcroppingsystem,JournalofCropImprovement.CropImprov.24,2010,142–152.

AroraV.K.,NathJ.C.andSinghC.B.,Analyzingpotatoresponsetoirrigationandnitrogenregimesinasub-tropicalenvironmentusingSUBSTOR-Ppotatomodel,AgriculturalWaterManagement.WaterManage.124,2013,69–76.

AssengS.,KeatingB.A.,FilleryI.R.P.,GregoryP.J.,BowdenJ.W.,TurnerN.C.,PaltaJ.A.andAbrechtD.G.,PerformanceoftheAPSIM-wheatmodelinWwesternAustralia,FieldCropsResearch.57,1998,163–179.

AssengS.,vanKeulenH.andStolW.,PerformanceandapplicationoftheAPSIMNwheatmodelintheNetherlands,EuropeanJournalofAgronomy.J.Agron.12,2000,37–54.

AssengS.,EwertF.,RosenzweigC.,JonesJ.W.,HatfieldJ.L.,RuaneA.C.,BooteK.J.,ThorburnP.J.,RotterR.P.,CammaranoD.,BrissonN.,BassoB.,MartreP.,AggarwalP.K.,AnguloC.,BertuzziP.,BiernathC.,ChallinorA.J.,DoltraJ.,GaylerS.,

GoldbergR.,GrantR.,HengL.,HookerJ.,HuntL.A.,IngwersenJ.,IzaurraldeR.C.,KersebaumK.C.,MuellerC.,KumarS.N.,NendelC.,O'LearyG.,OlesenJ.E.,OsborneT.M.,PalosuoT.,PriesackE.,RipocheD.,SemenovM.A.,ShcherbakI.

StedutoP.,StoeckleC.,StratonovitchP.,StreckT.,SupitI.,TaoF.,TravassoM.,WahaK.,WallachD.,WhiteJ.W.,WilliamsJ.R.andWolfJ.,Uncertaintyinsimulatingwheatyieldsunderclimatechange,NatureClimate.Clim.Change3,2013,

827–832.

AssengS.,EwertF.,MartreP.,RoetterR.P.,LobellD.B.,CammaranoD.,KimballB.A.,OttmanM.J.,WallG.W.,WhiteJ.W.,ReynoldsM.P.,AldermanP.D.,PrasadP.V.V.,AggarwalP.K.,AnothaiJ.,BassoB.,BiernathC.,ChallinorA.J.,DeSanctisG.,

DoltraJ.,FereresE.,Garcia-VileM.,GaylerS.,HoogenboomG.,HuntL.A.,IzaurraldeR.C.,JablounM.,JonesC.D.,KersebaumK.C.,KoehlerA.K.,MuellerC.,KumarS.N.,NendelC.,O'LearyG.,OlesenJ.E.,PalosuoT.,PriesackE.,Rezaei

E.E.,RuaneA.C.,SemenovM.A.,ShcherbakI.,StoeckleC.,StratonovitchP.,StreckT.,SupitI.,TaoF.,ThorburnP.J.,WahaK.,WangE.,WallachD.,WolfI.,ZhaoZ.andZhuY.,Risingtemperaturesreduceglobalwheatproduction,Nature

Climate.Clim.Change5,2015,143–147.

BachA.andNielsenS.,Vækstanalyseikartofler1981-–83,Planteavl89,1985,215–224.

BassoB.,GargiuloO.,PaustianK.,RobertsonG.P.,PorterC.,GraceP.R.andJonesJ.W.,ProceduresforInitializingSoilOrganicCarbonPinitializingsoilorganiccarbonpoolsintheDSSAT-CENTURYModelforAgriculturalSmodelforagriculturalsystems,Soil

ScienceSocietyofAmericaJournal.Soc.Am.J.75,2011,69–78.

BasuP.S.andMinhasJ.S.,Tuberizationathightemperaturesindifferentpotatogenotypes,JournaloftheIndianPotatoAssociation.IndianPotatoAssoc.26,1999,19–22.

BerlooR.v.,HuttenR.C.B.,EckH.J.v.,VisserR.G.F.,vanBerlooR.andvanEckH.J.,Anonlinepotatopedigreedatabaseresource,PotatoResearch.50,2007,45–57.

BobbinkR.,HicksK.,GallowayJ.,SprangerT.,AlkemadeR.,AshmoreM.,BustamanteM.,CinderbyS.,DavidsonE.,DentenerF.,EmmettB.,ErismanJ.W.,FennM.,GilliamF.,NordinA.,PardoL.andDeVriesW.,Globalassessmentofnitrogen

depositioneffectsonterrestrialplantdiversity:asynthesis,EcologicalApplications.Appl.20,2010,30–59.

Bowen,W.,Cabrera,H.,Barrera,V.,Baigorria,G.,1999.Simulatingtheresponseofpotatotoappliednitrogen.Impactonachangingworld.InternationalPotatoCenterProgramReport1997–1998,381–386.

Butzonitch,I.P.,Colavita,M.L.,Lucarini,O.,1994.PotatovirusMinArgentina.BoletinTecnico—EstacionExperimentalAgropecuaria,Balcarce,11pp.-11pp.

CIP,AnnualrReport1992,1992,InternationalPotatoCenter;Lima-Peru.

CarliC.,YuldashevF.,KhalikovD.,CondoriB.,MaresV.andMonneveuxP.,Effectofdifferentirrigationregimesonyield,wateruseefficiencyandqualityofpotato(SolanumtuberosumL.)inthelowlandsofTashkent,Uzbekistan:Aafieldandmodeling

perspective,FieldCropsResearch.163,2014,90–99.

ClavijoPonceN.L.,ValidacióndelmodelodesimulacionDSSATenelcultivodepapa(SolanumtuberosumL.)enlascondicionesdelcantonMontufarprovinciadDelModeloDeSimulacionDSSATEnElCultivoDePapa(SolanumTuberosumL.)EnLasCondiciones

DelCantonMontufarProvinciaDelCarchi,1999,Escueladeingenieriaagronómica,EscuelasuperiorpolitécnicadeChimborazo;Riobamba,Ecuador,p.80.

CondoriB.,HijmansR.J.,QuirozR.andLedentJ.F.,QuantifyingtheexpressionofpotatogeneticdiversityinthehighAndesthroughgrowthanalysisandmodeling,FieldCropsResearch.119,2010,135–144.

CurryR.B.,PeartR.M.,JonesJ.W.,BooteK.J.andAllenL.H.,Simulationasatoolforanalyzingcropresponsetoclimatechange,TransactionsoftheAsae.ASAE33,1990,981–990.

DatheA.,FleisherD.H.,TimlinD.J.,FisherJ.K.andReddyV.R.,Modelingpotatorootgrowthandwateruptakeunderwaterstressconditions,AgriculturalandForestMeteorology.ForestMeteorol.194,2014,37–49.

DeTemmermanL.,HacourA.andGunsM.,Changingclimateandpotentialimpactsonpotatoyieldandquality‘CHIP'’:introduction,aimsandmethodology,EuropeanJournalofAgronomy.J.Agron.17,2002a,233–242.

DeTemmermanL.,WolfJ.,CollsJ.,BindiM.,FangmeierA.,FinnanJ.,OjanperaK.andPleijelH.,Effectofclimaticconditionsontuberyield(SolanumtuberosumL.)intheEuropean‘CHIP'’experiments,EuropeanJournalofAgronomy.J.Agron.17,2002b,

243–255.

EdlefsenO.,Styringafkvælstofivandedekartofler,Science,1991,T.D.I.o.A.

EwingE.E.,Heat-stressandthetuberizationstimulus,AmericanPotatoJournal.PotatoJ.58,1981,31–49.

FAO,FAOSTAT,2010http://faostat.fao.org/.

FinnanJ.M.,DonnellyA.,JonesM.B.andBurkecJ.I.,Theeffectofelevatedlevelsofcarbondioxideonpotatocrops,JournalofCropImprovement.CropImprov.13,2008,91–111.

FleisherD.H.,BarnabyJ.,SicherR.,ResopJ.P.,TimlinD.J.andReddyV.R.,EffectsofelevatedCO2andcyclicdroughtonpotatoundervaryingradiationregimes,AgriculturalandForestMeteorology.For.Meteorol.171,2013,270–280.

ForeroHernandezD.andGarzonMontañoE.,In:MontanoE.G.,(Ed),ValidaciondelmodelodesimulaciondecDelModeloDeSimulacionDeCrecimientoSUBSTOR-potatoV.35paracuatrovariedadesmejoradasdepapa(Solanumtuberosumssp.andigena)

bajocondicionesdecultivocParaCuatroVariedadesMejoradasDePapa(SolanumTuberosumSsp.Andigena)BajoCondicionesDeCultivoComercial,2000,FacultadDeAgronomia,UniversidadNacionalDeColombia;Bogota,1–84.

FrankeA.C.,HaverkortA.J.andSteynJ.M.,ClimatechangeandpotatoproductionincontrastingSouthAfricanagro-ecosystems2:.Assessingrisksandopportunitiesofadaptationstrategies,PotatoResearch.56,2013,51–66.

FranzaringJ.,WellerS.,SchmidI.andFangmeierA.,Growth:senescenceandwateruseefficiencyofspringoilseedrape(BrassicanapusL.cv.Mozart)growninafactorialcombinationofnitrogensupplyandelevatedCO2EnvironmentalandExperimentalBotany2,

Environ.Exp.Bot.72,2011,284–296.

GaoJ.L.,LiuK.L.,ZhangB.L.andRenY.Z.,Accumulationanddistributionofdrymatterinpotato,ChinaPotato17,2003,209–212.

GaoJ.L.,LiuK.L.,ShengJ.H.,RenK.,WenX.J.,SuiQ.J.andJiangB.,Drymatteraccumulationanddistributionofpotatounderdryfarming,PotatoChina18,2004,9–15.

GawronskaH.,ThorntonM.K.andDwelleR.B.,Influenceofheat-stressondry-matterproductionandphotoassimilatepartitioningby4potatoclones,AmericanPotatoJournal.PotatoJ.69,1992,653–665.

GaylerS.,WangE.,PriesackE.,SchaafT.andMaidlF.X.,Modelingbiomassgrowth,N-uptakeandphenologicaldevelopmentofpotatocrop,Geoderma105,2002,367–383.

GobinA.,ModellingclimateimpactsoncropyieldsinBelgium,ClimateResearch.Res.44,2010,55–68.

GriffinT.S.,BradleyS.J.andRitchieJ.T.,AsimulationmodelforpotatogrowthanddSimulationModelforPotatoGrowthandDevelopment:SUBSTOR-potatovVersion2.0,1993,DepartmentofAggronomyandSoilScience,CollegeofTropicalAgricultureand

HumanResources,UniversityofHawai;Honolulu,p.29.

HaverkortA.andTopJ.,Thepotatoontology:delimitationofthedomain,modellingconcepts,andprospectsofperformance,PotatoResearch.54,2011,119–136.

HaverkortA.J.,FrankeA.C.,EngelbrechtF.A.andSteynJ.M.,ClimatechangeandpotatoproductionincontrastingSouthAfricanagro-ecosystems1:.Effectsonlandandwateruseefficiencies,PotatoResearch.56,2013,31–50.

HeidmannT.,ToftengC.,AbrahamsenP.,PlauborgF.,HansenS.,BattilaniA.,CoutinhoJ.,DolezalF.,MazurczykW.,RuizJ.D.R.,TakacJ.andVacekJ.,CalibrationprocedureforapotatocropgrowthmodelusinginformationfromacrossEurope,

EcologicalModelling.Modell.211,2008,209–223.

HijmansR.J.,Theeffectofclimatechangeonglobalpotatoproduction,AmericanJournalofPotatoResearch.J.PotatoRes.80,2003,271–279.

HodgesT.,JohnsonS.L.andJohnsonB.S.,Amodularstructureforcropsimulationmodels:IimplementedintheSIMPOTATOmodel,AgronomyJournal.J.84,1992,911–915.

HodgesT.,Waterandnitrogenapplicationsforpotato:Ccommercialandexperimentalratescomparedtoasimulationmodel,JournalofSustainableAgriculture.SustainableAgric.13,1998,79–90.

HoogenboomG.,JonesJ.W.,WilkensP.W.,PorterC.H.,BooteK.J.,HuntL.A.,SinghU.,LizasoJ.L.,WhiteJ.W.,UryasevO.,RoyceF.S.,OgoshiR.,GijsmanA.J.,TsujiG.Y.andKooJ.,DecisionSupportSystemforAgrotechnologyTransfer(DSSAT)

Version4.5[CD-ROM],2012,UniversityofHawaii;Honolulu,Hawaii.

HuamanZ.andSchmiedicheP.,ThepotatogeneticresourcesheldintrustbytheInternationalPotatoCenter(CIP)inPeru,PotatoResearch.42,1999,413–426.

HuamanZ.andSpoonerD.M.,Reclassificationoflandracepopulationsofcultivatedpotatoes(Solanumsect.Petota),AmericanJournalofBotany.J.Bot.89,2002,947–965.

IPCC,ThePhysicalScienceBphysicalsciencebasis,In:StockerT.F.,QinD.,PlattnerG.-K.,TignorM.,AllenS.K.,BoschungJ.,NauelsA.,XiaY.,BexV.andP.MM.,(Eds.),ContributionofWorkingGroupItotheFifthAssessmentReportofthe

IntergovernmentalPanelonClimateChange,2013,CambridgeUniversityPressCambridge;UnitedKingdomandNewYork,NY,USA,p.1535.

InceogluO.,SallesJ.F.,vanOverbeekL.andvanElsasJ.D.,EffectsofPlantGenotypeandGrowthStageontheBetaproteobacterialCommunitiesAssociatedwithDifferentPotatoCultivarsinTwoFieldsAppliedandEnvironmentalMicrobiologyplantgenotypeand

growthstageonthebetaproteobacterialcommunitiesassociatedwithdifferentpotatocultivarsintwofields,Appl.Environ.Microbiol.76,2010,3675–3684.

JørgensenV.andEdlefsenO.,Vandforsyningensindflydelsepåudbytteogkvalitetafindustrikartofler,Planteavl91,1987,329–347.

JørgensenV.,Vandforsyningensindflydelsepåudbytteogkvalitetafkartofler,Planteavl88,1984,453–468.

JaggardK.W.,QiA.andOberE.S.,Possiblechangestoarablecropyieldsby2050,PhilosophicalTransactionsoftheRoyalSocietyB-BiologicalSciences.Trans.R.Soc.B—Biol.Sci.365,2010,2835–2851.

JonesJ.W.,HoogenboomG.,PorterC.H.,BooteK.J.,BatchelorW.D.,HuntL.A.,WilkensP.W.,SinghU.,GijsmanA.J.andRitchieJ.T.,TheDSSATcroppingsystemmodel,EuropeanJournalofAgronomy.J.Agron.18,2003,235–265.

JonesJ.W.,JianqiangH.,BooteK.J.,WilkensP.,PorterC.H.andHuZ.,EstimatingDSSATcroppingsystemcultivar-specificparametersusingbayesiantechniques,In:AhujaL.R.andLiwangM.,(Eds.),Methodsofintroducingsystemmodelsinto

agriculturalrIntroducingSystemModelsintoAgriculturalResearch,2011,AmericanSocietyofAgronomy,CropScienceSocietyofAmerica,SoilScienceSocietyofAmericaMadison;WI,USA.

KaminskiK.P.,KorupK.,NielsenK.L.,LiuF.,TopbjergH.B.,KirkH.G.andAndersenM.N.,Gas-exchange,wateruseefficiencyandyieldresponsesofelitepotato(SolanumtuberosumL.)cultivarstochangesinatmosphericcarbondioxide

concentration,temperatureandrelativehumidity,AgriculturalandForestMeteorology.ForestMeteorol.187,2014,36–45.

KhanM.,vanEckH.andStruikP.,Model-BasedEvaluationofMaturityTypeofPotatoUsingaDiverseSetofStandardCultivarsandaSegregatingDiploidPbasedevaluationofmaturitytypeofpotatousingadiversesetofstandardcultivarsandasegregatingdiploid

population,PotatoResearch.56,2013,127–146.

KimH.Y.,LiefferingM.,KobayashiK.,OkadaM.,MitchellM.W.andGumpertzM.,Effectsoffree-airCO22enrichmentandnitrogensupplyontheyieldoftemperatepaddyricecrops,FieldCropsResearch.83,2003,261–270.

KoomanP.L.andHaverkortA.J.,Modellingdevelopmentandgrowthofthepotatocropinfluencedbytemperatureanddaylenght:LINTUL-POTATO,In:HaverkortA.J.andMacKerronD.K.L.,(Eds.),Potatoecologyandmodellingcropsunderconditionslimiting

gEcologyandModellingCropsUnderConditionsLimitingGrowth,1995,KluwerAcademicPublisher;Wageningen,TheNetherlands,41–59.

Lenz-WiedemannV.I.S.,KlarC.W.andSchneiderK.,Developmentandtestofacropgrowthmodelforapplicationwithinaglobalchangedecisionsupportsystem,EcologicalModelling.Modell.221,2010,314–329.

LevyD.andVeilleuxR.E.,Adaptationofpotatotohightemperaturesandsalinity—AreviewAmericanJournalofPotatoResearchareview,Am.J.PotatoRes.84,2007,487–506.

LiuK.L.,GaoJ.L.,RenK.,ShengJ.H.,SuiQ.J.andJiangB.,Nitrogenabsorption,accumulationanddistributionofpotatounderdryfarming,PotatoChina17,2003a,321–325.

LiuK.L.,GaoJ.L.,SunH.Z.andShengJ.H.,Thedynamicofsinkstructureinpotato,ChinaPotato17,2003b,267–272.

LutaladioN.andCastaidiL.,Potato:Tthehiddentreasure,JournalofFoodCompositionandAnalysis.FoodCompos.Anal.22,2009.

LutzW.andKCS.,Dimensionsofglobalpopulationprojections:whatdoweknowaboutfuturepopulationtrendsandstructures?,PhilosophicalTransactionsoftheRoyalSocietyB-BiologicalSciences.Trans.R.Soc.B—Biol.Sci.365,2010,2779–2791.

MacKerronD.K.L.,Necessityandsufficiencyorthebalancebetweenaccuracyandpracticality,In:MacKerronD.K.L.andHaverkortA.J.,(Eds.),Decisionsupportsystemsinpotatoproduction:bringingmodelstopSupportSystemsinPotatoProduction:Bringing

ModelstoPractice,2004,Wageningenacademicpublishers;TheNetherlands,213–223.

MackerronD.K.L.andDaviesH.V.,Markersformaturityandsenescenceinthepotatocrop,PotatoResearch.29,1986,427–436.

Marshall,B.,VanDenBroek,B.J.,1995.Fieldexperimentsandanalysisofdatausedinthecasestudy.Modellingandparameterizationofthesoil-plant-atmospheresystem:Aacomparisonofpotatogrowthmodels,179–210.

MigliettaF.,MagliuloV.,BindiM.,CerioL.,VaccariF.,LoducaV.andPeressottiA.,FreeairCO(2)enrichmentofpotato(SolanumtuberosumL.):developmentgrowthandyield,GlobalChangeBiology.4,1998,163–172.

MonfredaC.,RamankuttyN.andFoleyJ.A.,Farmingtheplanet:2.Geographicdistributionofcropareas,yields,physiologicaltypes,andnetprimaryproductionintheyear2000,GlobalBiogeochemical.Cycles22,2008,19.

NelsonD.G.,Lightinterception,drymatterproductionandpartitioningofthepotatocropintropicaleInterception,DryMatterProductionandPartitioningofthePotatoCropinTropicalEnvironments,1987,DepartmentofAgriculture,UniversityCollegeof

Wales,1–135.

O’BrienP.J.,AllenE.J.andFirmanD.M.,Areviewofsomestudiesintotuberinitiationinpotato(Solanumtuberosum)crops,JournalofAgriculturalScience.Agric.Sci.130,1998,251–270.

O’learyG.,ChristyB.,NuttallJ.,HuthN.,CammaranoD.,StöckleC.,BassoB.,ShcherbakI.,FitzgeraldG.,LouQ.,Farre-CodinaI.,PaltaJ.andAssengS.,Responseofwheatgrowth,grainyieldandwaterusetoelevatedCO2underaFreeAirCO2

Enrichment(FACE)experimentandmodellinginasemi-aridenvironment,GlobalChangeBiology.2014.

PorterC.H.,JonesJ.W.,AdikuS.,GijsmanA.J.,GargiuloO.andNaabJ.B.,Modelingorganiccarbonandcarbon-mediatedsoilprocessesinDSSATv4.5(vol10pg247,2010),OperationalResearch.Res.14,2014,471–471.

PrasadR.,HochmuthG.J.andBooteK.J.,EstimationofNitrogenPoolsinIrrigatedPotatoProductiononSandySoilUsingtheModelSUBSTORPloSonitrogenpoolsinirrigatedpotatoproductiononSandysoilusingthemodelSUBSTOR,PLoS

One10,2015,e0117891–e0117891.

RCoreTeam,,R:AlanguageandenvironmentforstatisticalcLanguageandEnvironmentforStatisticalComputing,2015,RFoundationforStatisticalComputing;Vienna,Austriahttp://www.R-project.org/.

RahmanM.H.,PatwaryM.M.A.,BaruaH.,HossainM.andNaharS.,Evaluationofyieldandyieldcontributingcharactersofheattolerantpotato(SolanumtuberosumL.)genotypesinBangladesh,TheAgriculturists12,2014,50–55.

RaymundoR.,AssengS.,CammaranoD.andQuirozR.,Potatosweetpotato,andyammodelsforclimatechange:Aareview,FieldCropsResearch.166,2014,173–185.

RitchieJ.T.,GriffinT.S.,JohnsonB.S.,1995.SUBSTOR:Functionalmodelofpotatogrowth,developmentandyield.Modellingandparameterizationofthesoil-plant-atmospheresystem:Aacomparisonofpotatogrowthmodels,

401–435.

SanthoshN.,IndireshK.M.,PadmarajaS.R.,PuneethM.andThyagarajG.N.,Evaluationofearlymaturingadvancedpotato(SolanumtuberosumL.)hybridsforgrowthandyieldparameters,TrendsinBiosciencesBiosci.7,2014,2778–2782.

SaueT.andKadajaJ.,PossibleeffectsofclimatechangeonpotatocropsinEstonia,BorealEnvironmentResearch.Res.16,2011,203–217.

StockleC.O.,NelsonR.L.,HigginsS.,BrunnerJ.,GroveG.,BoydstonR.,WhitingM.andKrugerC.,AssessmentofclimatechangeimpactoneasternWashingtonagriculture,Climatic.Change102,2010,77–102.

SupitI.,vanDiepenC.,deWitA.,WolfJ.,KabatP.,BaruthB.andLudwigF.,AssessingclimatechangeeffectsonEuropeancropyieldsusingtheCropGrowthMonitoringSystemandaweathergenerator,AgriculturalandForestMeteorology.For.

Meteorol.164,2012,96–111.

ThieleG.,HareauG.,SuarezV.,ChujoyE.andBonierbaleMaldonadoR.,VarietalchangeinpotatoesindevelopingcountriesandthecontributionoftheInternationalPotatoCenter:1972-2007SocialSciencesWorkingPaper2007–2007,Soc.Sci.Work.

Pap.2007,p.46.

TravassoM.I.,CaldizD.O.andSaluzzoJ.A.,YieldpredictionusingtheSUBSTOR-potatomodelunderArgentinianconditions,PotatoResearch.39,1996,305–312.

TrebejoI.andMidmoreD.J.,Effectofwaterstressonpotatogrowth,yieldandwateruseinahotandacooltropicalclimate,TheJournalofAgriculturalScienceJ.Agric.Sci.114,1990,321–334.

TubielloF.N.,RosenzweigC.,GoldbergR.A.,JagtapS.andJonesJ.W.,EffectsofclimatechangeonUScropproduction:simulationresultsusingtwodifferentGCMscenarios.PartI:Wwheatpotato,maize,andcitrus,ClimateResearch.Res.20,2002,

259–270.

VanderZaagP.,DemaganteA.L.andEwingE.E.,Influenceofplantspacingonpotato(SolanumtuberosumL.)morphology:growthandyieldundertwocontrastingenvironments,PotatoResearch.33,1990,313–323.

VanuytrechtE.,RaesD.andWillemsP.,ConsideringsinkstrengthtomodelcropproductionunderelevatedatmosphericCO2,AgriculturalandForestMeteorology.For.Meteorol.151,2011,1753–1762.

WallachD.andGoffinetB.,Meansquareerrorofpredictioninmodelsforstudyingecologicalandagronomicssytems,Biomestrics43,1987,561–573.

WarrenM.F.,MonitoringandmodelingwaterandnitrogentransportinthevadosezoneofavegetablefModelingWaterandNitrogenTransportintheVadoseZoneofaVegetableFarmintheSuwanneeRiverBasin,2003,DepartmentofAgriculturalandBiological

Engineering,UniversityofFlorida;Gainesville,Florida,pp.1–233.

WishartJ.,GeorgeT.S.,BrownL.K.,RamsayG.,BradshawJ.E.,WhiteP.J.andGregoryP.J.,Measuringvariationinpotatorootsinbothfieldandglasshouse:thesearchforusefulyieldpredictorsandasimplescreenforroottraits,Plant

andSoil368,2013,231–249.

WolfJ.andVanOijenM.,ModelsimulationofeffectsofchangesinclimateandatmosphericCO2andO3ontuberyieldpotentialofpotato(cvBintje)intheEuropeanUnion,AgricultureEcosystems&Environment.Ecosyst.Environ.94,2003,141–157.

QueriesandAnswersQuery:Theauthornameshavebeentaggedasgivennamesandsurnames(surnamesarehighlightedintealcolor).Pleaseconfirmiftheyhavebeenidentifiedcorrectly

Answer:Everythingiscorrect.

Query:Pleasechecktheeditsmadeinaffiliations"a","g"and"j"andcorrectifnecessary.

Answer:Correct.

Query:Highlightsmustbeprovidedas3-5bulletpoints,eachbulletpointhavingamaximumof125characters.Pleaserephrasethefollowinghighlightsorprovidenewhighlights.

Answer:Afileisattached.

Query:Pleasecheckthekeywordsandcorrectifnecessary.

Answer:OK.

Query:Pleasecheckthehierarchyofthesectionheadings.

Answer:Correct.

Query:“YourarticleisregisteredasbelongingtotheSpecialIssue/Collectionentitled“Modelingcropgenotypes”.IfthisisNOTcorrectandyourarticleisaregularitemorbelongstoadifferentSpecialIssuepleasecontact

[email protected].”

Answer:belongstoSIiscorrect.

Query:Thecitation"Ritchie,1998"hasbeenchangedto"Ritchieetal.,1995"tomatchthereferencelist.Pleasecheckandcorrectifnecessary.

Answer:OK

Query:Refs."GodwinandSingh,(1998);Singhetal.(1998);NgandLoomis(1984)"arecitedinthetextbutnotprovidedinthereferencelist.Pleaseprovidetheminthereferencelistordeletethesecitationsfromthetext.

Answer:PleasekeeponlyNgandLoomis(1984):

WolfJ.,Effectsofnutrientsupply(NPK)onspringwheatresponsetoelevatedatmosphericCO2,PlantandSoil185,1996,113–123.

YangJ.M.,YangJ.Y.,LiuS.andHoogenboomG.,Anevaluationofthestatisticalmethodsfortestingtheperformanceofcropmodelswithobserveddata,AgriculturalSystems.Syst.127,2014,81–89.

ZotarelliL.,RensL.R.,CantliffeD.J.,StoffellaP.J.,GergelaD.andFourmanD.,NitrogenFertilizerRateandApplicationTimingforChippingPotatoCfertilizerrateandapplicationtimingforchippingpotatocultivarAtlantic,AgronomyJournal.J.106,2014,

2215–2226.

vanOijenM.andEwertF.,TheeffectsofclimaticvariationinEuropeontheyieldresponseofspringwheatcv.MinarettoelevatedCO2andO-32andO3:ananalysisofopen-topchamberexperimentsbymeansoftwocropgrowthsimulationmodels,

EuropeanJournalofAgronomy.J.Agron.10,1999,249–264.

Highlights

• TheSUBSTOR−—potatomodelwastestedwith87experimentsand204treatments,including32cultivarsandthreepotatospecies.

• Themodel-observationcomparisonshowedthattheSUBSTOR-potatomodelcaningeneralsimulatetuberyieldsacrosscontrastingenvironments.

• However,theSUBSTOR-potatomodelcannotaccuratelysimulatetuberyieldunderelevatedCO2concentrationsandhightemperatureswhichneedstobeimproved.

Ng,E.,Loomis,R.S.,1984.Simulationofgrowthandyieldofthepotatocrop.Centreforagriculturalpublishinganddocumentation,Wageningen,TheNetherlands.

Remove:Singetal.1998andGodwinandSingh1998

Query:Oneormoresponsornamesandthesponsorcountryidentifiermayhavebeeneditedtoastandardformatthatenablesbettersearchingandidentificationofyourarticle.Pleasecheckandcorrectifnecessary.

Answer:OK.

Query:PleasechecktheauthornamesinRef."Finnanetal.,2008"andcorrectifnecessary.

Answer:Pleasereplacewith:

Finnan,J.M.,Donnelly,A.,Jones,M.B.,Burke,J.I.,2008.Theeffectofelevatedlevelsofcarbondioxideonpotatocrops.JournalofCropImprovement13,91-111.

Query:Pleasesupplythenameofthecityofpublication.

Answer:sorry,can'tfindthecityname.

Query:PleasecheckthepresentationofalltheTablesandcorrectifnecessary.

Answer:Pleasekeeptables1aand1bseparate,alsokeepthefootnotesbeloweachcorrespondingtable(assubmittedintheoriginaldocument).

Query:PleasechecktheeditsmadeinTables1aandbandcorrectifnecessary.

Answer:Pleasekeeptables1aand1bseparate,alsokeepthefootnotesbeloweachcorrespondingtable(assubmittedintheoriginaldocument).

Performance of the SUBSTOR-potato model across contrasting ...pure.iiasa.ac.at/12914/1/Performance...

Documents

Transcript of Performance of the SUBSTOR-potato model across contrasting ...pure.iiasa.ac.at/12914/1/Performance...