Food and Agriculture

Organization of the
United Nations
GAEZ 2009
Global Agro-ecological Zones
I I A S A
International Institute
for Applied
Systems Analysis
GAEZ ver 3.0
Global Agro-ecological Zones
Model Documentation
D o c u m e n t a t i o n D o c u m e n t a t i o n

ii

GlobalAgroEcologicalZones(GAEZv3.0)
ModelDocumentation

GntherFischer,FreddyO.Nachtergaele,SylviaPrieler,EdmarTeixeira,
GzaTth,HarrijvanVelthuizen,LucVerelst,DavidWiberg

iii

iv

Disclaimer
ThedesignationsemployedandthepresentationofmaterialsinGAEZdonotimplytheexpressionof
any opinion whatsoever on the part of the International Institute for Applied Systems Analysis
(IIASA) or the Food and Agriculture Organization of the United Nations (FAO) concerning the legal
status of any country, territory, city or area or its authorities, or concerning the delimitation of its
frontiersorboundaries.
IIASAandFAO
IIASA and FAO are the sole and exclusive owners of all rights, titles and interests, including
trademarks,copyrights,tradenames,tradesecretsandotherintellectualpropertyrights,contained
inthedataandsoftwareofGAEZ.
Acknowledgementsandcitation
Fullacknowledgementandcitationinanymaterialsorpublicationsderivedinpartorinwholefrom
GAEZdataisrequiredandmustbecitedasfollows:
IIASA/FAO, 2012. Global Agroecological Zones (GAEZ v3.0). IIASA, Laxenburg, Austria and FAO,
Rome,Italy.

vi
TableofContents
TableofContents...................................................................................................................................vi
Tables...................................................................................................................................................viii
Figures....................................................................................................................................................ix
Appendixes..............................................................................................................................................x
Appendixfigures.................................................................................................................................x
Appendixtables..................................................................................................................................xi
Preface..................................................................................................................................................xii
Acronymsandabbreviations...............................................................................................................xiv
1 Introduction....................................................................................................................................1
1.1 TheAgroEcologicalZonesMethodology..............................................................................1
1.2 StructureandoverviewofGAEZprocedures........................................................................2
1.2.1 ModuleI:Agroclimaticdataanalysis...............................................................................3
1.2.2 ModuleII:Biomassandyieldcalculation..........................................................................3
1.2.3 ModuleIII:Agroclimaticconstraints................................................................................4
1.2.4 ModuleIV:Agroedaphicconstraints...............................................................................4
1.2.5 ModuleV:Integrationofclimaticandedaphicevaluation...............................................5
1.2.6 ModuleVI:ActualYieldandProduction..........................................................................5
1.2.7 ModuleVII:YieldandProductionGaps...........................................................................6
2 DescriptionofGAEZinputdatasets................................................................................................7
2.1 Climatedata...........................................................................................................................7
2.1.1 Observedclimate..............................................................................................................7
2.1.2 ClimateScenarios..............................................................................................................8
2.1.3 UseofclimatedatainGAEZ..............................................................................................8
2.2 Soildata.................................................................................................................................8
2.3 Elevationdataandderivedterrainslopeandaspectdata....................................................9
2.4 Landcoverdata...................................................................................................................11
2.5 Protectedareas...................................................................................................................12
2.5.1 WDPA2009.....................................................................................................................12
2.5.2 Natura2000....................................................................................................................12
2.6 Administrativeareas............................................................................................................14
3 ModuleI(Agroclimaticanalysis)..................................................................................................17
3.1 OverviewModuleI..............................................................................................................17
3.2 Preparationofclimaticvariables.........................................................................................17
3.2.1 Temporalinterpolation...................................................................................................20
3.3 ThermalRegimes.................................................................................................................21
3.3.1 Thermalclimates.............................................................................................................21
3.3.2 ThermalZones.................................................................................................................22
3.3.3 Temperaturegrowingperiods(LGPt).............................................................................23
3.3.4 Temperaturesums(Tsum)..............................................................................................24
3.3.5 Temperatureprofiles......................................................................................................24
3.3.6 Permafrostevaluation.....................................................................................................25
3.4 Soilmoistureregime............................................................................................................26
3.4.1 Soilmoisturebalance......................................................................................................26
3.4.2 Soilmoisturebalanceswithsoilmoistureconservation.................................................27
3.4.3 Lengthofgrowingperiod(LGP)......................................................................................31
3.4.4 Multiplecroppingzonesforrainfedcropproduction...................................................32
3.4.5 Equivalentlengthofthegrowingperiod........................................................................35
3.4.6 NetPrimaryProductivity(NPP).......................................................................................35
3.5 GridcellanalysisModuleI...................................................................................................36
3.6 DescriptionofModuleIoutputs.........................................................................................36
4 ModuleII(Biomasscalculation)....................................................................................................37
vii
4.1 Introduction.........................................................................................................................37
4.2 LandUtilizationTypes..........................................................................................................37
4.3 ThermalsuitabilityscreeningofLUTs..................................................................................38
4.4 Biomassandyieldcalculation..............................................................................................41
4.5 Waterlimitedbiomassproductionandyields....................................................................42
4.5.1 Cropwaterrequirement.................................................................................................42
4.5.2 Yieldreductionduetowaterdeficits..............................................................................43
4.5.3 AdjustmentofLAIandHiinperennialcrops..................................................................43
4.6 Cropcalendar......................................................................................................................45
4.7 CO2fertilizationeffectoncropyields.................................................................................45
4.8 GridcellanalysisModuleII..................................................................................................47
4.9 DescriptionofModuleIIoutputs........................................................................................47
5 ModuleIII(Agroclimaticyieldconstraints).................................................................................49
5.1 Introduction.........................................................................................................................49
5.2 Conceptualbasis..................................................................................................................50
5.3 Calculationprocedures........................................................................................................52
5.1 DescriptionofModuleIIIoutputs.......................................................................................53
6 ModuleIV(Agroedaphicsuitability)............................................................................................55
6.1 Introduction.........................................................................................................................55
6.1.1 Levelsofinputsandmanagement..................................................................................56
6.1.2 Watersupplysystems.....................................................................................................57
6.1.3 Soilsuitabilityassessmentprocedures...........................................................................58
6.2 Soilcharacteristics...............................................................................................................60
6.2.1 Soilprofileattributes.......................................................................................................60
6.2.2 Soildrainage....................................................................................................................62
6.2.3 Soilphases.......................................................................................................................64
6.3 Soilsuitabilityratings...........................................................................................................67
6.3.1 Soilprofileattributesratings...........................................................................................67
6.3.2 Soiltextureratings..........................................................................................................68
6.3.3 Soildrainageratings........................................................................................................69
6.3.4 Soilphasesratings...........................................................................................................69
6.4 Soilqualityandsoilsuitability.............................................................................................71
6.4.1 Soilquality.......................................................................................................................71
6.4.2 Soilsuitability..................................................................................................................75
6.5 Terrainsuitability.................................................................................................................77
6.6 Soilandterrainsuitabilityassessmentforirrigatedconditions..........................................80
6.6.1 Soilsuitabilityforirrigatedconditions............................................................................80
6.6.2 Terrainsuitabilityforirrigatedconditions......................................................................80
6.7 Soil and terrain suitability assessment for rainfed conditions under water conservation
regimes83
6.8 Fallowperiodrequirements................................................................................................83
6.9 Suitabilityofwatercollectingsites.....................................................................................84
6.1 DescriptionofModuleIVoutputs.......................................................................................85
7 ModuleV(Integrationofclimaticandedaphicevaluation).........................................................86
7.1 Introduction.........................................................................................................................86
7.2 DescriptionofModuleVoutputs........................................................................................86
7.2.1 MainprocessingstepsinModuleV................................................................................86
7.2.2 ModuleVoutputresults.................................................................................................87
8 ModuleVI(ActualYieldandProduction)......................................................................................89
8.1 Introduction.........................................................................................................................89
8.2 Downscalingofagriculturalstatisticstogridcells..............................................................89
8.2.1 Estimationofcultivatedlandshares...............................................................................90
8.2.2 Allocationofagriculturalstatisticstocultivatedland.....................................................91
viii
8.1 DescriptionofModuleVIoutputs.......................................................................................91
9 ModuleVII(YieldandProductionGaps).......................................................................................95
9.1 Introduction.........................................................................................................................95
9.2 Yieldandproductiongapsassessmentprocedures............................................................96
9.3 DescriptionofModuleVIIoutputs......................................................................................97
References............................................................................................................................................99

Tables
Table21ClimaticinputvariablesfortheGAEZassessment.............................................................7
Table31Classificationofthermalclimates.....................................................................................22
Table32Temperatureprofileclasses.............................................................................................25
Table33ClassificationofpermafrostareasusedintheGAEZassessment....................................26
Table34Waterbalanceparametersbytemperatureandcover...................................................30
Table35Moistureregimes.............................................................................................................31
Table36Delineationofmultiplecroppingzonesunderrainfedconditionsinthetropics............34
Table37Delineationofmultiplecroppingzonesunderrainfedconditionsin
subtropicsandtemperatezones......................................................................................34
Table41Parameterizationusedtocorrectharvestindex(Hi)andleafareaindex(LAI)for
suboptimumlengthoftheeffectivegrowthperiod(LPGeff)...........................................44
Table42CropspecificcoefficientsforthecalculationofCO2fertilizationeffect...........................45
Table43YieldadjustmentfactorsforCO2fertilizationeffectaccordingtolandsuitabilityratings46
Table 44 The CO2 concentrations (ppm) used to model fertilization effect in GAEZ according to
differentIPCCscenariosandtimepoints..........................................................................46
Table51Agroclimaticconstraintsforrainfedwinterwheat.........................................................52
Table61Watersupplysystem/cropassociations............................................................................57
Table62Landqualities.....................................................................................................................58
Table63Soilqualitiesandsoilattributes........................................................................................60
Table64Soiltextureseparates........................................................................................................61
Table65Soilphases.........................................................................................................................64
Table66Soilprofileattributeratingsforrainfedwheat................................................................67
Table67Soiltextureratingsforrainfedwheat..............................................................................68
Table68Soildrainageclasses..........................................................................................................69
Table69Soildrainageratingsforrainfedwheat............................................................................69
Table610Soilphaseratingsforrainfedwheat.................................................................................70
Table611Terrainsloperatingsforrainfedconditions(Fm<1300)..................................................79
Table612Soilandterrainevaluationratingsbywatersupplysystem..............................................81
Table613FluvisolandGleysolsuitabilityratings...............................................................................85
Table91CommoditiesusedintheGAEZyieldgapanalysis.............................................................97
Table92CommoditiesforDownscalingandyieldgapassessments................................................98
TableA31ContentoffixedoutputrecordsfromGAEZModuleI....................................................121
TableA32ModuleIoutputfiledescribingthermalconditionsduringthegrowingperiod............122
TableA33ModuleIoutputforsoilmoistureconditionsandlengthofgrowingperiodcharacteristics
............................................................................................................................................................123

ix
Figures
Figure11OverallstructureanddataintegrationofGAEZvs3.0(Module17)................................3
Figure21HarmonizedWorldSoildatabase(HWSD)........................................................................9
Figure22Medianterrainslopes.....................................................................................................10
Figure23Exampleofcalculatedterrainslopeclasses(percentofgridcellwithslope>16%)......10
Figure24Exampleoflandcoverdata:dominantlandcoverpatternintheHWSD.......................12
Figure25Protectedareas..13
Figure26GAOULcountryboundarieslayerwithGAEZregionalization14
Figure31InformationflowinModuleIoftheGAEZmodelframework........................................15
Figure32Thermalclimates.............................................................................................................20
Figure33ThermalZones.................................................................................................................21
Figure34Frostfreeperiod(LGP
t10
)..............................................................................................22
Figure35TemperaturesumsforthefrostfreeperiodwithTa>10
o
C.........................................22
Figure36Referencepermafrostzones...........................................................................................24
Figure37Schematicrepresentationofwaterbalancecalculations...............................................25
Figure38ReferenceLengthofGrowingPeriodZones....................................................................26
Figure39Referencelengthofgrowingperiod...............................................................................29
Figure310Multiplecroppingzonesforrainfedconditions............................................................31
Figure310Multiplecroppingzonesforirrigatedconditions...........................................................31
Figure41InformationflowofModuleII.........................................................................................35
Figure42Schematicrepresentationofthermalsuitabilityscreening............................................37
Figure43Schematicrepresentationofkcvaluesfordifferentcropdevelopmentstages.............40
Figure44YieldresponsetoelevatedambientCO
2
concentrations...............................................44
Figure51InformationflowsofModuleIII......................................................................................46
Figure52Agroclimaticallyattainableyieldofwheat.....................................................................50
Figure61Regionaldistributionofsoildatasources.......................................................................51
Figure62InformationflowinModuleIV........................................................................................52
Figure63Soilsuitabilityratingprocedures.....................................................................................55
Figure64Soiltextureclassification................................................................................................57
Figure65Rainfedsoilsuitability,lowinputlevel...........................................................................73
Figure66Rainfedsoilsuitability,highinputlevel...........................................................................73
Figure67Rainfedsoilandterrainsuitability,lowinputlevel........................................................75
Figure68Rainfedsoilandterrainsuitability,highinputlevel.......................................................75
Figure69Watercollectingsites......................................................................................................82
Figure71InformationflowofModuleV.........................................................................................84
Figure72MappingandtabulationinModuleVresults.................................................................85
Figure73Agroecologicalsuitabilityandproductivitypotentialofwheat.....................................85
Figure81InformationflowofModuleVI........................................................................................86
Figure82Sharesofcultivatedlandby5arcminutegridcell.........................................................88
Figure83Harvestedareaofrainfedmaizein2000.......................................................................89
Figure84Yieldofrainfedmaizein2000........................................................................................89
Figure85Productionofrainfedmaizein2000..............................................................................90
Figure91SchematicrepresentationofModuleVII........................................................................90
Figure92Yieldgapestimationprocedures....................................................................................91
Figure93Yieldgapratios................................................................................................................92
x
Appendixes

Appendix21 CountryList(GAUL)andregionalizations.............................................................117
Appendix31 CalculationofReferenceEvapotranspiration.......................................................118
Appendix32 OutputsModuleI.................................................................................................121
Appendix33 SubroutinedescriptionsofModuleI....................................................................124
Appendix34ExampleofModuleIoutputatgridcelllevel.......................................................126
Appendix41 Cropsandlandutilizationtypes(LUTs)................................................................127
Appendix42 Parametersforcalculationofwaterlimitedyields..............................................136
Appendix43 TemperatureProfileRequirements......................................................................138
Appendix44 Cropvernalizationrequirements..........................................................................139
Appendix45 Biomassandyieldcalculation...............................................................................141
Appendix46 Biomassandyieldparameters.............................................................................143
Appendix47 OutputofModuleII..............................................................................................144
Appendix48 SubroutinedescriptionsofModuleII..................................................................146
Appendix49ExampleofModuleIIoutputatgridcelllevel......................................................152
Appendix51 Agroclimaticconstraintsfor.................................................................................152
Appendix52 OutputsModuleIII...............................................................................................152
Appendix53 SubroutinedescriptionsofModuleIII..................................................................153
Appendix61Soildrainageclasses..............................................................................................154
Appendix62 Soilprofileattributesuitabilityratings.................................................................154
Appendix63 Soiltexturesuitabilityratings...............................................................................154
Appendix64 Soildrainagesuitabilityratings............................................................................154
Appendix65 Soilphasesuitabilityratings.................................................................................154
Appendix66 Terrainslopesuitabilityratings............................................................................154
Appendix67 Fallowperiodrequirements.................................................................................154
Appendix68 Suitabilityofwatercollectingsites......................................................................155
Appendix69 OutputsofModuleIV...........................................................................................159
Appendix610SubroutinedescriptionsofModuleIV..................................................................161
Appendix71 OutputsofModuleV............................................................................................164
Appendix72SubroutinedescriptionsofModuleV...................................................................165
Appendix73Cropsummarytabledescription...........................................................................167
Appendix81 Estimationofsharesofcultivatedlandbygridcell.............................................168
Appendix82 Estimationofareayieldandproductionofcrops................................................169
Appendix83 OutputsofModuleVI...........................................................................................169
Appendix9 Globalterrainslopeandaspectdatadocumentation..........................................175

Appendixfigures
FigureA31StructureofsubroutinesandfunctionsinModuleI......................................................118
FigureA41DiagramofthesubroutinesandfunctionsofGAEZModuleII......................................141
FigureA51DiagramofthesubroutinesandfunctionsofGAEZModuleIII.....................................147
FigureA61DiagramofthesubroutinesandfunctionsofGAEZModuleIV.....................................155
FigureA71DiagramofthesubroutinesandfunctionsofGAEZModuleV......................................159

xi
Appendixtables
TableA31ModuleIoutputfileforthermalconditionsextractedovertheentireyear........................115
TableA32ModuleIoutputfiledescribingthermalconditionsduringthegrowingperiod...................116
TableA33ModuleIoutputforsoilmoistureconditionsandgrowingperiodlengthcharacteristics....117
TableA34SubroutinesandfunctionsofModuleI................................................................................118
TableA35FortransourcefilesforModuleIandincludedheaderfiles,subroutinesandfunctions......119
TableA41Cropgroups.........................................................................................................................121
TableA42Crops...................................................................................................................................121
TableA43Croptypes...........................................................................................................................122
TableA44Crop/LUTs............................................................................................................................123
TableA45Crops/commodities.............................................................................................................129
TableA46Parametersofbiomassyieldcalculations............................................................................130
TableA47Parameterizationforthecalculationoftherateofvernalization........................................133
TableA48ContentoffixedoutputrecordsfromGAEZModuleII........................................................138
TableA49InformationcontainedineachpixeldatarecordofModuleII.............................................139
TableA410SubroutinesandfunctionsofModuleII...............................................................................142
TableA411HeaderandfortransourcefilessubroutinesandfunctionsforGAEZModuleII145
TableA51SubroutinesandfunctionsofModuleIII..............................................................................147
TableA61ContentofoutputfilefromGAEZModuleIV......................................................................153
TableA62SubroutinesandfunctionsofGAEZModuleIV....................................................................156
TableA71InformationcontainedineachpixeldatarecordofModuleV.............................................158
TableA72SubroutinesandfunctionsofModuleV..............................................................................160
TableA73FortransourcefilesforModuleVandincludedheaderfiles,subroutinesandfunctions.....160
TableA111DescriptionoffilenamesoftheIIASALUCGlobalTerrainSlopesandAspectDatabase......168

xii
Preface
The International Institute for Applied Systems Analysis (IIASA) and the Food and Agriculture
Organization of the United Nations (FAO) have been continuously developing the AgroEcological
Zones (AEZ) methodology over the past 30 years for assessing agricultural resources and potential.
Rapid developments in information technology have produced increasingly detailed and manifold
globaldatabases,whichmadethefirstglobalAEZassessmentpossiblein2000.SincethenglobalAEZ
assessmentshavebeenperformedeveryfewyears,withthedatabeingpublishedonCDorDVD.
In the general context of preparing a global overview of prevailing and future conditions affecting
agricultural development and food security, the enlarged knowledge base on global agroecological
zoning (GAEZ), in particular the expanded number of crops and management techniques evaluated,
andnewdatasetsavailableforuseinthecropevaluation,asignificantupdateofGAEZ(Fischeretal.,
2002) is timely. This FAO sponsored project, here referred to as GAEZ v3.0, aims to include practical
applications such as a significantly updated version, including expanded crop coverage and dryland
managementtechniques.
In addition to the updating and expansion of GAEZ results, a novel methodology for spatially
downscaling of agricultural production statistics has been applied to produce a global gridded
inventoryofyear2000agriculturalyieldsandproduction.Thelatterinformation,inconjunctionwith
attainableyieldpotentialsfromGAEZv3.0,isusedtoquantifyyieldandproductiongapsworldwide
andatnationalandsubnationallevels.
GAEZv3.0includesthefollowingrevisionsandupdatesofprocedures:
Substantialupdatingandtuningofcroppotentialsimulationprocedures
Simulated crops now totaling some 280 cropLUTs combinations including all globally
important food, feed and fiber crops as well as number of important bioenergy
feedstocks.
Detailed water supply types including rainfed agriculture, rainfed agriculture with water
conservationandgravity,sprinkleranddripirrigationsystems.
Edaphicsuitabilityevaluationprocedures
Proceduresforspatiallydownscalingofagriculturalproductionstatistics.
Proceduresforestablishingyieldandproductiongapsformajorcropcommodities
Newandupdateddatabases:
Observedclimate:UpdatedCRUandGPCCclimatedata
Climate scenarios: Twelve GCMclimate IPCC_AR4 scenario combinations for the 2020s,
2050sand2080s
Soils:AnewspeciallydevelopedHarmonizedWorldSoilDatabase
Terrain:ElevationdataandderivedslopeandaspectdataderivedfromSRTM
Irrigatedareas:DigitalGlobalMapofIrrigatedAreas(GMIA)version4.01
Landcoverdata:Newdatabaseformajorlanduselandcovercategories
Protectedareas:WorldDatabaseofProtectedAreasAnnualRelease2009
Populationdensityinventoryforyear2000(FAOSDRN)
Administrativeareas:GlobalAdministrativeUnitLayers(GAUL)of2009.
Statisticaldata:
Forestresourceassessments(FRA2000,FRA2005,FRA2010)
FAOSTAT
AQUASTAT
UNPopulationStatistics

xiii
WitheachupdateofGAEZ,theissuesaddressed,thesizeofthedatabaseandthenumberofresults
havemultiplied.Anewsystem(GAEZv3.0DataPortal)wascreatedtomakethedataaccessibletoa
varietyofusers.
ThisreportonmodeldocumentationprovidesinformationonthestructureofGAEZmethodologyby
describing the conceptual framework by individual assessment modules in nine chapters. Relevant
data input parameters are provided in a voluminous appendix in printed or digital formats (CD
ROM).
ThisdocumentationisrecommendedforGAEZmodelersandusersofitsresultssuchasresearchers,
nationalandinternationalresearchinstitutesandmultilateralorganizationsdealingwithsustainable
utilizationoflandresources,agriculturaldevelopmentandfoodsecurity.

xiv
Acronymsandabbreviations
AEZ AgroecologicalZones
AR4 IPCCfourthAssessmentReport
AT2015/30 WorldAgricultureTowards2015/2030
C2A2 IPCCSRESA2ScenariofromtheCanadianCentreforClimateModellingandAnalysis'
SecondGenerationCoupledGlobalClimateModel(fullscenarioname:CCCma
CGCM2)
C2B2 IPCCSRESB2ScenariofromtheCanadianCentreforClimateModellingandAnalysis'
SecondGenerationCoupledGlobalClimateModel(fullscenarioname:CCCma
CGCM2B2)
CGCM2 CanadianGeneralCirculationModel
CGIAR ConsultativeGrouponInternationalAgriculturalResearch
CORINE CoordinateInformationontheEuropeanEnvironmemnt
CROPWAT Computerizedirrigationschedulingprogramme
CRU ClimateResearchUnitofEastAngliaUniversity
CSA2 AustralianCommonwealthScientificandResearchOrganizationMark2Model(full
name:CSIROMk2A2)
CSB1 AustralianCommonwealthScientificandResearchOrganizationMark2Model(full
scenarioname:CSIROMk2B1)
CSB2 AustralianCommonwealthScientificandResearchOrganizationMark2Model(full
name:CSIROMk2B2)
CSIRO Commonwealthscientificandindustrialresearchorganization,Australia
DSMW DigitalSoilMapoftheWorld
ECMWF EuropeanCentreforMediumRangeWeatherForecasts
EDC ErosDataCentre
EHA2 MaxPlanckInstitutfrMeterologieGCMmodel(fullscenarioname:MPIECHAM4
A2)
EHB2 MaxPlanckInstitutfrMeterologieGCMmodel(fullscenarioname:MPIECHAM4
B2)
EROS EartghResourcesObservationandScienceCenter
ESBN EuropeanSoilBureauNetwork
FACE Freeaircarbondioxideenrichment
FAO FoodandAgricultureOrganizationoftheUnitedNations
FAOSTAT FAOstatistics
fc0 Totalconstraint
fc1 Yieldconstraintfactorduetotemperatureconstraints

fc2 Yieldconstraintfactorduetomoisureconstraints

fc3 Yieldconstraintfactorduetoagroclimaticconstraints

fc4 Yieldconstraintfactorduetosoil anu teiiain constiaints

Fm Fournierindex
FRA2000 Globalforestresourcesassessment2000
FRA2005 Globalforestresourcesassessment2005
GAEZvs3.0 Global Agroecological Zones version 3.0 (Data access facility, research report and
documentation)
GAEZ2000 GlobalAgroecologicalZonesversion1.0(WebsiteandCDROM2000)
GAEZ2002 GlobalAgroecologicalZonesversion2.0(ResearchreportandCDROM2002)
GAUL GlobalAdministrativeUnitLayers
GCM Generalcirculationmodel
GLC2000 Globallandcover2000
GLCCD Globallandcovercharacteristicsdatabase
GMIA Globalmapofirrigatedaereas
GPCC GlobalPrecipitationClimatologyCentre
xv
GTOPO30 Global30arcsecondelevation
H3A1 UKMetOfficeHadleyCentrecoupledmodel(fullscenarioname:HadleyCM3A1FI)
H3A2 UKMetOfficeHadleyCentrecoupledmodel(fullscenarioname:HadleyCM3A2)
H3B1 UKMetOfficeHadleyCentrecoupledmodel(fullscenarioname:HadleyCM3B1)
H3B2 UKMetOfficeHadleyCentrecoupledmodel(fullscenarioname:HadleyCM3B2)
HadCM3 Headleycentre,UKMeteorologicalOffice(climatemodel3)
Hi Harvestindex
HWSD Harmonizedworldsoildatabase
IFPRI InternationalFoodPolicyResearchInstitute
IIASA InternationalInstituteforAppliedSystemsAnalysis
IPCC IntergovernamentalPanelonClimateChange
ISRIC InternationalsoilResearchandInformationCentreworldsoilinformation
ISSCAS InstituteofSoilScience,ChineseAcademyofScience
IUCN InternationalUnionforConservationofNature
JRC JointResearchcentreoftheEuropeanCommission
LAI Leafareasindex
LGP Lengthofgrowingperiod
LGPeq Equivalentgrowingperiod
LGPt Temperaturegrowingperiod
LUC LandUseChangeandAgriculturalprogramofIIASA
LUT Landutilizationtypes
mS Marginallysuitableland
MS Moderatelysuitableland
NATURA2000 EuropeanUnionNetworkofNatureProtectionAreas
NS Notsuitableland
PET Potentialevapotranspiration
S Suitableland
SOTER Soilandterraindatabase
SOTWIS Soterandwisederivedsoilpropertiesestimates
SQ1 Soilnutrientavailability
SQ2 Soilnutrientretentioncapacity
SQ3 Rootingconditions
SQ4 Oxygenavailabilitytoroots
SQ5 Excesssalts
SQ6 Toxicity
SQ7 Workability
SRES Specialreportonemissionscenarios
SRTM Shuttleradartopographymission
Tsumt Accumulatedtemperaturesforperiodwhentemperaturesexceedt
o
C
Unesco UnitedNationsEducational,ScientificandCulturalOrganization.
USGS UnitedStatesGeologicalSurvey
VASclimO Variabilityanalysisofsurfaceclimateobservations
vmS Marginallysuitableland
VS Verysuitableland
WCMC Worldconservationmonitoringcentre
WDPA Worlddatabaseofprotectedareas
WISE WorldInventoryofsoilemissionpotentials

1

1 Introduction
1.1 TheAgroEcologicalZonesMethodology
The quality and availability of land and water resources, together with important socioeconomic
and institutional factors, is essential for food security. Crop cultivation potential describes the
agronomically possible upper limit for the production of individual crops under given agroclimatic,
soilandterrainconditionsforaspecificlevelofagriculturalinputsandmanagementconditions.The
AgroEcologicalZones(AEZ)approachisbasedonprinciplesoflandevaluation(FAO1976,1984and
2007). The AEZ concept was originally developed by the Food and Agriculture organization of the
United Nations (FAO). FAO, with the collaboration of IIASA has over time, further developed and
appliedtheAEZmethodology,supportingdatabasesandsoftwarepackages.ThecurrentGlobalAEZ
(GAEZ v 3.0) provides a major update of data and extension of the methodology compared to the
release of GAEZ in 2002 (Fischer, et. al., 2002). GAEZ v 3.0 incorporates two important new global
datasetsonActualYieldandProductionandYieldandProductionGapsbetweenpotentialsand
actualyieldandproduction.
Georeferenced global climate, soil and terrain data are combined into a land resources database,
commonly assembled on the basis of global grids, typically at 5 arcminute and 30 arcsecond
resolutions. Climatic data comprises precipitation, temperature, wind speed, sunshine hours and
relativehumidity,whichareusedtocompileagronomicallymeaningfulclimateresourcesinventories
includingquantifiedthermalandmoistureregimesinspaceandtime.
Matching procedures to identify cropspecific limitations of prevailing climate, soil and terrain
resources and evaluation with simple and robust crop models, under assumed levels of inputs and
management conditions, provides maximum potential and agronomically attainable crop yields for
basic land resources units under different agricultural production systems defined by water supply
systems and levels of inputs and management circumstances. These generic production systems
usedintheanalysisarereferredtoasLandUtilizationTypes(LUT).
AttributesspecifictoeachparticularLUTincludecropinformationsuchascropparameters(harvest
index, maximum leaf area index, maximum rate of photosynthesis, etc.), cultivation practices and
input requirements, and utilization of main produce, crop residues and byproducts. For each LUT,
theGAEZproceduresareappliedforrainfedconditions,forrainfedconditionswithspecificwater
conservation practices, and for irrigated conditions. Calculations are done for different levels of
inputsandmanagementassumptions.
Severalcalculationstepsareappliedatthegridcellleveltodeterminepotentialyieldsforindividual
crop/LUTcombinations.Growthrequirementsofthecropspeciesarematchedagainstadetailedset
of agroclimatic and edaphic land characteristics derived from the land resources database.
Estimation of crop evapotranspiration and cropspecific soil moisture balance calculations are used
fordetailedassessmentsofcrop/LUTspecificsuitabilityandproductivity.
Global change processes raise new estimation problems challenging the conventional statistical
methods. These methods are based on the ability to obtain observations from unknown true
probability distributions, whereas the new problems require recovering information from only
partiallyobservableorevenunobservablevariables.Forinstance,aggregatedataexistatglobaland
nationallevelregardingagriculturalproduction.
Sequential rebalancing procedures that were developed in this project rely on appropriate
optimizationprinciples(Fischeretal.,2006a,2006b),e.g.,crossentropymaximization,andcombine
the available samples of real observations in the locations with other prior hard (statistics,
accountingidentities)andsoft(expertopinion,scenarios)data.
2
Actual yields and production are derived through downscaling year 2000 and 2005 agricultural
statistics of main food and fiber crops for all rainfed and irrigated cultivated areas. Results are
presented as (i) Crop production value, and (ii) crop area, production and yields for 23 major
commodities.
The comparison between simulated potential yields and production with observed yield and
production of crops currently grown, provides relevant yield and production gap information. For
the 23 main commodities, yield and production gaps are estimated by comparing potential
attainableyieldswithactualachievedyieldsandproduction(year2000and2005).
GAEZ generates large databases of (i) natural resources endowments relevant for agricultural uses
and(ii) spatially detailed resultsof individualLUT assessments intermsofsuitabilityandattainable
yields,(iii)spatiallydetailedresultsofestimate/actualyieldsofmainfoodandfibercommoditiesfor
all rainfed and irrigated cultivated areas, and (iv) spatially detailed yield and production gaps also
formainfoodandfibercommodities.
These databases provide the agronomic backbone for various applications including the
quantification of land productivity. Results are commonly aggregated for current major land
use/cover patterns and by administrative units, land protection status, or broad classes reflecting
infrastructureavailabilityandmarketaccessconditions.
1.2 StructureandoverviewofGAEZprocedures
The suitability of land for the cultivation of a given crop/LUT depends on crop requirements as
compared to the prevailing agroclimatic and agroedaphic conditions. GAEZ combines these two
components by successively modifying gridcell specific agroclimatic suitabilities according to
edaphicsuitabilitiesoflocationspecificsoilandterraincharacteristics.Thestructureallowsstepwise
reviewofresults.
Calculation procedures for establishing crop suitability estimates include five main steps of data
processing,namely:
(i) ModuleI:Climatedataanalysisandcompilationofgeneralagroclimaticindicators
(ii) Module II: Cropspecific agroclimatic assessment and waterlimited biomass/yield
calculation
(iii) ModuleIII:Yieldreductionduetoagroclimaticconstraints
(iv) ModuleIV:Edaphicassessmentandyieldreductionduetosoilandterrainlimitations
(v) ModuleV:IntegrationofresultsfromModulesIIVintocropspecificgridcelldatabases.
Twomainactivitieswereinvolvedinobtaininggridcelllevelarea,yieldandproductionofprevailing
maincrops,namely:
(vi) ModuleVI:Estimationofsharesofrainfedorirrigatedcultivatedlandby5gridcell,and
estimation of area, yield and production of the main crops in the rainfed and irrigated
cultivatedlandshares
Global inventories of yield gaps were created through comparison of potential rainfed yields with
yieldsofdownscaledstatisticalproduction.Theactivitiesinclude:
(vii) ModuleVII:Quantification of yield gaps between potential attainable crop yields and
downscaledcurrentcropyieldstatisticsoftheyear2000and2005;
TheoverallGAEZmodelstructureanddataintegrationareschematicallyshowninFigure11
3
Climate
resources
Land utilization
types
Module V
Crop
potentials
Module VI
Current crop
production
Module VII
Yield and production
gaps
Crop statistics
Suitable areas
and potential
crop yields
Module II
Biomass and yield
Module III
Agro-climatic
constraints
Module IV
Agro-edaphic
constraints
Module I
Agro-climatic
data analysis
Spatial data sets
soil and terrain resources, land cover, protected areas,
irrigated areas, population density, livestock density,
distance to market.
Agro-climatic
attainable
yields
Climatic
crop yields
crop constraints
crop calendars
Crop yield and
production gaps
Harvested area
crop yield and
production
Land
resources

Figure11OverallstructureanddataintegrationofGAEZv3.0(ModuleIVII)
1.2.1 ModuleI:Agroclimaticdataanalysis
Climatedataanalysisandcompilationofgeneralagroclimaticindicators
ModuleIcalculatesandstoresclimaterelatedvariablesandindicatorsforeachgridcell.Themodule
processes spatialgrids of historical,base lineand projected future climateto create layers of agro
climatic indicators relevant to plant production. First, available monthly climate data are read and
converted to variables required for subsequent calculations. Temporal interpolations are used to
transformmonthlydatatodailyestimatesrequiredforcharacterizationofthermalandsoilmoisture
regimes.Thelatterincludescalculationofreferencepotentialandactualevapotranspirationthrough
dailysoilwaterbalances.
Thermal regime characterization generated in Module I includes thermal growing periods,
accumulated temperature sums (for average daily temperature respectively above 0C, 5C and
10C),delineationofpermafrostzonesandquantificationofannualtemperatureprofiles.Soilwater
balance calculations (Section 3.4.1) determine potential and actual evapotranspiration for a
reference crop, length of growing period (LGP, days) including characterization of LGP quality,
dormancy periodsand cold brakes, and begin and end datesof one ormore LGPs.Based on a sub
setoftheseindicators,amultiplecroppingzonesclassificationisproducedforrainfedandirrigated
conditions.
1.2.2 ModuleII:Biomassandyieldcalculation
Cropspecificagroclimaticassessmentandpotentialwaterlimitedbiomass/yieldcalculation
In Module II, all land utilization types (LUT) are assessed for waterlimited biomass and yields,
currently280,cropandpasture,LUTsforeachoftheassumedinputlevels(Appendix41).TheLUT
conceptcharacterizesarangeofsubtypeswithinaplantspecies,includingdifferencesincropcycle
4
length(i.e.daysfromsowingtoharvest),growthanddevelopmentparameters.Subtypesdifferwith
assumed level of inputs. For instance, at low input level traditional crop varieties are considered,
whichmayhavedifferentqualitiesthatarepreferredbuthavelowyieldefficiencies(harvestindex)
and because of management limitations are grown in relatively irregular stands with inferior leaf
areaindex.Incontrast,withhighinputlevelhighyieldingvarietiesaredeployedwithadvancedfield
managementandmachineryprovidingoptimumplantdensitieswithhighleafareaindex.
Module II first calculates maximum attainable biomass and yield as determined by radiation and
temperature regimes, followed by the computation of respective rainfed crop water balances and
theestablishmentofoptimumcropcalendarsforeachoftheseconditions.Cropwaterbalancesare
usedtoestimate actual cropevapotranspiration, accumulated cropwaterdeficitduringthegrowth
cycle (respectively irrigation water requirements for irrigated conditions), and attainable water
limited biomass and yields for rainfed conditions. First, a window of time is determined when
conditions permit LUT cultivation (e.g. prevailing LGP in each grid cell). The growth of each LUT is
tested for the days during the permissible window of time with separate analysis for irrigated and
rainfed conditions. The growing dates and cycle length producing the highest (waterlimited or
irrigated)yielddefinetheoptimumcropcalendarofeachLUTineachgridcell.
DuetothedetailedcalculationsforaratherlargenumberofLUTs,ModuleIIrequiresaconsiderable
amount of computer time for its processing and is the most CPUdemanding component in GAEZ.
Results of Module II include LUTspecific temperature/radiation defined maximum yields, yield
reductionfactorsaccountingforsuboptimumthermalconditions,foryieldimpactsduetosoilwater
deficits, estimated amounts of soil water deficit, potential and actual LUT evapotranspiration,
accumulatedtemperaturesumsduringeachLUTcropcycle,andoptimumcropcalendars.
1.2.3 ModuleIII:Agroclimaticconstraints
Yieldreductionduetoagroclimaticconstraints
Module III computes for each grid cell specific multipliers, which are used to reduce yields for
various agroclimatic constraints as defined in the AEZ methodology. This step is carried out in a
separatemoduletomakeexplicittheeffectoflimitationsduetosoilworkability,pestanddiseases,
and other constraints and to permit timeeffective reprocessing in case new or additional
informationisavailable.Fivegroupsofagroclimaticconstraintsareconsidered,including:
a) Yieldadjustmentduetoyeartoyearvariabilityofsoilmoisturesupply;thisfactorisappliedto
adjustyieldscalculatedforaverageclimaticconditions
b) Yieldlossesduetotheeffectofpests,diseasesandweedconstraintsoncropgrowth
c) Yield losses due to water stress, pest and diseases constraints on yield components and yield
formationofproduce(e.g.,affectingqualityofproduce)
d) Yield losses due to soil workability constraints (e.g., excessive wetness causing difficulties for
harvestingandhandlingofproduce)
e) Yieldlossesduetooccurrenceofearlyorlatefrosts.
Agroclimatic constraints are expressed as yield reduction factors according to the different
constraintsandtheirseverityforeachcropandbylevelofinputs.Duetopaucityofempiricaldata,
estimatesofconstraintratingshavebeenobtainedthroughexpertopinion.
The results of Module III update for each grid cell the output file of Module II by filling in the
respective LUT agroclimatic constraints yield reduction factors. At this stage, the results of agro
climaticsuitabilitiescanbemappedforspatialverificationandfurtheruseinapplications.
1.2.4 ModuleIV:Agroedaphicconstraints
Yieldreductionduetosoilandterrainlimitations
This module evaluates cropspecific yield reduction due to limitations imposed by soil and terrain
conditions. Soil suitability is determined on the basis of the soil attribute data contained in the
5
Harmonized World Soil Database (FAO/IIASA/ISRIC/ISSCAS/JRC 2009). Soil nutrient availability, soil
nutrientretentioncapacity,soilrootingconditions,soiloxygenavailability,soiltoxicities,soilsalinity
and sodicity conditions and soil management constraints are estimated on crop by crop basis and
arecombinedinacropandinputspecificsuitabilityrating.
The soil evaluation algorithm assesses for soil types and slope classes the match between crop soil
requirementsandtherespectivesoilqualitiesasderivedfromsoilattributesoftheHWSD.Thereby
theratingproceduresresultinaquantificationofsuitabilityforallcombinationsofcroptypes,input
level,soiltypesandslopeclasses.
1.2.5 ModuleV:Integrationofclimaticandedaphicevaluation
Module V executes the final step in the GAEZ crop suitability and land productivity assessment. It
reads the LUT specific results of the agroclimatic evaluation for biomass and yield calculated in
Module II/III for different soil classes and it uses the edaphic rating produced for each soil/slope
combination in Module IV. The inventories of soil resources and terrainslope conditions are
integratedbyrankingallsoiltypesineachsoilmapunitwithregardtooccurrenceindifferentslope
classes. Considering simultaneously the slope class distribution of all grid cells belonging to a
particularsoilmapunitresultsinanoverallconsistentdistributionofsoilterrainslopecombinations
by individual soil association map units and 30 arcsec grid cells, soil and slope rules are applied
separatelyforrainfedandirrigatedconditions.
The algorithm in Module V steps through the grid cells of the spatial soil association layer of the
Harmonized World Soil Database and determines for each grid cell the respective makeup of land
units in terms of soil types and slope classes. Each of these component land units is separately
assigned the appropriate suitability and yield values and results are accumulated for all elements.
Processing of soil and slope distribution information takes place at 30 arcsecond grid cells. One
hundred of these produce the edaphic characterization at 5 arcminutes, the resolution used for
providingGAEZresults.
Cropping activities are the most critical in causing topsoil erosion, because of their particular cover
dynamicsandmanagement.TheterrainslopesuitabilityratingusedintheGAEZstudyaccountsfor
thefactorsthat influenceproductionsustainability andisachievedthrough: (i)definingpermissible
slope ranges for cultivation of various crop/LUTs and setting maximum slope limits; (ii) for slopes
withinthepermissiblelimits,accountingforlikelyyieldreductionduetolossoffertilizerandtopsoil;
and (iii) distinguishing among a range of farming practices, from manual cultivation to fully
mechanizedcultivation.Inaddition,theterrainslopesuitabilityratingisvariedaccordingtoamount
anddistributionofrainfall,whichisquantifiedinGAEZbymeansoftheFournierindex.
Application of the procedures in the modules described above result in an expected yield and
suitability distribution regarding rainfed and irrigation conditions for each 5minute gridcell and
eachcrop/LUT.Landsuitabilityisdescribedinfiveclasses:verysuitable(VS),suitable(S),moderately
suitable (MS), marginally suitable (mS), and not suitable (NS) for each LUT. Large databases are
created, which are used to derive additional characterization and aggregations. Examples include
calculationoflandwithcultivationpotential,tabulationofresultsbyecosystemtype,quantification
of climatic production risks by using historical time series of suitability results, impact of climate
change on crop production potentials, and irrigation water requirements for current and future
climates.
1.2.6 ModuleVI:ActualYieldandProduction
Global change processes raise new estimation problems challenging the conventional statistical
methods, which are based on the ability to obtain observations from unknown true probability
distributions. In contrast, problems such as downscaling of production require recovering
information from only partially observable or even unobservable variables. For instance, aggregate
data exist at global and national level regarding agricultural production and harvest areas.
Downscaling methods in this case should achieve plausible estimation of global distributions,
6
consistentwithlocaldataobtainedfromremotesensingandavailableaggregatestatistics,byusing
allavailableevidence.
This module estimates actual yields and production from downscaling year 2000 statistics of main
foodandfibercrops(statisticsderivedmainlyfromFAOSTATandtheFAOstudyAT2015/30).Results
are presented as (i) crop production value, and (ii) crop area, production and yields for 23 major
commodities.
Twomainactivitieswereinvolvedinobtaininggridcelllevelarea,yieldandproductionofprevailing
maincrops:
(i) Estimationofsharesofrainfedorirrigatedcultivatedlandby5gridcell,and
(ii) estimation of area, yield and production of the main crops in the rainfed and irrigated
cultivatedlandshares
Estimationofcultivatedlandshares
Landcoverinterpretationsschemesweredevisedthatallowaquantificationofeach5arcmin.grid
cell into seven main land use cover shares. Shares of cultivated land, subdivided into rainfed and
irrigatedland,wereusedforallocatingrainfedandirrigatedcropproductionstatistics.
Allocationofagriculturalstatisticstocultivatedland
Agricultural production statistics are available at national scale from FAO. Various layers of spatial
informationareusedtocalculateaninitialestimateoflocationspecificcropwiseproductionpriors.
The priors are adjusted in an iterative downscaling procedure to ensure that crop areas and
production are consistent with aggregate statistical data, are allocated to the available cultivated
landandreflectavailableancillarydata,e.g.,selectedcropareadistributiondata(Montfredaetal.,
2008)andagronomicsuitabilityofcropsestimatedinAEZ.
1.2.7 ModuleVII:YieldandProductionGaps
Yield gaps and production gaps have been estimated by comparing potential attainable yields and
production (estimated in GAEZ v3.0) and actual yields and production from downscaling year 2000
and2005statisticsofmainfoodandfibercrops(statisticsderivedmainlyfromFAOSTATandtheFAO
studyAT2015/30).
For main commodities, (see list in Appendix 41, Table A45), yield and production gaps are
estimated by comparing potential attainable yields and production (low and mixed input levels),
withactualachievedyieldsandproduction(year2000and2005).

7
2 DescriptionofGAEZinputdatasets
2.1 Climatedata
2.1.1 Observedclimate
FortheglobalagroecologicalzonesassessmenttimeseriesdataareusedfromtheClimateResearch
Unit (CRU) at the University of East Anglia, 10 arcminute latitude/longitude gridded average
monthlyclimatedata,versionCRUCL2.0(New2002),and30arcminutelatitude/longitudegridded
monthlyclimatedatatimeseriesfortheperiod19012002,versionCRUTS2.1(Mitchell2005).This
database revises and extends the earlier version CRU TS 1.0 (New 2000) used in the 2002 GAEZ
assessment(Fischer2002).SevenclimaticvariablesarerequiredforGAEZclimateanalysisasshown
inTable21.
For precipitation, an alternative data product was obtained from VASClimO (Variability Analysis of
Surface Climate Observations), a joint climate research project of the German Weather Service
(Global Precipitation Climatology Centre GPCC) and the Johann Wolfgang GoetheUniversity
Frankfurt(InstituteforAtmosphereandEnvironmentWorkingGroupforClimatology).VASClimOis
based on data being selected with respect to a (mostly) complete temporal data coverage and
homogeneity of the time series. The current version 1.1 of VASClimO uses timeseries of 9,343
stations covering the period 19512000 (Beck 2004). Results of gridded data (30 arcminute
latitude/longitude)wereavailablefromtheVASCLimWebsite(www.gpcc.dwd.de).Theselongterm
climatological analyses of homogenized areaaveraged precipitation timeseries are supported by
theIntergovernmentalPanelonClimateChange(IPCC)FourthAssessmentReport(AR4).
Original monthly CRU 10 arcminute and GPCC and CRU 30 arcminute latitude/longitude climatic
surfaces were interpolated at IIASA to a 5 arcminute grid for all years between 1960 and 2002.
Monthlyclimaticvariablesusedincludeprecipitation;numberofrainy days; mean minimum,mean
maximum temperature; diurnal temperature range; cloudiness; wind speed (only the average for
196190wasavailablefromCRUCL2.0);andvaporpressure.Forallvariablesexcepttemperaturea
bilinearinterpolationmethodwasappliedwithinArcGIS.Itusesthevaluesofthefournearestinput
cell centers to determine the value of the 5 arcminute output raster. The new value for the 5
output cell is a weighted average of these four values, adjusted to account for their distance from
thecenteroftheoutputcell.
In the case of temperature a lapse rate of 0.55
o
C per 100 meter elevation was applied using the
respectivedigitalelevationdata(DEM).First,a30arcminutesurfaceprovidedbyCRUwasusedto
calculate temperature values adjusted to sea level. Bilinear interpolation was performed for
temperatures at sea level. Second, a 5 arcminute DEM, derived from Shuttle Radar Topography
Mission (SRTM) data, was used to calculate temperatures for actual elevations. The 5 arcminute
DEMwascompiledfromdetailedSRTM3arcsecondelevationsusingthemedianofall3arcsecond
elevationdatawithineach5arcminutegridcell.
Table21ClimaticinputvariablesfortheGAEZassessment
Variable

Symbol Units Source

2
AverageTemperature T
a
o
C CRU
DiurnalTemperatureRange T
range
o
C CRU
Sunshinefraction n/N % CRU
Windspeedat10mheight U
10
m/s CRU
Relativehumidity RH % CRU
Wetdayfrequency WET days CRU
Precipitation P mm VASClimO
2
Seetextfordetails
8
2.1.2 ClimateScenarios
For the analysis of climate change impacts on agricultural production potential, available climate
predictions of General Circulation Models (GCM) were used for characterization of future climates.
The IPCC data distribution centre (http://www.ipccdata.org/) provides future climatic parameters
obtainedasoutputsofvariousGCMexperimentsforarangeofIPCCemissionscenarios.
ThefollowingfourGCMswereusedhereforcalculationoffuturepotentialagriculturalproductivity:
HadCM3(HadleyCentre,UKMeteorologicalOffice)
ECHAM4(MaxPlanckInstituteforMeteorology,Germany)
CSIRO(Australia'sCommonwealthScientificandIndustrialResearchOrganisation,Australia)
CGCM2(CanadianGeneralCirculationModel)
GCMmodeloutputsforindividualclimateattributeswereappliedasfollows:
Difference of the means for three 30year periods (the 2020s: years 20112040; the 2050s: years
20412070; and the 2080s: years 20712100) with the GCM baseline climate 19611990 were
calculatedforeachgridintherespectiveGCM.Aninversedistanceweightedinterpolationtoa30
arcminutegridwasperformedonthesedeltasofthecentrepointsofeachgridcellintheoriginal
GCM. Such changes (deltas) for monthly climatic variables, i.e. differences for maximum and
minimum monthly temperature, precipitation, total surface solar radiation and windrun, were
thenappliedtotheobservedclimateof19611990togeneratefutureclimatedata.Climatechange
induced alterations in agricultural productivity as a result of climate change can be calculated by
runningGAEZforfuturetimeslotsandcompareresultstotheoutcomesfortheclimaticbaseline.
2.1.3 UseofclimatedatainGAEZ
Theaverageclimateandyearbyyearhistoricaldatabaseswereusedtoquantify:
(i) Widely used agroclimatic indicators, such as the number of growing period days, thermal
climateclassification,aridityindices,and
(ii) to estimate for each gridcell by crop/LUT, average and individual years agroclimatically
attainablecropyieldsandvariability.
Monthly5arcminutelatitude/longitudegridsofaverageclimateandyearbyyearclimateattributes
forthesevenclimatevariables(Table21)werecombinedintobinaryrandomaccessfilesonefile
foreachclimatevariablecontainingallmonthlyvaluespergridcell,whichserveasinputtotheGAEZ
simulationprograms.
In a similar way, binary random access files were generated to hold monthly and annual climate
changedeltasderivedfromGCMoutputs.Inthiswayaveragefutureclimateconditionshavebeen
simulatedinGAEZ,aswellastimeseriesoffutureyears,bycombiningrespectivehistoricaldataand
GCMderiveddeltas.
2.2 Soildata
The Land Use Change and Agriculture Program of IIASA (LUC) and the Food and Agriculture
OrganizationoftheUnitedNations(FAO)havedevelopedanewcomprehensiveHarmonizedWorld
Soil Database (FAO/IIASA/ISRIC/ISSCAS/JRC 2009). Vast volumes of recently collected regional and
national updates of soil information were used for this stateoftheart database. The work was
carriedoutinpartnershipwith:
ISRICWorld Soil Information and FAO, which were responsible for the development of
variousregionalsoilandterraindatabasesandtheWISEsoilprofiledatabase;
theEuropeanSoilBureauNetwork,whichhadcompletedamajorupdateofsoilinformation
forEuropeandnorthernEurasia,and
the Institute of Soil Science, Chinese Academy of Sciences, which provided the 1:1,000,000
scaleSoilMapofChina.
9
TheHWSD(Figure21)iscomposedofageographicallayercontainingreferencetosome30,000soil
map units. This information is stored as a 30 arcsecond raster in a GIS, which is linked to an
attributedatabaseinMicrosoftAccessformatcontainingharmonizedsoilprofiledata.Fortheglobe
therasterhas21,600rowsand43,200columns,ofwhich221milliongridcellscovertheglobesland
territory.Over16,000differentsoilmappingunitsarerecognizedintheHWSDthatcombineexisting
regional and national updates of soil information worldwide with the information contained within
the1:5,000,000scaleFAOUNESCOSoilMapoftheWorld(FAO/UNESCO1974).

Figure21HarmonizedWorldSoildatabase(HWSD)
TheuseofastandardizedstructureinHWSDcreatesaharmonizeddataproductacrossthevarious
originalsoildatabases.Thisallowstheconsistentlinkageoftheattributedatawiththerastermapto
displayorquerythecompositionofsoilmappingunitsandthecharacterizationintermsofselected
soilparameters(organiccarbon,pH,soilwaterholdingcapacity,soildepth,cationexchangecapacity
of the soil and the clay fraction, total exchangeable nutrients, lime and gypsum contents, sodium
exchangepercentage,salinity,texturalclassandgranulometry).
Reliability of the information contained in the database is inevitably variable: the parts of the
databasethatmakeuseoftheSoil MapoftheWorldsuchas forNorthAmerica,Australia,mostof
West Africa and South Asia are considered less reliable, while most of the areas covered by SOTER
databasesaredeemedtohavethehighestreliability(CentralandSouthernAfrica,LatinAmericaand
theCaribbean,CentralandEasternEurope).
For the agroedaphic assessment GAEZ applies the most recent Version 1.1 of the HWSD (March
2009). A detailed description of HWSD and the latest version are available for download at:
www.iiasa.ac.at/Research/LUC/luc07/ExternalWorldsoildatabase/HTML/index.html.
GAEZproceduresinModuleIVandVmakeampleuseofthesoilinformationprovidedintheHWSD
inordertoassessvarioussoilqualitiesvisviscropsoilrequirements.
2.3 Elevationdataandderivedterrainslopeandaspectdata
Theglobalterrainslope(Figure22)andaspect(i.e.maindirectionthattheterrainfaces)databases
have been compiled using elevation data from the Shuttle Radar Topography Mission (SRTM). The
SRTMdataisavailableas3arcsecondDEMs(CGIARCSI,2006).
ThehighresolutionSRTMdatahavebeenusedforcalculating:
1. Terrainslopegradientsandclasses(foreach3arcsecgridcell);
2. Aspectofterrainslopes(foreach3arcsecgridcell);
3. Distributionsofslopegradientclassesandslopeaspectclassesfora30arcsecondgrid.
10
The SRTM data cover the globe for areas up to 60 latitude. For the areas north of 60 latitude, 30
arcseconds elevation data and derived slope class information were compiled from GTOPO30
(USGSGTOPO302002).
Theglobalterrainslopeandaspectdatabaseat30arcsecondsusedinGAEZcomprisesthefollowing
elements:
Medianelevation(m)of3arcsecondgridcellswithineach30arcsecondgridcell
Distributions (%) of eight slope gradient classes: 00.5%, 0.52%, 25%, 58%, 816%, 16
30%,3045%,and>45%.
Slope aspect information (%), compiled at 3 arcseconds and stored at 30 arcsecond in
distributionsoffiveclasses:slopesbelow2%(undefinedaspect;)slopesfacingNorth(315
45);East(45135);South(135225),andWest(225315).
AdetaileddescriptionoftheproceduresappliedcanbefoundinAppendix10.

Figure22Medianterrainslopes
Elevationdata,slopegradientsandslopeaspectsforbotha5arcminuteanda30arcsecondgrid
areavailablefordownload:(www.iiasa.ac.at/Research/LUC/luc07/ExternalWorldsoil
database/HTML/globalterrainslopedownload.html?sb=7).

Figure23Exampleofcalculatedterrainslopeclasses(percentofgridcellwithslope>16%)
11
2.4 Landcoverdata
Six geographic datasets were used for the compilation of an inventory of seven major land
cover/landusecategoriesata5arcminuteresolution.Thedatasetsusedare:
1. GLC2000landcover,regionalandglobalclassificationsat30arcseconds(JRC2006);
2. IFPRI Agricultural Extent database, which is a global land cover categorization providing 17
landcoverclasses at30 arcseconds (IFPRI2002),basedon areinterpretationoftheGlobal
LandCoverCharacteristicsDatabase(GLCCD2001),EROSDataCentre(EDC2000);
3. TheGlobalForestResourcesAssessment2000and2005(FRA2000andFRA2005)ofFAOat
30arcsecondsresolution;
4. DigitalGlobalMapofIrrigatedAreas(GMIA)version4.01(Siebert2007)at5arcminute
latitude/longituderesolution,providingbygridcellthepercentagelandareaequippedwith
irrigationinfrastructure;
5. IUCNWCMCprotectedareasinventoryat30arcseconds(http://www.unep
wcmc.org/worlddatabaseonprotectedareaswdpa_76.html),and
6. Spatialpopulationdensityinventory(30arcseconds)foryear2000developedbyFAOSDRN,
based on spatial data of LANDSCAN 2003, LandScanTM Global Population Database
(http://www.ornl.gov/landscan/),withcalibrationtoUN2000populationfigures.
An iterative calculation procedure has been implemented to estimate land cover class weights,
consistent with aggregate FAO land statistics and spatial land cover patterns obtained from (the
abovementioned)remotelysenseddata,resultinginthequantificationofmajorlanduse/landcover
sharesinindividual5arcminutelatitude/longitudegridcells.Theestimatedclassweightsdefinefor
eachlandcoverclassthepresenceofrespectivelycultivatedlandandforest.Startingvaluesofclass
weightsusedintheiterativeprocedurewereobtainedbycrosscountryregressionofstatisticaldata
ofcultivatedandforestlandagainstlandcoverclassdistributionsobtainedfromGIS,aggregatedto
national level. The percentage of urban/builtup land in a gridcell was estimated based on
occurrence of respective land cover classes as well as regression equations, obtained using various
subnational statistical data, relating builtup land with population density. Remaining areas, i.e.
areasthatarenotrepresentingcultivatedland,forestlandorbuiltupland,wereallocatedto:
1. Grasslandandothervegetatedareas,
2. Barrenorverysparselyvegetatedareas,and
3. Waterbodies
According to the land cover classes indicated at 3 arcseconds in GLC2000. Barren or very sparsely
vegetatedareasweredelineatedby(i)usingtherespectivelandcoverclassesinGLC2000and/or(ii)
aminimumbioproductivitythresholdof100kgDM/ha/year.
The resulting seven land use/land cover categories, used for land accounting and to characterize
each5arcminutegridcell,are:
1. Rainfedcultivatedland
2. Irrigatedcultivatedland
3. Forest
4. Grasslandandothervegetatedland
5. Barrenandverysparselyvegetatedland
6. Water
7. Urbanlandandlandusedforhousingandinfrastructure.
AnexampleoflandcoverdatabasefromtheHarmonizedWorldSoilDatabaseisshownonFigure24
below.
12

Figure24Exampleoflandcoverdata:dominantlandcoverpatternintheHWSD
2.5 Protectedareas
The World Database of Protected Areas Annual Release 2009 (henceforth WDPA 2009) and for the
territory of the European Union the NATURA 2000 network, were applied to identify broad
categoriesofprotectedareas,whicharedistinguishedintheGAEZanalysis:
1. Protectedareaswhererestrictedagriculturaluseispermitted
2. Strictlyprotectedareaswhereagriculturaluseisnotpermitted.
2.5.1 WDPA2009
The WDPA2009 includes both point and polygon data. The global polygon database was used to
delineate30arcsecondgridcellsofprotectedareasinGAEZ.WDPA2009identifies80,142different
mapping units (termed Siteids) with associated attribute data for over 450,000 polygons. The
majority of mapping units (51,556) refers to either an international or national convention. The
remaining mapping units record the type of protected area, e.g. national park, natural monument,
etc. (item DESIG_ENG in WDPA 2009). From these units, 77 designations were considered to be
strictly protected and therefore these categories are considered not available for agriculture. The
most important designations include National Parks, Forest Reserves, Zapovednik (a protected
area in Russia which is kept "forever wild"), Wildlife Management Area, Nature Park, Resource
Reserve,NatureReserve,andGameReserve.
TheEuropeanpartoftheWDPAinventorydoesnotincludeimportantprotectedareasfortheEU27,
which are however part of the NATURA 2000 network. WPDA 2009 grid and the NATURA 2000
networkinformationwerecombinedtoformtheGAEZprotectedarealayer.
2.5.2 Natura2000
The European Union has established a network of nature protection areas, known as the NATURA
2000 network, with the aim to assure the longterm survival of Europe's most valuable and
threatenedspeciesandhabitats.Italsofulfills anobligationundertheUNConventiononBiological
Diversity. The network is comprised of Special Areas of Conservation (SAC) designated by Member
StatesundertheHabitatsDirective,andalsoincorporatesSpecialProtectionAreas(SPAs)whichthey
designate under the Birds Directive. NATURA 2000 currently includes over 26,000 protected areas
coveringatotalareaofaround850,000km
2
,representingmorethan20%oftotalEUterritory.
To distinguish protected and strictly protected areas CORINE land cover 2000 (CLC2000;
http://etclusi.eionet.europa.eu/CLC2000) distributions of the NATURA 2000 sites were calculated.
CLC2000dataareavailableat100metersresolutionandcategorizedusingthe44landcoverclasses
of the 3level CORINE nomenclature. The spatialpolygon databaseof NATURA 2000 was converted
toa100mgridcellsizeandoverlaidwithCLC2000.Whereapplicable,theCORINElandcoverclasses
13
Arable land, Permanent crops and Heterogeneous agriculture were assigned to the protected
areas category, thus permitting restricted agricultural use. The remaining land cover classes were
consideredtorepresentstrictlyprotectedareas,wherecultivationofarablecropsisnotpossible.
The100metersresolutiongridmapshowingthetwotypesofprotectedareaswasprojectedtoa30
arcsecondlongitude/latitudegridmapandtherespectiveareasofthe27countriesoftheEuropean
Union(EU27)wereintegratedintheGAEZprotectedareaslayer.
Table 22 presents a summary of the various convention types used in the GAEZ protected areas
layer. The protected areas are subdivided in types which permit or do not permit agricultural use.
The GAEZ protected areas layer comprises 20% of protected areas where agriculture is
conditionally permitted and 80% strictly protected areas where agriculture is assumed not to be
permitted.
Table22GAEZprotectedareaslayer
Code Conventiontype Agriculturaluse Shareoftotal
protectedarea
1 IUCNIaStrictNatureReserve no 4.7%
2 IUCNIbWildernessArea no 7.2%
3 IUCNIINationalPark no 30.7%
4 IUCNIIINaturalMonument no 0.8%
5 IUCNIVHabitatManagement no 12.2%
6 IUCNVProtectedLandscape yes 8.9%
7 IUCNVIManagedResource yes 10.9%
8 RamsarConvention(Wetlands) no 3.1%
9 WorldHeritageConvention no 5.0%
10 UNESCOMABBiosphereReserves no 1.4%
11 ASEANHeritage no 0.2%
12 Natura2000(limitedagriculturaluse) yes 0.7%
13 Natura2000(noagriculturaluse) no 3.7%
14 National(Nonforest) no 7.9%
15 National(Forest)t no 2.5%
TOTAL(noagriculturaluse) 80%
TOTAL(limitedagriculturaluse)s 20%
TOTALprotected 100%

Figure25ProtectedAreas

14
2.6 Administrativeareas
The Global Administrative Unit Layers (GAUL) provides authoritative global spatial information on
administrativeunitsforallcountriesintheworld.GAULisaninitiativeimplementedbytheFoodand
Agriculture Organization (FAO) of the United Nations, which has significantly contributed to the
standardizationofcomprehensivelyrecordingspatialadministrativeunits.
The GAUL maintains global geographic layers with a unified coding system of national and sub
national administrative levels. Controversial and disputed boundaries are maintained such, that
national integrity for all disputing countries is preserved. Once a year, an updated version of the
GAUL set is released through Geonetwork (http://www.fao.org/geonetwork/srv/en/main.home).
TheversionofGAULappliedinGAEZv3.0wasobtainedin2009.
ForuseinGAEZtheGAULvectordatahasbeentransformedrespectivelytorastersof5arcminutes
and 30 arcsecond gridcells. For aggregating GAEZ country results and information at regional and
continentallevel,thecountriesincludedintheGAULhavebeencodifiedaccordingtothreelevelsof
supranationalregionalization,seeAppendix21.

Figure26GAULcountryboundarieslayerwithcontinentalGAEZregionalizations

15

Figure27GAULcountryboundarieslayerwithsubcontinentalGAEZregionalizations

Figure28GAULcountryboundarieslayerincomelevelregionalizations

16

17
3 ModuleI(Agroclimaticanalysis)
3.1 OverviewModuleI
ModuleIdealswithtemporal,interpolation,analysisandclassificationofclimatedataandcreation
of historical, base line and future gridded agroclimatic indicators relevanttoplant production. The
main objective in Module I is the compilation of georeferenced climatic resources inventory
containing relevant agroclimatic indicators. The inventory is used for the evaluation of land
suitabilityandestimationofcropyieldsandproductionin:ModuleII(biomassandyieldcalculation),
Module III (agroclimatic yield constraints) and Module V (integration of climatic and edaphic
evaluation).Figure31presentstheinformationflowinModuleI.

Figure31InformationflowinModuleIoftheGAEZmodelframework
Spatially explicit climatic databases provide the main input data for Module I. Available monthly
climate data and their spatial interpolation to a 5 arcminute grid for the globe are presented in
Section2.1.
3.2 Preparationofclimaticvariables
ClimaticvariablesarepreparedfortheuseinGAEZthroughconversionsandtemporalinterpolations
Temporalinterpolationsofthegriddedmonthlyclimaticvariablesintodailydata,providesthebasis
forthecalculationofsoilwaterbalancesandagroclimaticindicatorsrelevanttoplantproduction.

Windrunandwindspeed
Wind data is used for the estimation of evapotranspiration. For the agroclimatic calculations
observed, wind speed (U10) at 10 m height is converted to windspeed (U2) and wind run at 2 m
heightthatisstandardcropcanopyheightinagroclimatologicanalysis.(FAO1992)
18
Wetdayfrequency
Wet day frequency (WET) is used to derive daily precipitation events from monthly totals. For
historical or future time periods for which wet day frequency is not available as input data it is
establishedthroughtherelationship:
0.45
ref
ref
P
WET WET
P

=



where P and P
ref
are respectively the monthly precipitation of the historical or future time periods
andmonthlyprecipitationofthe19611990referenceclimateperiod.WET
ref

representsthemonthly
wet day frequencyin the reference climate. Additional climatic indicators, necessary to assess crop
suitability and yield in Module II, are calculated in the Module I of GAEZ. These are sunshine
duration, daylength, daytime and nighttime temperatures, temperature profiles and air frost
number.
Sunshineduration
Actual sunshine duration (n) is used for the calculation of incoming solar radiation, for
evapotranspiration and biomass calculations. Sunshine duration is calculated from the ratio actual
sunshinehoursovermaximumpossiblesunshinehours(n/N).
Daytimeandnighttimetemperatures
Thetemperatureduringdaytime(Tday,
o
C)andnighttime(Tnight,
o
C)arecalculatedasfollows:

+ =
0
0
0
0
11
11
sin
12
11
4 T
T
T
T Tn Tx
Ta Tday

Nighttimetemperatureiscalculatedas:

=
0
0
0
0
11
11
sin
11
4 T
T
T
T Tn Tx
Ta Tnight

where Ta is average 24 hour temperature, and T
0
is calculated as a function of daylength (DL,
hours).
0
12 0.5 T DL =

Daylength is calculated in the model and depends on the latitude of a gridcell and the day of the
year.

ReferenceEvapotranspiration(ETo)
The reference evapotranspiration (ETo) represents evapotranspiration from a defined reference
surface,whichcloselyresemblesanextensivesurfaceofgreen,wellwateredgrassofuniformheight
(12 cm), actively growing and completely shading the ground. GAEZ calculates ETo from the
attributes in the climate database for each gridcell according to the PenmanMonteith equation
(Monteith 1965; Monteith 1981; FAO 1992). A detailed description of the implementation of the
PenmannMonteithequationsinGAEZisprovidedinAppendix31.
Maximumevapotranspiration(ETm)
In Module I, the calculation of evapotranspiration (ETm) for a reference crop assumes that
sufficient water is available for uptake in the rooting zone. The value of ETm is related to ETo
through applying crop coefficients for water requirement (kc). The kc factors are related to
phenological development and leaf area. The kc values are crop and climate specific. They vary
generallybetween0.40.5atinitialcropstages(emergence)to1.01.2atreproductivestages.
19
ETo kc ETm =
ForthereferencecropasmodeledinGAEZ,valuesofkcdependonthethermalcharacteristicsofa
grid cell. For locations with a yearround temperature growing period (LGP
t5
equals 365 days), i.e.
when average daily temperature stays above 5
o
C for the entire year, the kc value applied for the
referencecropisalways1.0.WhenLGPt5<365days,thekcvalueincreaseslinearlyfrom0.4atthe
start of the temperature growing period until reaching the reference value 1.0 after 30 days to
account for increasing water demand as the crop canopy develops after the cold period. When
assessingspecificcrops,asisdoneinModuleII,empiricallydeterminedkcvaluesforthecalculation
of ETm are available from various sources (FAO 1998) and differ by the development stage of the
crop(seesection4.5.1).
Actualevapotranspiration(ETa)
Theactualuptakeofwaterforthereferencecropischaracterizedbytheactualevapotranspiration
(ETa,mm/day).ThecalculationofETadifferentiatestwopossiblecasesdependingontheavailability
ofwaterforplantextraction:
(i) Adequatesoilwateravailability(ETa=ETm)
(ii) Limitingsoilwateravailability(ETa<ETm)
Whenwaterisnotlimiting,theETavalueisequaltothemaximumevapotranspiration(ETm)ofthe
rewference crop. At limiting water conditions, ETa is a fraction of ETm, depending on soil water
availabilityasexplainedinfollowingsections.
ETaforadequatesoilwateravailability
The value of ETais set tobeequal to ETm as long asthewaterbalance (Wb) is above or equal the
thresholdofreadilyavailablesoilwater(Wr).Thischaracterizesasituationwhencropsareableto
easilyextractsufficientwaterandthereforenowaterstressoccurs.Thepotentiallytotalavailable
soilmoistureWxistheproductoftotalavailablesoilwaterholdingcapacity(Sa)androotingdepth
(D).TheshareofWxbelowwhichsoilmoisturestartstobecomedifficulttoextractisreferredtoas
p, the soil moisture depletion fraction. The fraction p varies with the evaporative demand of the
atmosphere, crop type, and soil characteristics. Estimates are available from various sources (FAO
1998). The value of p normally varies from 0.3 for shallow rooted plants at high rates of ETm (>8
mm/day)to0.7 fordeeprootedplantsatlowrates ofETm(<3mm/day).Ingeneral,thevalueofp
declines with increasing evaporative demand. The threshold of readily available soil moisture is in
turncalculatedfromWxandthesoilmoisturedepletionfraction(p).
D Sa Wx =
) 1 ( p Wx Wr =
A condition of adequate soil moisture availability is defined when (i) daily precipitation (P) is
greater or equal to ETmand/or (ii) precipitationP plusthedifference between water balance (Wb)
andthresholdofreadilyavailablewater(Wr)isgreaterthanETm.Theseconditionsimplythatthere
issufficienteasilyextractablewatertomeetthecropwaterdemand(ETm):
ETm ETa =

when
P ETm

orwhen
P ETm <

but

P Wb Wr ETm + > .
20
ETacalculationforlimitedsoilwateravailability
Whensoilwaterislimiting,i.e.whenaboveconditionsarenotmetandP+WbWr<ETm,thenETa
fallsshortofETm.Inthiscase,ETaiscalculatedasa fractionofETm.Thevariableistheratioof
currentwaterbalance(Wb)andthethresholdofreadilyavailablesoilwater(Wr).
Wr
Wb
=
ETaisthencalculatedasdailyprecipitationPplusthefractionofETm.
ETa P ETm = +
Thisprocedureassumesrainfallisimmediatelyavailabletoplantsonthedayofprecipitationpriorto
replenishingsoilmoisture.
Snowbalancecalculation
Thecalculationofasnowbalance(Sb,mm)affectsthewaterbalanceprocedureoutlinedabove.The
snow balance increases when precipitation falls as snow and decreases with snowmelt and snow
sublimation. All precipitation (P) falls as snow (P
snow
) when maximum temperature (Tx) is below a
temperaturethreshold(Ts).
Snowmelt(Sm)iscalculatedasafunctionofdailymaximumtemperature,thesnowmeltparameter
() and issubjecttothepreviouslyaccumulatedsnowbalance.Thesnowmeltfactorissetto5.5
mm/
o
C.
min( ( ), ) Sm Tx Ts Sb =
The sublimation factor (ks) is used to discount a fraction of maximum evapotranspiration as
sublimatedsnow.Thisfraction(ks*ETm)issubtractedfromthesnowbalance:
1
( )
snow
j j
Sb Sb Sm ks ETm P

= +
The sublimation factor (ks) is assumed to be 0, 0.1 or 0.2 of reference evapotranspiration (ETm,
mm),dependingtemperature:
ks=0.0,whenTx<Ts;Tsisassumedas0
o
CinGAEZ
ks=0.1,whenTx>TsandTa<0
o
C
ks=0.2,whenTx>Tsand0
o
C<Ta<5
o
C
Once the water balance for the reference cropis calculated, fivevariables are produced and used
forfurthercomputationsinGAEZmodules.Theseare:
1. Maximumevapotranspirationofreferencecrop(ETm)
2. Actualevapotranspirationofreferencecrop(ETa)
3. Waterbalanceforreferencecrop(Wb)
4. Snowbalance(Sb)
5. Excesswaterofreferencecropwaterbalance(We)
3.2.1 Temporalinterpolation
GAEZ uses quadratic spline interpolation to derive daily values from available monthly data (from
theCRUandGPCCclimatedata).
With data available for several periods ( N i : 1 = ), the goal of the interpolation procedure is to fill
intermediatedatapoints within the given observations. The quadratic spline interpolation assumes
that the data between subsequent data points ) , (
i i
y t and ) , (
1 1 + + i i
y t can be estimated by a
piecewisequadratic polynomial (
i i i
c t b t a y + + =
2
) connecting the given points where i denotes
the available monthly observations ( N i : 1 = ). As there are N observations and, therefore, N 3
unknown parameters (
i
a ,
i
b , and
i
c ), also N 3 equations are required to estimate these
21
parameters. N 2 equations are derived from the requirement of a continuous function, i.e. that
neighboringsegmentsmeetatthepointstheyhaveincommon(notethatthelastsegmentjoinsthe
firstone:
2 2
1 1 1 1 1 1 1 1 i i i i i i i i i i i
y a t b t c a t b t c
+ + + + + + + +
= + + = + +
Then another N equations are derived from the requirement that in each internal point
1 N : 2 i ( = ) as well as the last segment joining the first one, the first derivatives of the joining
quadraticfunctionsareequal:
1 1 1 1
2 2
i i i i i i
a t b a t b
+ + + +
+ = +
Thus, all parameters
i
a ,
i
b , and
i
c ( N i : 1 = ) of the polynomials
i i i
c t b t a y + + =
2
are
established.
Inthisway,thefollowingeightclimaticvariablesareconvertedbysplineinterpolationfrommonthly
topseudodailyvalues:
1. Minimumtemperature(Tn)
2. Daytimetemperature(Tday)
3. Maximumtemperature,(Tx)
4. Referenceevapotranspiration(ETo)
5. Windspeedat2mheight(U2)
6. Relativehumidity(RH)
7. Sunshinehours(n)
For distributing monthly precipitation (P) within a month, in addition the input data on observed
number of monthly precipitation events (wet day frequency) is used together with precipitation
trendsaccordingtothesplineinterpolationofmonthlyprecipitationvalues.
3.3 ThermalRegimes
Temperature is a major determinant of crop growth and development. In GAEZ the effect of
temperature on crops is characterized in each gridcell by thermal regimes. Thermal regimes are
represented by five types of indicators: (i) thermal climates; (ii) thermal zones; (iii) length of
temperaturegrowingperiods;(iv)temperaturesums,and(v)temperatureprofiles.
3.3.1 Thermalclimates
Latitudinal thermal climates provide a classification that is used in Module II for the assessment of
potentialcropLUTpresenceineachgridcell.Thedelineationofthermalclimatesisbasedon(i)the
average monthly temperature, (ii) proportions of respectively summer, winter rainfall
1
, and (iii) the
temperature amplitude as a measure of continentality (i.e. difference between temperatures of
warmestandcoldestmonth).Thermalclimatesarederivedfrommonthlytemperaturescorrectedto
sea level temperature with a fixed lapse rate of 0.55
o
C/100m. There is a further subdivision for
rainfall seasonality inthe subtropics andfortemperatureamplitude intemperateandborealzones
(Figure 32). In this way, latitudinal climates approximate temperature seasonality and ranges of
prevailing daylengths, which is used as a proxy for matching shortday, dayneutral and longday
croprequirements.

1
Rainfallhasbeenrepresentedwithsummerrespectively,winterP/Etoratios.
22
Table31Classificationofthermalclimates
ThermalClimateClassification
Thermalclimatesarederivedfrommonthlytemperaturescorrectedtosealevel.Thethermalclimateshavebeen
subdividedforrainfallseasonalityinthesubtropicsandfortemperatureseasonalityintemperateandborealzones.The
tropicshavebeensubdividedinlowlandandhighlandzones.
Climate RainfallandTemperatureSeasonality
Tropics
Allmonthswithmonthlymean
temperatures,correctedtosealevel,
above18C
Tropicallowland
Tropicswithactualmeantemperaturesabove20
o
C
Tropicalhighland
Tropicswithactualmeantemperaturesbelow20
o
C
Subtropics
Oneormoremonthswithmonthly
meantemperatures,correctedtosea
level,below18C,butallabove5C,
and812monthsabove10C
SubtropicsSummerRainfall
Northernhemisphere:P/EToinAprilSeptemberP/EToinOctoberMarch.
Southernhemisphere:P/EToinOctoberMarchP/EToinAprilSeptember
SubtropicsWinterRainfall
Northernhemisphere:P/EToinOctoberMarchP/EToinAprilSeptember.
Southernhemisphere:P/EToinAprilSeptemberP/EToinOctoberMarch
SubtropicsLowRainfall
Annualrainfalllessthan250mm
Temperate
Atleastonemonthwithmonthly
meantemperatures,correctedtosea
level,below5Candfourormore
monthsabove10C
OceanicTemperate
Seasonalitylessthan20C*
SubcontinentalTemperate
Seasonality2035C*
ContinentalTemperate
Seasonalitymorethan35C*
Boreal
Atleastonemonthwithmonthly
meantemperatures,correctedtosea
level,below5Cand13months
above10C
OceanicBoreal
Seasonalitylessthan20C*
SubcontinentalBoreal
Seasonality2035C*
ContinentalBoreal
Seasonalitymorethan35C*
Arctic
Allmonthswithmonthlymean
temperatures,correctedtosealevel,
below10C
Arctic
*Seasonalityreferstothedifferenceinmeantemperatureofthewarmestandcoldestmonth

Figure32Thermalclimates

3.3.2 ThermalZones
Thermal zones reflect the prevailing temperature regimes of major thermal climates. The
classificationispresentedinFigure33:
23

Figure33ThermalZones
(i) Warmintropicalzonesreferstoannualmeantemperaturesabove20
o
C,cool,cold,verycold
tropicsreferstoannualmeantemperaturebelow20
o
C;
(ii) Moderately cool refers to actual temperature conditions characterized by one or more
monthswithmonthlyaveragetemperaturesbelow18
o
Cbutallabove5
o
Cand812months
above10
o
C;
(iii) Coolreferstoconditionswithatleastonemonthwithmonthlymeantemperaturesbelow5

o
Candfourormoremonthsabove10
o
C;
(iv) Coldreferstoconditionswithatleastonemonthwithmonthlymeantemperaturesbelow5

o
Cand13monthsabove10
o
C,and
(v) Very cold refers to polar conditions i.e. allmonths with monthly mean temperatures below
10
o
C.
3.3.3 Temperaturegrowingperiods(LGPt)
The length of the temperature growing period (LGPt) is calculated as the number of days in the
year when average daily temperature (Ta) is above a temperature threshold t. In GAEZ three
standard temperature thresholds for temperature growing periods are used: (i) periods with Ta >
0
o
C, (ii) periods with Ta

> 5
o
C, which is considered as the period conducive to plant growth and
development,and(iii)periodswithTa

>10
o
C,whichisusedasaproxyfortheperiodoflowrisksfor
lateandearlyfrostoccurrences.
Therefore,GAEZcalculatesthefollowingthreeLGPts:
(i) LGPt
0
periodwhenTa>0
o
C
(ii) LGPt
5
periodwhenTa>5
o
C
(iii) LGPt
10
periodwhenTa>10
o
C

24

Figure34Frostfreeperiod(LGPt10)
3.3.4 Temperaturesums(Tsum)
Heat requirements of crops are expressed in accumulated temperatures. Reference temperature
sums (Tsum) are calculated for each gridcell by accumulating daily average temperatures (Ta) for
dayswhenTaisabovetherespectivethresholdtemperaturestasfollows:
(i) 0
o
C(Tsum
0
)
(ii) 5
o
C(Tsum
5
)
(iii) 10
o
C(Tsum
10
)

Figure35TemperaturesumsforthefrostfreeperiodwithTa>10oC
3.3.5 Temperatureprofiles
Temperatureprofiles(Table32)aredefinedintermsof9classesoftemperaturerangesfordays
withTa<5
o
Cto>30
o
C(at5
o
Cintervals)incombinationwithdistinguishingincreasinganddecreasing
temperature trends within the year. In Module II of GAEZ, these temperature profiles are matched
withcropspecifictemperatureprofilerequirementsprovidingeitheroptimummatch,suboptimum
matchorrenderingacropnotsuitablefortherespectivelocation.

25
Table32Temperatureprofileclasses
Averagetemperature Temperaturetrend
(Ta,
o
C) Increasing Decreasing
>30 A1 B1
2530 A2 B2
2025 A3 B3
1520 A4 B4
1015 A5 B5
510 A6 B6
05 A7 B7
50 A8 B8
<5 A9 B9
3.3.6 Permafrostevaluation
Occurrence of continuous or discontinuous permafrost conditions are used in the suitability
assessment.Permafrostareasarecharacterizedbysubsoilatorbelowthefreezingpointfortwoor
more years. Permafrost or gelic soils are considered unsuitable for crops and therefore their
identification is essential for the land resources assessment in GAEZ. Average air temperature and
thephysicalandchemicalcharacteristicsofthesoilsarethemainfeaturesinfluencingthepresence
of permafrost. Consequently, GAEZ considers permafrost in two ways: (i) it determines different
referencepermafrost zones basedon climaticconditions, and (ii)itrelieson soilclassification; soils
with a gelic connotation within or outside permafrost zones are considered to belong to the
continuouspermafrostzone.
InGAEZ,theproceduresproposedbyNelsonandOutcalt(1987)areappliedtocalculateanairfrost
index(FI)whichisusedtocharacterizeclimatederivedpermafrostconditionsintofourclasses:
(i) Continuouspermafrost
(ii) Discontinuouspermafrost
(iii) Sporadicpermafrost
(iv) Nopermafrost
Reference permafrost zones are determined based on prevailing daily mean air temperature (Ta).
The air frost index (FI) is calculated and used to characterize permafrost areas. For this calculation,
accumulateddegreedays,aboveandbelow0
o
C,areusedtocalculatethethawingindex(DDT)and
thefreezingindex(DDF).
ThethawingindexDDTiscalculatedas:

= Ta DDT ,whenTa>0
o
C
Thefreezingindex(DDF)iscalculatedas:

= Ta DDF ,when0
o
C
Thefrostindex(FI)isthencalculated(Nelson1987):
5 . 0 5 . 0
5 . 0
DDT DDF
DDF
FI
+
=
The value of FI is regarded a measure of the probability of occurrence of permafrost and used to
classifygridcellsinfourdistinctpermafrostclasses(Table33).
26
Table33ClassificationofpermafrostareasusedintheGAEZassessment
Permafrostclass Valueoffrost
Index
(FI)
Probabilityof
permafrost
*
(%)
Continuouspermafrost >0.625 >67
Discontinuouspermafrost 0.570<FI<0.625 3367
Sporadicpermafrost 0.495 < FI< 0.570 533
Nopermafrost <0.495 <5
*Probabilityofpermafrostoccurrencewascalculatedbasedondatasets
fromNelsonandOutcalt(1987)andanalyzedatIIASA.
In the GAEZ assessment, those gridcells characterized as continuous permafrost (class 1) or
discontinuouspermafrost(class2)areconsideredunsuitableforcropproduction.Regularyieldand
suitabilitycalculationsareperformedinclass3and4.Figure36presentsthereferencepermafrost
zonesmap.

Figure36Referencepermafrostzones
3.4 Soilmoistureregime
In Module I, GAEZ calculates a daily reference soil water balance for each gridcell and estimates
actual evapotranspiration (ETa) for a reference crop. In the Module II, soil moisture balance
calculationsareperformedconsideringspecificcrop/LUTs.
3.4.1 Soilmoisturebalance
Daily soil moisture balance calculation procedures follow the methodologies outlined in CROPWAT
(FAO 1986, 1992) and Crop Evapotranspiration (FAO, 1998). The quantification of a cropspecific
water balance determines crop actual evapotranspiration (ETa) used for waterconstrained crop
yieldcalculations.
The volume of water available for plant uptake iscalculated by means of a daily soil water balance
(Wb).TheWbaccountsforaccumulateddailywaterinflowfromprecipitation(P)orsnowmelt(Sm)
and outflow from actual evapotranspiration (ETa), and excess water lost due to runoff and deep
percolation(We).
) , min(
1
Wx ETa P Sm Wb Wb
j j j j j
+ + =


where j is the day of the year; Wx is the maximum water available to plants.The snowmelt (Sm) is
accountedwithinthesnowbalancecalculationproceduresandexcesswater(We)istheamountof
waterthatexceedsWx.
The upper limit Wx of the water available to plants depends on the soils physical and chemical
characteristicsthatinfluencetotalsoilwaterholdingcapacity(Sa).Bydefinition,Wxistheproductof
totalsoilwaterholdingcapacity(Sa)androotingdepth(D).
D Sa Wx =
The Sa value is a soilspecific attribute defined as the difference between soil moisture content at
fieldcapacity(Sfc)andpermanentwiltingpoint(Swp)overtherootingzone.Therefore,atanygiven
day, actual soil water content (Wb) will be available to plants if Swp < Wb < Sfc (Figure 39).
However, water extraction becomes more difficult as soil water content (Wb) is less than a critical
threshold (Wr) defined by p, the soil water depletion factor, and the soil water holding capacity
(Sa).

The values of Sa and rooting depth limitations due to soil are derived from soil information
contained in the Harmonized World Soil Database (FAO/IIASA/ISRIC/ISSCAS/JRC 2009). FAO has
developedproceduresfortheestimationofSa(FAO1995;Fischeretal.2002).Anywaterinputinto
the soil that exceeds Wx is lost as excess water (We) and considered not available in further
GAEZcalculations.Itaccountsforthewaterlosteitherbyrunoffordeeppercolation.
3.4.2 Soilmoisturebalanceswithsoilmoistureconservation
In arid and semi arid zones warer conservation management practices are used to copy with
marginal and unreliable, but where still sufficient rainfall occurs to build upadequate soil moisture
storageforsuccessfulgrowingofcrops.Theseareastypicalreceiveannualrainfallbetween300and
600 mm and reference growing periods of 30120 days. Figure 37 shows the occurrence of these
zones that have a total extent of 3.2 billion hectares, almost 24 % of the total world land surface
(excluding Antarctica). The majority of these areas occur in the United States, Argentina, Southern
Africa,NorthAfrica,theSahel,WestandCentralAsiaandAustralia.
Readilyavailable
soilmoisture
(p*Sa)
ETa=ETm
Easilyextractablewater

Unavailablewater
ETa<ETm
Lesseasilyextractablewater
Actualsoil
moisture(Wb)
Fieldcapacity(Sfc)
Permanentwilting
point(Swp)
Availablesoil
moisturestorage
capacity(Sa)
i (S )
Figure37Schematicrepresentationofwaterbalancecalculations
28

Figure38ReferenceLengthofGrowingPeriodZones
Forrainfedproductionwithsoilmoistureconservation,GAEZcalculatesanalternativesoilmoisture
balance with three major modificationscomparedto regular rainfed conditions (i.e.,regular water
balance without assuming specific soil moisture management practices), namely (i) land
managementreducingsoilevaporationoutsidetheperiodofcropgrowth(assumingsoilevaporation
rate of 20% of reference evapotranspiration during noncropped periods), (ii) different rooting
patterns with increased available soil moisture in the soil profile (increasing available soil moisture
holdingcapacitybyabout50%),and(iii)simulatingcropgrowthonlywhenstartseasonavailablesoil
moistureexceedsminimumthresholdlevels,otherwiseconservingsoilmoisturethroughfallow.

Enhancedsoilmoisturebalance
Reference evapotranspiration, stored soil moisture and rainfall are used together with crop
transpiration water requirements of dryland cropping systems and evaporation losses during clean
fallow(notillageorreducedtillage)inayearroundwaterbalance.
Theamountofsoilmoisturestoredinthesoilprofile,andavailabletoacrop,varies,e.g.,withdepth
of the soil profile, the soil physical characteristics, and the rooting pattern of the crop. For crops
relyingonrainfallduringthecropcycleandstoredsoilmoistureavailablesoilmoisture(AWC)isset
toamaximumof150mm,assumingthatthebulkofrootsoccurmainlyinthetop100cmofthesoil
profile(Fischeretal.,2002).
ForcropsforalargeextentrelyingonresidualsoilmoistureAWCis200250mmsoilmoisturecanbe
used since the bulk of the roots in deep soils may move with retracting soil moisture down to a
depth of about 150 cm. This implies an adjustment of the AEZ water balance model parameters
concernstheAWCforcropsgrownonresidualmoisture.
Water conservation by means of zero tillage and reduced tillage systems (which include the use of
herbicides or mechanical means for the removal of weeds to avoid additional transpiration losses)
leadstoreducedsoilevaporation.
In rainfed assessments conventional tillage systems are assumed. For the rainfed with water
conservationassessmentstillagesystemsareassumedthathelpconservingsoilmoistureinthesoil
profile.Anumberoffactorshavebeenconsiderednamely:(i)improvementofsoilmoistureintake,
(ii) reduction of soil evaporation losses, (iii) reduction of percolation losses and (vi) optimal use of
soilmoisture.
29
Box1Tillagesystems
Improvingsoilmoistureintake
Plantcover:slowsrunoff
Cropstubbleanddebris:slowsrunoffandcapturesdriftingsnow
Tillage:improvesinfiltrationintopoorlypermeablesoils
Reductionofevaporationlosses
Mulching:reducesevaporation,discouragesweedgrowth(transpiration)
Tillageoftopsoil:Mayreduceevaporationbybreakingsoilcapillarywatermovementtowardssoilsurface.
Weeding:reducesinterceptionlossesandevaporation.
Reducingofpercolationlosses
Increaseoforganicmatter:Improveavailablewaterholdingcapacitiesofsoilprofile
Optimizingsoilmoisture
Reduceseedrate/increasespacing:Increasesmoistureavailableperplant
Fallowing:(cleanfallowreducestranspirationofweeds)
Notillage:Reducesevaporationlosses
Reducedtillage:Reducesevaporationlosses
Subtillage:Reducesevaporationlossesandmayreducesoilcapillarywatermovementtowardssoilsurface.
Measures as described in box 1 have been accounted for. It has been assumed that depending on
soilandterrainconditionsadaptedmeasuresaretakentoachieveoptimalwaterconservationfrom
rainfallwhilepreventingsoilerosion.Wherepossible(forinstanceinabsenceofproblemslikerunoff
due to low soil infiltration rates because of heavy topsoils or sealing characteristics of the soil
surface)zero tillage withclean fallow is assumed. For soils with runoff due tolow infiltration rates,
prevalenceof topsoil sealing andother specific topsoil characteristics and unfavorable soil capillary
conditions,reducedtillageandsubtillagesystemsareassumed(FAO,1984).Alsoitisassumedthat
crop stubble, crop debris and mulching practices are used where practical and beneficial for soil
moisture conservation. In summary best practice visvis soil moisture conservation is assumed in
theAEZwaterbalance.
Soilmoistureconditionsatplanting
Theplantingofacropisassumedwhensufficientmoisturehasbeenaccumulatedinthesoilprofile
(AWC) to cover water requirements of a major part of the crop cycle. This latter value is a model
parameter, which may vary according to expected rainfall, specific crop water requirements and
evaporative demand of the atmosphere. The required AWC has been set to 175 mm. In case the
water balance shows that the 175 mm are not met the crop is assumed not planted and a fallow
periodtoconserveadditionalsoilmoistureisintroduceduntilanextfeasibleplantingdate.
Crop/LUTs for for rainfed with waterconservation assessments of suitability and productivity
includeLUTs ofwheat,barley, grain andsilagemaize,sorghum, millets,chickpea,cowpea,soybean
andrape.
A schematic overview of various temperature/cover phases of the water balance calculations as
performedinAEZforbothrainfalldependentcropgrowthaswellascropsgrowthrelyingonresidual
soil moisture is presented in Table 34. It shows how the various steps of AEZ water balance are
parameterizedandhowcalculationsareinfluencedbytemperatureandsnowcoverconditions.

30
Table34Waterbalanceparametersbytemperatureandcover
Periods P1a P1b P2a P2b P3 P4 P5 P6 P7 P8 P9 P10 P11
Cover(1) Fallow
Crop
Stage1
Crop
Stage2
Crop
Stage3
Fallow
MeanTemperature Tm<0 Tm<0 Tm<0 Tm<0 Tm05 Tm05 Tm>5 Tm>5 Tm>5 Tm>5 Tm>5 Tm05 Tm<0
Maximum
temperature
Tmax<0 Tmax<0 Tmax>0 Tmax>0 Tmax>0 Tmax>0 Tmax>0 Tmax>0 Tmax>0 Tmax>0 Tmax>0 Tmax>0 Tmax>0
Snowcover Snow Nosnow Snow Nosnow Snowmelt Nosnow Nosnow Nosnow
Evapo(transpi)ration EVsnow EVfroz1 EVsnow. EVfroz2 EVsnow
ETsoil/
EVsoil
ETsoil/
EVsoil
ETm ETm ETm
ETsoil/
EVsoil
ETsoil/
EVsoil
EVfroz
Rainfed/
conventionaltillage
0.2ETo 0.2Eto 0.2ETo 0.2ETo 0.2ETo 0.3ETo 0.4ETo kc*ETo kc*ETo kc*ETo 0.4ETo 0.3ETo 0.2ETo
Rainfed/zerotillage
+weedremovalor
reducedtillage
0.2ETo 0.2Eto 0.2ETo 0.2ETo 0.2ETo 0.2ETo 0.2ETo kc*ETo kc*ETo kc*ETo 0.2ETo 0.2ETo 0.2ETo

Cover(2) Fallow
Evapo(transpi)ration EVsnow EVfroz1 EVsnow EVfroz2 EVsnow
ETsoil/
EVsoil
ETsoil/
EVsoil
ETsoil/
EVsoil
ETsoil/
EVsoil
ETsoil/
EVsoil
ETsoil/
EVsoil
ETsoil/
EVsoil
EVfroz
Rainfed/
conventionaltillage
0.2ETo 0.2Eto 0.2ETo 0.2ETo 0.2ETo 0.3ETo 0.4ETo 0.4ETo 0.4ETo 0.4ETo 0.4ETo 0.3ETo 0.2ETo
Rainfed/zerotillage
+weedremovalor
reducedtillage
0.2ETo 0.2Eto 0.2ETo 0.2ETo 0.2ETo 0.2ETo 0.2ETo 0.2ETo 0.2ETo 0.2ETo 0.2ETo 0.2ETo 0.2ETo
Tm=meantemperature,Tmax=maximumtemperature;ETo=referenceevapotranspiration;EVsnow=sublimationrateofsnow(=0.2*ETo);EVfroz.=evaporationfromfrozensoil(=0.2*ETo);EVsoil=evaporation
fromnonfrozenbaresoil(=0.2*ETo);ETsoil=evapotranspirationfromnonfrozensoilandweeds(=0.3or0.4*ETo).ETm=maximumcropevapotranspiration(=kc*ETo,wherecropcoefficientkcrangesarecrop
stagedependent).
31
3.4.3 Lengthofgrowingperiod(LGP)
The agroclimatic potential productivity of land depends largely on the number of days during the
year when temperature regime and moisture supply are conducive to crop growth and
development. This period is termed the length of the growing period (LGP). The LGP is determined
based on prevailing temperatures and the above described water balance calculations for a
referencecrop.Inaformalsense,LGPreferstothenumberofdayswhenaveragedailytemperature
is above 5
o
C (i.e. within LGP
t5
) and ETa is above a specific fraction of ETo. In the current GAEZ
parameterization, LGP days are considered when ETa 0.5 ETo (FAO 197881; FAO 1992), which
aims to capture periods when sufficient soil moisture is available to allow the establishment of a
referencecrop.Figure39presentsamapofreferencelengthofgrowingperiod,whichisbasedon
soilmoistureholdingcapacityof100mm.

Figure39Referencelengthofgrowingperiod
The length of growing period data is also used for the classification of general moisture regimes
classes.TheGAEZmoistureregimesnomenclatureanddefinitionsarepresentedinTable35
Table35Moistureregimes
Lengthofgrowingperiod(days) MoistureRegime
0 Hyperarid
<60 Arid
60to119 Drysemiarid
120to179 Moistsemiarid
180to269 Subhumid
270to364 Humid
365(yearroundgrowingperiod) Perhumid
The moisture regime within a LGP is characterized by different water supply conditions as follows:
Growing period days without water stress (ETa=ETm): When ETa equals ETm crop water
requirementsarefullymet(i.e.nowaterstressforplantsoccurs).Fromasoilwaterbalancepointof
viewtheseLGPdayscanfurtherbedifferentiatedasfollows:
1. Dailyrainfallishigherthancropwaterrequirements(P>ETm)andstoredsoilmoistureislessthan
fieldcapacity(Wb<Sfc).Excessrainfallnowaddstoreplenishthesoilmoisturestorage.
2. Dailyrainfallishigherthancropwaterrequirements,P>ETm,andsoilmoistureisatfieldcapacity
(Wb=Sfc).Inthiscaseexcessprecipitationislosttosurfacerunoffand/ordeeppercolation.
3. Days when rainfall falls short of crop water requirements (P<ETm) but easily available soil
moisture exceeds crop water requirements (Wb>(ETmP)+Wr. In this case ETa equals ETm and
thesoilmoisturecontentinthesoilprofileisdecreasing.
Growing period days with water stress (ETa<ETm): ETa falls short of ETm. The crop experiences
waterstressasnotenoughreadilyavailablewatercanbeobtainedfromrainfallormoisturestored
inthesoilprofile.Waterstressimpliesthatcropgrowthandyieldformationarereduced.
32
Discontinuousgrowingperiods
TotalannualLGPdaysmaybeinonecontinuousperiodormayoccurastwoormorediscontinuous
growing periods. When moisture becomes insufficient (ETa < 0.5 ETo), LGP ends and/or is
interruptedbyadryperiod.Inthecaseoftemperaturelimitations(Ta<5C),LGPisinterruptedby
eitheradormancybreakoracoldbreak.Thisdistinctionisdeterminedonthebasisoftemperature
limits for survival of hibernating crops. During a dormancy period hibernating crops can survive as
opposedtoacoldbreakwhentemperaturedropsbelowacropspecificcriticaltemperaturelimit.
GAEZdeterminesindividualcontinuousLGPs.VarioussoilmoisturesupplystagesduringtheLGPare
recordedandvariousindicatorsarecalculatedasfollows:
1. Totalnumberofgrowingperioddays
2. Numberofgrowingperioddays,duringwhichETa=ETm
3. NumberofgrowingperioddayswhenP>ETm
4. Numberofindividualgrowingperiods
5. Lengthofindividualgrowingperiods
6. Numberofgrowingperioddaysinindividualgrowingperiods
7. NumberofdayswhenP>ETminindividualgrowingperiods
8. Begindateofindividualgrowingperiods
9. Enddateofindividualgrowingperiods
10. Temperatureprofile,i.e.,numberofgrowingperioddaysoccurringin5
o
Cstepsbetween<
5
o
Cto>30
o
C,withincreasingtemperaturetrend.
11. Temperatureprofile,i.e.,numberofgrowingperioddaysoccurringin5
o
Cstepsbetween<
5
o
Cto>30
o
C,withdecreasingtemperaturetrend.
12. Temperaturesumsforgrowingperioddaysaboverespectively0
o
C,5
o
Cand10
o
C
13. TemperaturessumsduringthelongestLGPaboverespectively0
o
C,5
o
Cand10
o
C
3.4.4 Multiplecroppingzonesforrainfedcropproduction
IntheGAEZcropsuitabilityanalysis,theLUTsconsideredrefertosinglecroppingofsolecrops,i.e.,
eachcropispresumedtooccupythelandonlyonceayearandinpurestand.Consequently,inareas
where the growing periods are sufficiently long to allow more than one crop to be grown in the
same year or season, single crop yields do not reflect the full potential of total time and space
available per unit area of land for rainfed production. To assess the multiple cropping potential, a
number of multiple cropping zones have been defined through matching both growth cycle and
temperaturerequirementsofindividualsuitablecropswithtimeavailableforcropgrowth.Forrain
fed conditions this period is approximated by the LGP, i.e., the number of days during which both
temperatureandmoistureconditionspermitcropgrowth.
For the definition of multiple cropping zones four types of crops are distinguished: thermophilic
crops requiring warm temperatures, cryophilic crops performing best under cool and moderately
cool conditions, hibernating crops, and wetland crops with specific water requirements.
Furthermore,thecropsaresubdividedaccordingtogrowthcyclelength,namelyoflessormorethan
120 days duration, respectively. According to the above criteria, the following nine zones were
classifiedandmapped(seeFigure310):
A. Zoneofnocropping(toocoldortoodryforrainfedcrops)
B. Zoneofsinglecropping
C. Zoneoflimiteddoublecropping(relaycropping;singlewetlandricemaybepossible)
D. Zone of double cropping (sequential cropping; double cropping with wetland rice not
possible)
E. Zoneofdoublecropping(sequentialcropping;wetlandricecroppossible)
F. Zone of limited triple cropping (partly relay cropping; no third crop possible in case of two
wetlandricecrops)
33
G. Zone of triple cropping (sequential cropping of three shortcycle crops; two wetland rice
cropspossible)
H. Zoneoftriplericecropping(sequentialcroppingofthreewetlandricecropspossible)
Delineation of multiple cropping zones for rainfed conditions is solely based on agroclimatic
attributes calculated during AEZ analysis. The following attributes were used in the definition of
croppingzones:
LGP length of growing period, i.e., number of days when temperature and soil moisture permit
cropgrowth.
LGP
t=5
numberofdayswithmeandailytemperaturesabove5
o
C.
LGP
t=10
numberofdayswithmeandailytemperaturesabove10
o
C.
TS
t=0
accumulatedtemperature(degreedays)ondayswhenmeandailytemperature0
o
C.
TS
t=10
accumulatedtemperature(degreedays)ondayswhenmeandailytemperature10
o
C.
TSG
t=5
accumulatedtemperatureduringgrowingperiodwhenmeandailytemperature5
o
C.
TSG
t=10
accumulatedtemperatureduringgrowingperiodwhenmeandailytemperature10
o
C.
Table 35 and 36 summarize the delineation criteria for multiple cropping zones under rainfed
conditionsinrespectivelythetropicsandthesubtropics/temperatezones.

Figure310Multiplecroppingzonesforrainfedconditions

Figure311Multiplecroppingzonesforirrigatedconditions
34
Table36Delineationofmultiplecroppingzonesunderrainfedconditionsinthetropics
Zone LGP LGP
t=5
LGP
t=10
TS
t=0
TS
t=10
TSG
t=5
TSG
t=10
A
19)

B
20)
45 120 90 1600 1000
C
21)

220 220 5500

200 200 120 6400 n.a. 3200 2700
180 200 7200
D
22)

270 270 5500

240 240 165 6400 n.a. 4000 3200
210 240 7200
E n.a. n.a. n.a. n.a. n.a. n.a. n.a.
F 300 300 240 7200 7000 5100 4800
G n.a. n.a. n.a. n.a. n.a. n.a. n.a.
H 360 360 360 7200 7000

Table37Delineationofmultiplecroppingzonesunderrainfedconditionsinsubtropicsandtemperatezones
Zone LGP LGP
t=5
LGP
t=10
TS
t=0
TS
t=10
TSG
t=5
TSG
t=10
A
19)

B
20)
45 120 90 1600 1000
C 180 200 120 3600 3000 3200 2900
D 210 240 165 4500 3600 4000 3200
E 240 270 180 4800 4500 4300 4000
F 300 300 240 5400 5100 5100 4800
G 330 330 270 5700 5500
H 360 360 330 7200 7000
19)
AppliesifconditionsforzoneB(singlecropping)arenotmet.

20)
Theprogramtestsifatleastoneofthecrop/LUTsisagroclimaticallysuitableintherespectivegridcell.
21),22)
Refersto,respectively,highland,midhighland,andlowlandareasinthetropics.
35

3.4.5 Equivalentlengthofthegrowingperiod
ToaccountforsignificantdifferencesinwetnessconditionsoflongLGPs(>225days)equivalentThe
reference LGP accounts for both temperature and soil moisture conditions. Therefore, the wetness
conditions in different locations can be better compared by the socalled equivalent LGP (LGPeq,
days) which is calculated on the basis of regression analysis of the correlation between reference
LGPandthehumidityindexP/ETo.
A quadratic polynomial is used to express the relationship between the number of growing period
days and the annual humidity index. Parameters were estimated using data of all gridcells with
essentiallyyearroundtemperaturegrowingperiods,i.e.withLGPt
5
=365.
2
14.0 293.66 61.25 ; 2.4;
366 ; 2.4;
P P P
when
ETo ETo ETo
LGPeq
P
when
ETo

>

The equivalent LGP is used in the assessment of agroclimatic constraints which relate
environmental wetness with the occurrences of pest and diseases and workability constraints for
harvestingconditionsandforhighmoisturecontentofcropproduceatharvesttime.
3.4.6 NetPrimaryProductivity(NPP)
Net primary production (NPP) is estimated as a function of incoming solar radiation and soil
moistureattherhizosphere.Actualcropevapotranspiration(ETa)hasacloserelationshipwithNPP
of natural vegetation as it is quantitatively related to plant photosynthetic activity which is also
driven by radiation and water availability. In GAEZ, NPP is estimated according to Zhang (1995) as
follows:
d
A
ETa NPP
0
=

TheETaareaccumulatedestimatesofdailyETafromtheGAEZwaterbalancecalculationsforthe
specific water holding capacity of individual soil types. The variable A
0
is a proportionality constant
dependingondiffusionconditionsofCO
2
anddisanexpressionofsensibleheat.TheratioA
0
/dcan
beapproximatedbyafunctionoftheradiativedrynessindex(RDI)(Uchijima,1988).
( ) ( ) RDI RDI RDI f
d
A
+ = 25 . 6 87 . 9 exp
0

with:
P
Rn
RDI
j
j
j
12
1
12
1
=
=

=
wheretheRnisaccumulatednetradiationfortheyearandPisprecipitationfortheyear.
InGAEZ,twoseparateevaluationsoftheNPPfunctionareperformed:
a. For NPP estimates under natural, i.e rainfed conditions, RDI is calculated from prevailing
net radiation and precipitation of a grid cell and ETa is determined by the GAEZ reference
waterbalance:
36
( ) RDI RDI ETa NPP
rf
+ =

25 . 6 87 . 9 exp

b. ForanNPPestimateapplicableunderirrigationconditions,ETa=ETmisassumedandaRDI
of1.375isused,whichresultsinamaximumforthefunctiontermapproximatingtheA
0
/d
ratio:
( ) 375 . 1 25 . 6 87 . 9 exp 375 . 1 + =

ETa NPP
ir

3.5 GridcellanalysisModuleI
Results of the calculation procedures of Module I are presented for a sample gridcell in Appendix
34.Theexampleprovidesoutputdataoftheagroclimaticdataanalysisforreferenceclimate(1962
1990)foragridcellnearIlonga,Tanzania.
3.6 DescriptionofModuleIoutputs
ModuleIproducestwodetailedoutputfiles,whichrespectivelycontainthecalculatedindicatorsof
thermalandmoistureconditionsineachgridcell.ThesefilesarethenusedtogeneratevariousGIS
raster maps of the agroclimatic analysis results for visualization and download, but primarily as
inputtothecomputationsinModulesII,III,andV.

The output variables from Module I are described in Appendix 32. Subroutine descriptions of
ModuleIaredescribedinAppendix33.

37
4 ModuleII(Biomasscalculation)
4.1 Introduction
ThemainpurposeofModuleIIisthecalculationofagroclimaticallyattainablebiomassandyieldfor
specific land utilization types (LUTs) under various input/management levels for rainfed and
irrigatedconditions.
ModuleIIconsistsoftwosteps:
(i) Calculation of crop biomass and yield potentials considering only prevailing radiation and
temperatureconditions,and
(ii) Computationofyieldlossesduetowaterstressduringthecropgrowthcycle.Theestimation
is based on rainfed crop water balances for different levels of soil water holding capacity,
with and without water conservation measures. Yield estimation for irrigation conditions
assumesthatnocropwaterdeficitswilloccurduringthecropgrowthcycle.
TheactivitiesandinformationflowofModuleIIareshowninFigure41.

Figure41InformationflowofModuleII
4.2 LandUtilizationTypes
Differences in crop types and production systems are empirically characterized by the concept of
Land UtilizationTypes (LUTs). A LUT consists of a set of technical specifications for crop production
within a given socioeconomic setting. Attributes specific to a particular LUT include agronomic
information,natureofmainproduce,watersupplytype,cultivationpractices,utilizationofproduce,
and associated crop residues and byproducts. The GAEZ v3.0 framework distinguishes nearly 900
crop/LUT and management combinations, which are separately assessed forr rainfed with and
without moisture conservation and iirigated conditions. These LUts are madeup of 49 different
food, feed, fiber, and bioenergy crops (Appendix 41, Table A42). The calculated yield of each
38
crop/LUT is affected by water source and the intensity of input and management assumed to be
applied. In GAEZ, three generic levels of input/management are defined: (i) low, intermediate, and
highinputlevel.

Lowlevelinputs
Under a low level of inputs (traditional management assumption), the farming system is largely
subsistence based. Production is based on the use of traditional cultivars (if improved cultivars are
used, they are treated in the same way as local cultivars), labor intensive techniques, and no
applicationofnutrients,nouseofchemicalsforpestanddiseasecontrolandminimumconservation
measures.

Intermediatelevelinputs
Under an intermediate level of input (improved management assumption), the farming system is
partlymarketoriented.Productionforsubsistencepluscommercialsaleisamanagementobjective.
Productionisbasedonimprovedvarieties,onmanual laborwithhandtools and/or animaltraction
and some mechanization, is medium labor intensive, uses some fertilizer application and chemical
pestdiseaseandweedcontrol,adequatefallowsandsomeconservationmeasures.

Highlevelinputs
Under a high level of input (advanced management assumption), the farming system is mainly
market oriented. Commercial production is a management objective. Production is based on
improved or high yielding varieties, is fully mechanized with low labor intensity and uses optimum
applicationsofnutrientsandchemicalpest,diseaseandweedcontrol.
InGAEZ,thisvarietyinmanagementandinputlevelsistranslatedintoyielddifferencesbyassigning
differentparametersforLUTsdependingontheinput/managementlevel,e.g.suchasharvestindex
andmaximumleafareaindex.
LUTsareparameterizedtoreflectenvironmentalandecophysiologicalrequirementsforgrowthand
developmentofdifferentcroptypes.Numericalvaluesofcropparametersarevarieddependingon
theassumedinput/managementleveltowhichLUTsaresubjected.
4.3 ThermalsuitabilityscreeningofLUTs
As initial criteria to screen the suitability of gridcells for the possible presence of individual LUTs,
GAEZteststhematchofprevailingconditionswiththeLUTstemperaturerequirements.
ThereareseveralstepsappliedtotestthematchbetweenthermalconditionsandLUTtemperature
(and relative humidity) requirements: (i) Thermal (latitudinal) climatic conditions; (ii) permafrost
conditions; (iii) length of temperature growing period (LGP
t=5
); (iv) length of frost free period
(LGP
t=10
);(v)temperaturesums(Tsum
t
);(vi)temperatureprofiles;(vii)vernalizationconditions;(viii)
diurnal temperature ranges (for selected tropical perennials); and (ix) relative humidity conditions
(for selected tropical perennials). LUT specific requirements are individually matched with
temperatureregimes(andrelativehumidity)prevailinginindividualgridcells.Matchingistestedfor
the full range of possible starting dates and resulting in optimum match, suboptimum match and
not suitable conditions. The optimum and suboptimum match categories are considered for
further biomass and yield calculations. The thermal suitability screening procedure is sketched in
Figure42.
39
T Thermalclimate
T
T
Temperaturesum
Temperatureprofile
Permafrost
Diurnaltemperature
range
T
Temperature
growingperiod
T Frostfreeperiod
Optimalconditions
Suboptimal
conditions
Notsuitable
conditions
Relativehumidity*
OptimalMatch NoMatch SuboptimalMatch
*Relativehumidityrequirementsforselectedperennialsarescreenedinthisprocedure
Vernalization
Temperature
growingperiod
Frostfreeperiod
Temperaturesum
Temperatureprofile
Vernalization
T
Diurnaltemperature
range
RelativeHumidity*

Figure42Schematicrepresentationofthermalsuitabilityscreening
Thermalclimate
In Module II, the GAEZ model first checks if an LUT is deemed suitable to grow in the climate
prevailing in a gridcell. The procedure aims to capture compatibility of the LUT requirements in
terms of overall temperature requirements, climatic seasonality and seasonal daylength enabling
thescreeningforrespectivelylongday,dayneutralandshortdayscropLUTs.
The screening of crop/LUTs with regard to prevailing climate results in a yes/no filter for further
calculationstobeperformedforanLUTinindividualgridcells.
Permafrost
Areas with reference continuous and discontinuous permafrost are considered not suitable. Gelic
soils, indicating permafrost, that occur outside the reference continuous and discontinuous
permafrostzonesaredealtwithintheagroedaphicsuitabilityassessment.
Temperaturegrowingperiod
The period during the year when temperatures are conducive to crop growth and development is
representedbythetemperaturegrowingperiod,whichisdefinedastheperiodduringtheyearwith
meandailytemperatureabove5
o
C,alsoreferredtoasLGP
t=5
.Growthcyclelengthsofcrop/LUTsare
40
matched with LGP
t=5
. The result of the matching provides optimum match when the growth cycle
cangenerouslybeaccommodatedwithinLGP
t=5
.Otherwisethematchisconsideredsuboptimumor
notsuitable.
Hibernating crops survive low temperatures, e.g. during a winter season, by entering into a
dormancyperiod.GAEZconsidersfourhibernatingcropspecies:winterwheat,winterbarley,winter
ryeandwinterrape.Thesearetheonlycrop/LUTsallowedtoprevailatdailyaveragetemperatures
<5
o
C. A dormancy period is considered when Ta ranges between 5
o
C and the cropspecific critical
low temperature for coldbreak. If the dormancy period is longer than 200 days, or daily average
temperatures drop bellow critical thresholds (see below), the LUT is considered to be not suitable.
Foreffectofsnowcoveronlowtemperaturethresholdsforcoldbreak,seeFischeretal.,2002.
Frostfreeperiod
Difference in sensitivityof crop/LUTs forearly and latefrost is accountedforthroughthematching
ofcrop/LUTgrowthcycleswithprevailingfrostfreeperiods.Thefrostfreeperiodisapproximatedby
the period duringtheyear when mean daily temperatures are above 10
o
C (LGP
t=10
).Depending on
thesensitivityofaspecificcrop/LUTthematchingofgrowthcyclelengthwiththeavailablefrostfree
periodprovidesoptimummatch,suboptimummatchornotsuitableconditions.
Temperaturesum
Individual crop/LUT heat unit requirements are matched with temperature sums during the
crop/LUT growth cycle duration (Tsum
c
). The Tsum
c
is defined as the sum of mean daily
temperaturescalculatedfromabasetemperatureof0
o
C.
ThematchofthecropLUTheatunitrequirementswiththeprevailingTSUMareoptimum,whenthe
requirementsarefallingwithintheoptimumTsum
c
range,suboptimumwhenfallinginTsum
c
range
conditions and not suitable when prevailing Tsum
c
s are too high are too low. Optimum and sub
optimumTsum
c
rangesarepresentedforallcrop/LUTsintheAppendix43.
Temperatureprofile
The temperature profile requirements are crop/LUTspecific rules that take into account classes of
mean daily temperatures (Ta). These classes in 5
o
C intervals are defined separately by days with
increasingordecreasingtemperaturetrends(Fischeretal.,2002).GAEZhasdefinedindetailforall
crop/LUTs temperature profile requirements. Two temperature profile requirements data sets for
respectivelyoptimumconditionsandforsuboptimumconditionhavebeenspecifiedforuseinGAEZ
(Appendix4.3)
Potential crop calendars of each LUT are tested for the match of crop/LUT temperature profile
requirements and prevailing temperature profiles, while considering growth cycle starting days
within the length of the growing period for rainfed conditions, and within the year for irrigated
conditionsseparately.ForallfeasiblecropcalendarswithintheLGP(rainfed)orwithintheyearthe
prevailing temperature profile conditions are tested against optimum and suboptimum crop
temperatureprofilerequirementsandineachcaseanoptimum,suboptimumornotsuitable
matchisestablished.
Vernalization
Somecropsrequireavernalizationperiod(i.e.dayswithcoldtemperatures)forperformingspecific
phenological development phases such as flowering. The production of flowers and grains, which
directly influences crop yield, is dependenton theextent and intensity of exposure to periodswith
cold temperature. This cold temperature requirement is measured in vernalization days. In GAEZ,
therearefourhibernatingcropsthatneedtofulfillvernalizationrequirementsinordertoproduce:
winterwheat,winterbarley,winterryeandwinterrape.DetailsareprovidedinAppendix44.

41
Diurnaltemperaturerangeandrelativehumidityconditions
For a number of tropical perennial crops such as coconut, cacao and oil palm diurnal temperature
ranges as well as relative humidity levels affect crop growth and yield. For these perennials
requirements visvis optimum, suboptimum and not suitable diurnal temperature ranges as well
aspermissiblerangesofrelativehumidityhavebeendefined.
Combiningtemperaturerelatedconstraints.
In case of a suboptimum conditions for crop cultivation, the degree of suboptimality is derived
through quantifying for each tested requirement a constraint factor fc
1k
, k=1, , K, based on the
distance of the calculated indicator from respectively the thresholds for optimum and sub
optimum matches. At the threshold defining suboptimum conditions it is assumed that crop
growth and yield are reduced by 25% whereas no reduction is applied for values exceeding the
thresholdforoptimumconditions.Themostlimitingtemperaturerelatedconstraintfactoristhen
used to reduce potential yields calculated in Module II. For that a yield reduction factor
c
1
=
min
k
{c
1k
, k = 1, , K ] is calculated representing the minimum, i.e., most severe, of the
individualtemperaturereductionfactors.
4.4 Biomassandyieldcalculation
In this section the calculation procedures of constraintfree biomass and yield (i.e. carbon
accumulation driven mainly by prevailing radiation and temperature regimes in a gridcell) are
explained.TheproceduresusedarebasedontheecophysiologicalmodeldevelopedbyA.H.Kassam
(1977)
The constraintfree crop yields calculated in the AEZ biomass model reflect yield potentials with
regard to temperature and radiation regimes prevailing in the respective gridcells. The model
requiresthe following cropcharacteristics: (a) Length ofgrowth cycle (days from emergenceto full
maturity); (b) minimum temperature requirements for emergence; (c) maximum rate of
photosynthesis, (d) respiration rates for leguminous and non leguminous crops as functions of
temperature;(e)lengthofyieldformationperiod;(f)leafareaindex(LAI)atmaximumgrowthrate;
(g) harvest index (Hi); (h) crop adaptability group, and (i) sensitivity of crop growth cycle length to
heat provision. The biomass calculation also includes simple procedures to account for different
levels of atmospheric CO2 concentrations. Appendix 45 presents details of the calculation
proceduresandAppendix46providesthemodelparameters.
The results of the biomass and yield calculation depend on timing of crop growth cycle (crop
calendar). Maximum biomass and yields are separately calculated for irrigated and rainfed
conditions,asfollows:
Irrigation:
Foreachdaywithinthewindowoftimewhencroptemperatureandradiationrequirementsaremet
optimallyoratleastsuboptimally,theperiodresultinginthe highestbiomassandyield isselected
tosetthecropcalendaroftherespectivecrop/LUTforaparticulargridcell.
Rainfed:
Within the window with optimumor suboptimum temperature conditions, and starting within the
duration of the moisture growing period, the period resulting in the highest expected (moisture
limited) yield is selected to represent maximum biomass and yield for rainfed conditions of the
respectivecrop/LUTforaparticulargridcell.
In other words, for each crop type and gridcell the starting and ending dates of the crop growth
cycle are determined optimally to obtain best crop yields, separately for rainfed and irrigated
conditions. This procedure also entails adaptation of crop calendar (smart farmer) in simulations
with yearbyyear historical weather conditions, or under climate distortions applied in accordance
withvariousclimatechangescenarios.
NetbiomassandyieldsformostLUTsinGAEZareexpressedinkilosofdrymatter(DM)perhectare
with the exception ofsome oil crops (yield expressed as oil), sugar crops (yield expressedas sugar)
andcotton(yieldexpressedaslint).FordetailsseeTable911.
4.5 Waterlimitedbiomassproductionandyields
Under rainfed conditions, water stress may occur during different stages of the crop development
reducingbiomassproductionandtheyieldsachieved.InGAEZ,waterrequirementsforeachLUTare
calculated and taken into account in the calculation of LUTspecific waterbalance and actual
evapotranspirationinagridcell.Awaterstressyieldreductionfactor(fc
2
)iscalculatedandapplied
tothenetbiomass(Bn)andpotentialyield(Yp)calculated.
4.5.1 Cropwaterrequirement
The total water requirement of a crop without any water stress is assumed to be the cropspecific
potential evapotranspiration (ETm). ETm is calculated in proportion to reference potential
evapotranspiration(ETo),asinModuleI,multipliedbycropandcropstagespecificparameterskc.
Thevaluesofkcfordifferentstagesofcropdevelopmentaregivenasinputparameters.

Figure43Schematicrepresentationofkcvaluesfordifferentcropdevelopmentstages
Thefourstagesofcropdevelopment(days)aredenotedasinitial(d1),vegetative(d2),reproductive
(d3) and maturation stage (d4). For each stage, input parameters define the length of each crop
stageasapercentageoftotalcyclelength(GC).Threeinputparametersdefinethecropcoefficient
forwaterrequirement(kc,fractional)throughoutstagesd1(kc1)andd3(kc2)andattheendpoint
of stage d4 (kc3). The values of kc throughout periods d2 and d4 are then calculated by linear
interpolation.Alternatively,anaveragekcparameterrepresentativefortheentiregrowthcycle(kc0)
canbespecifiedtocalculateanoverallwaterrequirementofthecrop.

d1 d2 d3 d4
kc1

kc2
kc3

0.5

kc

1.0

Cycle

initial maturation vegetative reproductive

43
Thevalueofkcforaparticulardayjisdefinedby:
( )
( ) ( )
1
2
3
4
1
2 1
1 1
2
2
3 2
2 1 2 3
4
j
kc
j D
kc kc
kc j d
j D
d
kc
j D kc
j D kc kc
kc j d d d
d

+


=

+ + +

4.5.2 Yieldreductionduetowaterdeficits
Yieldreductioninresponsetowaterdeficits iscalculated as a functionoftherelationshipbetween
actual crop evapotranspiration (ETa, mm/day) and maximum crop evapotranspiration (ETm,
mm/day), both accumulated within the four crop stages. Daily ETa is calculated from the water
balanceasdescribedalsoinModuleI,withthedifferenceofbeingLUTspecificinModuleII.Also,in
ModuleII,thevalueofsoilwaterdepletionfraction(p)varieswiththeparticularcrop.
Thesensitivityofeachcroptowaterstressisexpressedbythevalueofthewaterstresscoefficient
(ky,fractional),anLUTspecificparameterwhichchangeswithcropdevelopmentstage.Thereareky
values for each of the four development stages (ky1, ... , ky4) and also an average ky value for the
overall crop growth cycle (ky0). GAEZ uses both the crop stage specific coefficients and estimated
waterdeficitsandtheoverallvalueofkc0tocalculateawaterstressyieldreductionfactor(fc
2
).
1
2
1
1 0 1
TCL
T
TCL
ETa
fc ky
ETm


=

, , 1,...,4
j j
j k j k
k D k D
TETa ETa TETm ETm j

= = =


j
4
1 j j
j
j
CS
2
TETm
TETa
1 ky 1 fc

=
) , min(
2 2 2
T CS
fc fc fc = ,
where TETa
j
and TETm
j
are respectively total actual evapotranspiration and total potential
evapotranspirationfordaysduringcropstaged
j
.
The weighting coefficients
j
add to one and are taken as the relative length of each crop
development stage, Hence, fc
2
is taken as the minimum of factor fc
T
2
representing the effect of
overallwaterdeficit andthe factorfc
cs
2
represents theweightedeffectofcropstage specificwater
stress.
Water limited yield (Y
w
) is then calculated as potential yield (Y
p
) multiplied by the waterstress
reductionfactorfc
2
.
2
fc Y Y
p w
=

4.5.3 AdjustmentofLAIandHiinperennialcrops
Perennial crops have limited opportunity to express their genetic potential to expand canopy (i.e.
develop leaf area index, LAI) and to complete formation of yield components (e.g. fill grains) if the
periodforgrowthistooshortinagivenlocation.Thesetwoaspectsofperennialcropsarecaptured
44
in GAEZ by adjustment factors for LAI (fp
LAI
) and for harvest index (fp
Hi
) which are related to the
lengthoftheeffectivegrowingperiod(LGP
eff
,days).
eff HI
HI
HI
LGP
fP

=
eff LAI
LAI
LAI
LGP
fP

=
Foreachrespectivevariable,twoparametersareusedtocalculatetheadjustmentfactorsforHIand
LAIofperennials.Theseparametersrelatetothecriticalandlimitingeffectivelengthofthegrowing
periodbelowwhichayieldreducingadjustmentisappliedornoyieldisobtained.Also,notethata
perennialcropmaybeconsiderednotsuitableforlevelsofLGP
eff
wellabove
Hi
or
LAI
.Theeffective
growingperiod(LGP
eff
,days) accountsforthedays intheyearwhenperennialcrops areeffectively
growing. Under rainfed conditions it falls within the LGP determined for a particular grid cell and
therefore the period of vigorous growth may be limited by rainfall and soil moisture availability. It
excludestheperiodofdormancyorrestingofperennialcrops.
TheparameterizationforperennialcropsusedinGAEZisgiveninTable41.

Table41Parameterizationusedtocorrectharvestindex(Hi)andleafareaindex(LAI)forsuboptimum
lengthoftheeffectivegrowthperiod(LPGeff)
fP
LAI
fP
HI
Crop
LAI
.
LAI
.
Hi

Hi

Cassava 0 240 40 140

Sugarcane 0 250 90 270
Banana 0 330 210 120
Oilpalm 0 360 210 150
Olive 0 216 0 0
Yellowyam 0 270 110 220
Cocoyam 0 270 110 220
Citrus 0 150 90 60
Cocoa 0 270 120 150
Tea 0 270 120 150
Coffee(arabica) 0 270 120 120
Coffee(robusta) 0 270 120 150
Alfalfa 0 180 30 150
Miscanthus 30 135 30 135
Switchgrass 30 135 30 135
Reedcanarygrass 0 135 0 135

ThefinalHiandLAIforperennialsarethencalculatedas:
Hi max per
fP Hi Hi =
LAI max per
fP LAI LAI =

45
4.6 Cropcalendar
The crop calendar (i.e. sowing and harvesting dates) for a given LUT and gridcell is determined by
identifying the sowing date that leads to the highest attainable yield. GAEZ tests all possible
LUT/sowingdatescombinationswithineachgridcell.
ForeachLUT,thetotalcropcycleexpectedfortheaverageclimate(30yeartimeperiodfrom1961
1990) is given in days as an input parameter. For the average base climate, an accumulated
temperature sum (Tsum
5
) is calculated during each crop LUT. This cropspecific value of Tsum
5
is
assumedtorepresentforalocationthespecificcropcyclerequirementoftheLUT.Whensimulating
individual years, the crop cycle is adjusted until the specific Tsum
5
is reached, as calculated for
averageclimateconditions,e.g.isshortenedinyearswarmerthannormal.
ForrainfedproductionGAEZcalculatespotentialcropyieldsbyshiftingcomputedcalendarswithin
thepermissiblepartoftheLGP,andselectsthestartdateofthecropwhenyieldisthehighest.This
optimum crop calendar for rainfed conditions is reflecting, for a particular crop/LUT, the optimum
combinationofradiationregime,temperatureregimeandsoilmoistureavailability.
For irrigated production GAEZ tests all possibilities of crop yield performance in LGP
t5
(i.e., in the
period during the year when Ta >5
o
C) and selects the period with highest attainable yields, thus
driven mainly by radiation and temperature regime. Alternatively, GAEZ could also use a selection
criterion which would account for the tradeoff between additional water use and additional
additionalyieldgenerated.
4.7 CO2fertilizationeffectoncropyields
ThefertilizationeffectofincreasingatmosphericCO
2
oncropyieldisaccountedinGAEZbytheCO
2

yieldadjustment factor (f
CO2
). Crop species respond differently to CO
2
depending on physiological
characteristicssuchasphotosyntheticpathway(e.g.C3orC4plants).Thesecropspecificresponses
areaccountedintheparameterizationoff
CO2:

2 2
CO _ sui 2
2
2 CO
xf ) c ] CO [ x ) b ] CO [ ax ( 1 f + + + =
Wherea,bandcareparameters(bybroadcropgroups)usedtocapturethedifferentCO
2
responses
of four crop groups (Table 49). The factor f
sui_CO2
is an empirical correction accounting for land
suitabilityasexplainedbelow.
Table42CropspecificcoefficientsforthecalculationofCO2fertilizationeffect
CropGroup
(*)
Coefficients
(1)
1 2 3 4
a 0.000029051 0.00002408 0.000035537 0.000053184
b 0.075951 0.06933 0.062189 0.11551
c 21.9 20.26 16.652 32.327
I:wheat,barley,rye,oat,buckwheat,potato,sugarbeet,highland/temperatebeans.chickpea,dry
pea, temperate sunflower, rape, temperate cotton, flax, olive, coffee arabica, temperate onion,
temperatetomato,cabbage,carrot,tea,alfalfa,reedcanarygrass.
II: rice, cassava, sweet potato, lowland beans, cowpea, gram, pigeon pea, groundnut, tropical
sunflower, tropical cotton, banana oilpalm, yam, cocoyam, tobacco, citrus, cocoa, coffee robusta,
subtropicalonions,subtropicaltomato,subtropicalcarrots,coconut,jathropa.
III:maize,sorghum,millet,sugarcane,switchgrass,miscanthus.
IV:soybean.
V:pasturelegume,grass(averageC3andC4).
The local environment also influences the impact that CO
2
has on crop growth. Realization of the
fertilization effect of CO
2
is adjusted when suboptimum growth conditions are indicated by the
suitability classification for a LUT in a given gridcell. Under very suitable conditions it is assumed
thatafertilizationeffectoftwothirdsthatderivedfromlaboratoryexperimentscouldberealizedin
46
farmers fields. For marginally suitable conditions this share is set to onethird see Table 44)). On
averagethisresults inabouthalfoftheCO
2
fertilizationeffectmeasured in laboratoryexperiments
to be applied in GAEZ, as is broadly consistent with results reported in freeair CO
2
enrichment
(FACE)experiments.
Table43YieldadjustmentfactorsforCO2fertilizationeffectaccordingtolandsuitabilityratings
VS S MS mS
f
sui_CO2
0.667 0.555 0.444 0.333
Land suitability classes are very suitable (VS), suitable (S), moderately suitable (MS),
marginallysuitable(mS).
In GAEZ various scenarios were simulated as published by IPCC (Nakicenovic et al. 2000) in the
special reports on emission scenarios (SRES) and quantified by different climate modeling groups.
GAEZrunswereperformedwithdifferentCO
2
concentrationsforeachscenarioforthreefuturetime
periods(2020s,2050sand2080s)asshowninTable44.
ThecorrectionincrementforCO
2
withoutlandsuitabilityconstraintsisshowninFigure44.

Figure44YieldresponsetoelevatedambientCO2concentrations
Table44TheCO2concentrations(ppm)usedtomodelfertilizationeffectinGAEZaccordingto
differentIPCCscenariosandtimepoints

0.00
0.05
0.10
0.15
0.20
0.25
350 450 550 650 750
CO2 concentration (ppm)
C
O
2

a
d
j
u
s
t
m
e
n
t

f
a
c
t
o
r
Crop Group 1
Crop Group 2
Crop Group 3
Crop Group 4
Scenario
(1)
Year
(2)

2020s 2050s 2080s

A2 430 547 721
B2 417 488 568
B1 422 494 534
A1b 440 547 649
A1f 434 594 834
(1)
SRESscenariosfromIPCC
(2)
Corresponds to the CO
2
concentration at the midpoint
of a 30year period (e.g. year 2025 represents the 2020s
and corresponds to mid point of the period from 2011 to
2040).
47
4.8 GridcellanalysisModuleII
Results of the calculation procedures of Module II are presented for a sample gridcell in Appendix
49.Theexampleprovidesoutputdataofthebiomass andyieldcalculations forrainfedhigh input
cropproductionforreferenceclimate(19621990)foragridcellnearIlonga,Tanzania.
4.9 DescriptionofModuleIIoutputs
TheoutputofModuleIIrequireslargeamountsoffilestorageasitrecordsforeachgridcellandLUT
therelevantresultsofthebiomasscalculation,includingpotentialyields,yieldreducingfactors,and
actualcropevapotranspiration,accumulatedtemperatures,waterdeficitsandcropcalendar.
ThemainoutputinformationprovidedbyModuleIIisgiveninAppendix47and48.
48

49
5 ModuleIII(Agroclimaticyieldconstraints)
5.1 Introduction
Atthestageofcomputingpotentialbiomassandyields,noaccountistakenoftheclimaticrelated
effects operating through pests and diseases, and workability. Such effects need to be included to
arrive at realistic estimates of attainable crop yields. Precise estimates of their impacts are very
difficult to obtain for a global study. Here it has been achieved by quantifying the constraints in
terms of reduction ratings, according to different types of constraints and their severity for each
crop, varying by length of growing period zone and by level of inputs. The latter subdivision is
necessarytotakeaccountofthefactthatsomeconstraints,suchasbollwormoncotton,arepresent
under low input conditions but are controllable under high input conditions in certain growing
period zones. While some constraints are common to all input levels, others (e.g., poor workability
throughexcessmoisture)aremoreapplicabletohighinputconditionswithmechanizedcultivation.
Agroclimatic constraints cause direct or indirect losses in the yield and quality of produce. Yields
losses in a rainfed crop due to agroclimatic constraints have been formulated based on principles
and procedures originally proposed in FAO197881a. Details of the conditions that are influencing
yieldlossesarelistedbelow.
Therelationshipsbetweentheseconstraintswith general agroclimaticconditionssuch asmoisture
stress and excess air humidity, and risk of early or late frost are varying by location, between
agriculturalactivities aswell asbythe useofcontrolmeasures.Ithasthereforebeenattemptedto
approximate the impact of these yieldconstrains on thebasis of prevailing climatic conditions.The
efficacy of control of these constraints (e.g. pest management) is accounted for through the
assumed three levels of inputs. Due to the relatively high level of uncertainty, this assessment of
agroclimaticconstraintshasbeenappliedseparatelyinModuleIII,suchthateffectsaretransparent,
wellseparatedandGAEZassessmentscanbemadewithandwithouttheseconstraints(Figure51).

Figure51InformationflowsofModuleIII
50
In Module III, yield losses caused by agroclimatic constraints are subtracted from the yield
calculated in Module II. Five different yield constrains (i.e. yieldreducing factors) are taken into
account:
a. Longtermlimitationtocropperformanceduetoyeartoyearrainfallvariability
b. Pests,diseasesandweedsdamageonplantgrowth
c. Pests,diseasesandweedsdamageonqualityofproduce
d. Climaticfactorsaffectingtheefficiencyoffarmingoperations
e. Frosthazards
Althoughtheconstraintsofgroupdarenotdirectyieldlossesinreality,suchconstraintsdomean,
for example, that the high input level mechanized cultivator cannot get onto the land to carry out
operations. In practice, these limitations operate like yield reductions. Similarly for the low input
cultivator, for example, excessive wetness could mean that the produce is too wet to handle and
remove,andagainlosseswouldbeincurredeventhoughtheproducemaybestandinginthefield.
Also included in this group, are constraints due to the cultivator having to use longer duration
cultivarstoenableharvestingindryconditions.Theuseofsuchcultivarsincursyieldrestrictions,and
suchcircumstancesunderwetconditionshavethereforebeenincorporatedintheseverityratingsof
agroclimaticconstraintsingroupd.
In general, with increasing length of growing period and wetness, constraints due to pests and
diseases (groupsb andc)become increasinglysevereparticularlyto low inputcultivators.Asthe
length of growing period gets very long, even the high input level cultivator cannot keep these
constraintsundercontrolandtheybecomesevereyieldreducingfactorsatallthreelevelsofinputs.
Other factors, such as poor pod set in soybean or poor quality in short lengths of growing period
zones, are of similar severity for all three levels of inputs. Difficulties in lifting root crops under dry
soil conditions (short lengths of growing periods group d) are rated more severely under the high
level of inputs (mechanized) than under intermediate and low level of inputs. For irrigated
productionthecconstraintisappliedonlyatthewetend,i.e.,above300daysintheexample.
Inthis sense,agroclimaticconstraints areassumedtorepresentanydirectorindirectlosses inthe
yield and quality of produce. An explanation of the main yieldreducing components addressed by
agroclimaticconstraintsisprovidedinthefollowingparagraphs.
5.2 Conceptualbasis
Matchingcropgrowthcycleandthelengthofgrowingperiod
Whenthegrowingperiodisshorterthanthegrowthcycleofthecrop,fromsowingtofullmaturity,
there is loss of yield. The biomass and yield calculations account for direct losses by appropriately
adjusting LAI and harvest index. However, the loss in the marketable value of the produce due to
poorqualityoftheyieldasinfluencedbyincompleteyieldformation(e.g.,incompletegrainfillingin
grain crops resulting in shriveled grains or yield of a lower grade, incomplete bulking in root and
tuber leading to a poor grade of ware), is not accounted for in the biomass and yield calculations.
This loss is to be considered as an agroclimatic constraint in addition to the quantitative yield loss
duetocurtailmentoftheyieldformationperiod.Yieldlossescanalsooccurwhenthelengthofthe
growing period is much longer than the length of the growth cycles. These losses operate through
yieldandqualityreducingeffectsof(i)pests,diseasesandweeds,(ii)climaticfactorsaffectingyield
components and yield formation, and (iii) climatic conditions affecting the efficiency of farming
operations.
Waterstressduringthegrowingperiod
Waterstress generally affects crop growth, yield formation and quality of produce. The yield
reducingeffectsofwaterstressvariesfromcroptocrop.Thetotalyieldimpactcanbeconsideredin
termsof(i)theeffectongrowthofthewholecrop,and(ii)theeffectonyieldformationandquality
51
ofproduce.Forsomecrops,thelattereffectcanbemoreseverethantheformer,particularlywhere
theyieldisareproductivepart(e.g.,cereals)andyieldformationdependsonthesensitivityoffloral
partsandfruitsettowaterstress(e.g.,silkdryinginmaize).
Pests,diseasesandweeds
Toassesstheagroclimaticconstraintsofpest,diseaseandweedcomplex,theeffectsonyieldsthat
operate through loss in crop growth potential (e.g., pest and diseases affecting vegetative parts in
graincrops)havebeenseparatedfromeffectsonyieldthatoperatedirectlyonyieldformationand
quality of produce (e.g., cotton stainer affecting lint quality, grain mould in sorghum affecting both
yieldandgrainquality).
Climaticfactorsdirectlyorindirectlyreducingyieldandqualityofproduce
Theseincludeproblemsofpoorseedsetand/ormaturityundercoolorlowtemperatureconditions,
problems of seed germination in the panicle due to wet conditions at the end of grain filling,
problemsofpoorqualitylintduetowetconditionsduringthetimeofbollopeningperiodincotton,
problems of poor seed set in wet conditions at the time of flowering in some grain crops, and
problemsofexcessivevegetativegrowthandpoorharvestindexduetohighnighttimetemperature
orlowdiurnalrangeintemperature.
Climaticfactorsaffectingtheefficiencyoffarmingoperationsandcostsofproduction
Farming operations include those related to land preparation, sowing, cultivation and crop
protectionduringcropgrowth,andharvesting(includingoperationsrelatedtohandlingtheproduce
during harvest and the effectiveness of being able to dry the produce). Agroclimatic constraints in
this category are essentially workability constraints, which primarily account for excessive wetness
conditions.Limitedworkabilitycancausedirectlossesinyieldandqualityofproduce,and/orimpart
adegreeofrelativeunsuitabilitytoanareaforagivencropfromthepointofviewofhoweffectively
cropcultivationandproducehandlingcanbeconductedatagivenlevelofinputs.
Frosthazard
Theriskofoccurrenceoflateandearlyfrostincreasessubstantiallywhenmeantemperaturesdrop
below10C.Hence,lengthofthethermalgrowingperiodwithtemperaturesabove10C(LGPT10)in
a gridcell has been compared with growth cycle length of frost sensitive crops. When the crop
growth cycle is slightly shorter than LGPT10 the constraints related to frost risk are adjudged
moderate, when the growth cycle is very close or equal to LGPT10, the constraints have been
adjudgedassevere.
Box51
In general, with increasing length of growing period and wetness, constraints due to pests and diseases
(groupsbandc)becomeincreasinglysevereparticularlytolowinputcultivators.Asthelengthofgrowing
periodgets verylong,eventhehighinputlevelcultivatorcannotkeeptheseconstraintsundercontroland
theybecomesevereyieldreducingfactorsatallthreelevelsofinputs.Otherfactors,suchaspoorpodsetin
soybeanorpoorqualityinshortlengthsofgrowingperiodzones,areofsimilarseverityforallthreelevelsof
inputs.Difficultiesinliftingrootcropsunderdrysoilconditions(shortlengthsofgrowingperiodsgroupd)
areratedmoreseverelyunderthehighlevelofinputs(mechanized)thanunderintermediateandlowlevel
ofinputs.Forirrigatedproductionthecconstraintisappliedonlyatthewetend,i.e.,above300daysinthe
exampleforwinterwheatshowninTable51.
Although the constraints of group d are not direct yield losses in reality, such constraints do mean, for
example,thatthehighinputlevelmechanizedcultivatorcannotgetontothelandtocarryoutoperations.In
practice,thisresultsinyieldreductions.Similarlyforthelowinputcultivator,forexample,excessivewetness
could mean that the produce is too wet to handle and remove, and again losses would be incurred even
though the produce may be standing in the field. Also included in this group are constraints due to the
cultivator having to use longer duration cultivars to enable harvesting in dry conditions. The use of such
cultivars incurs yield restrictions, and such circumstances under wet conditions have therefore been
incorporatedintheseverityratingsofagroclimaticconstraintsingroupd.
52
Theavailabilityofhistoricalrainfalldatahasmadeitpossibletoderivetheeffectofrainfallvariability
through yearbyyear calculation of yield losses due to water stress. Therefore the a constraint,
relatedtorainfallvariabilityisnolongerapplied.Howevertheaconstrainthasbeenretainedinthe
agroclimatic constraints database for use with data sets containing average rainfall data and for
comparisonwithresultsofthepresentlyusedyearbyyearanalysis.
Theb,anddconstraintsandpartofthecarerelatedtowetness.Theratingsoftheseconstraints
havebeenlinkedtotheLGP.Itappearshowever,thatindifferentclimatezones,wetnessconditions,
traditionallyexpressed as P/EToratios, varyconsiderably forsimilarLGPs.LongLGPswithrelatively
low P/ETo ratios occur generally in subtropical, temperate and boreal zones, while relatively high
ratiosoccurinthetropics.
To account for these significant differences in wetness conditions of long LGPs (> 225 days), agro
climatic constraints have been related to P/ETo ratios by calculating equivalent LGPs, i.e.,
adjustments where P/ETo ratios where below average. The equivalent LGPs are then used in the
applicationoftheb,c,anddconstraints(Seesection3.4.4).
Table51presentsanexampleofagroclimaticconstraintsforwinterwheat.Forirrigatedproduction
only the agroclimatic constraints related to excess wetness apply. A listing of the agroclimatic
constraintparametersconsideredforallthecrop/LUTsarepresentedinAppendix51
Table51Agroclimaticconstraintsforrainfedwinterwheat
SUBTROPICS,TEMPERATEANDBOREAL
Growthcycle
40dayspredormancy+120dayspostdormancy
LGP/LGP
eq

6089 90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365

365
+
Lowinputs
a* 50 50 25 25 0 0 0 0 0 0 0 0
b 0 0 0 0 0 0 25 25 25 25 25 25
c 25 25 25 0 0 0 0 0 25 25 50 50
d 0 0 0 0 0 0 0 0 0 25 50 50
IntermediateInputs
a 50 50 25 25 0 0 0 0 0 0 0 0
b 0 0 0 0 0 0 0 25 25 25 25 25
c 25 25 25 0 0 0 0 0 25 25 50 50
d 0 0 0 0 0 0 0 0 0 25 50 50
Highinputs
a 50 50 25 25 0 0 0 0 0 0 0 0
b 0 0 0 0 0 0 0 0 25 25 25 25
c 25 25 0 0 0 0 0 0 0 25 25 50
d 0 0 0 0 0 0 0 25 25 25 50 50
LGP
t=10
6089 90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365
Allinputlevels
e 100 50 25 0 0 0 0 0 0 0 0
* The a constraint (yield losses due to rainfall variability) is not applied in the current assessment. This constraint has become
redundantduetoexplicitquantificationofyieldvariabilitythroughtheapplicationofhistoricalrainfalldatasets.
The application of the agroclimatic constraints to the combined results of temperature suitability
andthebiomassandyieldcalculationsprovidesagroclimaticattainableyields.
5.3 Calculationprocedures
Thevaluesoftheyieldreducingfactorsforagroclimaticconstraintsaresystematicallyorganizedin
lookuptables(Appendix51)accessedbyGAEZaccordinglyto:
53
(i) Landutilizationtype,LUT
(ii) Thermalclimate
(iii) Inputlevel
(iv) Length of the growing period, LGP, length of the equivalent LGP (LGP
eq
), and the
frostfreeperiod(LGP
t=10
)
Bycombiningthefiveagroclimaticyieldreducingfactors , ,
a e
fct fct K forconstrainttypesatoe,
anoverallyieldreducingfactor(fc
3
)iscalculated:
3
min{(1 ) (1 ) (1 ) (1 ),1 }
a b c d e
fc fct fct fct fct fct =
With agroclimatic constraints quantified, the agronomically atttainable crop yields have been
calculatedbyapplyingthefactor(fc
3
)totheagroclimaticyieldsascalculatedinModuleII.Notethat
theevaluationisdoneseparatelyforrainfedandirrigatedconditions.
5.1 DescriptionofModuleIIIoutputs
TheoutputformatofModuleIIIisthesameasforModuleII.TheinformationprovidedbyModuleIII
is described in Appendix 52 and 53. Various utility programs have been developed to map the
contents of Module III crop databases in terms of agroclimatically attainable yield, agroclimatic
reduction factor and overall yield reduction factor. Figure 52 shows the agroclimatically attainale
yieldsforrainfed,highinputwheat.

Figure52Agroclimaticallyattainableyieldofwheat

54

55
6 ModuleIV(Agroedaphicsuitability)
6.1 Introduction
In the context of this complete update of the global agroecological zones study, FAO and IIASA
recognizedthattherewasanurgentneedtocombineexistingregionalandnationalupdatesofsoil
information worldwide and incorporate these with the information contained within the FAO
UNESCOSoil MapoftheWorldwhichwasinlarge partsnolongerreflectingtheactual stateofthe
soil resource. In order to do this, partnerships were sought with the International Soil Resources
Information Centre (ISRIC) who had been largely responsible for the development of regional Soil
and Terrain databases and with the European Soil Bureau Network (ESBN) who had undertaken a
majorupdateofsoilinformationforEuropeandnorthernEurasiainrecentyears.Theincorporation
of the 1:1,000,000 scale Soil Map of China was an essential addition obtained through the
cooperationwiththeAcademiaSinica.Inordertoestimatesoilpropertiesinaharmonizedwaythe
use of actual soil profile data and the development of pedotransfer rules was undertaken in
cooperation with ISRIC and ESBN drawing on the WISE soil profile database and earlier work of
Batjesetal.andVanRanstetal.
Theresultingglobaldatabaseusesrastergridsat30arcsecondswhicharelinkedtoaharmonized
attribute database quantifications of composition of soil units within soil associations and
characterization of these soil units by the following soil parameters: Organic carbon, pH, water
storage capacity, soil depth, cation exchange capacity of the soil and the clay fraction, total
exchangeable nutrients, lime and gypsum contents, sodium exchange percentage, salinity, textural
classandgranulometry.
The four sourcedatabasesusedinthisHarmonized World SoilDatabase(HWSD),aretheEuropean
Soil Database (ESDB), the CHINA 1:1 million soil map, various regional SOTER databases (SOTWIS
Database), and the Soil Map of the World of FAO/Unesco. Figure 61 presents the regional
distributionofthedatasources.

Figure61Regionaldistributionofsoildatasources
This Module IV of GAEZ estimates for yield reductions caused by constraints induced by prevailing
soil and terrainslope conditions. Crop yield impacts from suboptimum soil and terrain conditions
are assessed separately. The soil suitability is assessed through crop/LUT specific evaluations of
seven major soil qualities. Terrain suitability is estimated from terrainslope and rainfall
concentration characteristics. Soil and terrain characteristics are read from 30 arcsecond gridcells
in which prevailing soil and terrain combinations have been quantified. This module calculates
suitability distributions for each gridcell by considering all occurring soilunit and terrain slope
56
combinationsseparately.Thecalculationsarecrop/LUTspecificandareperformedforallthreebasic
inputlevelsandfivewatersupplysystemsseparately.
The agroedaphic assessment, which is an integral part of the GAEZ modeling framework is
schematicallypresentedbelow.

Figure62InformationflowinModuleIV
6.1.1 Levelsofinputsandmanagement
Individualsoilandterraincharacteristicshavebeenrelatedtorequirementsandtolerancesofcrops
atthreebasiclevelsofmanagementandinputscircumstances,high,intermediateandlow.
Lowlevelinputs/traditionalmanagement
Underthelowinput,traditionalmanagementassumption,thefarmingsystemislargelysubsistence
basedandnotnecessarilymarketoriented.Productionisbasedontheuseoftraditionalcultivars(if
improved cultivars are used, they are treated in the same way as local cultivars), labor intensive
techniques, and no application of nutrients, no use of chemicals for pest and disease control and
minimumconservationmeasures.
Intermediatelevelinputs/improvedmanagement
Under the intermediate input, improved management assumption, the farming system is partly
market oriented. Production for subsistence plus commercial sale is a management objective.
Productionisbasedonimprovedvarieties,onmanual laborwithhandtools and/oranimaltraction
andsomemechanization.Itismediumlaborintensive,usessomefertilizerapplicationandchemical
pest,diseaseandweedcontrol,adequatefallowsandsomeconservationmeasures.
Highlevelinputs/advancedmanagement
Under the high input, advanced management assumption, the farming system is mainly market
oriented.Commercialproductionisamanagementobjective.Productionisbasedonimprovedhigh
yielding varieties, is fully mechanized with low labor intensity and uses optimum applications of
nutrientsandchemicalpest,diseaseandweedcontrol.

57
Mixedlevelofinputs
Under mixed level of inputs only the best land is assumed to be used for high level input farming,
moderately suitable and marginal lands are assumed to be used at intermediate or low level input
andmanagementcircumstances.Thefollowingprocedureswereappliedtoindividualgridcells.
(1) Determinealllandverysuitableandsuitableathighlevelofinputs.
(2) of the balance of land after (1), determine all land very suitable, suitable or moderately
suitableatintermediatelevelofinputs,and
(3) of the balance of land after (1) and (2), determine all suitable land (i.e. very suitable,
suitable,moderatelysuitableormarginallysuitable)atlowlevelofinputs.
6.1.2 Watersupplysystems
Five water supply systems have been separately evaluated. Apart from evaluating crop production
systems based on rainfed cultivation and rainfed with water conservation, specific soil
requirementsforthreemajorirrigationsystemshavebeenestablishednamelyforgravity,sprinkler
anddripirrigation.Table61presentsthewatersupplysystem/cropassociationsthatareconsidered
intheassessment.
Table61Watersupplysystem/cropassociations

WaterSupplySystems
Rainfed Rainfedwith
soilmoisture
conservation
Irrigation
Gravity Sprinkler Drip
InputLevels H,I,L H,I.
2
H,I H,I H
Crops
Wheat corrugation/border
Wetland_Rice basin
Dryland_Rice
Maize furrow
Barley corrugation/border
Sorghum furrow
Rye corrugation/border
Pearl_Millet furrow
Foxtail_Millet furrow
Oat corrugation/border
Buckwheat corrugation/border
White_Potato furrow
Sweet_Potato furrow
Cassava
Yam
Cocoyam(Taro)
Sugarcane basin/furrow
Sugarbeet furrow
Phaseolus_Bean
Chickpea

v
furrow
furrow

Cowpea furrow
DryPea furrow
Gram furrow
Pigeonpea furrow
Soybean v furrow
Sunflower furrow
Rape furrow
Groundnut furrow
OilPalm
Olive basin/furrow

58

WaterSupplySystems
Rainfed Rainfedwith
soilmoisture
conservation
Irrigation
Gravity Sprinkler Drip
InputLevels H,I,L H,I.
2
H,I H,I H
Crops
Jatropha furrow
Cabbage furrow
Carrot furrow
Onion furrow
Tomato furrow
Banana_Plantain basin/furrow
Citrus basin/furrow
Coconut furrow
Cacao furrow
Cotton furrow
Flax furrow
Coffee furrow
Tea
Tobacco furrow
Alfalfa corrugation/border
Switchgrass
ReedCanaryGrass

H:Highinputs,I:Intermediateinputs,L:Lowinputs
6.1.3 Soilsuitabilityassessmentprocedures
IntheGAEZapproach,landqualitiesareassessedinseveralstepsinvolvingspecificprocedures.The
landqualitiesrelatedtoclimateandclimatesoilinteractions(floodingregimes,soilerosionandsoil
nutrient maintenance) are treated separate from those land qualities specifically related to soil
propertiesandconditionsasreflectedintheHarmonizedWorldSoilDatabaseandtheGAEZterrain
slopedatabase.
Table62Landqualities
LandQualities AEZProcedures
Climateregime(temperature,moisture,radiation) Climaticsuitabilityclassification
Floodingregime Moistureregimeanalysisofwatercollectingsites
Soilerosion Assessmentofsustainableuseofslopingterrain
Soilnutrientmaintenance Fallowperiodrequirementassessments
Soilphysicalandchemicalproperties Soilsuitabilityclassification
Proceduresandactivitiesemployedareschematicallyrepresentedbelow:

2
AllLUTsofmarkedcropsexceptfortropicalhighlandmaizeandsorghum.Onlyaridandsemiaridmoisture
and the dryer part subhimid moisture regimes are considered. (LGP < 210 and P/PET between 20 and 80%)
Coldareasareexcluded(LGP
t=5
<165days).

59

Figure63Soilsuitabilityratingprocedures
The individual soil profile attributes, soil drainage conditions and soil phases prevalence, that have
been related to requirements and tolerances of crops at three generic levels of management and
inputs circumstances, high , intermediate and low, for five different water supply systems, need to
becombinedultimatelyintolandutilizationspecificsoilsuitabilityratings.
In the GAEZ approach, first individual soil qualities are defined and quantified. Table 63 below
provides an overview of the seven soil qualities in relation to relevant soil profile attributes,
includingsoildrainageconditionsandsoilphaseprevalence.

60
Table63Soilqualitiesandsoilattributes
SoilQualities Soilqualityrelatedsoilprofileattributes,soildrainageconditions
andsoilphasecharacteristics
SQ1 Nutrientavailability. Soiltexture,soilorganiccarbon,soilpH,totalexchangeablebases.
SQ2 Nutrientretentioncapacity. Soiltexture,basesaturation,cationexchangecapacityofsoilandof
clayfraction.
SQ3 Rootingconditions. Soil textures, coarse fragments, vertic soil properties and soil
phasesaffectingrootpenetrationandsoildepthandsoilvolume.
SQ4 Oxygenavailabilitytoroots. Soildrainageandsoilphasesaffectingsoildrainage
SQ5 Excesssalts. Soilsalinity,soilsodicityandsoilphasesinfluencingsoilsalinityand
sodicityconditions.
SQ6 Toxicity. Calciumcarbonateandgypsum.
SQ7 Workability (constraining
fieldmanagement).
Soil texture, effective soil depth/volume, and soil phases
constraining soil management (soil depth, rock outcrop, stoniness,
gravel/concretionsandhardpans).

6.2 Soilcharacteristics
Thyesevensoilqualities(SQ17)areestimatedfromsoilcharacteristics(e.g.organiccarboncontent,
soil pH, texture) read from the Harmonized World Soil Database. The soil qualities influencing crop
performance considered in the assessment include: nutrient availability (SQ1); nutrient retention
capacity (SQ2); rooting conditions (SQ3); oxygen availability to roots (SQ4); toxicities (SQ5); salinity
and sodicity (SQ6), and workability (SQ7). Each of the seven SQ ratings is derived from specific soil
characteristics.
6.2.1 Soilprofileattributes
Soil profile attributes considered for both topsoil (030 cm) and subsoil (30100cm) separately
include: soiltexture;organiccarboncontent;pH,cationexchange capacityof soil andclay fraction;
base saturation; total exchangeable bases; calcium carbonate contents; gypsum content; sodicity
andsalinity.Inadditionprevalenceofsoilphases,soildrainagecharacteristics,verticsoilproperties
andgelicsoilconditionsareconsidered.
Soiltexture
Soil texture influences nutrientavailability, nutrient retention, rootingconditions, drainage and soil
workability.
Soil texture is a soil property used to describe the relative proportion of different grain sizes of
mineral particles in a soil. Particles are grouped according to their size into what are called soil
separates (clay, silt, and sand). The soil texture class (e.g., sand, clay, loam, etc) corresponds to a
particular range of separate fractions, and is diagrammatically represented by the soil texture
triangle.Coarsetexturedsoilscontainalargeproportionofsand,mediumtexturesaredominatedby
silt,andfinetexturesbyclay(seediagram)andtable64.
61

Figure64Soiltextureclassification
Table64Soiltextureseparates
Soilseparates Diameterlimits(mm)(USDAclassification)
Clay lessthan0.002
Silt 0.002 0.05
Sand 0.052.00

Organiccarboncontent
Organic carbon is, together with pH, the best simple indicator of the health status of the soil.
Moderatetohighamountsoforganiccarbonareassociatedwithfertilesoilswithagoodstructure.
Soilsthatareverypoorinorganiccarbon(<0.2%),needorganicorinorganicfertilizerapplicationto
beproductive.Soilswithanorganicmattercontentoflessthan0.6%areconsideredpoorinorganic
matter.
Soilacidity(pHvalue)
The pH,measured ina soilwater solution,is ameasure forthe acidity andalkalinityofthe soil.pH
hasastrongontheavailabilityofnutrientstotheplant.OptimumpHvaluesrangebetween5.5and
7.0.
Cationexchangecapacityofclay
The type of clay mineral dominantly present in the soil often characterizes a specific set of
pedogenetic factors in which the soil has developed. Tropical, leaching climates produce the clay
mineral kaolinite, while confined conditions rich in Ca and Mg in climates with a pronounced dry
seasonencouragetheformationoftheclaymineralsmectite(montmorillonite).
Clay minerals have typical exchange capacities, with kaolinites generally having the lowest at less
than16cmol/kg,whilesmectiteshaveoneofthehighestwith80cmol/kgormore.
Cationexchangecapacityofsoil
The total nutrient fixing capacity of a soil is well expressed by its Cation Exchange Capacity (CEC).
Soils with low CEC have little resilience and can not build up stores of nutrients. Many sandy soils
have CEC less than 4 cmol/kg. The clay content, the clay type and the organic matter content all
62
determine the total nutrient storage capacity. Values in excess of 10 cmol/kg are considered
satisfactoryformostcrops.
Basesaturation
The base saturation measures the sum of exchangeable cations (nutrients) Na, Ca, Mg and K as a
percentageoftheoverallexchangecapacityofthesoil(includingthesamecationsplusHandAl).
Totalexchangeablebases
Total exchangeable bases represent for the sum of exchangeable cations in a soil: sodium (Na),
calcium(Ca),magnesium(Mg)andPotassium(K).
Calciumcarbonate
Calciumcarbonateisachemicalcompound(asalt),withthechemicalformulaCaCO
3
.Itisacommon
substance found as rock in all parts of the world and is the main component of shells of marine
organisms,snails,andeggshells.Calciumcarbonateistheactiveingredientinagriculturallime,andis
usuallytheprincipalcauseofhardwater.Itisquitecommoninsoilsparticularlyindrierareasandit
may occur in different forms as myceliumlike threads, as soft powdery lime, as harder concretions
orcementedinpetrocalcichorizons.Lowlevelsofcalciumcarbonateenhancesoilstructureandare
generally beneficial for crop production. At higher concentrations they may induce iron deficiency
andwhencementedlimitthewaterstoragecapacityofsoils.
Calciumsulphate(gypsum)
Gypsumisachemicalcompound(asalt)whichoccursoccasionallyinsoilsparticularlyindryerareas.
Research indicates that up to 2% gypsum in the soil favors plant growth, between 2 and 25% has
littleornoadverseeffectifinpowderyform,butmorethan25%cancausesubstantialreductionin
yields.
Exchangeablesodiumpercentage
The exchangeable sodium percentage (ESP)has been usedto indicate levelsof sodium in soils. It is
calculatedastheratioofNainCEC:
CEC
100 Na
ES

=
Sodiumadsorptionratio(SAR)hasbeenalsousedtoindicatelevelsofsodiumhazardsforcrops:
2
Mg Ca
Na
SAR
+
=
Electricalconductivity
Coastal and desert soils in particular can be enriched with watersoluble salts or saltsmore soluble
than gypsum. Crops vary considerably in their resistance and response to salt in soils. Some crops
willsufferatvaluesaslittleas2dS.m
1
otherscanstandupto16dS.m
1
.
6.2.2 Soildrainage
Ratings based on FAO'95 "Guidelines to estimation of drainage classes based on soiltype, texture,
soilphaseandterrainslope".Ratingshavebeenappliedtoallsoiltype,texture,soilphaseandbroad
slope classes and results have been distributed over eight GAEZ slope classes. The drainage classes
aredefinedasfollows(FAO1995).
Excessivelydrained(E):
Water is removed from the soil very rapidly The soils are commonly very coarse textured or rocky,
shalloworonsteepslopes.
63
Somewhatexcessivelydrained(SE):
Waterisremovedfromthesoilrapidly.Thesoilsarecommonlysandyandverypervious.
Welldrained(W):
Waterisremovedfromthesoilreadilybutnotrapidly.Thesoilscommonlyretainoptimumamounts
ofmoisture,butwetnessdoesnotinhibitrootgrowthforsignificantperiods.
Moderatelywelldrained(MW):
Waterisremovedfromthesoilsomewhatslowlyduringsomeperiodsoftheyear.Forashortperiod
thesoilsarewetwithintherootingdepth.Theycommonlyhaveanalmostimperviouslayer.
Imperfectlydrained(I):
Water is removed slowly so that the soil is wet at a shallow depth for significant periods. Soils
commonlyhaveanimperviouslayer,ahighwatertable,oradditionsofwaterbyseepage.
Poorlydrained(P):
Water is removed so slowly that the soils are commonly wet at a shallow depth for considerable
periods. The soils commonly have a shallow water table which is usually the result of an almost
imperviouslayer,orseepage.
Verypoorlydrained(VP):
Waterisremovedsoslowlythatthesoilsarewetatshallowdepthsforlongperiods.Thesoilshavea
veryshallowwatertableandarecommonlyinlevelordepressedsites.
ThealgorithmusedareasfollowsfortheFAO/UnescosoilclassificationusedintheDigitalSoilMap
oftheWorld(DSMW)(FAO1995):
1. Histosols(O)areconsideredtobeverypoorlydrained(100%VP).
2. AllGleysols(G)andFluvisols(J)withclayeytopsoiltexturesareconsideredtobepartlyvery
poorlyandpartlypoorlydrained(50%VPand50%P).
3. AllPlanosols(W),andthegleyicsoilunitsofotherFAOsoilgroups,suchasZg,Sg,Hg,Mg,
Lg,Dg,Pg,andAgareconsideredtobepartlypoorlydrainedandpartlyimperfectlydrained
(50%Pand50%I).
4. AllVertisols(V)areconsideredfor2/3oftheirareatobeimperfectlydrained,theremainder
isconsideredpoorlydrained(66%Iand34%P).
5. Plinthic Ferralsols (Fp), Plinthic Acrisols (Ap), and Gleyic Cambisols (Bg) are considered
partlyimperfectlydrainedandpartlymoderatelywelldrained(50%Iand50%MW).
6. All Arenosols (Q), nongleyic Podzols (P) and Regosols (R) with coarse topsoil textures and
occurring on gentle slopes (<8%) are considered to be partly excessively, and partly
somewhatexcessivelydrained(50%Eand50%SE).Ifthesamesoilsoccuronsteeperslopes
(>8%)thentheyareconsideredtobeexcessivelydrained(100%E).
7. Lithosols(I),Rankers(U)andRendzinas(E)withsandytopsoilandwhenoccurringongentle
slopes are considered moderately well drained (100% MW). When the same soils have a
loamyorclayeytopsoil andoccurongentleslopestheyareconsideredimperfectlydrained
(100% I). When these soils occur on steeper slopes they are considered to be partly well
drainedandpartlysomewhatexcessivelydrained(50%Wand50%SE).
8. All other soils with an argic horizon such as Luvisols (L), Acrisols (A), Podzolluvisols (D),
Nitosols (N) and luvic soil units in other FAO soil groups such as Xl, Yl, Kl, Cl and Hl when
havingasandytopsoiltextureandoccurringonflat(<8%slope)terrainareconsideredtobe
partlywelldrainedandpartlymoderatelywell drained(50%MWand50%W).Whenthese
64
soils have a finer topsoil texture and occur on slopes of less than 8% they are considered
dominantly moderately well drained and partly well drained (75% MW and 25% W). When
thesesoilsoccuronsteeperslopestheyareconsideredtobedominantlywelldrained,partly
moderately well drained and partly somewhat excessively drained (25% MW, 50% W and
25%SE).
9. All other soils with sandy topsoil textures occurring on flat terrain are considered to be
dominantly well drained, partly excessively and partly somewhat excessively drained (50%
W,25%SEand25%E).Whenthesesoilsarefinerandoccuronflatorgentlyslopingterrain
they are considered these soils have a sandy topsoil texture and occur on steeper slopes
(>8%) they are considered to be partly moderately well and partly well drained (50% MW
and50%W)>.Whenthesesoilsareloamyorclayeyandoccuronsteeperslopes(>8%)they
are considered to be dominantly well drained, partly somewhat excessively drained and
partlymoderatelywelldrained(50%W,25%SEand25%MW).
10. Soils discussed under 8. and 9. above and having a Petrocalcic, Petrogypsic Petroferric, or
Duripanarelessdeepthanthetypicalsoilunits.Thereforehalfoftheirareaisconsideredto
havesimilardrainageasthoseconsideredunderrule7.(Lithosolsetc.).
Results of the soil drainage evaluation for the FAO 1974 and the FAO 1990 soil classification are
presentedintheAppendix61.
6.2.3 Soilphases
Phases are subclassifications of soil units based on characteristics which are significant for the use
ormanagementofthelandbutarenotdiagnosticfortheseparationofthesoilunitsthemselves.In
HWSDsoilphasesnumbered1to12wereusedintheSoilMapoftheWorld(FAO74),phases13to
22 were used in association with the Revised Legend of the Soil Map of the World (FAO90), while
phases23to33arespecificfortheEuropeanSoilDatabase.
Table65Soilphases
Code Phase Code Phase
0 Nophase 17 Placic
1 Stony 18 Rudic
2 Lithic 19 Salic
3 Petric 20 Skeletic
4 Petrocalcic 21 Takyric
5 Petrogypsic 22 Yermic
6 Petroferric 23 Erosion
7 Phreatic 24 Nolimitationtoagriculturaluse
8 Fragipan 25 Gravelly
9 Duripan 26 Concretionary
10 Saline 27 Glaciers
11 Sodic 28 Soilsdisturbedbyman
12 Cerrado 29 Excessivelydrained(setto0)
13 Anthraquic 30 Flooded
14 Gelundic 31 Obstaclestoroots
15 Gilgai 32 ImpermeableLayer
16 Inundic 33 Wetness
Eachsoilphaseisexplainedinthefollowingparagraphs.
Stonyphase:
Marksareaswherethepresenceofgravel,stones,bouldersorrockoutcropsinthesurfacelayersor
at the surface makes the use of mechanized agricultural equipment impracticable. Hand tools can
normally be used and also simple mechanical equipment if other conditions are particularly
65
favorable.Fragmentsupto7.5cm are considered asgravel; larger fragments arecalled stonesand
boulders.
Lithicphase:
Thisphaseisusedwhencontinuouscoherentandhardrockoccurswithin50cmofthesoilsurface.
ForLeptosolsthelithicphaseisnotshownasitisimpliedinthesoilunitname.
Petricphase:
Thepetricphasemarkssoilswithalayerconsistingof40%ormore,byvolume,ofoxidicconcretions
orofhardenedplinthite,orironstoneorothercoarsefragmentswithathicknessof atleast25cm,
the upper part of which occurs within 1 m of the surface. The petric phase differs from the
petroferricphaseinthattheconcretionarylayerofthepetricphaseisnotcemented.
Petrocalcicphase:
Marks soils in which the upper part of a petrocalcic horizon (> 40% lime, cemented,usuallythicker
than10cm)occurswithin100cmofthesurface.
Petrogypsicphase:
Used for soils in which the upper part of a petrogypsic horizon (> 60% gypsum, cemented, usually
thickerthan10cm)occurswithin100cmofthesurface.
Petroferricphase:
The petroferric phase marks soils in which the upper part of the petroferric horizon occurs within
100 cm from the soil surface. A petroferric horizon is a continuous layer of indurated material in
whichironisimportantcementandorganicmatterisabsent.
Phreaticphase:
Thephreaticphasemarkssoilswhichhaveagroundwatertablebetween300and500cmfromthe
surface.
Fragipanphase:
Thefragipanphasemarkssoilswhichhavetheupperlevelofthefragipanoccurringwithin100cmof
the surface. The fragipan is a loamy subsurface horizon with a high bulk density relatively to the
horizon aboveit.It is hard orvery hard and seeminglycemented when dry. Dry fragments slakeor
fractureinwater.Afragipanislowinorganicmatterandisonlyslowlypermeable.
Duripanphase:
Theduripanphasemarkssoilsinwhichtheupperlevelofaduripanoccurswithin100cmofthesoil
surface. A duripan is a subsurface horizon that is cemented by silica and contains often accessory
cementsmainlyironoxidesorcalciumcarbonate.
Salinephase:
The saline phase marks soils in which in some horizons within 100 cm of the soil surface show
electric conductivity values higher than 4 dS m
1
. The saline phase is not shown for Solonchaks
becausetheirdefinitionimpliesahighsaltcontent.

66
Sodicphase:
Thesodicphasemarkssoilswhichhavemorethan6%saturationwithexchangeablesodiuminsome
horizonswithin100cmofthesoilsurface.ThesodicphaseisnotshownforSolonetzbecausetheir
definitionimpliesahighESP.
Cerradophase:
Cerrado is the Brazilian name for level open country of tropical savannas composed of tall grasses
and low contorted trees. This type of vegetation is closely related to the occurrence of nutrient
depletedsoilsonoldlandsurfaces.
Anthraquicphase:
The anthraquic phase marks soils showing stagnic properties within 50 cm of the surface due to
surfacewaterloggingassociatedwithlongcontinuedirrigation,particularlyofrice.
Gelundicphase:
Thegelundicphasemarkssoilsshowingformationofpolygonsontheirsurfaceduetofrostheaving.
Gilgaiphase:
Gilgai is a microrelief typical of clayey soils, mainly Vertisols. The microrelief consists of either a
succession of enclosed microbasins and microknolls in nearly level areas, or of microvalleys and
microridgesthatrunupanddowntheslope.
Inundicphase:
The inundic phase is used when standing or flowing water is present on the soil surface for more
than10daysduringthegrowingperiod.
Placicphase:
Theplacicphasereferstothepresenceofathinironpan,ablacktodarkreddishlayercementedby
ironwithmanganeseororganicmatter.Itsthicknessvariesfrom2to10mm.
Rudicphase:
Therudicphasemarksareaswherethepresenceofgravel,stones,bouldersorrockoutcropsinthe
surfacelayersoratthesurfacemakestheuseofmechanizedagriculturalequipmentimpracticable.
Skeleticphase:
Theskeleticphasereferstosoilmaterialwhichcontainsmorethan40%coarsefragmentsoroxidic
concretions.
Takyricphase:
The takyric phase applies to heavy textured soils with cracks into polygonal elements that form a
platyormassivesurfacecrust.
Yermicphase:
Theyermicphaseappliestosoilswhicharelowinorganiccarbonandhavefeaturesassociatedwith
deserts or very arid conditions (desert varnish, presence of palygorskyte, cracks filled with sand,
presenceofblownsandsonastablesurface.
Gravellyphase:
ThegravellyphaseisusedinESDBandindicatesover35%gravelswithdiameter<7.5cm.

67
Concretionaryphase:
TheconcretionaryphaseisusedinESDBandindicatesover35%concretions,diameter<7.5cmnear
thesurface.
Glaciers(Miscellaneousunit):
Permanentsnowcoveredareasandglaciers.
Soilsdisturbedbymanphase:
Areasfilledartificiallywithearth,trash,orboth,occurmostcommonlyinandaroundurbanareas.
Obstaclestoroots
Soils which have physical or chemical obstacles to root penetration are classified in relation to the
depthofthelayer.
Impermeablelayerphase
Soils which have an impermeable layer impeding drainage and root penetration are classified in
relationtothedepthofthelayer.
Wetnessphase
Soils that have severe wetness conditions in the soil profile. Wetness is classified by depth
occurrenceanditsdurationduringtheyear.
6.3 Soilsuitabilityratings
The soil suitability assessment considers soil profile attributes, soil texture, soil drainage and soil
phases.
6.3.1 Soilprofileattributesratings
The soil profile attribute suitability ratings are empirical coefficients. They have been compiled by
input level (high, Intermediate and low) and by five water supply systems (rainfed, rainfed with
waterconservation,gravityirrigation,sprinklerirrigationanddripirrigationsystems).Thesoilprofile
attribute ratings iaccount for gelic soil conditions and vertic soil properties. The ratings presented
below(Table63)refertotherainfedproductionofwheat.TheratingsystemisadaptedfromSyset
al1993andusessixclassesnamely:
S0 Noconstraint(100%)
S1 Slightconstraint(90%)
S2 Moderateconstraint(70%)
S3 Severeconstraint(50%)
S4 Verysevereconstraint(30%)
N Notsuitable(<10%)
The characteristics and properties are organized by soil quality to which they apply and by level of
inputandmanagementwhereapplicable.
Table66Soilprofileattributeratingsforrainfedwheat
Soilprofileattributes,
verticsoilproperties
andgelicsoil
SoilQualityand
InputLevel
SoilprofileattributeRatings
S0 S1 S2 S3 S4 N
100% 90% 70% 50% 30% 10%
OrganicCarbon SQ1,SQ2 1.5 0.8
LowpH(H2O) SQ1/SQ2(LOW+INT) 6.5 6 5.6 5.2 4.7 4.2
LowpH(H2O) SQ2(HIGH) 6 5.6 5.2 4.7 4.5 3.9
HighpH(H2O) SQ1,SQ2(LOW+INT) 7.5 8.2 8.3 8.5 8.6
HighpH(H2O) SQ2(HIGH) 7.5 8.2 8.3 8.5 8.6
68
Soilprofileattributes,
verticsoilproperties
andgelicsoil
SoilQualityand
InputLevel
SoilprofileattributeRatings
S0 S1 S2 S3 S4 N
100% 90% 70% 50% 30% 10%
TEB SQ1 8 5 3.5 2
TEB SQ1,SQ2 5 3.5 2
CEC(clay) SQ2(LOW+INT) 24 16
CEC(clay) SQ2(HIGH) 16
BaseSaturation(%) SQ2(LOW+INT) 80 50 35 0
BaseSaturation(%) SQ2(HIGH) 50 35
Rootingdepth(cm) SQ3,SQ7b 90 70 35 30 0
Rootingdepth(cm) SQ7a 70 35 30 0
Coarsefragments SQ3,SQ7a,SQ7b 15 35 55 100
ElectricConductivity SQ5 1 3 5 6 10 100
ESP(%) SQ5 15 20 35 45 100
CaCO
3
(%) SQ6 20 30 40 60 500
Gypsum(%) SQ6 3 5 10 20 100
Verticproperties SQ3 x
CECsoil SQ2(LOW+INT) 10 8 4 2
CECsoil SQ2High 8 4 2
Gelic SQ3,SQ7a X
Verticproperties SQ7aHIIGH X
Verticproperties SQ7b(LOW+INT) X
SoilprofileattributeratingsforallcropsareprovidedbythefivewatersupplysystemsinAppendix
62.
6.3.2 Soiltextureratings
Soiltextureconditionsareinfluencingthevarioussoilqualities(SQ1,SQ2,SQ3andSQ7).Inaddition,
textureisusedinthedeterminationofsoildrainageconditionsandthereforeindirectlyusedforSQ4
as well. The table below provides example soil texture ratings for rainfed production of wheat for
individual soil qualities. Soil workability ratings differ for high (H) and intermediate and low inputs
(L+I) and are provided separately. Soil texture ratings are compiled for individual water supply
systems.Table67presentssoiltextureratingsfor13textureclassesfortheproductionofrainfed
wheat.
Table67Soiltextureratingsforrainfedwheat
SoilTextureRatingsforRainfedProductionofWheat
SoilQualitiesandInputLevel
c
l
a
y

(
h
e
a
v
y
)

s
i
l
t
y

c
l
a
y

c
l
a
y

(
l
i
g
h
t
)

s
i
l
t
y

c
l
a
y

l
o
a
m

c
l
a
y

l
o
a
m

s
i
l
t

s
i
l
t

l
o
a
m

s
a
n
d
y

c
l
a
y

l
o
a
m

s
a
n
d
y

c
l
a
y

l
o
a
m

s
a
n
d
y

l
o
a
m

l
o
a
m
y

s
a
n
d

s
a
n
d

1 2 3 4 5 6 7 8 9 10 11 12 13
Nutrient Availability, SQ1 100 100 100 100 100 100 100 100 100 100 90 70 30
Nutrient retention capacity, SQ2) 100 100 100 100 100 100 100 100 100 100 90 70 30
Rooting conditions, SQ3 90 100 100 100 100 100 100 100 100 100 100 100 100
Workability Constraints, SQ7b (H) 100 100 100 100 100 100 100 100 100 100 100 100 100
Workability Constraints, SQ7a (L+I) 50 100 100 100 100 100 100 100 100 100 100 100 100
SoiltextureratingsforallcropsareprovidedbythefivewatersupplysystemsintheAppendix63.

69
6.3.3 Soildrainageratings
SoildrainageischaracterizedintheHarmonizedWorldSoilDatabasein7classes:
Table68Soildrainageclasses
Code Drainagelevel
VP VeryPoor
P Poor
I Imperfectly
MW Moderatelywell
W Well
SE Somewhatexcessive
E Excessive
Soildrainageratingsarevaryingbycropandmaybyvarybyprevalentsoiltextureconditions.Table
69presentssoildrainageratingsfortheproductionofrainfedwheat.Assumptionsforartificialsoil
drainage differ by input levels. High level inputs assumes that a that full and adequate artificial
drainagesystemsareinstalledwhilelowandintermediateinputsassumenoartificialdrainage.
Table69Soildrainageratingsforrainfedwheat
Fine,mediumandcoarsetextures
Drainageclasses VP P I MW W SE E
Lowinputs* 10 50 90 100 100 100 100
IntermediateInputs** 10 50 90 100 100 100 100
HighInputs*** 90 100 100 100 100 100 100
* Lowinputdrainageratingsassumenoartificialdrainage
** Intermediate input drainage ratings assume no artificial drainage (For organic farming or other sophisticated
managementtypeswithreducedagrochemicalinputs,highinputdrainageratingsaretobeappliedinthemodel
*** Highinputdrainageratingsassumethatfullandadequateartificialdrainagesystemsareinstalled
SoildrainageratingsforallcropsareprovidedbythefivewatersupplysystemsintheAppendix64.
6.3.4 Soilphasesratings
The soil phase ratings available from published and unpublished data sets have been compiled by
input level (high, intermediate and low) and by five water supply systems (rainfed, rainfed with
waterconservation,gravityirrigation,sprinklerirrigationanddripirrigationsystems).
The ratings presented below (Table 610) refer to the rainfed production of wheat. The ratings
represent constraints implied by the occurrence of soil phases in percentage (100% rating no
consttraintto0%renderingasoiltotallyunsuitable).
The soil phases are organized by soil quality to which they apply and by level of input and
managementandwatersupplysystem.Tworatingtypeshavebeenused:Socalledfullindicating
thatthesoilphaseratingwouldapplyto100%oftheextentofthesoilunittowhichthesoilphaseis
attributedandSplit,wherethesoilphaseratingisassumedtoaffect50%ofthesoiltowhichitis
attributedwhiletheother50%isassumednottobeaffected.

70
Table610Soilphaseratingsforrainfedwheat
Soil Rating
SoilPhases(HWSD)
INPUTLEVEL
Quality Type* HIGH INT LOW
SQ3 Full Stony 75 75 75
Lithic 50 50 50
Split Petric 60 60 60
Petrocalcic 60 60 60
Petrogypsic 60 60 60
Petroferric 60 60 60
Fragipan 100 95 85
Duripan 100 95 85
Full Anthraquic 85 85 85
Split Placic 100 95 85
Full Rudic 75 75 75
Split Skeletic 60 60 60
Full Erosion 100 100 100
SQ3 Split Gravelly 60 60 60
Concretionary 60 60 60
Full 0Noinformation(ROO=0) 100 100 100
1Noobstacletorootsbetween0and80cm(ROO=1) 100 100 100
2Obstacletorootsbetween60and80cmdepth(ROO=2) 90 90 90
3Obstacletorootsbetween40and60cmdepth(ROO=3) 80 80 80
4Obstacletorootsbetween20and40cmdepth(ROO=4) 50 50 50
5Obstacletorootsbetween0and80cmdepth(ROO=5) 70 70 70
6Obstacletorootsbetween0and20cmdepth(ROO=6) 0 0 0
0Noinformation(IL=0) 100 100 100
1Noimpermeablewithin150cm(IL=1) 100 100 100
2Impermeablebetween80and150cm(IL=2) 100 100 100
3Impermeablebetween40and80cm(IL=3) 80 80 80
4Impermeablewithin40cm(IL=4) 30 30 30
SQ4 Full Phreatic 100 100 100
Anthraquic 100 100 100
Inundic 100 100 100
Split Placic 100 100 100
Full Excessivelydrained 100 100 100
Flooded 100 100 100
0Noinformation(IL=0) 100 100 100
1Noimpermeablewithin150cm(IL=1) 100 100 100
2Impermeablebetween80and150cm(IL=2) 100 100 100
3Impermeablebetween40and80cm(IL=3) 100 100 100
4Impermeablewithin40cm(IL=4) 100 100 100
0Noinformation(WR=0) 100 100 100
1Notwetwithin80cmforover3months,norwetwithin40
cmforover1month(WR=1)
100 100 100
2Wetwithin80cmfor3to6months,butnotwetwithin40
cmforover1month(WR=2)
100 100 100
3Wetwithin80cmover6months,butnotwetwithin40
cmforover11month(WR=3)
100 100 100
4Wetwithin40cmdepthforover11month(WR=4) 100 100 100
SQ5 Split Saline 20 20 20
Sodic 35 35 35
Salic 20 20 20
SQ6 Petrocalcic 50 50 50
Petrogypsic 35 35 35

71
Soil Rating
SoilPhases(HWSD)
INPUTLEVEL
Quality Type HIGH INT LOW
SQ7 Full Stony 50 75 75
Lithic 30 50 75
Split Petric 50 50 50
Petrocalcic 50 50 50
Petrogypsic 50 50 50
Petroferric 50 50 50
Fragipan 100 100 100
Duripan 100 100 100
Placic 100 100 100
Full Rudic 50 75 75
Split Skeletic 50 50 50
Full Erosion 100 100 100
Nolimitationtoagriculturaluse 100 100 100
Split Gravelly 50 50 50
Concretionary 50 50 50
Full Noinformation(ROO=0) 100 100 100
Noobstacletorootsbetween0and80cm(ROO=1) 100 100 100
Obstacletorootsbetween60and80cmdepth(ROO=2) 100 100 100
Obstacletorootsbetween40and60cmdepth(ROO=3) 50 75 100
Obstacletorootsbetween20and40cmdepth(ROO=4) 30 50 75
Obstacletorootsbetween0and80cm depth(ROO=5) 50 75 75
Obstacletorootsbetween0and20cmdepth(ROO=6) 0 0 0
Noinformation(IL=0) 100 100 100
Noimpermeablewithin150cm(IL=1) 100 100 100
Impermeablebetween80and150cm(IL=2) 100 100 100
Impermeablebetween 40and80cm(IL=3) 50 75 100
Impermeablewithin40cm(IL=4) 30 50 75
*Rating type: Full = Total area affected by constraints as indicated; Split = 50% of area with constraints as indicated and
50%withoutconstraints
SoilphaseratingsforallcropsareprovidedbythefivewatersupplysystemsintheAppendix65.
6.4 Soilqualityandsoilsuitability
Thissectiondealswithsoilsuitabilityclassificationprocedures,followingatwostepapproach:
1) Crop responses to individual soil attribute conditions and relevant soil drainage and phase
conditionsarecombinedintosoilquality(SQ)ratings.
2) Soilqualitiesarecombinedincropspecific,inputandmanagementlevelspecificandwater
supplyspecificsoilsuitabilityratings.
6.4.1 Soilquality
The procedures used to derive the soil qualities
3
: (SQ17) from various combinations of soil
attributesaredescribedbelow.
Let(x
1
,.,x
m
)beavectorofsoilattributesrelevantforaparticularsoilqualitySQand((x
1
),,(x
m
)
thevectorofrespectivesoilattributeratings,0(x
j
)100.

3
The soil qualities are separately estimated for topsoil (030 cm) and subsoil (30100 cm) and
combinedbyweightingfashionaccordingtoprevalenceofactiveroots.
72
Further,letj
o
denotethesoilattributewiththelowestratingsuchthat:
(x
jo
)(x
j
),j=1,,m.
ThenwedefinesoilqualitySQasaweightedsumofsoilattributeratings,asfollows:
S =
S
(x
1
, , x
m
) =
(x
]
0
) +
1
m - 1
(x
]
)
]=]
0
2

Nutrientavailability(SQ1)
Naturalavailabilityofnutrientsisdecisiveforsuccessfullowlevelinputfarmingandtosomeextent
alsoforintermediateinputlevels.Diagnosticsrelatedtonutrientavailabilityaremanifold.Important
soil profile attributes for the topsoil (030 cm) are: soil texture/mineralogy/structure (TXT), soil
organic carbon (OC), soil pH and total exchangeable bases (TEB). For the subsoil (30100 cm) these
are:texture/mineralogy/structure,pHandtotalexchangeablebases.
Thesoilprofileattributesrelevanttosoilnutrientavailabilityarerelated.ForSQ1theattributewith
the lowest suitability rating is combined with the average of the remaining ones. The relationships
shown below represent topsoil and subsoil separately using the soil attributes and ratings for the
respectivesoillayersandinputlevels.
SQ1
topsoil
=f
SQ
(TXT,OC,pH,TEB)
SQ1
subsoil
=f
SQ
(TXT,pH,

TEB)

Nutrientretentioncapacity(SQ2)
Nutrientretentioncapacityisofparticularimportancefortheeffectivenessoffertilizerapplications
andisinparticularrelevantforintermediateandhighinputlevels.
Nutrientretentioncapacityreferstothecapacityofthesoiltoretainaddednutrientsagainstlosses
caused by leaching. Plant nutrients are held in the soil on the exchange sites provided by the clay
fraction, organic matter and the clayhumus complex. Losses vary with the intensity of leaching
which is determined by the rate of drainage of soil moisture through the soil profile. Soil texture
affectsnutrientretentioncapacityintwoways,throughitseffectsonavailableexchangesitesonthe
claymineralsandbysoilpermeability.
Thesoilcharacteristicsusedfortopsoilarerespectivelysoiltexture/mineralogy/structure(TXT),base
saturation(BS),cationexchangecapacityofsoil(CEC
soil
),andforsubsoilsoilTXT,pH,BS,andcation
exchangecapacityofclayfraction(CEC
clay
).SoilpHservesasindicatorforaluminumtoxicityandfor
micronutrientdeficiencies.
For SQ2 the attribute with the lowest suitability rating is combined with the average of the
remaining ones. Separately for high and intermediate inputs and management, and for topsoil and
subsoil,thefollowingrelationshipsareused:
SQ2
topsoil
=f
SQ
(TXT,BS
.
,CEC
soil
)
SQ2
subsoil
=f
SQ
(TXT,pH,BS,CEC
clay
)
Rootingconditions(SQ3)
Rootingconditionsincludeeffectivesoildepth(cm)andeffectivesoilvolume(vol.%)accountingfor
presenceofgraveland stones.Rootingconditionsmaybeaffectedbythe presenceofasoil phase,
either limiting the effective rooting depth or decreasing the effective volume accessible for root
penetration.Rootingconditionsinfluencecropgrowthinvariousways:
Adequacyoffoothold,i.e.,sufficientsoildepthforthecropforanchoring;
Availablesoilvolumeandpenetrabilityofthesoilforrootstoextractnutrients;
73
Space for root and tuber crops for expansion where the economic yield is produced in the
soil,and
Absence of shrinking and swelling properties (vertic) in particular affecting root and tuber
crops
Soildepthandvolumelimitationsaffectrootpenetrationandconstrainyieldformationforrootsand
tubers.Rootingconditions(SQ3)areestimatedbycombiningthereferencesoildepthratingwiththe
soil peoperty or soil phase that is most severely rated with regard to soil depth and volume
conditions.
Relevant soil properties considered are: Reference soil depth, soil properties i.e., soil
texture/mineralogy/structure, vertic properties, gelic properties, petric properties
4
and presence of
coarsefragments.
Relevantsoilphasesconsideredare:
FAO74soilphases:Stony,lithic,petric,petrocalcic,petrogypsic,petroferric,fragipanandduripan.
FAO90soilphases:Rudic,lithic,pertroferric,placic,skeletic,fragipanandduripan.
ESB (FAO85) soil phases and other soil depth/volume related characteristics include: Stony, lithic,
petrocalcic, petroferric, fragipan and duripan, and presence of gravel or concretions, obstacles to
roots(sixclasses)andimpermeablelayers(fourclasses).SQ3isevaluatedseparatelyfortopsoiland
subsoilattributes.
SQ3=(RSD)*min[((SPR),(SPH),(OSD)]
where,(RSD)isreferencesoildepthrating,(SPR)issoilpropertyrating,(SPH)issoilphaserating
and(OSD)isothersoildepth/volumerelatedcharacteristicsrating.
Oxygenavailability(SQ4)
Oxygen availability in soils is largely defined by soil drainage characteristics of soils. The
determinationof soil drainage classes is based on proceduresdeveloped at FAO (FAO 1995). These
proceduresaccountforsoiltype,soiltexture,soilphasesandterrainslope.
Assumptions regarding artificial drainage vary with input level. For low and intermediate input
drainage ratings assume no artificial drainage. For for high input, drainage ratings assume that
adequateartificialdrainagesystemsareinstalled.
Apart from drainage characteristics, oxygen availability may be influenced by soil and terrain
characteristicsthataredefinedthroughtheoccurrenceofspecificsoilphases.Theseincludeforthe
FAO74classificationsoilphasesindicatingphreaticconditions,andfortheFAO90classificationsoil
phasesindicatingrespectivelyphreatic,anthraquic,inundic,andplacicconditions.
SQ4 has been defined as the most limiting rating for a specific crop of either soil drainage or soil
phase. Soil quality differs between farming input levels due to the different assumptions regarding
artificialdrainage.SQ4isevaluatedseparatelyfortopsoilandsubsoilattributes.
SQ4

min[(DRG),(SPH)]
where,()istherespectiveinputlevelspecificattributeratingfunctionfordrainageandsoilphase.
Excesssalts(SQ5)
Accumulationofsaltsmaycausesalinity.Excessoffreesalts,referredtoassoilsalinity,measuredas
electric conductivity (EC) or as saturation of the exchange complex with sodium ions. This then is
referred to as sodicity or sodium alkalinity and is measured as exchangeable sodium percentage
(ESP).

4
PetricCalcisolsandPetricGypsysols
74
Salinity affects crops through inhibiting the uptake of water. Moderate salinity affects growth and
reducesyields;highsalinitylevelsmightkillthecrop.Sodicitycausessodiumtoxicityandaffectssoil
structure leading to massive or coarse columnar structure with low permeability. Apart from soil
salinityandsodicity,saline(salic)andsodicsoilphasesaffectcropgrowthandyields.
In case of simultaneous occurrence of saline (salic) and sodic soils the limitations are combined.
Subsequently the most limiting of the combined soil salinity and/or sodicity conditions and
occurrence of saline (salic) and/or sodic soil phase is selected. This soil quality is assumed
independentof level of input and management. SQ5 isevaluated separately fortopsoil andsubsoil
attributes.
SQ5

=min[(ESP)*(EC),(SPH)]
where, ( ) is the respective attribute rating function evaluated separately for topsoil and subsoil
attributes.
Toxicities(SQ6)
Low pH leads to acidity related toxicities e.g., aluminum, iron, manganese toxicities and to
deficiencies of, for instance, phosphorus and molybdenum. Calcareous soils exhibit generally
micronutrient deficiencies of, e.g., iron, manganese, and zinc and in some cases toxicity of
molybdenum. Gypsum (GYP) strongly limits available soil moisture. Tolerance of crops to calcium
carbonate(CCB)andgypsumvarieswidely(FAO,1990;Sys,1993).
Low pH and high CCB and GYP are mutually exclusive. The acidity (pH) related toxicities and
deficiencies are accounted in SQ1, nutrient availability, and SQ2, nutrient retention capacity
respectively.
InSQ6,themostlimitingofthecombinationofexcesscalciumcarbonateandgypsuminthesoiland
occurrence of petrocalcic and petrogypsic soil phases is selected. This soil quality is assumed
independentof level of input and management. SQ6 isevaluated separately fortopsoil andsubsoil
attributes.
SQ6
topsoil/subsoil
=min[(CCB)*(GYP),(SPH)].
where,()istherespectiveattributeratingfunction.
Workability(SQ7)
Diagnostic characteristics that can be related to soil workability vary by type of management
applied. Workability or ease of tillage depends on interrelated soil characteristics such as texture,
structure, organic matter content, soil consistence/bulk density, the occurrence of gravel or stones
intheprofileoratthesoilsurfaceandthepresenceofcontinuoushardrockatshallowdepthaswell
asrockoutcrops.Somesoilsareeasytoworkindependentofmoisturecontent,othersoilsareonly
manageable at a specific moisture status, in particular for manual cultivation or light machinery.
Irregular soil depth, gravel and stones in the profile and rock outcrops, might prevent the use of
heavy farm machinery.The soilconstraintsrelated tosoiltextureandsoilstructureareparticularly
affecting low and intermediate input farming LUTs, while the constraints related to irregular soil
depth and stony and rocky soil conditions are foremost affecting mechanized land preparation and
harvesting operations of highlevel input mechanized farming LUTs. Workability constraints are
thereforehandledseparatelyforlow/intermediateandhighinputs.
In the GAEZ rating procedure, the SQ7 is influenced by (i) physical hindrance to cultivation and (ii)
limitations to cultivation imposed by texture/clay mineralogy and bulkdensity. In all cases, SQ7 is
derived by combining the most limiting soil/soil phase attribute with the average of the remaining
attribute response ratings. Soil phases considered are from FAO 74 classification: stony, lithic,
petric, petrocalcic, petroferric, fragipan and duripan, and from FAO 90 classification: duripan,
75
fragipan, lithic, petroferric, rudic and skeletic. SQ7 is evaluated by input level separately for topsoil
andsubsoilattributes.
SQ7

=f
SQ
((RSD),(GRC),(SPH),(TXT),(VSP))
where, ( ) is the respective input level specific attribute rating function, GRC is soil gravel content
ratingandVSPisverticsoilpropertiesrating;otherattributesasdefinedbefore.
In addition, for FAO74 soil classification system: Shifting sand, Rock debris, Outcrops, Dunes, Salt
flats, Lakes and Ice caps miscellaneous units are considered to render soils unsuitable for crop
production, and for FAO90 soil classification system these are: Gelundic, Takyric, Yermic, Desert
andGobimiscellaneousunits.
6.4.2 Soilsuitability
Functionalrelationshipsofsoilqualitieshavebeenformulatedtoquantifycrop/LUTsuitabilityofsoil
units.Thefollowingguidingprinciplesformedthebasisforthewaysoilqualitieswerecombinedfor
differentlevelsofinputsandmanagement:
Nutrientavailabilityandnutrientretentioncapacityarekeysoilqualities;
Nutrient availability is of utmost importance for low level input farming; nutrient retention
capacityismostimportantforhighlevelinputs;
Nutrientavailability and nutrientretentioncapacityareconsideredofequal importancefor
intermediatelevelinputsfarming;
Nutrient availability and nutrient retention capacity are strongly related to rooting depth
andsoilvolumeavailable,and
Oxygen available to roots, excess salts, toxicity and workability are regarded as equally
importantsoilqualities,andthecombinationofthesefoursoilqualitiesisbestachievedby
multiplication of the most limiting rating with the average of the ratings of the remaining
threesoilqualities.
Followingtheaboveprinciplesforindividualcropsbythreelevelsofinputsandfivedifferentwater
supplysystems,eachsoilunitsuitabilityrating(SR)hasbeenestimated.Thefunctionalrelationships
forrespectivelylow,intermediateandhighinputfarmingarepresentedbelow.
Lowinputfarming:
SRlow=SQ1*SQ3*fSQ(SQ4,SQ5,SQ6,SQ7)
Intermediateinputfarming:
SRint.=0.5*(SQ1+SQ2)*SQ3*fSQ(SQ4,SQ5,SQ6,SQ7)
Highinputfarming:
SRhigh=SQ2*SQ3*fSQ(SQ4,SQ5,SQ6,SQ7)
The results of soil unit suitability assessment have been tabulated by each crop/soilunit/slope
class/input level/water supply system combination for integration with the results of the agro
climaticsuitabilityassessment.
Inmodule V (see below agroecologicalcropsuitability and corresponding agronomicallyattainable
yieldsisgeneratedforeachgridcell,throughassessingalldominantandassociatedsoilcomponent
sharesofallsoilassociationsasquantifiedintheHarmonizedWorldSoilDatabase.

76

Figure65Rainfedsoilsuitability,lowinputlevel

Figure66Rainfedsoilsuitability,highinputlevel

77
6.5 Terrainsuitability
Theinfluenceoftopographyonagriculturallanduseismanifold.Farmingpracticesarebynecessity
adapted to terrain slope, slope aspect, slope configuration and microrelief. For instance, steep
irregular slopes are not practical for mechanized cultivation, while these slopes might very well be
cultivatedwithadaptedmachineryandhandtools.
Sustainable agricultural production on sloping land is foremost concerned with the prevention of
erosion of topsoil and decline of fertility. Usually this is achieved by combining special crop
management and soil conservation measures. Cultivated sloping land may provide inadequate soil
protection and without sufficient soil conservation measures, cause a considerable risk of
acceleratedsoilerosion.Intheshortterm,cultivationofslopesmightleadtoyieldreductionsdueto
loss of applied fertilizer and fertile topsoil. In the long term, this will result in losses of land
productivity due to truncation of the soil profile and consequently reduction of natural soil fertility
andofavailablesoilmoisture.
Rainfedannualcropsarethemostcriticaltocausetopsoilerosion,becauseoftheirparticularcover
dynamicsandmanagement.TheterrainslopesuitabilityratingusedintheGlobalAEZstudycaptures
thefactorsdescribedabovewhichinfluenceproductionandsustainability.Thisisachievedthrough:
(i)definingforthevariouscropspermissiblesloperangesforcultivation,bysettingmaximumslope
limits;(ii)forslopeswithinthepermissiblelimits,accountingforlikelyyieldreductionduetolossof
fertilizer and topsoil, and (iii) distinguishing among farming practices ranging from manual
cultivationtofullymechanizedcultivation.
Ceteris paribus, i.e., under similar crop cover, soil erodibility and crop and soil management
conditions, soil erosion hazards largely depend on amount and intensity of rainfall. Data on rainfall
amountisavailableonamonthlybasisforallgridcellsintheclimateinventory.Rainfallintensityor
energy,asisrelevantforsoilerosion,isnotestimatedinthesedatasets.
Toaccountforclearlyexistingdifferencesinbothamountandwithinyeardistributionofrainfall,use
has been made of the modified Fournier index (Fm), which reflects the combined effect of rainfall
amountanddistribution(FAO/UNEP,1977),asfollows:
i
i
i
i
P
P
Fm

=
=
=
12
1
2
12
1
12

where,P
i
=precipitationofmonthi
When precipitation is equally distributed during the year, i.e., in each month onetwelfth of the
annualamountisreceived,thenthevalueofFmisequaltotheannualprecipitation.Ontheother
extreme, when all precipitation is received within one month, the value of Fm amounts to twelve
times the annual precipitation. Hence, Fm is sensitive to both total amount and distribution of
rainfallandislimitedtotherange1to12timestheannualprecipitation.
The Fm index has been calculated for all gridcells of the climatic inventory. The results have been
groupedinsixclasses,namely:Fm<1300,13001800,18002200,22002500,25002700,andFm>
2700.Theseclassesweredeterminedonthebasisofregressionanalysis,correlatingdifferentranges
oflengthofgrowingperiodzoneswithlevelsoftheFournierindexFm.Thiswasdonetoincorporate
the improved climatic information on within year rainfall distribution into GAEZ while keeping
consistencywithearlierproceduresofthemethodology,whichweredefinedbyLGPclasses.
Slope ratings are defined for the eight slope range classes used in the land resources database,
namely: 00.5% very flat, 0.52% flat, 25% gently sloping, 58 % undulating, 816% rolling, 1630%
hilly,3045%steep,and>45%verysteep.Thefollowingsuitabilityratingclassesareemployed:

78
S1 Optimum conditions
S2 Suboptimum conditions
S1/S2 50%optimum and50%suboptimum conditions
S2/N 50%suboptimum and50%notsuitableconditions
N Notsuitableconditions
Table611presentsterrainsloperatingsforrainfedconditionsforeightcropgroupsatthreelevels
of inputs and management as used for the lowest class of the Fournier index, i.e., Fm< 1300.
Appendix66presentsterrainsloperatingsfortheotherclasses ofFm,namely:Fm 13001800,Fm
18002200,Fm22002500,Fm25002700andFm>2700.

Figure67Rainfedsoilandterrainsuitability,lowinputlevel

Figure68Rainfedsoilandterrainsuitability,highinputlevel

79
Table611Terrainsloperatingsforrainfedconditions(Fm<1300)
HighInputs
Slope Gradient
Classes
0
0.5%
0.52% 25% 58% 816% 1630% 3045% >45%
Annuals1
S1 S1 S1 S1 S1/S2 N N N
Annuals2
S1 S1 S1 S1 S1/S2 N N N
Annuals3
S1 S1 S1/S2 S2/N N N N N
Perennials1
S1 S1 S1 S1/S2 S2/N N N N
Perennials2
S1 S1 S1 S1 S1/S22 S2/N N N
Perennials3
S1 S1 S1 S1 S1/S2 S2/N N N
Perennials4
S1 S1 S1 S1 S1 S2 N N
Perennials5
S1 S1 S1 S1 S1/S2 N N N
IntermediateInputs
Slope Gradient
Classes
00.5% 0.52% 25% 58% 816% 1630% 3045% >45%
Annuals1
S1 S1 S1 S1 S1 S2 N N
Annuals2
S1 S1 S1 S1 S1/S2 S2 N N
Annuals3
S1 S1 S1/S2 S2 S2/N N N N
Perennials1
S1 S1 S1 S1 S1 S1/S2 N N
Perennials2
S1 S1 S1 S1 S1 S1/S2 N N
Perennials3
S1 S1 S1 S1 S1 S1/S2 S2/N N
Perennials4
S1 S1 S1 S1 S1 S1/S2 S2/N N
Perennials5
S1 S1 S1 S1 S1 S1/S2 S2/N N
LowInputs
Slope Gradient
Classes
00.5% 0.52% 25% 58% 816% 1630% 3045% >45%
Annuals1
S1 S1 S1 S1 S1 S1/S2 N N
Annuals2
S1 S1 S1 S1 S1 S1/S2 N N
Annuals3
S1 S1 S1 S1/S2 S2 S2/N N N
Perennials1
S1 S1 S1 S1/S2 S2 S2/N N N
Perennials2
S1 S1 S1 S1 S1 S1/S2 S2/N N
Perennials3
S1 S1 S1 S1 S1 S1/S2 S2/N N
Perennials4
S1 S1 S1 S1 S1 S1 S1 S1/N
Perennials5
S1 S1 S1 S1 S1 S1/S2 S2/N N
CropGroups:
Annuals1: wheat,barley,rye,oat,buckwheat
Annuals2: maize, sorghum, pearl millet, foxtail millet, potato, white potato, sweet potato, beans,
phaseolus bean, chickpea, cowpea, gram, dry pea, pigeon pea, rapeseed, soybean and
groundnut, sunflower, cotton, sugar beet, rape, flax, white yam, greater yam, tobacco,
cabbage,carrot,onion,tomato)
Annuals3: wetlandrice
Perennials1: sugarcane
Perennials2: olive,citrus
Perennials3: cassava,oilpalm,banana,yellowyam,cocoyam,cocoa,coffee,coconut,jatropha.
Perennials4: pasturelegumes,grasses,tea
Perennials5: alfalfa,switchgrass,miscanthus,reedcanarygrass
80
6.6 Soilandterrainsuitabilityassessmentforirrigatedconditions
Fivewatersupplysystemshavebeenseparatelyevaluated.Apartfromevaluatingrainfedandrain
fed with water conservation based crop production systems, specific soil requirements for three
majorirrigationsystemshavebeenestablishednamelyforgravity,sprinkleranddripirrigation.
6.6.1 Soilsuitabilityforirrigatedconditions
Evaluation of rainfed with water conservation systems follows procedures as outlined for rainfed
production.Thesuitabilityevaluationproceduresforirrigatedcropproductioncoverdrylandcrops
andwetlandrice,atintermediateandhighlevelsofinputs.Cropspecificsoillimitationsforrainfed
production, such as limitations imposed by soil rooting conditions, soil nutrient availability, soil
nutrient retention capacity, soil toxicity are similar to those for rainfed suitability. Examples of
water supply system specific soil evaluation criteria are soil salinity and soil alkalinity that are
separately evaluated for drip irrigation systems and gypsum content, which is separately evaluated
forgravityirrigation(Fischeretal.,2002).
The following land and soil characteristics have been interpreted specifically for the irrigation
suitability classification: topography; soil drainage; soil texture; surface and subsurface stoniness;
calcium carbonate levels; gypsum status; and salinity and alkalinity conditions. The main literature
sourcesusedintheinterpretationincludeSysetal.(1993),SysandRiquier(1980),FAO(1985),FAO
(1996),FAO(1976b),FAO/Unesco(1974),andFAO/Unesco/ISRIC(1990).Detailsoftheapplicationof
standard or adapted ratings are presented by water supply system in Table 612. Soil profile
attribute ratings, soil texture ratings, soil drainage ratings and soil phase ratings for all crops, all
relevantlevelsofinputsandthefivewatersupplysystemsarepresentedintheAppendix62.
6.6.2 Terrainsuitabilityforirrigatedconditions
The dominant terrain factor governing the suitability of an area for any water supply system is
terrainslope.Othertopographicfactors,suchasmicrorelief,havepartlybeenaccountedforinthe
soilunitandsoilphasesuitabilityclassifications.
Permissible slopes depend on type of water supply system and assumed level of inputs and
management. Terrain suitability ratings for individual water supply systems and input levels, for
eight slope classes and eight crop groups, are presented by the six Fournier index classes varying
fromFm<1300toFm>2700,intheAppendix66.
81

Table612Soilandterrainevaluationratingsbywatersupplysystem
SOILANDTERRAINEVALUATION
Watersupplysystems Rainfed Irrigated
WithoutSoil
Moisture
Conservation
WithSoilMoisture
Conservation
GravityIrrigation Sprinkler
Irrigation
DripIrrigation
InputLevels H,I.,L H,I H,I. H,I H
Deviationsfromrainfedsoilparameterrating.
SQ7
Texture/minerarology standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
SQ3
Rootingdepth standard(rainfed) adaptedratings standard(rainfed) standard(rainfed) standard(rainfed)
SQ4
Drainage standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
SQ6
CaCO3 standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed)
SQ6
CaSO4 standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
SQ5
Salinity standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) adaptedratings
SQ5
Sodicity standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed)
Deviationsfromrainfedslopeparameterrating.
Other Slopes standard(rainfed) standard(rainfed) standard(irrigated) standard(rainfed) standard(rainfed)
Deviationsfromrainfedphaseparameterrating
SQ4
Phreatic n,a. n,a. standard(rainfed) n,a. n,a.
Anthraquic standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
Inundic standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
Placic standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
Excessivelydrained standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard (rainfed)
Flooded standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
0Noinformation(IL=0) n,a. n,a. n,a. n,a. n,a.
1Noimpermeablewithin150cm(IL=1) n,a. n,a. n,a. n,a. n,a.
2Impermeablebetween80and150cm(IL=2) standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed)
3Impermeablebetween40and80cm(IL=3) standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
4Impermeablewithin40cm(IL=4) standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
0Noinformation(WR=0) n,a. n,a. n,a. n,a. n,a.
1Notwetwithin80cmforover3months,norwetwithin
40cmforover1month(WR=1)
standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed)
2Wetwithin80cmfor3to6months,butnotwetwithin
40cmforover1month(WR=2)
standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
3Wetwithin80cmover6months,butnotwetwithin40
cmforover11month(WR=3)
standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
4Wetwithin40cmdepthforover11month(WR=4) standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
SQ5
Saline standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) adaptedratings
Sodic standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed)
Salic standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) adaptedratings
82

SOILANDTERRAINEVALUATION
Watersupplysystems Rainfed Irrigated
Without Soil
Moisture
Conservation
WithSoil
Moisture
Conservation
GravityIrrigation Sprinkler
Irrigation
DripIrrigation
InputLevels H,I,L H,I H,I H,I. H
SQ6
Petrocalcic standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed) standard(rainfed)
Petrogypsic standard(rainfed) standard(rainfed) adaptedratings standard(rainfed) standard(rainfed)
SQ7
Stony standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Lithic standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Petric standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Petrocalcic standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Petrogypsic standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Petroferric standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Fragipan standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Duripan standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Placic standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard (rainfed)
Rudic standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Skeletic standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Erosion n.a. n.a. n.a. n.a. n.a.
Nolimitationtoagriculturaluse n.a. n.a. n.a. n.a. n.a.
Gravelly standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Concretionary standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Noinformation(ROO=0) n.a. n.a. n.a. n.a. n.a.
Noobstacletorootsbetween0and80cm(ROO=1) n.a. n.a. n.a. n.a. n.a.
Obstacletorootsbetween60and80cmdepth(ROO=2) standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Obstacletorootsbetween40and60cmdepth(ROO=3) standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Obstacletorootsbetween20and40cmdepth(ROO=4) standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Obstacletorootsbetween0and80cmdepth(ROO=5) standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Obstacletorootsbetween0and20cmdepth(ROO=6) standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Noinformation(IL=0) n.a. n.a. n.a. n.a. n.a.
Noimpermeablewithin150cm(IL=1) n.a. n.a. n.a. n.a. n.a.
Impermeablebetween80and150cm(IL=2) standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Impermeablebetween40and80cm(IL=3) standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)
Impermeablewithin40cm(IL=4) standard(rainfed) adaptedratings adaptedratings standard(rainfed) standard(rainfed)

83

6.7 Soilandterrainsuitabilityassessmentforrainfedconditionsunder
waterconservationregimes

Selected annual crop/LUTs have separately been evaluated for and rainfed condition with water
conservation management. The waterbalance implications are presented in section 3.4.13.4.2.
Crop/LUTsforforrainfedwithwaterconservationassessmentsincludeLUTsofwheat,barley,grain
and silage maize, sorghum, millets, chickpea,cowpea, soybean and rape. The assessmenthas been
carriedoutforaridandsemiaridareas,whichcorrespondwithLGPsoflessthan180days.

6.8 Fallowperiodrequirements
In their natural state, many soils, in particular in the tropics, cannot be continuously cultivated
without undergoing degradation. Such degradation is marked by a decrease in crop yields and a
deteriorationofsoilstructure,nutrientstatusandotherphysical,chemicalandbiologicalattributes.
Undertraditionallowinputfarmingsystems,thisdeteriorationiskeptincheckbyalternatingsome
yearsof cultivationwith periodsof fallow.The length ofthenecessary rest period is dependenton
inputs applied, soil and climate conditions, and crops. Hence, the main reason for incorporating
fallow into crop rotations is to enhance sustainability of production through maintenance of soil
fertility.
Regeneration of nutrients and maintenance of soil fertility under low input cultivation is achieved
through natural bush or grass fallow. At somewhat higher inputs to soils, the soil fertility is
maintained through fallow, which may include for a portion of time a grass, grasslegume ley or a
greenmanure crop. Factors affecting changes in soil organic matter are reviewed in Nye and
Greenland (1960) and Kowal and Kassam (1978). They include temperature, rainfall, soil moisture
and drainage, soil parent material, and cultivation practices. The fallow factors used in the present
GAEZ land potentials assessments are based on earlier work done in the context of FAOs regional
assessments(YoungandWright,1980)andtheKenyaAEZstudy(FAO/IIASA,1991).
The fallow factors have been established by main crop groups and environmental conditions. The
cropgroupsincludecereals,legumes,rootsandtubers,andamiscellaneousgroupconsistingoflong
termannuals/perennials.Theenvironmentalframeconsistsofindividualsoilunits,thermalregimes
andmoistureregimes.Thethermalregimesareexpressedintermsofannualmeantemperaturesof
>25C,2025C,1520Cand<15C.ThemoistureregimesareexpressedintermsoffivebroadLGP
ranges:<60days,60120days,120180days,180270days,and>270days.
The fallow factors are expressed as percentage of time during the fallowcropping cycle the land
mustbeunderfallow.
Forthefourcropgroups:cereals,legumes,rootsandtubers,andamiscellaneousgroupconsistingof
longtermannuals/perennials,atintermediatelevelofinputs,thefallowrequirementsaresetatone
thirdofthelevelsrequiredunderlowlevelofinputs(seeAppendix67),andathighlevelsofinputs
andmanagementfallowrequirementsareuniformlysetat10%.
Exceptionstotheaboveare:
(i) for Fluvisols and Gleysols fallow factors are lower because of their special moisture and
fertilityconditions;
(ii) forwetlandriceonFluvisols,fallowrequirementsforallthreeinputlevelsaresetto10%;
(iii) forwetlandriceonGleysols,athighandintermediateinputsthefallowrequirementsareset
to10%andatlowinputsto20%;
84
(iv) forwetlandriceonothersoilsthenFluvosolsandGleysols fallowrequirementaresetas for
cropgroup1(cereals),and
(v) fallow requirements have been assumed to be negligible for the perennial crops oilpalm,
olive, citrus, cocoa, tea, coffee, jatropha, coconut, miscanthus, switchgrass, reed canary
grassandalfalfa.Fortheseperennialsnofallowrequirementshavebeenset.
In GAEZ vs3.0 the fallow requirement factors have been applied for the estimations of average
annualpotentialproduction.

6.9 Suitabilityofwatercollectingsites
Inwatercollecting sites substantially more watercan be available to plantsas compared toupland
situations. Watercollecting sites are difficult to locate in a global study but can be approximately
determined on the basis of prevalence of specific soil types. Fluvisols
5
and to a lesser extent
Gleysols
6
aretypicallyrepresentingtheflatterrainofalluvialvalleysandotherwatercollectingsites.
The cultivation of Fluvisols (under unprotected natural conditions) is determined by frequency,
duration and depth of flooding. The flooding attributes are generally controlled by external factors
such as a rivers flood regime which inturn is influenced by hydrological features of the catchment
areaandcatchment/siterelations,ratherthanbytheamountofonsiteprecipitation.
Therefore, with the exception of wetland crops, the cultivation of these soils is mainly confined to
postflood periods, with crops growing on residual soil moisture. The flooding regime in arid and
semiaridzonesiserratic.Someyears,severeflashfloodsmayoccur,inotheryearsnofloodsoccur
at all. In subhumid and humid zones flooding is more regular but duration and depth of flooding
mayvarywidelyfromyeartoyear.Gleysolsarenotdirectlyaffectedbyriverflooding.Thesesoilsare
howeverfrequentlysituatedinlowlyingwatercollectingsitesandwhennotartificiallydrained,the
Gleysols may be subject to waterlogging or even inundation as result from combinations of high
groundwater tables and ponding rainwater. In arid and semiarid areas these soils are cultivated in
thelaterpartandafterrainyseasons;thecropsgrowandmatureonresidualsoilmoisture.Insub
humid and humid areasGleysols without artificial drainage oftenremain waterlogged for extensive
periods,renderingthemunsuitableforcultivationofdrylandcrops.
Onboth,Fluvisols andGleysols,crops ofshortdurationthatareadaptedtogrowing andproducing
yields on residual soil moisture and which are tolerant to flooding, waterlogging and high
groundwatertables,canbefoundproducingsatisfactorilyoutsidethegrowingperioddefinedbythe
local rainfall regime. Therefore, a separate crop suitability classification for watercollecting sites is
required.Incompilingthisclassification,thelogicoftheoriginalAEZstudy(FAO,197881a)hasbeen
followed. This includes accounting for cropspecific tolerances to excess moisture (high
groundwater,waterloggingandflooding/inundation)andtheuseofavailableestimatesofflooding
regimes of the Fluvisols. Since Gleysols are mostly, but not necessarily, subjected to waterlogging
andinundationjustlikethenaturalFluvisols,itwasdecidedtotreatGleysolswithterrainslopesof
lessthan2%thesameasFluvisols.
In many parts of the world the flooding of Fluvisols is increasingly being controlled with dikes and
other protection means. Fluvisols, in protected conditions, do not benefit additional water supply

5
Fluvisols are by definition flooded by rivers. Fluvisols are young soils where sedimentary structures are
clearlyrecognizableinthesoilprofile.
6
Gleysolsaregenerallynotfloodedbyrivers.However,thesoilprofilesindicateregularoccurrenceofhigh
groundwatertablesthroughreduction(gley)features.LowlyingGleysolsmaybeponded/waterloggedby
highgroundwaterandrainfallduringtherainyseason.
85
and regular fresh sediment deposits, nor do they suffer from flooding. The moisture regime of
FluvisolsundertheseprotectedconditionsissimilartoothersoilsandthereforeprotectedFluvisols
aretreatedaccordingtotheproceduresusedforcropsinuplandconditions.
In a similar way, Gleysols may be artificially drained, thereby diminishing a major limitation for the
cultivationofthesesoils.ForareaswheretheGleysolshavebeendrained,arevised(i.e.,lesssevere)
set of soil ratings is used and the rules for natural Fluvisols are not applied. Since spatial details of
the occurrence of protected Fluvisols and artificial drainage of Gleysols are not available at the
global scale these factors are assumed to be linked to the level of inputs/management. The
applicationofFluvisolsuitabilityratingsandsoilunitsuitabilityratingsofartificiallydrainedGleysols
arepresentedbelow:
Table613FluvisolandGleysolsuitabilityratings
Fluvisols Gleysols
natural protected natural artificiallydrained
RAINFED
Highlevelinputs no yes no yes
Intermediatelevelinputs 50% 50% 50% 50%
Lowlevelinputs yes no yes no

IRRIGATION
Highlevelinputs no yes no yes
Intermediatelevelinputs 50% 50% 50% 50%
The moisture suitability ratings devised for unprotected Fluvisols and Gleysols without artificial
drainage are organized in ten groups of crops with comparable growth cycle lengths and similar
tolerancestohighgroundwaterlevels,waterloggingandflooding.Theratingtablesarepresentedin
Appendix68.

Figure69Watercollectingsites
6.1 DescriptionofModuleIVoutputs
ThemainoutputinformationprovidedbyModuleIVisgiveninAppendix69and610.

86
7 ModuleV(Integrationofclimaticandedaphicevaluation)
7.1 Introduction
Module V executes the final step in the GAEZ crop suitability and land productivity assessment. It
reads the LUT specific results of the agroclimatic evaluation for biomass and yield calculated in
Module II/III for different soil classes and it uses the edaphic rating produced for each soil/slope
combination in Module IV. The inventories of soil resources and terrainslope conditions are
integratedbyrankingallsoiltypesineachsoilmapunitwithregardtooccurrenceindifferentslope
classes. Considering simultaneously the slope class distribution of all grid cells belonging to a
particularsoilmapunitresultsinanoverallconsistentdistributionofsoilterrainslopecombinations
byindividual soilassociationmapunitsand30arcsecgridcells. Soilevaluationand sloperules are
appliedseparatelyforeachwatersupplysystems.TheinformationflowinModuleVissummarized
inFigure71.

Figure71InformationflowinModuleV
7.2 DescriptionofModuleVoutputs
7.2.1 MainprocessingstepsinModuleV
The algorithm in Module V steps through the grid cells of the spatial soil association layer of the
Harmonized WorldSoilDatabase (HWSD)anddetermines foreachgridcelltherespectivemakeup
of land units in terms of soil types and slope classes. Each of these component land units is
separately assigned the appropriate suitability and yield values and results are accumulated for all
elements. Processing of soil and slope distribution information takes place at 30 arcsecond grid
cells. One hundred of these produce the edaphic characterization at 5 arcminutes, which is the
resolutionusedforprovidingGAEZresults.Asaresult,informationstoredfor5arcminutegridcells
containsdistributionsoftheindividualsubgridevaluations.
87
ThemainpurposeofModuleVistocompileagridcelldatabaseforeachcroporcropgroupstoring
evaluation results that summarize the processed subgrid information. Computations include the
followingsteps:

Reading agroclimatic yields calculated for separate crop water balances of six broad soil
AWCclasses(fromModuleII/III);
applying AEZ rules for watercollecting sites (defined as Fluvisols and Gleysols on flat
terrain);
applying reduction factors due to edaphic evaluation for the specific combinations of soil
types/slopeclassesmakingupagridcell;
aggregatingresultsovercomponentlandunits(soiltype/slopecombinations),and
calculating applicable fallow requirement factors depending on climate characteristics, soil
typeandcropgroup.

7.2.2 ModuleVoutputresults
TheresultsofcropevaluationsinModuleVarestoredasalargenumberofseparatedatabaseseach
organized by grid cells. Separate files are generated by crop, input level, water supply system and
scenario/timeperiod, eachcontaining subgriddistribution information intermsofsuitableextents
andpotentialproductionbysuitabilityclasses.
AdetaileddescriptionoftheinformationprovidedbyModuleVisgivenintheAppendix71and
72.

Figure72MappingandTabulationinModuleVresults

88
Various utility programs have been developed to aggregate and tabulate results by administrative
unitsortomapthecontentsofModuleVcropdatabasesintermsofsuitabilityindexandpotential
gridcell output. Crop summary tables provide standardized information on distributions of crop
suitabilityandcropyielddata,whicharediscussedatlengthinAppendix73.Figure73belowshows
theagroecologicallysuitabletotalproductionofrainfed,higyinputwheat.

Figure73Agroecologicalsuitabilityandproductivitypotentialofwheat

89
8 ModuleVI(ActualYieldandProduction)
8.1 Introduction
Global change processes raise new estimation problems challenging the conventional statistical
methods. These methods are based on the ability to obtain observations from unknown true
probability distributions, whereas the new problems require recovering information from only
partiallyobservableorevenunobservablevariables.Forinstance,aggregatedataexistatglobaland
national level regarding agricultural production. Downscaling methods in this case should achieve
plausible estimation of spatial distributions, consistent with local data obtained from remote
sensing,availableaggregateagriculturalstatistics,andotheravailableevidence.
For this purpose a flexible sequential downscaling method, based on iterative rebalancing, was
developed at IIASA and implemented for use in GAEZ. The information flow associated with the
spatialallocationofagriculturalstatisticsissketchedinFigure81.

Figure81InformationflowofModuleVI
8.2 Downscalingofagriculturalstatisticstogridcells
Agricultural production and land statistics are available at national scale from FAO, but these
statistical data do not include the spatial heterogeneity of agricultural production at finer
resolutions, e.g., grid cells, within country boundaries. In this case a downscaling method is
needed for allocation of aggregate national production values to individual spatial units (gridcells)
byapplyingformalmethodsthataccontforlandcharacteristics,assesspossibleproductionoptions
and using available evidence from observed or inferred geospatial information, e.g. remotely
sensedlandcover,soil,climateandvegetationdistribution,populationdensityanddistribution,etc.
Land cover data products contain classifications providing detailed geographical information of the
distribution of cultivated land. Besides these there exists other important information on factors,
which significantly affect the patterns and intensities of crop production. For example, spatially
explicit biophysical data related to land constraints or crop suitabilities for specific agricultural
activities, farming systems data, human population distribution, locations of markets and
90
infrastructure etc. Such data in combination with GAEZ crop suitability layers is used in the
downscaling procedures to construct a prior distribution for allocation of agricultural cropping
activitiesandproduction.
To achieve consistency of available data and estimates across scales, the sequential rebalancing
proceduresthatweredeveloped atIIASArelyon appropriateoptimizationprinciples(Fischeretal.,
2006a, 2006b) and combine the available statistics with other prior hard (accounting identities)
andsoft(expertopinion)data.
To guidethespatial allocation of crops, GAEZ procedures for the calculation of potential yields and
production have been applied to, respectively, rainfed and irrigated cultivated land shares of
individual 5 arcminute gridcells. Rather than taking an average yield for the entire grid cell it is
assumed that the cultivated land will occupy the better part of the suitability distribution
determined in each grid cell. To estimate consistent spatial yield patterns of currently cultivated
crops by gridcells requires joint downscaling of agricultural statistics for all crops simultaneously.
The sequential downscaling consists of efficient iterative rebalancing procedures (Fischer et al.,
2006) based on cross entropy maximization principles, thereby allocating cropping activities to
appropriatetractsofrainfedrespectivelyirrigatedlandwhileprovidingrealisticestimatesofcurrent
yieldandproductionforthecultivatedlandinindividualgridcells,consistentwiththelandsspatial
distributionandagronomiccapabilities.
In summary, two main steps were involved in obtaining downscaled gridcell level area, yield and
productionofmaincrops:
(i) Estimationofsharesofrainfedorirrigatedcultivatedlandby5arcminutegridcell(for
explanationsseeAppendix81),and
(ii) Estimationofcropspecificharvestedarea,yieldandproductionofcropswithintherain
fedandirrigatedcultivatedlandofeachgridcell(fordetailsseeAppendix82).
Figure 82 presents the example results of the estimations of shares of cultivated land by grid cell
andFigure83showsresultsofharvestedareaforwheatproductionin2000.
8.2.1 Estimationofcultivatedlandshares
Fortheestimationoflandsharesbymajorlandusesinindividual5arcminutegridcells,datafrom
severallandcoverdatasetswasused.Fortheyear2000thedatabasecombines(i)theGLC2000land
cover regional and global classifications (http://wwwgvm.jrc.it/glc2000), (ii) a global land cover
categorization,compiledbyIFPRI(IFPRI,2002),basedonareinterpretationoftheGlobalLandCover
Characteristics Database (GLCC) ver. 2.0, EROS Data Centre (EDC, 2000), and (iii) a special layer of
forest land from the Forest Resources Assessment of FAO (FAO, 2001). Furthermore, global 5 arc
minuteinventoriesofirrigatedland(GMIAversion4.0;FAO/UniversityofFrankfurt,2006)wereused
andaninterpretationoftheIUCNWCMCprotectedareasinventory(WPDA,2009)(alongwithother
conventiontypesoflegallyprotectedareas)todistinguishprotectedlandintwocategories,namely
areas where some restricted agricultural use is permitted and protected areas where cultivation is
strictly prohibited. Finally, a population inventory for year 2000 has been used to estimate land
requiredforhousingandinfrastructure(populationdensitymapdevelopedbyFAOSDRN,basedon
spatialdataofLANDSCAN2003,withcalibrationtoUN2000populationfigures).
Instep(i)variouslandcoverinterpretationsarecombinedtoproduceaquantificationofeachgrid
cell in the spatial raster in terms of seven main land use/land cover shares. These shares are:
cultivated land, subdivided into (i) rainfed and (ii) irrigated land; (iii) forest; (iv) pasture and other
vegetatedland;(v)barrenandverysparselyvegetatedland;(vi)water,and(vii)urbanlandandland
requiredforhousingandinfrastructure.
An iterative calculation procedure was used to estimate land cover class weights, consistent with
aggregate FAO land statistics (of arable land and forest land) and spatial land cover patterns
91
obtained from remotely sensed data. The estimated class weights define for each land cover class
and spatial allocation unit (e.g., country) the contents in terms of respectively cultivated land and
forest. Starting values of class weights used in the iterative procedure were obtained by cross
country regression of statistical data of cultivated and forest land against aggregated extents of
nationallandcoverclassdistributionsobtainedfromGIS.
Theoccurrenceofcultivatedland(share)in2000ispresentedinFigure82.

Figure82Sharesofcultivatedlandby5arcminutegridcell
8.2.2 Allocationofagriculturalstatisticstocultivatedland
AgriculturalcropproductiondataareavailableatnationalscalefromFAO.Subnationalinformation
was collected and compiled by Montfreda et al. (2008) and FAOs AgroMAPS: Global spatial
database of agricultural landuse statistics, version 2.5. With spatial occurrence of rainfed and
irrigatedcultivatedlandestablishedinthepreviousstep,themainobjectiveofthesecondstepisto
allocate crop production statistics to the respective spatial land units while meeting statistical
accountsand respecting cropsuitability and land capabilitiesreflectedin the spatial land resources
inventory.
The algorithm can be summarized as follows: The potential suitability of individual crops in the
cultivated land of each grid cell is available from geographically detailed GAEZ assessments for
different input levels and water sources (i.e., rainfed and irrigated) including estimates of
agronomically attainable crop yields. Additional spatially explicit information can be used in
estimating crop distributions, for example, spatially explicit farming systems information (including
purpose of production in terms of subsistence/market orientation), distance to nearest market,
livestockdensity,populationdensity,etc.
The crop production statistics and the spatial information available for each country were used to
calculateaninitialestimateofcropwiseareaallocationandproduction,asocalledprior.Thepriors
aresubsequentlyrevisedinaniterativeproceduretoensurethatcropdistributionandproductionis
consistentwithaggregatestatisticaldataofcropharvestedareaandproduction,isallocatedtothe
availablerainfedandirrigatedcultivatedland,includingitscapacitytosupportmulticropping,and
is in agreement with ancillary subnational data, in particular selected crop area distribution data
(Montfredaetal.,2008)andagronomicsuitabilityofcropsasestimatedinAEZ.
8.1 DescriptionofModuleVIoutputs
The downscaling procedures and implementation for the year 2000 respectively 2005 agricultural
statisticshaveresultedinthefollowingdatasets:
92
(i) Global inventory of shares of cultivated land, forest land, grass and other vegetated
land, barren and very sparsely vegetated land, infrastructure and builtup urban areas
andwaterbygridcell.Thecultivatedlandsharesaresubdividedinrainfedandirrigated
land;
(ii) Area, yield and production for major crops in rainfed cultivated land, based on year
2000and2005statistics,and
(iii) Area,yieldandproductionformajorcropsinirrigatedlandbasedonyear2000and2005
statistics.
(iv) Estimates of the spatial distribution of total crop production and production of major
cropgroups(cereals,rootcrops,oilcrops),valuedatyear2000internationalprices.
For illustration, maps of cultivated land, harvested area, yield and production of rainfed maize are
presentedinFigure82,Figure83,Figure84andFigure85respectively.

Figure83Harvestedareaofrainfedmaizein2000

Figure84Yieldofrainfedmaizein2000

93

Figure85Productionofrainfedmaize
Theresultsofspatialallocationofcropstatisticsfortheyear2000and2005inModuleVIarestored
as a large number of separate GIS rasters of 5 arcminute gridcells, separately by 23 crops/crop
groups,rainfedandirrigatedconditions,forharvestedarea,productionandassociatedyield.

94

95
9 ModuleVII(YieldandProductionGaps)
9.1 Introduction
Apparent yield and production gaps have been estimated by comparing potential attainable yields
and production (estimated in GAEZ v3.0) and actual yields and production from downscaling year
2000 and 2005 statistics of main food and fiber crops (statistics derived from FAOSTAT and
AQUASTAT.
Numerical yield gap analysis relies on results of both crop suitability analysis (Module V) and
downscaling of base year agricultural statistics (Module VI). A schematic representation of Module
VIIispresentedinFigure91below.

Figure91SchematicrepresentationofModuleVII
Yieldandproductiongapsareestimatedbycomparingsimulatedpotentialand(downscaled)actual
yieldandproductionofmainfoodandfibercrops.Theunderlyingmetyhologicalframeworkofyield
andproductiongapestimationispresentedinFigure92.
96
Quantification
ofspatiallyexplicitagricultural
productionactivities
GAEZ
Spatiallyexplicitcommodity
specificyieldandproduction
potentialsincultivatedland
Gridcelllandshares
ofrainfedandirrigated
cultivatedland
ACTUAL
Spatiallyexplicitcommodity
specificyieldandproduction
incultivatedland
Yieldand
productiongapestimation
procedures
Nationalandsubnational
agriculturalstatistics
ofmaincrops
(Year2000and2005)
Yieldandproductiongap
database
formaincommoditiesbygridcell
(Year2000and2005)
Crop
distribution
priors

Figure92Yieldgapestimationprocedures
9.2 Yieldandproductiongapsassessmentprocedures
For 18ofthe 23maincommoditiescomprisingacountrystotal cropproduction,downscaledcrop
area, yield and production statistics can be compared with potential crop yield and production
resultsforbothrainfedandirrigatedcultivatedland.ThesecommoditiesarepresentedinTable91
below.
Note that for comparison of FAOSTAT statistical production (usually in harvested area or fresh
weight) with GAEZ simulated potential production (yield calculated in dry weight of main produce)
anappropriateconversionfactorisappliedtotheGAEZestimates.ConversionfactorsbetweenGAEZ
andFAOSTATdataaregiveninTable91.

97
Table91CommoditiesusedintheGAEZyieldgapanalysis
FAOSTAT
Crop/commodity
GAEZCrop/LUTequivalents
Wheat WheatLUTs(20)
Rice WetlandriceLUTs (8)
Maize GrainmaizeLUTs (18)
Sorghum SorghumLUTs (18)
Millet PearlmilletandfoxtailmilletLUTs(6)
Tubercrops PotatoandsweetpotatoLUTs(11)
Cassava/otherroots Cassava,yamandcocoyamLUTs(7)
Sugarbeet SugarbeetLUTs(8)
Sugarcane SugarcaneLUT (1)
Pulses PhaseolusbeanChickpea,Cowpea,Drypea,Grams,PigeonpeaLUTs(35)
Soybean SoybeanLUTs(6)
Rape RapeLUTs(10)
Sunflower SunflowerLUTs(6)
Groundnut GroundnutLUTs(3)
Oilpalm OilpalmLUT (1)
Olive OliveLUT(1)
Cotton CottonLUTs(7)
Comparisons of FAOSTAT compatible GAEZ area, yield and production with downscaled FAOSTAT
2000 respectively 2005 area yield and production statistics is presented in ratios and absolute
differences. The comparison isperformed on the cultivated land occurring within 5 arcminute grid
cells.
GAEZpotentialsrespectproductionpotentialsfromrainfedandirrigatedcultivatedareas.Twoinput
levels are used for the yield gap analysis low input potentials and mixed input potentials. The
definition of mixed input (see section 6.1.1) assumes high agricultural inputs and management on
the best land, intermediate inputs moderately suitable land and low inputs on marginal land. This
assumption is regarded as a reasonal reflection of actual agricultural input and management
circumstances.
Figure 93 presents apparent yield gap ratios (actual/potential production) comparing high input
levelpotentialswithactualyields.

Figure93Yieldgapratios
9.3 DescriptionofModuleVIIoutputs

98

Table92CommoditiesforDownscalingandyieldgapassessments
CROPS FAOSTAT(HARVESTEDWEIGHT) GAEZ(DRYWEIGHT) FAOSTAT
GAEZ
Code Name Commodities Produce Unit Price(GK$/t) Crop/LUTs Produce Unit Conversion
factor
1 Wheat Wheat grain tons 155 WheatLUTs grain tons 0.875
2 Rice Rice grain tons 200 WetlandriceLUTs grain tons 0.875
3 Maize Maize grain tons 125 GrainmaizeLUTs grain tons 0.87
4 Sorghum Sorghum grain tons 130 SorghumLUTs grain tons 0.88
5 Millet Millet grain tons 140,170 PearlmilletandfoxtailmilletLUTs grain tons 0.9
6 Othercereals Othercereals grain tons 92250 Barley,rye,oat,buckwheat,dryriceLUTs grain tons 0.8750.9
7 Tubers Potato,Sweetpotato tuber tons 105,85 PotatoandsweetpotatoLUTs tuber tons 0.25,0.3
8 Roots Cassava,Yams,otherrootsandPlantain root tons 75,95,120 Cassava,yam,cocoyamandplantainLUTs root tons 0.35
9 Sugarbeet Sugarbeet root tons 32 SugarbeetLUTs sugar tons 0.14
10 Sugarcane Sugarcane stalk tons 20 SugarcaneLUT sugar tons 0.1
11 Pulses Pulses grain tons 235500 Ph.bean,chickpea,cowpea,drypea,grams,pigeonpeaLUTs grain GK$ 1
12 Soybean Soybean grain tons 250 SoybeanLUTs grain tons 0.9
13 Rape Rapeseed seed tons 330 RapeLUTs seed tons 0.9
14 Sunflower Sunflower seed tons 300 SunflowerLUTs seed tons 0.9
15 Groundnut Groundnutsinshells grain tons 436 GroundnutLUTs grain tons 0.67
16 Oilpalm Oilpalm fruit tons 75 OilpalmLUT oil tons 0.225
17 Olive Olive fruit tons 500 OliveLUT oil tons 0.22
18 Cotton Cotton seed+lint tons 525,1430 CottonLUTs lint tons 0.35
19 Cashcrops1 Banana,Coconut fruit tons 150,105 Banana&coconutLUTs fruit,copra GK$ 0.35,0.175
20 Vegetables Vegetables various tons 1001650 VegetablesLUTs(cabbage,carrot,oinion,tomato) various GK$ 0.1250.175
21 Cashcrops2 Coffee,Tea,Cocoa beans,
leaves
tons 1000,1500,
750
CoffeeLUTs,teaLUTs,cocoaLUTs beans,cd.
leaves
GK$ 0.35,0.3,0.5
22 Fodder Fodder AGB tons 25 FodderLUTs AGB GK$ 0.1
23 Residual Othercropsnotlistedabove various tons 904500 n.a. n.a. n.a. n.a.
Cropsavailableinyieldandproductiongapassessments
PulsesinFAOSTATinclude:Drybeans,Drybroadbeans,Drypeas,Chickpeas,Cowpeas,Pigeonpeas,Lentils,Bambarabeans,otherpulses.
AGB=Abovegroundbiomass
99

References
Batjes, N.B. (199S). WorldInventoryofSoilEmissionPotentials(WISE2.1),ProfileDatabase,User
Manual, and Coding Protocols. Technical Papei 26, Inteinational Soil Refeience anu
Infoimation Centie, Wageningen, Netheilanus.
Batjes, N.B. (2uu2). Soil parameter estimates for the soil types of the world for use in global and
regional modeling (Version 2.1). ISRIC Repoit 2uu2u2c, Inteinational Foou Policy
Reseaich Institute (IFPRI) anu Inteinational Soil Refeience anu Infoimation Centie
(ISRIC), Wageningen, Netheilanus.
Batjes, N.B. (2uuS). A taxotiansfei iulebaseu appioach foi filling gaps in measuieu soil uata in
piimaiy S0TER uatabases (uEFS0C Pioject). Repoit 2uuSuS, ISRIC Woilu Soil
Infoimation, Wageningen, Netheilanus.
Batjes, N.B. (2uu6). ISRICWISE derived soil properties on a 5 by 5 arcminutes global grid (ver.
1.1). Repoit 2uu6u2, ISRIC Woilu Soil Infoimation, Wageningen, Netheilanus.
Batjes, N.B. (2uu7). SOTERbased soil parameter estimates for Central Africa DR of Congo,
Burundi and Rwanda (SOTWIScaf, version 1.0). ISRIC Woilu Soil Infoimation,
Wageningen, Netheilanus.
Batjes, N.B. (2uu8). SOTERparameterestimatesforSenegalandtheGambiaderivedfromSOTER
andWISE(SOTWISSenegal,version1.0).ISRIC Woilu Soil Infoimation, Wageningen,
Netheilanus.
Batjes, N.B., R. AlAuamat, T. Bhattachaiyya, N. Beinoux, C.E.P. Ceiii, P. uicheiu, P. Kamoni, E.
Nilne, B.K. Pal anu Z. Rawajfih (2uu7). PreparationofconsistentsoildatasetsforSOC
modelling purposes: secondary SOTER data sets for four case studyareas. Agiicultuie,
Ecosystems anu Enviionment 112, 26S4.
Batjes, N.B., E.N. Biiuges anu F.0. Nachteigaele (199S). World Inventory of Soil Emission
Potentials: development of a global soil data base of processcontrolling factors. In:
Peng S, KT Ingiam, B0 Neue anu LB Ziska (euitois), Climate Change anu Rice.
Spiingeiveilag, Beiuelbeig, pp. 1u211S.
Batjes, N.B., u. Fischei, F.0. Nachteigaele, v.S. Stolbovoy anu B.T. van velthuizen (1997). Soil
dataderivedfromWISEforuseinglobalandregionalAEZstudies(ver.1.0). Inteiim
Repoit IR97u2S, FA0 IIASA ISRIC, Laxenbuig
(http:www.iiasa.ac.atAuminP0BBocumentsIR97u2S.puf).
Beck, C., }. uiiesei, anu B. Ruuolf (2uu4). ANewMonthlyPrecipitationClimatologyfortheGlobal
Land Areas for the Period 1951 to 2000. Climate Status Repoit 0. ueiman Weathei
Seivice, ueimany. pp. 18119u.
Boei, u.}., u.N. Flato, N.C. Reauei, anu B. Ramsuen (2uuu). Atransientclimatechangesimulation
withhistoricalandprojectedgreenhousegasandaerosolforcing:Experimentaldesign
and comparison with the instrumental record for the 20th century. Climate Bynamics
16:4274Su.
BoonsPiins, E.R., u.B.}. ue Koning, C.A. van Biepen, anu F.W.T. Penning ue viies (199S). Crop
Specific Simulation Parameters for Yield Forecasting across the European Community.
Simulation Repoits CAB0TT, No. S2, Wageningen, Netheilanus.
Biammei, B., }. Antoine, A. Kassam, anu B.T. van velthuizen (1988). Land Resources of
BangladeshforAgriculturalDevelopment(Repoit 17), FA0, Rome, Italy.
100

BieuningNausen, B.B., anu R.}.A. }ones (1998). Towards a European soil profile analytical
database. In: Beineke B}, W Eckelmann, A} Thomasson, R}A }ones, L Nontanaiella anu
B Buckley (euitois), Lanu Infoimation Systems: Bevelopments foi planning the
sustainable use of lanu iesouices. 0ffice foi 0fficial Publications of the Euiopean
Community, Luxembouig, pp 4S49.
Biuinsma, }. (2uu9). The resource outlook to 2050: By how much do land, water use and crop
yields need to increase by 2050? Papei piesenteu at the Expeit meeting on how to
feeu the woilu in 2uSu. Foou anu Agiicultuie 0iganization of the 0niteu Nations
(FA0). Rome, Italy.
Biuggeman, B.Y., anu F.0. Nachteigaele (1986). MethodologyforIdentificationofPhysical
ConstraintstoAgriculturalDevelopment. Au:uCPRAS1u7}PN, Fielu Bocument S,
RAPA, FA0, Bangkok, Thailanu.
Buaique ue Bollanua, }., anu A.B. Poole (2uu1). SugarcaneasanenergysourceinBrazil.Rio ue
}aneiio, Biazil.
Bucklanu, B. (2uuS). The Oil for Ape Scandal: How Palm Oil Is Threatening OrangUtan Survival.
Fiienus of the Eaith, The Ape Alliance, The Boineo 0iangutan Suivival Founuation,
The 0iangutan Founuation (0K) anu the Sumatian 0iangutan Society.
Campbell, }.E., B.B. Lobell, R.C. uenova, anu C.B. Fielu, (2uu8). The global potential of bioenergy
onabandonedagriculturelands. Enviion. Sci. Technol. 42, S791 (2uu8).
CuIARCSI (2uu6). NASA Shuttle Rauai Topogiaphic Nission (SRTN). The SRTN uata is
available as S aic seconu (appiox. 9um iesolution) BENs. The uataset is available foi
uownloau at: http:sitm.csi.cgiai.oig
Cheng Chunshu (eu.) (199S). Climate and Agriculture in China. China Neteoiological Piess,
Beijing, China.
CIESIN (2uuu). Gridded Population of the World (GPW v. 2). CIESIN, Columbia 0niveisity, New
Yoik, NY, 0SA.
Colchestei, N., N. }iwan, et al. (2uu6). PromisedLand.PalmOilandLandAcquisitioninIndonesia:
Implications for Local Communities and Indigenous Peoples. Foiest Peoples
Piogiamme, Sawit Watch, BiNA anu ICRAu (Bisg.). Bogoi.
Conab (2uu8). Companhia Nacional de Abastecimento. Siies Bistoiicas.
http://www.conab.gov.br.
C0NCAWE, E0CAR, }RC (2uu6). Welltowheels analysis of future automotive fuels and
powertrains in the European context, WELLtoWBEELS iepoit, veision 2b; Nay
2uu6.
Coiley, R.B.v. anu P.B. Tinkei (2uuS). TheOilPalm, Blackwell Science.
Cramer,Jaqueline(groupchair),(2006).Criteriaforsustainablebiomassproduction. Available at:
http:www.globalpioblemsglobalsolutions
files.oigunf_websitePBFciiteiia_sustainable_biomass_piou.puf. Final Repoit of
pioject gioup sustainable piouuction of biomass. accesseu }anuaiy 2uu9.
Ciutzen, P.}., A.R. Nosiei, K.A. Smith, anu W. Winiwaitei (2uu8). N2O release from agrobiofuel
production negates global warming reduction by replacing fossil fuels. Atmospheiic
Chemistiy anu Physics 8: S89S9S.
ue Wit, C.T. (196S). Photosynthesis of Leaf Canopies. Agiicultuial Reseaich Repoit 66S, Puuoc,
Wageningen, Netheilanus.
101

Eastman, }. R. (2uu6). IBRISI Anues. Guide to GIS and Image Processing, Claik Labs. Claik
0niveisity, Woicestei, NA, 0SA.
EBI0 (Euiopean Biouiesel Boaiu) (2uu8). Statistics. The E0 Biouiesel Inuustiy. Available at:
http:www.ebbeu.oigstats.php anu http://www.ebbeu.org/EBBpressreleases/EBB%
2uBiochuie%2uFINAL%2u2S.u7.u8.puf, accesseu }anuaiy 2uu9.
EC (Euiopean Commission) (2uu8). Proposal for a directive of the European Parliament and of
theCouncilonthepromotionoftheuseofenergyfromrenewablesources. C0N(2uu8)
19 final, }anuaiy 2uu8.
EC Bu AuRI (Euiopean Commission Biiectoiate ueneial Agiicultuie anu Ruial Bevelopment)
(2uu8). BioeneryandtheCAP. Available at: http://ec.europa.eu/agriculture/bioenergy/
inuex_en.htm, accesseu }anuaiy 2uu9.
Elia Neto, A. (2uu8). gua na indstria da Canadeacar. In: Woikshop Piojeto PPP: Apecto
ambientais ua Caueia uo etanol ue canaue acai, Painel I, Sao Paulo, Biazil.
Eliasson, A., R.}.A. }ones, F. Nachteigaele, B.u. Rossitei, }N. Teiies, }. van 0ishoven, B. van
velthuizen, K. Boettchei, P. Baastiup, C. Le Bas (2u1u). Common criteria for the
redefinition of Intermediate Less Favoured Areas in the European Union.
Enviionmental Science anu Policy, 1S(8):766777 (Becembei 2u1u).
ER0S Bata Centei, (1998). Global 30 arcsecond Digital Elevation Model.
http://edcwww.cr.usgs.gov Anonymous ftp access at eucftp.ci.usgs.gov.
ER0S Bata Centei, (2uuu). LandCoverCharacteristicsDatabase,Version2.0,
http://edcwww.cr.usgs.gov. Anonymous ftp access at eucftp.ci.usgs.gov.
ESB, (2uu4). EuropeanSoilDatabase(vs2.0). Euiopean Commission }RC Institute foi
Enviionment anu Sustainability, Euiopean Soil Buieau. Ispia, Italy.
Eschweilei, B. (199S). Biaft Physiogiaphic Nap of Latin Ameiica. AuLS Woiking Papei. FA0,
Rome, Italy.
ESRI, (1984). Final Report UNEP/FAO World and Africa GIS Database. Enviionmental Systems
Reseaich, Institute, Reulanus, Califoinia, 0SA.
ESRI, (1999). ExtendableImage Formats for ArcView GIS 3.1 and 3.2. An ESRI White Papei, }uly
1999.
Eulei, B. anu B. uoiiitz (2uu4). CasestudyonJatrophacurcas. Stuuy commissioneu by the ulobal
Facilitation 0nit foi 0nueiutilizeu Species (uF0) anu the Beutsche uesellschaft fi
Technische Zusammenaibeit (uTZ).
Eui0bseivER (2uu8). Biofuels Baiometei. 7.7 consumeu in E0 in 2uu7. Available at:
http://www.energiesrenouvelables.org/obseiveistat_baioobseivbaio18S.puf,
accesseu in Becembei 2uu8.
Euiostat (2uu8). Agiicultuial statistics: Nain iesults 2uu62uu7. Available at:
http://epp.eurostat.ec.europa.eu/cache/ITY_OFFP 0BKSEBu8uu1ENKSEBu8
uu1EN.PBF
Faaij, A., Alfieu Szwaic anu Ainaluo Waltei (2uu8). Demand for bioethanol for transport. In:
Zuuibiei anu van ue vooien (eus.) (2uu8). Sugaicane ethanol: Contiibutions to
climate change mitigation anu the enviionment. Wageningen Acauemic Publisheis,
ISBN 9789u8686u9u6.
102

FA0 (1976). Irrigation and Drainage Paper 46. Lanu anuWatei Bevelopment Bivision. FA0,
Rome, Italy.
FA0 (1976a). Aframeworkforlandevaluation. FA0 Soils Bulletin S2. FA0, Rome, Italy.
FA0 (1976b). Water Quality for Irrigation. FA0 Biainage anu Iiiigation Papei 29, FA0, Rome,
Italy.
FA0 (197881). Report on the AgroEcological Zones Project. Woilu Soil Resouices Repoit 48,
FA0, Rome, Italy.
FA0 (197881b). Crop adaptability inventory. In: Repoit on the AgioEcological Zones Pioject.
Woilu Soil Resouices Repoit 48, FA0, Rome, Italy.
FA0 (1979). YieldResponsetoWater. Biainage anu Iiiigation Papei SS, FA0, Rome, Italy.
FA0 (198u). LandResourcesforPopulationsoftheFuture. FA0, Rome, Italy.
FA0 (198u). Report on the Second FAO/UNFPA Expert Consultation on Land Resources for
PopulationsoftheFuture, iepiinteu 1984. FA0, Rome, Italy.
FA0 (1984). Tillage systems for soil and water conservation. FA0 Soil Bulletin S4. FA0, Rome,
Italy.
FA0 (1984a). LandEvaluationforDevelopment. FA0, Rome, Italy.
FA0 (1984b). Guidelines: Land Evaluation for Rainfed Agriculture. FA0 Soils Bulletin S2, FA0,
Rome, Italy.
FA0 (198S). Guidelines: Land Evaluation for Irrigated Agriculture. FA0 Soils Bulletin SS, FA0,
Rome, Italy.
FA0 (1988). LandResourcesAppraisalofBangladesh. Repoit 6, FA0, Rome, Italy.
FA0 (199u). GuidelinesforSoilDescription. FA0, Rome, Italy.
FA0 (1992). Crop water requirements. FA0 Iiiigation anu Biainage Papei no. 24. Lanu anu
Watei Bevelopment Biv. , 1992 , 144 p. Accession No: 412881. }. Booienbos, W. 0.
Piuitt, A. Aboukhaleuet al. Rome, Italy.
FA0 (1992). CROPWAT: A Computer Program for Irrigation Planning and Management. FA0
Iiiigation anu Biainage Papei 46. Lanu anu Watei Bevelopment Bivision, FA0, Rome,
Italy.
FA0 (1992b). ReportontheExpertConsultationonRevisionofFAOMethodologiesforCropWater
Requirements. Lanu anu Watei Bevelopment Bivision, FA0, Rome, Italy.
FA0 (199S). Agroecological Assessment for National Planning: The Example of Kenya. FA0,
Rome, Italy.
FA0 (199S). DigitalSoilMapoftheWorldandDerivedSoilProperties(Version3.5). CBR0N, FA0,
Rome, Italy.
FA0 (199Sa). Planning for Sustainable Use of Land Resources: Towards a New Approach. FA0
Lanu anu Watei Bulletin 2, Lanu anu Watei Bevelopment Bivision, FA0, Rome, Italy.
FA0 (199Sb). FAOCLIM Users Manual: A CDROM with Worldwide Agroclimatic Database.
Agiometeoiology Seiies Woiking Papei No. 11, FA0, Rome, Italy.
103

FA0 (199Sc). Digital Soil Map of the World and Derived Soil Properties (Version 3.5). CBR0N,
FA0, Rome, Italy.
FA0 (199Su). World Agriculture: Towards 2010, AnFAOStudy. Euiteu by N. Alexanuiatos, Foou
anu Agiicultuial 0iganization of the 0niteu Nations anu }. Wiley & Sons, Englanu, pp.
488.
FA0 (1996). ECOCROP1,CropEnvironmentalRequirementsDatabase. FA0, Rome, Italy.
FA0 (1998). Crop Evapotranspiration: Guidelines for computing crop water requirements. FA0
Biainage anu Iiiigation Papei S6. FA0, Rome, Italy.
FA0 (2uuu). Agriculture: Towards 2015/30. Technical Inteiim Repoit, ulobal Peispective
Stuuies 0nit, FA0, Rome, Italy.
FA0 (2uu1). ulobal Foiest Resouices Assessment 2uuu. Foiestiy Papei No. 14u. Foou anu
Agiicultuie 0iganization of 0niteu Nations, Rome, Italy.
FA0 (199S, 2uuS). TheDigitizedSoilMapoftheWorldandDerivedSoilProperties. (veision S.S)
FA0 Lanu anu Watei Bigital Neuia Seiies 1. FA0, Rome, Italy.
FA0 (2uuS). World Agriculture: Towards 2015/1030. A FAO Perspective. Biuinsma } (Eu.), Foou
anu Agiicultuial 0iganization of the 0niteu Nations, Rome, Italy anu Lonuon.,
Eaithscan, Lonuon, 0K. 4S2 pp.
FA0(2uuS). Mappingglobalurbanandruralpopulationdistributions, by N. Salvatoie, F. Pozzi, E.
Ataman, B. Buuuleston & N. Bloise. Enviionmental anu Natuial Resouices Woiking
Papei 24. Rome, Italy.
FA0 (2uu6). Food Insecurity, Poverty and Environment Global GIS Database (FGGD) and Digital
Atlas for the Year 2000, by E. Ataman, N. Salvatoie, B. Buuuleston, N. Zanetti, N.
Bloise, N. Ascione, }.F. Booley, B. van velthuizen, u. Fischei & F.0. Nachteigaele.
Enviionmental anu Natuial Resouices Woiking Papei No. 26. Rome, Italy.
FA0 (2uu6a). World Agriculture Toward 2030/2050. Foou anu Agiicultuie 0iganization of
0niteu Nations, Rome, Italy.
FA0 (2uu6b). GlobalForestResourcesAssessment2005. Foiestiy Papei No. 147. Foou anu
Agiicultuie 0iganization of 0niteu Nations, Rome, Italy.
FA0 (2uu7). GlobalAdministrativeUnitLayers(GAUL), available at:
http://www.fao.org/geonetwork/srv/en/metadata.show?id=12691&currTab=simple
FA0 (2uu8). TheStateofFoodandAgriculture. Biofuels: piospects, iisks anu oppoitunities. Foou
anu Agiicultuie 0iganization of 0niteu Nations, Rome, Italy. 1S8 p.
FA0 (2uu9). The State of Food Insecurity in the World 2009. Foou anu Agiicultuie 0iganization
of the 0niteu Nations, Rome, Italy.
FA0CAS (199S). Introduction on the Attributes of Legends in Soil Map of China. Compileu anu
euiteu by uong Zitong, Zhou Buizhen, Shi Xuezheng, anu Luo uuobao. Nanjing, China.
FA0 IuABB Italian Coopeiation, (1998). SoilandterraindatabasefornortheasternAfricaand
Cropproductionzones. Lanu anu Watei Bigital Neuia Seiies # 2. FA0, Rome, Italy.
FA0IIASA (1991). AgroEcological Land Resources Assessment for Agricultural Development
Planning A Case Study for Kenya, Main Report and Annex (Vols. 19). Woilu Soil
Resouices Repoit 71, FA0, Rome, Italy.
104

FA0IIASA (1999). Soil and Terrain Database for North and Central EURASIA (Version 1.0), CB
R0N, FA0, Rome. (FA0 Lanu anu Watei Bigital Neuia Seiies #7 (1999)).
FA0IIASA (2uuS). Global agroecological assessment for citrus, cocoa, coffee, tea, tobacco, yam
and selected vegetables: methodology and results, by u. Fischei, B. van velthuizen &
F.0. Nachteigaele. Inteiim iepoit of woik in piogiess.
FA0IIASABokuchaiev InstituteAcauemia Sinica Soil anu Teiiain (1999). Database for north
and 1999 central Eurasia at 1:5 million scale. FA0 Lanu anu Watei Bigital Neuia
seiies 7. FA0, Rome, Italy.
FA0IIASAISRICISSCAS}RC (2uu9). Harmonized World Soil Database (version 1.1). FA0 anu
IIASA. Rome, Italy anu Laxenbuig, Austiia.
FA0IIASA0NFPA (1982). PotentialPopulationSupportingCapacitiesofLandsintheDeveloping
World. Technical Repoit on Pioject INTS1S, FA0, Rome, Italy.
FA0ISRIC (2uuu) Soil and Terrain Database, Land Degradation Status and Soil Vulnerability
Assessment for Central and Eastern Europe Version 1.0 (1:2.5 million scale). Lanu anu
Watei Bigital Neuia Seiies # 1u. FA0, Rome, Italy.
FA0ISRIC (2uuS) SoilandTerrainDatabaseforSouthernAfrica. Lanu anu Watei Bigital Neuia
Seiies # 26. FA0, Rome, Italy.
FA0ISRIC0niveisity of uent (2uu7). SoilandTerrainDatabaseofCentralAfrica BR of Congo,
Buiunui anu Rwanua (S0TERCAF 1.u).
FA0I0SSISRIC (2uu6). Woilu Refeience Base (WRB) foi Soil Resouices. A fiamewoik foi
inteinational classification anu communication. Woilu Soil Resouices Repoit 1uS.
FA0, Rome, Italy.
FA00NEP (1977). AssessingSoilDegradation. FA0 Soils Bulletin S4, FA0, Rome, Italy.
FA00NEPISRICCIP (1998). Soil and terrain digital database for Latin America and the
Caribbeanat1:5Millionscale. FA0 Lanu anu Watei Bigital Neuia seiies S. FA0, Rome,
Italy.
FA00NESC0 (19711981). The FAOUnesco Soil Map of the World. Legenu anu 9 volumes.
0NESC0, Paiis, Fiance.
FA00NESC0 (1981). Soil Map of the World, 1:5,000,000. Vol. 5 Europe. 0niteu Nations
Euucational, Scientific, anu Cultuial 0iganization, Paiis, Fiance.
FA00NESC0 (1974). LegendoftheSoilMapoftheWorld. FA0, Rome, Italy.
FA00NESC0ISRIC (199u). Revised Legend of the Soil Map of the World. Woilu Soil Resouices
Repoit 6u, FA0, Rome, Italy.
FA0 Woilu Bank (2uu1). Farming systems and poverty, by }. Bixon, A. uullivei & B. uibbon.
Rome anu Washington B.C.
FA0STAT (Timeseiies anu cioss sectional uata ielating to foou anu agiicultuie). FA0. Rome.
Available at: http:faostat.fao.oiguefault.aspx.
Finke, P., R. Baitwich, R. Buual, }. Ibanez, N, }amagne, B, King, L, Nontanaiella anu N, Yassoglou
(1998). Georeferenced soil database for Europe. Manual of procedures (ver. 1.0).
Repoit E0R 18u92, Euiopean Soil Buieau, }oint Reseaich Centie, Ispia.
105

Fischei, u. anu B. van velthuizen (199S). World agriculture: Towards 2010. An FAO study. In:
Woilu Agiicultuie: Towaius 2u1u, N. Alexanuiatos (eu.), }ohn Wiley & Sons,
Chichestei, 0K.
Fischei, u. anu B. van velthuizen (1996). Climate Change and Global Agricultural Potential
Project: A Case Study of Kenya. Inteinational Institute foi Applieu Systems Analysis,
Laxenbuig, Austiia. 96 pp.
Fischei, u. anu u.K. Beilig (1997). Population momentum and the demand on land and water
resources. Philosophical Tiansactions of the Royal Society of LonuonB, SS2:869889.
Fischei, u., u. uianat, anu N. Nakowski (1998). AEZWIN.AnInteractiveMultipleCriteriaAnalysis
Tool for Land Resources Appraisal. Inteiim Repoit IR98uS1, Inteinational Institute
foi Applieu Systems Analysis, Laxenbuig, Austiia.
Fischei, u., anu B. van velthuizen (1999). AgroecologicalzonesofChina,theformerSovietUnion
and Mongolia. Inteinational Institute foi Applieu Systems Analysis, Laxenbuig,
Austiia.
Fischei, u., B. van velthuizen, F.0. Nachteigaele (2uuu). Global AgroEcological Zones
Assessment: Methodology and Results. IIASA Inteiim Repoit IRuuu64 |Novembei
2uuu, SS8 ppj.
Fischei, u., B. van velthuizen, N. Shah anu F.0. Nachteigaele (2uu2a). Global AgroEcological
AssessmentforAgricultureinthe21stCentury:MethodologyandResults. IIASA RRu2
u2, IIASA, Laxenbuig, Austiia.
Fischei, u., N. Shah anu B. van velthuizen (2uu2b). Climate Change and Agricultural
Vulnerability. Special Repoit as contiibution to the Woilu Summit on Sustainable
Bevelopment, }ohannesbuig 2uu2. Inteinational Institute foi Applieu Systems
Analysis, Laxenbuig, Austiia. pp 1S2.
Fischei, u., N. Shah, F.N. Tubiello, B. van velhuizen (2uuS). Socioeconomic and climate change
impactsonagriculture:anintegratedassessment,19902080.Phil. Tians. Royal Soc. B,
uoi:1u.1u98istb.2uuS.1744.
Fischei, u., S. Piielei, B. van velthuizen (2uuS). Biomass potentials of miscanthus, willow and
poplar:Results and policy implications for Eastern Europe, Northern and CentralAsia.
Biomass anu Bioeneigy, 28(2):1191S2 |2uuSj.
Fischei, u., N.N.Shah, B. van velthuizen, F.0. Nachteigaele, (2uu6).Agroecologicalzones
assessment. In: Lanu 0se, Lanu Covei anu Soil Sciences E0LSS, W.B. veiheye (eu.),
Eolss Publisheis, 0xfoiu, 0K |2uuSj.
Fischei, u., T. Eimolieva, Y. Eimoliev, B. van velthuizen, (2uu6). Spatialrecoveringof
agriculturalvaluesfromaggregateinformation:Sequentialdownscalingmethods.
Inteinational }ouinal of Knowleuge anu Systems Sciences, S(1): |2uu6j.
Fischei, u., F.N. Tubiello, B. van velthuizen, B. Wibeig (2uu7). Climate change impacts on
irrigation water requirements: Effects of mitigation, 19902080. Technological
Foiecasting & Social Change (Special Issue: uieenhouse uases Integiateu
Assessment), 74(7):1u8S11u7 (2uu7).
Fischei, u., E. Teixeiia, E. BizsnyikTothne, B. van velthuizen (2uu8). Landusedynamicsand
sugarcaneproduction. In: Sugaicane Ethanol. Contiibutions to Climate Change anu
the Enviionment, P. Zuuibiei anu }. van ue vooien (eus), Wageningen Acauemic
Publisheis, Wageningen, The Netheilanus pp.2962 |2uu8j.
106

Fischei, u. (2uu9). Worldfoodandagricultureto2030/50:Howdoclimatechangeandbioenergy

alterthelongtermoutlookforfood,agricultureandresourceavailability?Pioceeuings
of the Expeit Neeting on Bow to Feeu the Woilu in 2uSu, 2426 }une 2uu9. Foou anu
Agiicultuie 0iganization of the 0niteu Nations, Rome, Italy. Available online at
ftp:ftp.fao.oiguociepfaou12akS42eakS42euu.puf
Fischei, u., E. Bizsnyik, S. Piielei, N. Shah, B. van velthuizen (2uu9a). BiofuelsandFoodSecurity.
0FIBIIASA, viennaLaxenbuig. <Available at:
http:www.iiasa.ac.atReseaichL0CBomepageNews
BighlightsBiofuels%2uRepoit%2uFinal.puf. accesseu in Nai 2u1u>.
Fischei, u., B. van velthuizen , E. Bizsnyik , B. Wibeig (2uu9b). Potentially obtainable yields in
the semiarid tropics. ulobal Theme on Agioecosystems, Repoit no. S4. Inteinational
Ciops Reseaich Institute foi the SemiAiiu Tiopics, Patancheiu Su2 S24, Anuhia
Piauesh, Inuia.
Flato, u.N., u.}. Boei, W.u. Lee, N.A. NcFailane, B. Ramsuen, N.C. Reauei, anu A.}. Weavei (2uuu).
TheCanadianCentreforClimateModellingglobalcoupledmodelanditsclimate.
Flato, u.N. u.}. Boei, W.u. Lee, N.A. NcFailane, B.Ramsuen, N.C. Reauei, anu A.}. Weavei (1998)
The Canadian Centre for Climate Modellingand Analysis GlobalCoupled Model and its
Climate, submitteu to Climate Bynamics.
FRA, (2uuu). ulobal Foiest Resouices Assessment. Foou anu Agiicultuial 0iganization of the
0niteu Nations, FA0 Foiestiy Papei 14u, Nain Repoit. Rome, Italy. ISSN u2S861Su.
FRA, (2uuS). Global Forest Resources Assessment Progress towards sustainable forest
management. Foou anu Agiicultuial 0iganization of the 0niteu Nations, FA0 Foiestiy
Papei 147, Nain Repoit. Rome, Italy. ISBN 92S1uS4819.
FRA, (2u1u). GlobalForestResourcesAssessment2010.Foou anu Agiicultuial 0iganization of the
0niteu Nations, FA0 Foiestiy Papei 16S, Nain Repoit. Rome, Italy. ISBN 97892S
1u66S46
uLC, (2uuu) GlobalandRegionalLandCoverMaps.Euiopean Commission, }oint Reseaich Centie,
ulobal Enviionment Nonitoiing 0nit.
uLCCB (2uu1). Global Land Cover Characteristics Database. veision 2.u. available online at:
http:eucuaac.usgs.govglccglcc.html.
uleick, P (2uu9). The World's Water 20082009: The Biennial Report on Freshwater Resources.
Islanu Piess, Washington, BC.
uoufiay, C., }. Beuuington }, I. Ciute I, L. Bauuau L, B. Lawience B, }. Nuii }, }. Pietty }, S. Robinson
S, S. Thomas S, C. Toulmin C (2u1u). FoodSecurity:TheChallengeofFeeding9Billion
People. Science S27: 812818.
uiaham, S.T., }.S., Famiglietti, B.R., Naiument, (1999) Fiveminute, 1/2 and 1 data sets of
continental watersheds and river networks for use in regional and global hydrologic
andclimatesystemmodelingstudies. Watei Resoui. Res. SS: S8SS87.
Backett, C., (1991). PLANTGRO:ASoftwarePackageforCoarsePredictionofPlantGrowth. CSIR0,
Nelbouine, Austialia.
107

van Beemst B.B.}., (1988). PlantDataValuesrequiredforSimpleCropGrowthSimulationModels:

Review and Bibliography. Simulation Repoit CaboTT, Centie foi Agiobiological
Reseaich (CAB0) anu Bepaitment of Theoietical Piouuction Ecology, Agiicultuial
0niveisity Wageningen, 1uup. Netheilanus. http://www
eosdis.ornl.gov/SOILS/igbp.html.
Beilig u.K., u. Fischei , B. van velthuizen (2uuu). CanChinafeeditself?AnanalysisofChina'sfood
prospects with special reference to water resources. The Inteinational }ouinal of
Sustainable Bevelopment anu Woilu Ecology, 7(S):1SS172 |2uuuj.
Biggins u.N., A.B. Kassam, }.N. Kowal, S. Saiiaf, B.}.N. Ainoluussen, N. Fieie, }. Biabovszki, B.
van velthuizen (1978). ReportontheAgroecologicalzonesproject.Vol.1Methodology
andresultsforAfrica. Woilu soil iesouices iepoit No 48, AuLS, FA0, Rome, Italy.
Biggins u.N., A.B. Kassam, B. van velthuizen, N.F. Puinell (1987). Methods Used by FAO to
Estimate Environmental Resources, Potential Outputs of Crops, and Population
Supporting Capacities in Developing Nations. In Agiicultuial Enviionments, C.A.B.
Wallingfoiu, 0K.
IEA (Inteinational Eneigy Agency) (2uu9). Woilu Eneigy 0utlook 2uu9. 0ECBIEA. Paiis. 691 p.
IFPRI (2uu2). GlobalAgriculturalExtentv2.0.Reinterpretationofgloballandcover
characteristicsdatabase(GLCCDv.2.0). Available online at:
http:www.asb.cgiai.oigBNPPphase2bnpp. Phase 2 uatasets. ulobal agiicultuial
extent. ER0S uata centei (EBC), 2uuu. Washington BC, Inteinational Foou Policy
Reseaich Institute.
IuBPBIS (2uuu). GlobalSoilDataProductsCBR0N (IuBPBIS). Available at:
IPCC (2uu7). ClimateChange2007:ThePhysicalScienceBasis. Contiibution of Woiking uioup I
to the Fouith Assessment Repoit of the Inteigoveinmental Panel on Climate Change.
Cambiiuge 0niveisity Piess, Cambiiuge, 0K.
}RC (2uu6). Global2000LandCover(GLC2000). Retiieveu Becembei 2uu9, fiom
http:bioval.jic.ec.euiopa.eupiouuctsglc2uuuglc2uuu.php
Kassam, A.B. (1977). NetBiomassProductionandYieldofCrops. FA0, Rome, Italy.
Kassam, A.B., B. van velthuizen, u.N. Biggins, A. Chiistofoiiues, R.L. vooitman, B. Spieis (198S).
AssessmentofLandResourcesforCropProductioninMozambique. (6 volumes)
FA00NBP, Rome, Italy.
Kassam A.B, anu B. van velthuizen (1984). EnvironmentalConstraintstoRainfedCereals
ProductioninAfrica,aCountryAssessment. In Auvancing Agiicultuial Piouuction in
Afiica, Pioceeuings of CAB's fiist scientific confeience, Aiusha, Tanzania.
Kassam A.B, B. van velthuizen, u. Fischei, N.N. Shah (1991). Agroecological Land Resources
Assessment for Agricultural Development Planning, Woilu Soil Resouices Repoits
(71vol. 18) FA0IIASA, Rome.
Kassam A.B, B. van velthuizen, u. Fischei, N.N. Shah (199S). Agroecological Assessments for
National Development Planning: The Example of Kenya, FA0 Soils Bulletin 67, Rome,
Italy.
Kassam A.B, B. van velthuizen, K.B. Singh (1991). LandEvaluationforForestResource
DevelopmentPlanningatNationalLevel. FA0, Rome.
108

Kineman, }.}. anu N.A. 0hienschall (1992). GlobalEcosystemsDatabaseVersion1.0(onCDROM)

DiscA, Documentation Manual. Key to ueophysical Recoius Bocumentation No. 27,
0SB0CN0AA, Bouluei, C0, 0SA.
Kowal, }. anu A.B. Kassam (eus.) (1978). Agricultural Ecology of Savanna. Claienuon Piess,
0xfoiu, 0K.
Lambeit }}., }. Baioussin, N. Eimbeick, C. Le Bas, N. }amagne, B. King, L. Nontanaiella (2uuS).
Soil Geographical Database for Eurasia & The Mediterranean. Instructions Guide for
Elaborationatscale1:1,000,000version4.0. E0R 2u422 EN. }RC, Ispia, Italy. 64 p.
Leemans, R. anu W. Ciamei, (1991). The IIASA Database for Mean Monthly Values of
Temperature, Precipitation and Cloudiness of a Global Terrestrial Grid. Reseaich
Repoit RR9118, Inteinational Institute foi Applieu Systems Analysis, Laxenbuig,
Austiia.
Licht, F.0. (Licht Inteiactive Bata) (2uu7). WorldEthanolandBiofuelsReport,0ct 2S, 2uu7
Licht, F.0. (Licht Inteiactive Bata) (2uu8). WorldEthanolandBiofuelsReport, Nay S, 2uu8
Lonuo, N., S. Lensink, E. Beuiwaaiuei, A. Wakkei, S. Piielei, u. Fischei, E. Bizsnyik, B. van
velthuizen, N. ue Wit, A.P.C. Faaij, N. }ungingei, B. Buei, }. Lunubaek, P. Chiistensen,
A. uainak; u. Beinues; }. Bansson; A. Egeskog; S. Weinei; u. Wisniewski; A. Kupczyk;
B. Ruciski, K. Konighofei, u. }ungmeiei, (2uu8) Eyesonthetrack,Mindonthe
horizon.Frominconvenientrapeseedtocleanwood:AEuropeanroadmapforbiofuels,
FinalreportoftheREFUELproject. Repoit commissioneu anu publisheu by ECN
Eneigy ieseaich Centie of the Netheilanus, Petten, the Netheilanus, 2uu8, Su pp.
Naceuo, I.C., }.E.A. Seabia anu }.E.A.R. Silva (2uu8). Greenhousegasesemissionsintheproduction
anduseofethanolfromsugarcaneinBrazil:The2005/2006averagesandaprediction
for2020. Biomass anu Bioeneigy S2(7): S82S9S.
Naitinelli, L. A. anu S. Filoso, (2uu8). Expansion of sugarcane ethanol production in Brazil:
environmentalandsocialchallenges.Ecological Applications 18(4): 88S898.
NcCoinick, P.u., S.B. Awulachew anu N. Abebe, (2uu8). Waterfoodenergyenvironmentsynegies
and tradeoffs: Major issues and case studies. In: Watei Policy volume 1u Supplement
(2uu8).
NcCiee, K.}., (1974). Equationsforrateofdarkrespirationofwhitecloverandgrainsorghum,as
functions of dry weight, photosynthesis rate and temperature. Ciop Science, 14:Su9
S14.
NEA, (2uuS). Ecosystems and human wellbeing: Synthesis. Nillennium Ecosystem Assessment.
Washington, BC: Islanu Piess, Washington, 0SA.
NcLaughlin, S., }. Bouton, B. Biansby, B. Congei, W. 0cumpaugh, B. Paiiish, C. Taliafeiio, K.
vogel, anu S. Wullschlegei (1999). Developingswitchgrassasabioenergycrop. p. 282
299. In: }. }anick (eu.), Peispectives on new ciops anu new uses. ASBS Piess,
Alexanuiia, vA. 0nline at:
http://www.hort.purdue.edu/newcrop/pioceeuings1999v4282.html .
Nitchell, T. B., anu P.B. }ones, (2uuS). Animprovedmethodofconstructingadatabaseofmonthly
climate observations and associated high resolution grids. Noiwich, 0K Climatic
Reseaich 0nit, School of Enviionmental Sciences, 0niveisity of East Anglia.
Nittelbach N., (2uu8). InnovativeBiofuelsFeedstock.2nu National IEA Bioeneigy Task S9 Liquiu
Biofuels Woikshop. 9 Septembei 2uu8, vienna.
109

Available at: http://www.abcenergy.at/pdfs/ieaworkshop/Mittelbach.pdf, accesseu

}anuaiy 2uu9.
Nonfieua et al. (2uu8). Farming the planet: 2. Geographic distribution of crop areas, yields,
physiological types, and net primary production in the year 2000, ulobal
Biogeochemical Cycles, vol.22, uB1u22, uoi:1u.1u292uu7uBuu2947.
Nonteith, }. L., (196S). Evapotranspiration andthe environment. In: The State anu Novement of
Watei in Living 0iganisms. XIXth Symposium. Cambiiuge, 0K, Society foi Xp. Biology,
Swansea: 2uS2S4.
Nonteith, }.L., (1981). Evapotranspiration and surface temperature. Quaiteily }ouinal Royal
Neteoiological Society, 1u7:127.
Nuiphy, }.N., (199S). TransientresponseoftheHadleyCentreCoupledModeltoincreasingcarbon
dioxide.PartI:Controlclimateandfluxadjustment. Climate, 8:SSS6, 496S14.
Nuiphy, }.N. anu }.F.B. Nitchell, (199S). TransientresponseoftheHadleyCentreCoupledModelto
increasingcarbondioxide.PartII:Temporalandspatialevolutionofpatterns. Climate,
8:S78u.
Nyeis, Noiman (1998). Thenextgreenrevolution:Itsenvironmentalunderpinnings.Piesenteu at
the Inteiuisciplinaiy Bialogue on Nalthus anu Nenuel: Population, Science anu
Sustainable Secuiity, helu at N. S. Swaminathan Reseaich Founuation, Chennai, Inuia,
28 }anuaiy 1998. Cuiient Science 76 (4), pp. Su7S1S. Available at:
http:www.ias.ac.incuiiscifeb2Saiticles16.htm.
Nachteigaele, F.0., (1988). Methodology for Regional Analysis of Physical Potential for Crop
Production. Au:uCPRAF2Su}PN, Fielu Bocument 1, KILIN0, Bai es Salaam.
Niciofilm, FA0, Rome, Italy.
Nachteigaele F.0., B. van velthuizen, L. veielst (2uu9). Harmonized World Soil Database. FA0,
Rome, Italy, anu IIASA, Laxenbuig, Austiia (Naich 2uu9).
Nakicenovic, N. anu R. Swait (eus.), (2uuu). Special Report on Emission Scenarios.
Inteigoveinmental Panel on Climate Change, Cambiiuge 0niveisity Piess, Cambiiuge,
0K |ISBN uS21 8uu811j. p. S7u.
Nakicenovic, N., }. Alcamo, A. uiueblei, K. Riahi, R.A. Roehil, B.B. Rognei, N. victoi, anu otheis,
(2uuu). Special Report on Emissions Scenarios (SRES). A Special Repoit of Woiking
uioup III of the Inteigoveinmental Panel on Climate Change. Emission Scenaiios.
Inteigoveinmental Panel on Climate Change. R. Swait. Cambiiuge, 0K, Cambiiuge
0niveisity Piess: S9S.
Nelson, F. E., anu S.I. 0utcalt, (1987). A computational methou foi pieuiction anu iegionalization
of peimafiost. Aictic anu Alpine Reseaich 19(S): 279288.
Nelson, A., (2uu8). Estimatedtraveltimetothenearestcityof50,000ormorepeopleinyear2000.
ulobal Enviionment Nonitoiing 0nit }oint Reseaich Centie of the Euiopean
Commission, Ispia Italy. Available at http:bioval.jic.ec.euiopa.eupiouuctsgam.
New, N.u., N. Bulme, anu P. B. }ones, (1998). Representing 20th century spacetime climate
variability.PartI:Developmentofa19611990meanmonthlyterrestrialclimatology. }.
Climate. (http://www.cru.uea.ac.uk).
New, N., B. Listei, N. Bulme, anu I. Nakin, (2uu2). A highresolution data set of surface climate
overgloballandareas.Climate Reseaich 21.
110

New, N., N. Bulme, P.B. }ones, (2uuu). Representing twentieth century spacetime climate
variability.Part2:developmentof190196monthlygridsofterrestrialsurfaceclimate.
}ouinal of Climate 1S: 221722S8.
Nielsen, B.C., F.N.F. vigil, R.L. Anueison, R.A. Bowman, }.u. Benjamin anu A.B. Balvoison.,
(2uu2). CroppingInfluenceonplantingwatercontentandyieldofwinterwheat. Agion
}. 94: 962967.
Nye, P.B. anu B.}. uieenlanu, (196u). The Soil under Shifting Cultivation. Comm. Bui. Soils,
Baipenuen, 0K.
0beihubei, }.N., (199S). SimulationoftheAtlanticcirculationwithacoupledseaicemixedlayer
isopycnal general circulation model. Part I: Model description. }ouinal of Physical
0ceanogiaphy, 1S:8u8829.
0lson, }.S., (1994). Global ecosystem frameworkdefinitions: USGS EROS Data Center Internal
Report, Sioux Falls, SB, 0SA. S7p.
Papauakis, }., (197ua). Climates of the world. Their classification, similitudes, differences and
geographicdistribution. Buenos Aiies, Aigentina, 1966, 17up.
Papauakis, }., (197ub). Agriculturalpotentialitiesoftheworldclimates. Buenos Aiies, Aigentina
197u 7up.
Pelkmans, L., E. Poitouli, A. Papageoigiou anu P. ueoigopoulos (2uu6). Impact assessment of
measurestowardstheintroduction ofbiofuelsintheEuropeanUnion,PRENIA iepoit,
vIT0, Belgium, Septembei 2uu6.
Pelkmans, L., K. Kessels anu L. uovaeits (2uu8). Inventory of biofuel policy measures and their
impactonthemarket.Repoit B2.1 of EL0BI0 subtasks 2.12.2. vIT0 Belgium.
Platou, S.W., A.B. Nii, B.B. Nausen, (1989). DigitizationoftheECSoilMap. In: Computeiization
of lanu use uata. }ones R. }. A., Biagi B. (eus.). E0R 111S1 EN. CEC, 0ffice foi 0fficial
Publications of the EC, Luxembouig, 1S6.
Pleijsiei, K., (1989). Variabilityinsoildata. In: Bouma } anu AK Biegt (euitois), Lanu Qualities in
Space anu Time. P0B0C, Wageningen, pp 8998.
Renewable Fuels Association (RFA), (2uu8a). EthanolIndustryStatistics. Available at:
www.ethanolifa.oigpagesstatistics , accesseu in Becembei 2uu8.
Renewable Fuels Association (RFA), (2uu8b). Biofuelcoproducts.Available at:
http:ethanolifa.oigpagesinuustiyiesouicescopiouucts , accesseu in Becembei
2uu8.
Riahi, K., A. uiueblei anu N. Nakicenovic, (2uu6). Scenarios of longterm socioeconomic and
environmental development under climate stabilization. Technological Foiecasting &
Social Change. uoi:1u.1u16j.techfoie.2uu6.uS.u26.
Robeitson, u.P., v.B. Bale, 0.C. Boeiing, S.P. Bambuig, }.N. Nelillo, N.N. Wanuei, W.}.Paiton,
W.W. Wilhelm, (2uu8). Agriculture: Sustainable biofuels redux. Science, S22 (S898),
pp. 49Su. uoi: 1u.1126science.1161S2S.
Roecknei, E., K. Aipe, L. Bengtsson, S. Biinkop, L. Bmenil, N. Esch, E. Kiik, F. Lunkeit, N.
Ponatei, B. Rockel, R. Suasen, 0. Schlese, S. Schubeit, anu N. Winuelbanu, (1992).
Simulation of the presentday climate with the ECHAM 4 model: impact of model
physics and resolution. NaxPlanck Institute foi Neteoiology, Repoit No. 9S,
Bambuig, ueimany, 171 pp.
111

Roecknei, E., K. Aipe, L. Bengtsson, N. Chiistoph, N. Claussen, L. Bmenil, N. Esch, N. uioigetta,

0. Schlese, anu 0. Schluzweiua, (1996). The atmospheric general circulation model
ECHAM4: model description and simulation of presentday climate. NaxPlanck
Institute foi Neteoiology, Repoit No. 218, Bambuig, ueimany, 9u pp.
Rosenzweig, C. anu N.L. Paiiy, (1994). ImpactsofClimateChangeonWorldFoodSupply. Natuie
S67: 1SS1S8.
Saxton, K.E., W.}. Rawls, }.S. Rombeigei anu R.I. Papenuick, (1986). Estimating Generalized Soil
waterCharacteristicsfromTexture.Soil Sci. Soc. Am. } Su, 1uS11uS6.
Schailemann, }.P.W. anu W.F. Lauiance, (2uu8). EnvironmentalScience:HowGreenAreBiofuels?
Science S19(S8S9): 4S44.
Schlegel, A.}., K.C. Bhuyvettei, C.R. Thompson anu }.L. Bavlin, (1999). Agronomic impacts of
tillageandrotationonwheatandsorghum.}.Piou Agiic., vol 12, no 4, 1999: 6296SS.
Schmiut, }.B. anu B.P. Weiuema, (2uu7). Shiftinthemarginalsupplyofvegetableoil.Inteinational
}ouinal of Life Cycle Assessment 1S(S), pp. 2SS2S9.
Seaichingei, T., (2uu8). Theimpactsofbiofuelsongreenhousegases:howlandusechangealters
the equation. Policy Biief. Washington, BC, The ueiman Naishall Funu of the 0niteu
States.
Seaichingei, T., R. Beimlich, R.A. Boughton, F. Bong, A. Elobeiu, }. Fabiosa, S. Tokgoz, B. Bayes
anu T. Yu, (2uu8). UseofU.S.croplandsforbiofuelsincreasesgreenhousegasesthrough
emissionsfromlandusechange.Sciencexpiess, 7 Febiuaiy.
Shah, N., (Cooiuinating Authoi), (1992). The Global Partnership for Environment and
DevelopmentAGuidetoAgenda21. Eaith Summit Repoit, The 0niteu Nations Piess,
New Yoik, }une 1992.
Shah, N., (Cooiuinating Leau Authoi) et al., (2uuS). Millennium Ecosystems Assessment,
EcosystemsandHumanWellbeing.volume S Policy Responses, Chaptei S Foodand
Ecosystems, pp 17S212. Islanu piess, ISBN1SS96S269u, 2uuS.
Shi, X.Z., B.S. Yu, E.B. Wainei, X.Z. Pan, u.W. Peteisen, Z.u. uong, anu B.C. Weinuoif, (2uu4). Soil
Databaseof1:1,000,000DigitalSoilSurveyandReferenceSystemoftheChineseGenetic
SoilClassificationSystem. Soil Suivey Boiizons 4S:1291S6.
Shi, X.Z., B.S. Yu, E.B. Wainei, X.Z. Pan, X. Sun, u.W. Peteisen, Z.u. uong, anu B. Lin, (2uu6a). Cross
Reference System for translating Between Genetic Soil Classification of China and Soil
Taxonomy. Soil Science Society of Ameiica, }ouinal. 7u: 788S.
Shi, X.Z., B.S. Yu, u.X. Yang, B.}. Wang, W.X. Sun, u.B. Bu, Z.T. uong, (2uu6b). Crossreference
Benchmarks for Correlating the Genetic Soil Classification of China and Chinese Soil
Taxonomy. Peuospheie 16(2):1471SS.
Shi, X.Z., B.}. Wang, E.B. Wainei, B.S. Yu, W.X. Sun anu Y.C. Zhao, (2uu7). CrossReferenceSystem
forInterpretingGeneticSoilClassificationofChinatoWRB. ueoueima (in ieviewing).
Siebeit, S., P. Boll, S. Feick, }. Boogeveen, K. Fienken, (2uu7). Global Map of Irrigation Areas
version 4.0.1. Fiankfuit am Nain, ueimany; Rome, Italy, }ohann Wolfgang uoethe
0niveisity Foou anu Agiicultuie 0iganization of the 0niteu Nations.
Silvius, N., A. Kaat anu B. van ue Bunu, (2uu6). Peatland degradation fuels climate change.
Wetlanus Inteinational. The Netheilanus.
112

Singh, S., (2uu7). IndiaBiofuelsAnnual2007,GainReport,New Belhi. The Eneigy anu Resouices

Institute (TERI) (2uu7), vulneiability to Climate vaiiability anu Change in Inuia:
Assessment of Auaptation Issues anu 0ptions, TERI, New Belhi, Inuia.
Smeets, E.N.W., B.N. }ungingei, A.P.C Faaij, A. Waltei, P. Bolzan, W.C. Tuikenbuig, (2uu8). The
sustainability of Brazilian ethanol An assessment of the possibilities of certified
production.Biomass anu Bioeneigy S2(8): 78181S.
Smyth, A.}. anu }. Bumanski, (199S). AnInternationalFrameworkforEvaluatingSustainableLand
Management. Woilu Soil Resouices Repoit 7S, FA0, Rome, Italy.
Souhi, N.S. anu B.W. Biook, (2uu6). Southeast Asian Biodiversity in Crisis. Cambiiuge 0niveisity
Piess.
Souhi, N.S., etal.,(2uu4). SoutheastAsianbiodiversity:animpendingdisaster,Trends.Ecol. Evol.
19 pp. 6S466u. Aiticle | PBF (4u7 K) | view Recoiu in Scopus | Citeu in Scopus (78).
Sombioek, W.u., (1984). Towards a Global Soil Resources Inventory at Scale 1:1 Million.
Biscussion Papei. ISRIC, Wageningen, The Netheilanus.
State Lanu Auministiation of the People's Republic of China 0niteu Nations Bevelopment
Piogiam State Science anu Technology Commission of the People's Republic of
China Foou anu Agiicultuial 0iganization of the 0niteu Nations, (1994) Land
Resources Use and Productivity Assessment in China. Pioject CPR87u29 (Nain
Repoit), Beijing, China.
Steenblick, R., (2uu7). Biofuels At What Cost? Government support for ethanol and biodiesel in
selectedOECDcountries.Inteinational Institute foi Sustainable Bevelopment. ueneva.
Stewait, R.B., (198S). ModelingMethodologyforAssessingCropProductionPotentialsinCanada.
Agiicultuie Canaua, Reseaich Bianch, 0ttawa, Canaua. 29 pp.
Stolbovoi, v., (1996). Biaft Physiogiaphic Nap of the Foimei Soviet 0nion anu Nongolia at Scale
1:S Nillions. AuLS Woiking Papei. FA0, Rome.
Stolbovoi, v., (1998). SoilsofRussia.CorrelatedwiththeRevisedLegendoftheFAOSoilMapofthe
World. IIASAFA0 Inteiim Repoit IR98uS7, Inteinational Institute foi Applieu
Systems Analysis, Laxenbuig, Austiia.
Sys, C. anu }. Riquiei, (198u). Ratings of FAO/UNESCO soil units for specific crop production. In:
Repoit on the Seconu FA00NFPA Expeit Consultation on Lanu Resouices foi
Populations of the Futuie. FA0, Rome, Italy.
Sys, C., (199u). Consultants Report for the UNDP/SSTC/FAO/SLA Land Resources, Use and
ProductivityAssessmentProjectCPR/89/029. Beijing, China.
Sys, C., E. van Ranst, }. Bebaveye, anu F. Beeinaait, (199S). Land Evaluation Parts I, II and III.
Inteinational Tiaining Centie foi Postuiauuate Soil Scientists (ITC), 0niveisity of
uhent, Agiicultuial Publications No. 7, ueneial Auministiation foi Bevelopment
Coopeiation, Biussels, Belgium.
Tan, K.T., K.T. Lee, A.R. Nohameu, S. Bhatia (2uu9). Palm Oil: Addressing issues and towards
sustainabledevelopment.Renewable anu Sustainable Eneigy Reviews 1S (2uu9) 42u
427 anu ScienceBiiect www.scienceuiiect.com
113

Tainocai, C., }.N. Kimble, B. Swanson, S. uoiyachkin, Ye.N. Naumov, v. Stolbovoi, B. }akobsen, u.
Bioll, L. Nontanaiella, A. Ainoluussen, 0. Ainalus, anu N. YliBalla. (2uu2). Northern
Circumpolar Soils. 1:10,000,000 scale map. 0ttawa, Canaua: Reseaich Bianch,
Agiicultuie anu AgiiFoou Canaua, 0SBANRCS, Bokuchaev Institute. Bistiibuteu by
the National Snow anu Ice Bata CenteiWoilu Bata Centei foi ulaciology, Bouluei,
C0, 0SA.
Teixeiia E, u. Fischei, B. van velthuizen, F. Eweit, R. van Bingenen, F. Bentenei, u. Nills (2uu9).
Surfaceozoneimpactonglobalfoodsupply:Potentialdamageandadaptationfor
soybeancrops (Abstiact). I0P Confeience Seiies: Eaith anu Enviionmental Science,
6(S7):S72u4S (1 Febiuaiy 2uu9).
The 0ffice foi the Seconu National Soil Suivey of China (199S1996). Soil Species of China.
volume IvI. Chinese Agiicultuie Piess, Beijing, China.
The 0ffice foi the Seconu National Soil Suivey of China (199S). Soil map of People's Republicof
China. Napping Piess, p. 16u.
The Royal Society (2uu8). Sustainable Biofuels: Prospects and Challenges. Clyveuon Piess Ltu,
0K, ISBN 978 u8S4uS 662 2.
Tsuji, u.Y., u. 0ehaia, anu S. Balas (Eus.). (1994). DSSAT v3. 0niveisity of Bawaii, Bonolulu,
Bawaii.
Tubiello, F.N., anu u. Fischei (2uu6). Reducingclimatechangeimpactsonagriculture:Globaland
regional effects of mitigation, 20002080. Technological Foiecasting & Social Change
(2uu6), uoi:1u.1u16j.techfoie.2uu6.uS.u27.
0chijima, Z., anu B. Seino, (1988). Anagroclimaticmethodofestimatingnetprimaryproductivity
ofnaturalvegetation. }. Agi. Res. Quat. 21: 2442Su.
0itto anu Akiko 0no (Euitois), (1996). Population,landmanagement,andenvironmentalchange.
The 0niteu Nations 0niveisity, Tokyo. Available at:
http:www.unu.euuunupiessunupbooksuuuSpeuuuSpeuu.htm#Contents.
0NBPSSTCFA0SLA, (1994). Lanu Resources, Use and Productivity Assessment in China. Nain
Repoit anu Technical annex (vol.19) of Pioject CPR87u29Bu1R. 0niteu
Nations Bevelopment Piogiam, State Science anu Technology Commission of the
Peoples Republic of China, Foou anu Agiicultuie 0iganization of the 0niteu Nations,
State Lanu Auministiation of the Peoples Republic of China, Beijing.
0NEPWCNC, (2uu1). ProtectedAreasDatabase. Cambiiuge, 0K.
0niteu Nations, (1998). World Population Estimates and Projections, 1998 Revision, 0niteu
Nations Population Bivision (uigital uatasets).
0niteu Nations, (2uu1). World Population Prospects: The 2000 Revision, 0niteu Nations, New
Yoik, NY, 0SA.
0niteu Nations, (2uu6). HumanDevelopmentReport2006. 0NBP, New Yoik
0niteu Nations, (2uu9). World Population Prospects: The 2008 Revision. 0niteu Nations, Naich
2uu9.
0SBA (2uu6). TheEconomicFeasibilityofEthanolProductionfromSugarintheUnitedStates.
Available at:
http://www.usda.gov/oce/reports/eneigyEthanolSugaiFeasibilityRepoitS.puf,
accesseu }anuaiy 2uu9.
114

0SBA, (2uu7). Indonesia:PalmOilProductionProspectsContinuetoGrow. Available at:

http://www.pecad.fas.usda.gov/highlights/2007/12Inuonesia_palmoil. accesseu
Febiuaiy 2uu9.
(0SBA), (2uu8a). FarmBill.
http:www.usua.govwpspoitalusuafaimbill2uu8.naviu=FARNBILL2uu8
0SBA FAS (2uu8b). Oilseeds:WorldMarketandTrade:
http://www.fas.usda.gov/oilseeds/ciiculai2uuSuS1uoctcov.puf.
0SBA (2uu8c). Tiostle, R., Global Agricultural Supply and Demand: Factors Contributing to the
RecentIncreaseinFoodCommodityPrices.0SBA, Nay 2uu8.
0SBANRCS (2uu4). SoilSurveyLaboratoryManualSoil Suivey Investigations Repoit 42 (vei.
4.u) |Available at: ftp:ftp
c.sc.egov.usua.govNSSCLab_Nethous_NanualSSIR42_2uu4_ view.puf. veiifieu 11
}an 2uu7j, 0SBANational Resouices Conseivation Seivice, Washington.
0SuSuT0P0Su (2uu2). Global30arcsecondelevationdata. 0.S. ueological Suivey, National
Napping Bivision, ER0S Bata Centei. Available at:
http:eucuaac.usgs.govgtopoSugtopoSu.html. Retiieveu Becembei 2uu8.
van Engelen v.W.P., N.B. Batjes, K. Bijkshooin anu }. Buting, (2uuS). Harmonized Global Soil
Resources Database (Final Report). Repoit 2uuSu6, FA0 anu ISRIC Woilu Soil
Infoimation, Wageningen, the Netheilanus.
van Engelen, v. anu W. Tingtiang (eus.), (199S). ulobal anu National Soils anu Teiiain Bigital
Batabases (S0TER). Inteinational Soil Refeience anu Infoimation Centie,
Wageningen, the Netheilanus.
van Lynuen, u.W.}., (1994). DraftPhysiographicMapofAsia. FA0ISRIC Woiking Papei.
van Ranst, E., L. vanmechelen, A.}. Thomasson, }. Baioussin, }.N. Bollis, R.}.A. }ones, N. }amagne,
anu B. King, (199S). Elaboration of an extended knowledge database to interpret the
1:1,000,000 EU Soil Map for environmental purposes. In: B. King, R.}.A. }ones & A.}.
Thomasson (eus.). Euiopean Lanu Infoimation Systems foi Agioenviionmental
Nonitoiing. E0R 162S2 EN, p. 7184. 0ffice foi 0fficial Publications of the Euiopean
Communities, Luxembouig.
van Reeuwijk, L.P., (198S). Onthewaytoimproveinternationalsoilclassificationandcorrelation:
thevariabilityofsoilanalyticaldata. Annual Repoit 198S, ISRIC, Wageningen
http:www.isiic.oigisiicwebuocsBocsannual_iepoit_198S.puf, p. 71S.
van velthuizen, B.T., anu A.B. Kassam, (198S). ClimaticResourcesInventoryofEastAsia. FA0,
Rome, Italy.
van velthuizen B, (1988). AssessmentofLandResourcesforRainfedProductionofMaize,Bunded
RiceandWheatinTanzania. Woilu Bank, Washington BC.
van velthuizen B, (1991). ClimaticResourcesInventoryofKadunaState. FA0, Kauuna, Nigeiia.
van velthuizen, B.T., (1992). AgroecologicalSuitabilityofChickpeaandEuropeanOlive. Inteinal
iepoit. FA0, Rome, Italy.
van velthuizen, B., (199S). GuidelinesforClimaticResourcesInventoryCompilationforthe
CaspianAreainIran, Tehian, Iian.
van velthuizen, B.,(199S). ClimaticResourcesInventoryandClimaticSuitabilitiesforNuwakot
District, Katmanuu, Nepal.
115

van velthuizen, B., anu L veielst, (199S). CropProductionSystemZonesoftheIGADRegion.

Technical iepoit with maps anu electionic atlas anu uata annex volumes foi Bjibouti,
Eiitiea, Ethiopia, Kenya, Suuan, anu 0ganua, FA0, Rome.
van velthuizen, B., (1997). AgroecologicalZoningofJiangxiProvince, MountainRiverLake
Organization, Nanchang, China.
van velthuizen, B.,(1999). GeospatialmodelingforAgroecologicalAssessmentsandDecision
SupportforDevelopmentPlanning; Atlas of Lanu Resouices anu Lanu Suitability of
uhana, EISBSRI, Accia.
van velthuizen, B., (2uuu). GlobalAgroecologicalZones. Repoit anu CBR0N, FA0AuLIIASA,
Rome, Italy.
van velthuizen, B., (2uuu). HydroclimaticResourcesofBangladesh. FA0BARC, Bhaka.
van velthuizen, B., (2uu1). AgroecologicalZonationReports(2)fortheInventoryofpostwarof
landresourcesinBosniaandHerzegovina. FA0, Saiajevo.
van velthuizen, B., B. Buuuleston, u. Fischei, N. Salvatoie, E. Ataman , F.0. Nachteigaele, N.
Zanett, N. Bloise (2uu7).MappingBiophysicalFactorsthatInfluenceAgricultural
ProductionandRuralVulnerability. Enviionment anu Natuial Resouices Seiies No.
11; FA0, Rome, Italy |2uu7j.
veiheye, W., (1987). Definition and Delineation of AgroClimatic Zones in the European
Community. CEE0niveisity Instelling, Antweipen, Belgium, 2up. + ann.WNI.
vogel A.W., (1994). Comparability of soil analytical data: determinations of cation exchange
capacity,organiccarbon,soilreaction,bulkdensity,andvolumepercentageofwaterat
selected pF values by different methods. Woik. Pap. 94u7, ISRIC, Wageningen, the
Netheilanus.
vooitman, R.L. ,(2u1u). Explorations into African land resource ecology on the chemistry
between soils, plants and fertilizers. Centie foi Woilu Foou Stuuies, Fiee 0niveisity
Amsteiuam.
vooitman, R.L., B.u.}.S. Sonnevelu, }.W.A. Langevelu, u. Fischei, anu B.T. van velthuizen, (1999).
Climate Change and Global Agricultural Potential: A Case Study of Nigeria. Woiking
Papei WP99u6, Centie foi Woilu Foou Stuuies, Fiee 0niveisity Amsteiuam, anu
Inteinational Institute foi Applieu Systems Analysis, Laxenbuig, Austiia.
voiosmaity, C.}., B.N. Fekete, N. Neybeck, anu R. Lammeis, (2uuu). A simulated topological
networkrepresentingtheglobalsystemofriversat30minutespatialresolution(STN
30). ulobal Biogeochemical Cycles 14: S99621.
Wen, Tingtiang, (199S). Draft Physiographic Map of Latin America. AuLS Woiking Papei. FA0,
Rome, Italy.
Wissenschaftlichei Beiiat uei Bunuesiegieiung ulobale 0mweltveinueiungen (WBu0), (2uu8)
WeltimWandelZukunftsfhigeBioenergieundnachhaltigeLandnutzung.
http://www.wbgu.de, Beilin, ueimany.
ue Wit, C.T., (196S). Photosynthesis of leaf canopies. Agiicultuial Reseaich Repoits 66S, Puuoc,
Wageningen. 88p.
Woilu Bank, (2uu7). WorldDevelopmentReport2008: AgricultureforDevelopment. Woilu Bank,
Washington B.C.
116

Woilu Resouices Institute, (2uu7). Earth Trends: Environmental Information. Available at:
http:eaithtienus.wii.oig. Washington BC: Woilu Resouices Institute.
Woiluwatch Institute, (2uu7). Biofuels for transport. Global potential and implications for
sustainableenergyandagriculture. Eaithscan, Lonuon, 0K, ISBN: 978844u74228.
WBPA, (2uu9). World Database of Protected Areas, Annual Release 2009, Available at:
http:www.wupa.oigAnnualRelease.aspx.
Young, A. anu A.C.S. Wiight, (198u). Rest period requirements of tropical and subtropical soils
under annual crops. In: Repoit on the Seconu FA00NFPA Expeit Consultation on
Lanu Resouices foi Populations of the Futuie. FA0, Rome, Italy.
Zah, R., B. Boni, N. uauch, R. Bischiei, N. Lehmann, anu P.Wgei, (2uu7). kobilanz von
Energieprodukten: kologische Bewertung von Biotreibstoffen. St uallen, Switzeilanu,
ENPA www.empa.ch.
Zhang, X., u. Zhou, (199S). ANewNPPModel. China Contiibution to ulobal Change Stuuies, China
ulobal Change Repoit No. 2. Y. Buzheng anu L. Bai. Beijing China, Science Piess. ISBN
7uS99S9SSSP.
Zoblei, L., (1986). A world soil file for global climate modeling. Technical Nemoianuum 878u2,
NASA uouuaiu Institute foi Space Stuuies (uISS), New Yoik, NY.

117

APPENDIXES

Appendix21 CountryList(GAUL)andregionalizations
Thisdocumentisavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/GAEZ_Regions.pdf

118

Appendix31 CalculationofReferenceEvapotranspiration
The calculation of reference evapotranspiration (ET
o
), i.e., the rate of evapotranspiration from a
hypothetic reference crop with an assumed crop height of 12 cm, a fixed canopy resistance of
70ms
1
andanalbedoof0.23(closelyresemblingtheevapotranspirationfromanextensivesurface
of green grass), is done according to the PenmanMonteith equation (Monteith, 1965, 1981; FAO,
1992b).Thecalculationprocedureusesastandardizedsetofinputparameters,asfollows:
T max maximumdailytemperature(
o
C)
T min minimumdailytemperature(
o
C)
RH meandailyrelativehumidity(%)
U2 windspeedmeasurement(ms
1
)
SD brightsunshinehoursperday(hours)
A elevation(m)
L latitude(deg)
J Juliandate,i.e.,numberofdayinyear
The PenmanMonteith combination equation can be written in terms of an aerodynamic and a
radiationterm(FAO,1992b):
ET ET ET o ar ra = + (1)
wheretheaerodynamictermcanbeapproximatedby
ET
T
U e e ar
a
a d =
+

+

*
( )
900
273
2 (2)
andtheradiationtermby
ET R G ra n =
+

*
( )
1
(3)
wherevariablesin(2)and(3)areasfollows:
psychrometricconstant(kPa
o
C
1
)
* modifiedpsychrometricconstant(kPa
o
C
1
)
slopeofvaporpressurecurve(kPa
o
C
1
)
Ta averagedailytemperature(
o
C)
ea saturationvaporpressure(kPa)
ed vaporpressureatdewpoint(kPa)
( ) e e a d vaporpressuredeficit(kPa)
U2 windspeedmeasurement(ms
1
)
Rn netradiationfluxatsurface(MJm
2
d
1
)
G soilheatflux(MJm
2
d
1
)
latentheatofvaporization(MJkg
1
)
In the calculation procedure for the reference crop we use the following relationships to define
termsin(2):
Averagedailytemperature:
T T T a = + 05 . ( ) max min (4)
Latentheatofvaporization:
= 2501 0002361 . . Ta (5)
119

Atmosphericpressure(kPa)atelevationA:
P
A
=

1013
293 00065
293
5256
.
.
.
(6)
Psychrometricconstant:

= 00016286 .
P
(7)
Aerodynamicresistance:
r
U
a =
208
2
(8)
Cropcanopyresistance:
r
R
LAI
c
l
=
05 .
(9)
where under ambient CO
2
concentrations the average daily stomata resistance of a single leaf, Rl
(sm
1
), is set to Rl = 100, and leaf area index of the reference crop is assumed as
LAI = = 24 012 288 . . .
Modifiedpsychrometricconstant:
* = +

1
r
r
c
a
(10)
Saturationvaporpressuree
a
forgiventemperaturesT minandT max
e
T
T
ax =
+

06108
1727
2373
. exp
.
.
max
max
(11)
e
T
T
an =
+

06108
1727
2373
. exp
.
.
min
min
(12)
e e e a ax an = + 05 . ( ) (13)
Vaporpressureatdewpoint,ed:
e
RH
e e
d
ax an
=
+

100
05
1 1
.
(14)
Slopeofvaporpressurecurve,,forgiventemperaturesT maxandT min:
x
ax e
T
=
+
4096
2373
2
( . ) max
(15)
n
an e
T
=
+
4096
2373
2
( . ) min
(16)
( ) = + x n
(17)
Using (4)(17) all variables in (2) can be calculated from the input parameters. To determine the
remainingvariables Rn andGusedintheradiationterm ETra ofequation(3),weproceedwiththe
followingcalculationsteps:
120

Latitudeexpressedinrad:

=
L
180
(18)
Solardeclination(rad):

= 405 . 1
365
2
sin 4093 . 0 J

(19)
RelativedistanceEarthtoSun:

+ = J d
365
2
cos 033 . 0 1

(20)
Sunsethourangle(rad):
= arccos( tan tan ) (21)
Extraterrestrialradiation(MJm
2
d
1
):
R d a = + 37586 . ( sin sin cos cos sin ) (22)
Maximumdaylighthours:
DL =
24

(23)
Shortwaveradiation Rs (MJm
2
d
1
)
R
SD
DL
R s a = +

025 05 . .
(24)
For a reference crop with an assumed albedo coefficient = 023 . net incoming shortwave
radiation Rns (MJm
2
d
1
)is:
R R ns s = 077 . (25)
Netoutgoinglongwaveradiation Rnl (MJm
2
d
1
)isestimatedusing:
( )
( ) ( )
2
16 . 273 16 . 273
139 . 0 34 . 0 9 . 0 1 . 0 10 903 . 4
4
min
4
max
9
T T
e
DL
SD
R d nl
+ + +

+ =

(26)
Using(25)and(26),netradiationfluxatsurface, Rn ,becomes
R R R n ns nl = (27)
Finally,soilheatfluxisapproximatedusing
G T T a n a n = 014 1 . ( ) , , (28)
where Ta n , and Ta n , 1 are average monthly temperatures of current and previous month,
respectively. With equations (5), (10), (17), (27) and (28) all variables in (3) are defined and can be
calculatedfromtheinputparametersdescribedatthebeginningofthisAppendix.

121

Appendix32 OutputsModuleI
OutputscalculatedinModuleIarestoredintwoseparatebinaryfiles,oneholdingvariablesrelated
to temperature profiles and thermal growing periods and one file storing moisture related
characteristics. Each file begins some 18 header records holding a copy of the main control
parameters used to run the model. The output variables from Module I are described in the tables
below.
TableA31ContentoffixedoutputrecordsfromGAEZModuleI
Variables Description
Record
number
Type of
variable
Length of
variable
(inbytes)
btext Explanatorytextstring 1 Character 16
version Programversionstring 2 Character 14
datestr datestring(whenfilewascreated) 3 Character 9
Mrow Numberofrowsofgrid 4 Integer 2
Mcol Numberofcolumnsofgrid 4 Integer 2
Lenmin ControlparameterLENMIN 4 Integer 2
Itflg ControlparameterITFLG 4 Integer 2
Rlps Lapsetheapplied(degreeCperm) 4 Real 4
dRl Parameter of change in winter use efficiency under
elevatedCO2
4 Real 4
Sa0 AWClevel(mm/m) 4 Real 4
Sdep0 Maximumapplicablesoildepth(m) 4 Real 4
Rplim1 WaterbalancecontrolparameterRPLIM1 4 Real 4
Rplim2 WaterbalancecontrolparameterRPLIM2 4 Real 4
Samin WaterbalancecontrolparameterSAMIN 4 Real 4
Kc1 WaterbalancecontrolparameterKC1 5 Real 4
Kc2 WaterbalancecontrolparameterKc2 5 Real 4
Kc3 WaterbalancecontrolparameterKc3 5 Real 4
Kc4 WaterbalancecontrolparameterKc4 5 Real 4
KC5 WaterbalancecontrolparameterKc5 5 Real 4
Kc6 WaterbalancecontrolparameterKc6 5 Real 4
Kc7 WaterbalancecontrolparameterKc7 5 Real 4
dTx ClimatesensitivityruncontrolparameterdTx 5 Real 4
dTn ClimatesensitivityruncontrolparameterdTn 5 Real 4
dP ClimatesensitivityruncontrolparameterdP 5 Real 4
dS ClimatesensitivityruncontrolparameterdS 5 Real 4
dW ClimatesensitivityruncontrolparameterdW 5 Real 4
flnmap1 Inputfilename:gridcelllandmask 6 Character 64
flninp Inputfilename:landpixelfile 7 Character 64
flntc1 Outputfilename:thermalregimepixelvalues 8 Character 64
flnlgp Outputfilename:moistureregimepixelvalues 9 Character 64
flntmx Inputfilename:averagemonthlytemperature 10 Character 64
flntmn Inputfilename:averagemonthlytemperaturerange 11 Character 64
flnpcp Inputfilename:monthlyprecipitation 12 Character 64
flnwnd Inputfilename:monthlywindrun 13 Character 64
flnsol Inputfilename:monthlysunshinefraction 14 Character 64
flnrhu Inputfilename:monthlyrelativehumidity 15 Character 64
flnwdf Inputfilename:monthlywetdayfrequency 16 Character 64
flngcm Outputfilename:climatechangedistortionsfromGCM 17 Character 64
EoH EndofheaderstringEOH 18 Character 3

Followingtheheaderrecords,thereisonerecordsavedineachfileforeverygridcellmarkedasland
inthebinarylandmaskfile,aslistedinTable32andTable33.

122

TableA32ModuleIoutputfiledescribingthermalconditionsduringthegrowingperiod
Variables Description
Type of
variable
Length of
variable
(inbytes)
irow Pixelreference:rownumber Integer 2
icol Pixelreference:columnnumber Integer 2
alt Pixelreference:medianelevation[m] Integer 2
itcc Thermalclimateclass Integer 2
Itcc2 Thermalzonesclass Integer 2
iscold Coldbreakindicator(i.e.nohibernatingcropspermitted) Integer 2
cidx Indexofcontinentality Integer 2
tmean Meanannualtemperature[C*100] Integer 2
tamin Meanannualminimumtemperature[C*100] Integer 2
tamax Meanannualmaximumtemperature[C*100] Integer 2
cbtlim Minimumsnowadjustedmonthlytemperature[C] Real 4
tadif Annualtemperatureamplitude(=warmestmonthminuscoldestmonth) Real 4
ndtr(19,1) Number of days above (30, 25, 20, 15, 10, 5, 0, 5, < 5) C for period when
temperaturetrendisup
Integer 9*2
ndtr(19,2) Number of days above (30, 25, 20, 15, 10, 5, 0, 5, < 5) C for period when
temperaturetrendisdown
Integer 9*2
Tsum(13) Accumulated temperature sums for periods with average daily temperature
above0,5,10C(averagetemperature)[Cd]
Real 3*4
Tsumh(13) Averagetemperaturefordayswithaveragedailytemperatureabove0,5,10
C[hours]
Real 3*4
lgpt(13) Numberofdays,beginningday,endingdaywithaveragedailytemperature>
0,5,10C[days]
Integer 3*3
ndx35 Numberofdayswithmaximumtemperature>35C Integer 2
ndx30 Numberofdayswithmaximumtemperature>30C Integer 2
ndx00 Numberofdayswithminimumtemperature>0C Integer 2
ndxa00 Numberofdayswithaveragetemperature>0C Integer 2
ndxa05 Numberofdayswithaveragetemperature>5C Integer 2
ndxa10 Numberofdayswithaveragetemperature>10C Integer 2
frost1 Airfrostindex Real 4
frost2 Snowadjustedairfrostindex Real 4
ndtr2(16,1) NumberofdaysinlongestLGPwithaveragedailytemperatureabove(30,25,
20,15,10,5,0,5,elseC)fortheperiodwhentemperaturetrendisup
Integer 6*2
ndtr2(16,2) NumberofdaysinlongestLGPwithaveragedailytemperatureabove(30,25,
20,15,10,5,0,5,elseC)fortheperiodwhentemperaturetrendisdown
Integer 6*2
Tsum2(13) Accumulated temperature sums in longest LGP for days above 0, 5, 10 C
[Cd]
Real 3*4
Tsum2h(13) Accumulated temperature sums in longest LGP for days above 0, 5, 10 C
[hours]
Real 3*4

123

TableA33ModuleIoutputforsoilmoistureconditionsandlengthofgrowingperiodcharacteristics
Variables Description
Type of
variable
Length of
variable
(inbytes)
irow Pixelreference:rownumber Integer 2
icol Pixelreference:columnnumber Integer 2
sP Annualrainfall[mm] Integer 2
sETo Annualreferencepotentialevapotranspiration[mm] Integer 2
sETa Annual(actual)evapotranspirationofreferencecrops[mm] Integer 2
sWex Annualexcessmoistureinreferencewaterbalance[mm] Integer 2
ridx Annualaridityindex(100*Pcp/ETo) Integer 2
ridx2 AridityindexduringLGP
t=5
Integer 2
NPP1 Annualnetprimaryproductionunderirrigationconditions Real 4
NPP2 Annualnetprimaryproductionunderrainfedconditions Real 4
ishum NumberofmonthswithP>ETo Integer 2
ishum05 NumberofmonthswithP>EToandTa>5 Integer 2
nmon05 NumberofmonthswithTa>5 Integer 2
lgptot Totalnumberofgrowingperioddays Integer 2
ndwtot NumberofgrowingperioddayswithP>ETo,referencecrop Integer 2
ndhtot NumberofgrowingperioddayswithETaETo,referencecrop Integer 2
nlgp Numberofcomponentgrowingperiods Integer 2
begdrm Beginningofdormancyperiod(0,ifnodormancy)[day] Integer 2
enddrm, Endofdormancyperiod,(0,ifnodormancy)[day] Integer 2
ndw2 NumberofdaysduringLGP
t=5
withETa09ETo Integer 2
ndw1 NumberofdaysduringLGP
t=5
withETa04ETo Integer 2
ndw0 NumberofdaysduringLGP
t=5
withEta<04ETo Integer 2
ndwb90 NumberofdaysduringLGP
t=5
withwaterbalanceW
b
09S
a
Integer 2
ndwb50 NumberofdaysduringLGP
t=5
withwaterbalanceW
b
05S
a
Integer 2
ndwb10 NumberofdaysduringLGP
t=5
withwaterbalanceW
b
01S
a
Integer 2
ndwb00 NumberofdaysduringLGP
t=5
withwaterbalanceW
b
<01S
a
Integer 2
ridxw Seasonalacidityindex,OctoberMarch Integer 2
ridxs Seasonalacidityindex,AprilSeptember Integer 2
ridq1 Seasonalacidityindex,month13 Integer 2
ridq2 Seasonalacidityindex,month46 Integer 2
ridq3 Seasonalacidityindex,month79 Integer 2
ridq4 Seasonalacidityindex,month1012 Integer 2
Lgplen,ndpet,ndwet,beglgp,endlgp(1nact*) Integer 5*nact*2
*nactmin(nlgp,5)

124

Appendix33 SubroutinedescriptionsofModuleI

ModuleI:Climatedataanalysisandreferencecropassessment

ModuleIdealswiththereading,conversion,interpolation,analysisandclassificationofclimatedata
andthecreationofhistorical,baselineandfuturegriddedclimatedatalayersrelevanttoagronomic
suitability analysis. The main objective in Module I is the compilation of georeferenced climatic
indicators,includingagroclimaticindicatorsandtheestimationofyearroundsoilmoisturebalance
and evapotranspiration for FAO reference crops (similar to grass). A diagram of the functions and
routines of Module I is shown in Figure 31 shows the subroutines and functions in alphabetical
order with a short description of what the routine or function is used for, which file it is contained
within,andwhichotherroutinesorfunctionsitlinksto.Table34liststhesourcefilesinalphabetical
orderandwhichroutinesandfunctionsarecontainedwithinthesourcefile.

TableA34SubroutinesandfunctionsofModuleI
Program Subroutines Function Calledfrom Callsto
LGPSUB.F AMAT5 generatescoefficientsofasystemoflinear
equationsforgeneratingdailyvaluesbyspline
interpolation
P01 GAUSSJ
LGPSUB.F DAILY convertsmonthlyordecadaldatatodailyvalues
andcallsthenexttwofunctions.
P01 F365S1,
F365S2
LGPSUB.F F365S1 fillsindailyvaluesbysplineinterpolation DAILY
LGPSUB.F F365S2 fillsindailyvaluesbysplineinterpolationwhen
thevaluemustbegreaterorequalto0
DAILY
FROST FROSTNUM Calculatesairfrostindexandsnowadjustedair
frostindex
POI
125

Program Subroutines Function Calledfrom Callsto

LGPSUB.F GAUSSJ GaussJordaneliminationwithfullpivoting AMAT5
P01.F GCMDAT readsGCMclimatechangedataforcorrectionof
currentposition
P01 DAYHR,
SCALEFCT
LGP.F LGPS determinesthenumber,lengthanddatesof
growingperiods
P01
P01.F RDMPU utilityroutinetoreadthesoilmappingunit
availablewatercontentdatafromaninputfile
GCMDAT
P01.F SCALEFCT algorithmusedtocalculateascalingfractionfor
scalingmonthlyclimaticdatavariablestomatch
theannualchangesfromGCM
LGPS
LGP.F SETDAT shiftsacalculateddayoftheyearbyamultipleof
365tofitwithintherange1365
P01
P01.F TCC calculatesthermalclimateclassandthermalzone
class
P01 THRIDX
P01.F TRP calculatestemperaturegrowingperiodsand
thermalregimeparameters
P01
P01.F WDFRQ distributesrainaccordingtowetdayfrequency
LGPSUB.F CYCSUM integratesanattributeovergrowthcycle
LGPSUB.F CYCVAL averagesanattributeovergrowthcycle P01,GCMDAT
P01.F DAYHR calculatesdaylengthforagivenlatitudeandday
oftheyear
P01
ET0.F ET0 calculatespotentialevapotranspirationbythe
PenmanMonteithmethod
LGPS
ETAM.F ETA calculatesactualevapotranspirationby
simulatingadailywaterbalanceforanFAO
referencecrop(similartograss)
P01
P01.F GETBUF utilityfunctiontogetnextlinefromaninputfile,
skippingcomments
LGPS
LGP.F LENDAT calculatesthenumberofdaysbetweentwo
datesincludingthestartandenddate.
LPGS
ETAM.F PSH calculatesthesoilwaterdepletionfraction(p)for
agivencroptypeandlevelofdailyETo
P01,TRP
LGPSUB.F THRIDX determinestemperatureprofileclassindex AMAT5

TableA35FortransourcefilesforModuleIandincludedheaderfiles,subroutinesandfunctions
Fortranfile AssociatedHeaderFiles Subroutines Functions
ET0.F et0
ETAM.F eta,psh
FROST.F clim.h,tcc.h Frostnum
LGP.F lgps,setdat lendat
LGPSUB.F amat5,daily,f365s1,f365s2,gaussj cycval,cycsum,trcidx
P01.F clim.h,tcc.h gcmdat,rdmpu,scalefct,tcc,trp,wdfrq dayhr,getbuf

126

Appendix34ExampleofModuleIoutputatgridcelllevel
Example of information generated at grid cell level is given for Ilonga, Tanzania is available for
downloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/Example_grid_cell_output_Module_I.docx

127

Appendix41 Cropsandlandutilizationtypes(LUTs)
Suitabilityandpotentialyieldassessmentsareavailablefor11cropgroups(Table41),49crops
(Table42),92croptypes(Table43)and280Crop/LUTs(Table44).Resultsfordownscalingof
crops/commoditiesareavailablefor23crop/commoditiesandresultsofyieldandproductiongap
analysisareavailablefor17crops/commodities(Table45).
TableA41Cropgroups
Code Cropgroup
1 Cereals
2 Rootsandtubers
3 Sugarcrops
4 Pulses
5 Oilcrops
6 Vegetables
7 Fruits
8 Fibrecrops
9 Narcoticsandstimulants
10 Foddercrops
11 Bioenergyfeedstocks
TableA42Crops
Code Commonname Scientificname Cropgroup
1 Wheat Triticumspp. Cereals
2 Wetlandrice Oryzasativa Cereals
3 Drylandrice Oryzasativa Cereals
4 Maize Zeamays Cereals
5 Barley Hordeumvulgare Cereals
6 Sorghum Sorghumbicolor Cereals/Sugarcrops
7 Rye Secalecereale Cereals
8 Pearlmillet Pennisetumglaucum Cereals
9 Foxtailmillet Setariaitalica Cereals
10 Oat Avenasativa Cereals
11 Buckwheat Fagopyrumesculentum Cereals
12 Whitepotato Solanumtuberosum Rootsandtubers
13 Sweetpotato Ipomoeabatatas Rootsandtubers
14 Cassava Manihotesculenta Rootsandtubers
15 YamandCocoyam Dioscoreaspp.andColocasiaesculenta Rootsandtubers
16 Sugarcane Saccharumspp. Sugarcrops
17 Sugarbeet BetavulgarisL. Sugarcrops
18 Phaseolusbean PhaseolusvulgarisandPh.lunatus Pulses
19 Chickpea Cicerarietinum Pulses
20 Cowpea Vignaunguiculata Pulses
21 Drypea PisumsativumL. Pulses
22 Gram Vignaradiata Pulses
23 Pigeonpea Cajanuscajan Pulses
24 Soybean Glycinemax Oilcrops
25 Sunflower Helianthusannuus Oilcrops
26 Rape Brassicanapus Oilcrops
27 Groundnut Arachishypogaea Oilcrops
28 Oilpalm Elaeisoleifera Oilcrops
128

Code Commonname Scientificname Cropgroup

29 Olive Oleaeuropaea Oilcrops
30 Jatropha Jatrophacurcas. Bioenergyfeedstocks
31 Cabbage Brassicaoleracea Vegetables
32 Carrot Daucuscarota Vegetables
33 Onion Alliumcepa Vegetables
34 Tomato Lycopersiconlycopersicum Vegetables
35 Banana/Plantain Musaspp. Fruits
36 Citrus CitrusSinensis Fruits
37 Coconut Cocosnucifera Fruits
38 Cacao Theobromacacao Narcoticsandstimulants
39 Cotton Gossypiumhirsutum. Fibre
40 Flax Linumusitatissimum Fibrecrops
41 Coffee Coffeaarabica Narcoticsandstimulants
42 Tea CamelliaSinensesvar.Sinensis Narcoticsandstimulants
43 Tobacco Nicotianatobacum Narcoticsandstimulants
44 Alfalfa Medicagosativa Foddercrops
45 Pasturelegume various Foddercrops
46 Grass various Foddercrops
47 Miscanthus Miscanthusspp Bioenergyfeedstocks
48 Switchgrass Panicumvirgatum Bioenergyfeedstocks
49 Reedcanarygrass Phalarisarundinacea Bioenergyfeedstocks

TableA43Croptypes
Code Commonname Scientificname Cropgroup
1 Winterwheat Triticumspp. Cereals
2 Springwheat Triticumspp. Cereals
3 Wheat(subtropicalcultivars) Triticumspp. Cereals
4 Wheat(tropicalcultivars) Triticumspp. Cereals
5 Japonicawetlandrice Oryzajaponica Cereals
6 Indicawetlandrice Oryzaindica Cereals
7 Indicadrylandrice Oryzasativa Cereals
8 Maize(tropicallowlandcultivars) Zeamays Cereals
9 Maize(tropicalhighlandcultivars) Zeamays Cereals
10 Maize(temperateandsubtropicalcult.) Zeamays Cereals
11 Silagemaize(temperateandsubtropicalcult.) Zeamays Foddercrops
12 Winterbarley Hordeumvulgare Cereals
13 SpringBarley Hordeumvulgare Cereals
14 Barley(subtropicalcultivars) Hordeumvulgare Cereals
15 Barley(tropicalcultivars) Hordeumvulgare Cereals
16 Sorghum(tropicallowlandcultivars) Sorghumbicolor Cereals
17 Sorghum(tropicalhighlandcultivars) Sorghumbicolor Cereals
18 Sorghum(temperateandsubtropicalcult.) Sorghumbicolor Cereals
19 Sweetsorghum(temperateandsubtropicalcult.) Sorghumbicolor Sugarcrops
20 Winterrye Secalecereale Cereals
21 Springrye Secalecereale Cereals
22 Pearlmillet Pennisetumglaucum Cereals
23 Foxtailmillet Setariaitalica Cereals
24 Springoat Avenasativa Cereals
25 Buckwheat Fagopyrumesculentum Cereals
26 Whitepotato Solanumtuberosum Rootsandtubers
27 Sweetpotato Ipomoeabatatas Rootsandtubers
28 Cassava Manihotesculenta Rootsandtubers
29 Whiteyam Dioscoreaspp. Rootsandtubers
129

Code Commonname Scientificname Cropgroup

30 Greateryam Dioscoreaspp. Rootsandtubers
31 Yellowyam Dioscoreaspp. Rootsandtubers
32 Cocoyam Colocasiaesculenta Rootsandtubers
33 Sugarcane Saccharumspp. Sugarcrops
34 Sugarbeet BetavulgarisL. Sugarcrops
35 Phaseolusbean(tropicallowland) PhaseolusvulgarisandPh.lunatus Pulses
36 Phaseolusbean(tropicalhighland) PhaseolusvulgarisandPh.lunatus Pulses
37 Phaseolusbean(temperateandsubtropicalcult.) PhaseolusvulgarisandPh.lunatus Pulses
38 Chickpea Cicerarietinum Pulses
39 Chickpea(coldtolerant) Cicerarietinum Pulses
40 Cowpea Vignaunguiculata Pulses
41 Drypea PisumsativumL. Pulses
42 Gram Vignaradiate Pulses
43 Pigeonpea Cajanuscajan Pulses
44 Soybean(tropicalandsubtropicalcult.) Glycinemax Oilcrops
45 Soybean(temperateandsubtropicalcult.) Glycinemax Oilcrops
46 Sunflower(tropicalandsubtropicalcult.) Helianthusannuus Oilcrops
47 Sunflower(temperateandsubtropicalcult.) Helianthusannuus Oilcrops
48 Winterrape Brassicanapus Oilcrops
49 Springrape Brassicanapus Oilcrops
50 Rabirape Brassicanapus Oilcrops
51 Groundnut Arachishypogaea Oilcrops
52 Oilpalm Elaeisoleifera Oilcrops
53 Olive Oleaeuropaea Oilcrops
54 Jatropha Jatrophacurcas Oilcrops
55 Cabbage Brassicaoleracea Vegetables
56 Carrot(temperateandsubtropicalcultivars) Daucuscarota Vegetables
57 Carrot(temperateandsubtropicalcultivars) Daucuscarota Vegetables
58 Carrot(tropicalcultivars) Daucuscarota Vegetables
59 Onion(temperateandsubtropicalcultivars) Alliumcepa Vegetables
60 Onionhibernatingcultivar Alliumcepa Vegetables
61 Onion(tropicalcultivars) Alliumcepa Vegetables
62 Tomato(temperateandsubtropicalcultivars) Lycopersiconlycopersicum Vegetables
63 Tomato(tropicalandsubtropicalcultivars) Lycopersiconlycopersicum Vegetables
64 Banana/Plantain Musaspp. Fruits
65 Citrus Citrussinensis Fruits
66 Coconut1(tall) Cocosnucifera Fruits
67 Coconut2(hybridtall) Cocosnucifera Fruits
68 Coconut3(dwarf) Cocosnucifera Fruits
69 Cacao(comun) Theobromacacao Narcoticsandstimulants
70 Cacao(hybrid) Theobromacacao Narcoticsandstimulants
71 Cotton(tropicalcultivars) Gossypiumspp. Fibrecrops
72 Cotton(temperateandsubtropicalcult.) Gossypiumspp. Fibrecrops
73 Flax Linumusitatissimum Fibrecrops
74 Coffeearabica Coffeaarabica Narcoticsandstimulants
75 Coffeerobusta Coffearobusta Narcoticsandstimulants
76 Tea(chinatea) CamelliaSinensesvar.Sinensis Narcoticsandstimulants
77 Tea(hybridtea) SinensisandAssamica Narcoticsandstimulants
78 Tea(assamtea) Camelliasinensisvar.assamica Narcoticsandstimulants
79 Tobacco(tropicalcultivars) Nicotianatobacum Narcoticsandstimulants
80 Tobacco(temperateandsubtropicalcult.) Nicotianatobacum Narcoticsandstimulants
81 Alfalfa(temperateandsubtropicalcult.) Medicagosativa Foddercrops
82 Alfalfa(tropicalcultivars) Medicagosativa Foddercrops
83 Pasturelegumes(temp.andsubtropicalcult.) various Foddercrops
130

Code Commonname Scientificname Cropgroup

84 Pasturelegumes(tropicalandsubtropicalcult.) various Foddercrops
85 Pasturegrasses(C3/Icultivars) various Foddercrops
86 Pasturegrasses(C3/IIcultivars) various Foddercrops
87 Pasturegrasses(C4/IIcultivars) various Foddercrops
88 Pasturegrasses(C4/Icultivars) various Foddercrops
89 Miscanthus(C4/II) Miscanthusspp Bioenergyfeedstocks
90 Miscanthus(C4/I) Miscanthusspp Bioenergyfeedstocks
91 Switchgrass Panicumvirgatum Bioenergyfeedstocks
92 Reedcanarygrass Phalarisarundinacea Bioenergyfeedstocks

TableA44Crop/LUTs
Code Croptype Growthcycle Harvestedpart
1 Winterwheat 35+105days Grain
2 Winterwheat 40+120days Grain
3 Winterwheat 45+135days Grain
4 Winterwheat 50+150days Grain
5 Springwheat 90days Grain
6 Springwheat 105days Grain
7 Springwheat 120days Grain
8 Springwheat 135days Grain
9 Springwheat 150days Grain
10 Wheat(subtropicalcultivars) 105days Grain
11 Wheat(subtropicalcultivars) 120days Grain
12 Wheat(subtropicalcultivars) 135days Grain
13 Wheat(subtropicalcultivars) 150days Grain
14 Wheat(tropicalhighlandcultivars) 100days Grain
15 Wheat(tropicalhighlandcultivars) 115days Grain
16 Wheat(tropicalhighlandcultivars) 130days Grain
17 Wheat(tropicalhighlandcultivars) 145days Grain
18 Wheat(tropicalhighlandcultivars) 160days Grain
19 Wheat(tropicalhighlandcultivars) 175days Grain
20 Wheat(tropicalhighlandcultivars) 190days Grain
21 Japonicawetlandrice 105days Grain
22 Japonicawetlandrice 120days Grain
23 Japonicawetlandrice 135days Grain
24 Japonicawetlandrice 150days Grain
25 Indicawetlandrice 105days Grain
26 Indicawetlandrice 120days Grain
27 Indicawetlandrice 135days Grain
28 Indicawetlandrice 150days Grain
29 Indicadrylandrice 105days Grain
30 Indicadrylandrice 120days Grain
31 Indicadrylandrice 135days Grain
32 Maize(tropicallowlandcultivars) 90days Grain
33 Maize(tropicallowlandcultivars) 105days Grain
34 Maize(tropicallowlandcultivars) 120days Grain
35 Maize(tropicallowlandcultivars) 135days Grain
36 Maize(tropicalhighlandcultivars) 120days Grain
37 Maize(tropicalhighlandcultivars) 150days Grain
38 Maize(tropicalhighlandcultivars) 180days Grain
39 Maize(tropicalhighlandcultivars) 210days Grain
40 Maize(tropicalhighlandcultivars) 240days Grain
41 Maize(tropicalhighlandcultivars) 270days Grain
42 Maize(tropicalhighlandcultivars) 300days Grain
131

Code Croptype Growthcycle Harvestedpart

43 Maize(temperateandsubtropicalcultivars) 90days Grain
44 Maize(temperateandsubtropicalcultivars) 105days Grain
45 Maize(temperateandsubtropicalcultivars) 120days Grain
46 Maize(temperateandsubtropicalcultivars) 135days Grain
47 Maize(temperateandsubtropicalcultivars) 150days Grain
48 Maize(temperateandsubtropicalcultivars) 165days Grain
49 Maize(temperateandsubtropicalcultivars) 180days Grain
50 Silagemaize(temperateandsubtropicalcultivars) 105days Fodder
51 Silagemaize(temperateandsubtropicalcultivars) 120days Fodder
52 Silagemaize(temperateandsubtropicalcultivars) 135days Fodder
53 Silagemaize(temperateandsubtropicalcultivars) 150days Fodder
54 Silagemaize(temperateandsubtropicalcultivars) 165days Fodder
55 Silagemaize(temperateandsubtropicalcultivars) 180days Fodder
56 Winterbarley 35+105days Grain
57 Winterbarley 40+120days Grain
58 Winterbarley 45+135days Grain
59 Winterbarley 50+150days Grain
60 Springbarley 90days Grain
61 Springbarley 105days Grain
62 Springbarley 120days Grain
63 Springbarley 135days Grain
64 Barley(subtropicalcultivars) 90days Grain
65 Barley(subtropicalcultivars) 105days Grain
66 Barley(subtropicalcultivars) 120days Grain
67 Barley(subtropicalcultivars) 135days Grain
68 Barley(tropicalhighlandcultivars) 100days Grain
69 Barley(tropicalhighlandcultivars) 115days Grain
70 Barley(tropicalhighlandcultivars) 130days Grain
71 Barley(tropicalhighlandcultivars) 145days Grain
72 Barley(tropicalhighlandcultivars) 160day) Grain
73 Barley(tropicalhighlandcultivars) 175days Grain
74 Barley(tropicalhighlandcultivars) 190days Grain
75 Sorghum(tropicallowlandcultivars) 90days Grain
76 Sorghum(tropicallowlandcultivars) 105days Grain
77 Sorghum(tropicallowlandcultivars) 120days Grain
78 Sorghum(tropicallowlandcultivars) 135days Grain
79 Sorghum(tropicalhighlandcultivars) 120days Grain
80 Sorghum(tropicalhighlandcultivars) 150days Grain
81 Sorghum(tropicalhighlandcultivars) 180days Grain
82 Sorghum(tropicalhighlandcultivars) 210days Grain
83 Sorghum(tropicalhighlandcultivars) 240days) Grain
84 Sorghum(tropicalhighlandcultivars) 270days Grain
85 Sorghum(tropicalhighlandcultivars) 300days Grain
86 Sorghum(temperateandsubtropicalcultivars) 90days Grain
87 Sorghum(temperateandsubtropicalcultivars) 105days Grain
88 Sorghum(temperateandsubtropicalcultivars) 120days Grain
89 Sorghum(temperateandsubtropicalcultivars) 135days Grain
90 Sorghum(temperateandsubtropicalcultivars) 150days Grain
91 Sorghum(temperateandsubtropicalcultivars) 165days Grain
92 Sorghum(temperateandsubtropicalcultivars) 180days Grain
93 Sweetsorghum(temperateandsubtropicalcultivars) 90days Supra
94 Sweetsorghum(temperateandsubtropicalcultivars) 105days Supra
95 Sweetsorghum(temperateandsubtropicalcultivars) 120days Supra
96 Sweetsorghum(temperateandsubtropicalcultivars) 135days Supra
97 Sweetsorghum(temperateandsubtropicalcultivars) 150days Supra
98 Sweetsorghum(temperateandsubtropicalcultivars) 165days Supra
99 Sweetsorghum(temperateandsubtropicalcultivars) 180days Supra
132

Code Croptype Growthcycle Harvestedpart

100 Winterrye 30+90days Grain
101 Winterrye 35+105days Grain
102 Winterrye 40+120days Grain
103 Winterrye 45+135days Grain
104 Springrye 90days Grain
105 Springrye 105days Grain
106 Springrye 120days Grain
107 Springrye 135days Grain
108 Pearlmillet 70days Grain
109 Pearlmillet 90days Grain
110 Foxtailmillet 75days Grain
111 Foxtailmillet 90days Grain
112 Foxtailmillet 105days Grain
113 Foxtailmillet 120days Grain
114 Springoat 90days Grain
115 Springoat) 105days Grain
116 Springoat 120days Grain
117 Buckwheat 75days Grain
118 Buckwheat 90days Grain
119 Whitepotato 90days Tuber
120 Whitepotato 105days Tuber
121 Whitepotato 120days Tuber
122 Whitepotato 135days Tuber
123 Whitepotato 150days Tuber
124 Whitepotato 165days Tuber
125 Whitepotato 180days) Tuber
126 Sweetpotato 120days Tuber
127 Sweetpotato 135days Tuber
128 Sweetpotato 150days Tuber
129 Sweetpotato 165days Tuber
130 Cassava perennial Root
131 Whiteyam 195days Tuber
132 Whiteyam 225days Tuber
133 Greateryam 240days Tuber
134 Greateryam 270days Tuber
135 Yellowyam 330days Tuber
136 Cocoyam 330days Tuber
137 Sugarcane 330days Sugar
138 Sugarbeet 120days Sugar
139 Sugarbeet 135days Sugar
140 Sugarbeet 150days Sugar
141 Sugarbeet 165days Sugar
142 Sugarbeet 180days Sugar
143 Sugarbeet 195days Sugar
144 Sugarbeet 210days Sugar
145 Phaseolusbean(tropicallowlandcultivars) 90days Grain
146 Phaseolusbean(tropicallowlandcultivars) 105days Grain
147 Phaseolusbean(tropicallowlandcultivars) 120days Grain
148 Phaseolusbean(tropicallowlandcultivars) 135days Grain
149 Phaseolusbean(tropicallowlandcultivars) 150days Grain
150 Phaseolusbean(tropicalhighlandcultivars) 120days Grain
151 Phaseolusbean(tropicalhighlandcultivars) 135days Grain
152 Phaseolusbean(tropicalhighlandcultivars) 150days Grain
153 Phaseolusbean(tropicalhighlandcultivars) 165days Grain
154 Phaseolusbean(tropicalhighlandcultivars) 180days Grain
155 Phaseolusbean(temperateandsubtropicalcultivars) 90days Grain
156 Phaseolusbean(temperateandsubtropicalcultivars) 105days Grain
133

Code Croptype Growthcycle Harvestedpart

157 Phaseolusbean(temperateandsubtropicalcultivars) 120days Grain
158 Phaseolusbean(temperateandsubtropicalcultivars) 135days Grain
159 Phaseolusbean(temperateandsubtropicalcultivars) 150days Grain
160 Chickpea 90days Grain
161 Chickpea 105days Grain
162 Chickpea 120days Grain
163 Chickpea(coldtolerant) 150days Grain
164 Chickpea(coldtolerant) 165days Grain
165 Chickpea(coldtolerant) 180days Grain
166 Cowpea 80days Grain
167 Cowpea 100days Grain
168 Cowpea 120days Grain
169 Drypea 90days Grain
170 Drypea 105days Grain
171 Drypea 120days Grain
172 Greengram 60days Grain
173 Greengram 80days Grain
174 Greengram 100days Grain
175 Pigeonpea 135days Grain
176 Pigeonpea 150days Grain
177 Pigeonpea 165days Grain
178 Pigeonpea 180days Grain
179 Pigeonpea 195days Grain
180 Soybean(tropicalandsubtropicalcultivars) 105days Grain
181 Soybean(tropicalandsubtropicalcultivars) 120days Grain
182 Soybean(tropicalandsubtropicalcultivars) 135days Grain
183 Soybean(temperateandsubtropicalcultivars) 105days Grain
184 Soybean(temperateandsubtropicalcultivars) 120days Grain
185 Soybean(temperateandsubtropicalcultivars) 135days Grain
186 Sunflower(tropicalandsubtropicalcultivars) 135days Seed
187 Sunflower(tropicalandsubtropicalcultivars) 150days Seed
188 Sunflower(temperateandsubtropicalcultivars) 105days Seed
189 Sunflower(temperateandsubtropicalcultivars) 120days Seed
190 Sunflower(temperateandsubtropicalcultivars) 135days Seed
191 Sunflower(temperateandsubtropicalcultivars) 150days Seed
192 Winterrape 35+105days Seed
193 Winterrape 40+120days Seed
194 Winterrape 45+135days Seed
195 Winterrape 45+150days Seed
196 Springrape 105days Seed
197 Springrape 120days Seed
198 Springrape 135days Seed
199 Springrape 150days Seed
200 Rabirape 135days Seed
201 Rabirape 150days Seed
202 Groundnut 90days Kernel
203 Groundnut 105days Kernel
204 Groundnut 120days Kernel
205 Oilpalm perennial Oil
206 Olive perennial Oil
207 Jatropha perennial Oil
208 Cabbage 90days Head
209 Cabbage 105days Head
210 Cabbage 120days Head
211 Cabbage 135days Head
212 Cabbage 150days Head
213 Cabbage 165days Head
134

Code Croptype Growthcycle Harvestedpart

214 Carrot(freshearly)(temperateandsubtropicalcultivars) 60days Root
215 Carrot(freshearly)(temperateandsubtropicalcultivars) 75days Root
216 Carrot(freshearly)(temperateandsubtropicalcultivars) 90days Root
217 Carrot(storagelate)(temperateandsubtropicalcultivars) 135days Root
218 Carrot(storagelate)(temperateandsubtropicalcultivars) 165days Root
219 Carrot(storagelate)(temperateandsubtropicalcultivars) 195days Root
220 Carrot(fresh)(tropicalcultivars) 75days Root
221 Carrot(fresh)(tropicalcultivars) 90days Root
222 Carrot(fresh)(tropicalcultivars) 105days Root
223 Onion(temperateandsubtropicalcultivars) 120days Bulb
224 Onion(temperateandsubtropicalcultivars) 135days Bulb
225 Onion(temperateandsubtropicalcultivars) 150days Bulb
226 Onion(temperateandsubtropicalcultivars) 165days Bulb
227 Onion(temperateandsubtropicalcultivars) 180days Bulb
228 Onion(hybernating)(temperateandsubtropicalcultivars) 45+105days Bulb
229 Onion(hybernating)(temperateandsubtropicalcultivars) 60+120days Bulb
230 Onionhybernating)(temperateandsubtropicalcultivars) 75+135days Bulb
231 Onion(tropicalcultivars) 90days Bulb
232 Onion(tropicalcultivars) 105days Bulb
233 Onion)(tropicalcultivars) 120days Bulb
234 Onion(tropicalcultivars) 135days Bulb
235 Tomato(temperateandsubtropicalcultivars) 90days Fruit
236 Tomato(temperateandsubtropicalcultivars) 105days Fruit
237 Tomato(temperateandsubtropicalcultivars) 120days Fruit
238 Tomato(temperateandsubtropicalcultivars) 135days Fruit
239 Tomato(tropicalandsubtropicalcultivars) 105days Fruit
240 Tomato(tropicalandsubtropicalcultivars) 120days Fruit
241 Tomato(tropicalandsubtropicalcultivars) 135days Fruit
242 Banana/Plantain perennial Fruit
243 Citrus perennial Fruit
244 Coconut1(tall) perennial) Copra
245 Coconut2(hybridtall) perennial Copra
246 Coconut3(dwarf) perennial Copra
247 Cacao(comun) perennial Beans
248 Cacao(hybrid) perennial Beans
249 Cotton(tropicalcultivars) 135days Fiber
250 Cotton(tropicalcultivars) 150days Fiber
251 Cotton(tropicalcultivars) 165days Fiber
252 Cotton(tropicalcultivars) 180days Fiber
253 Cotton(temperateandsubtropicalcultivars) 135days Fiber
254 Cotton(temperateandsubtropicalcultivars) 150days Fiber
255 Cotton(temperateandsubtropicalcultivars) 165days Fiber
256 Flax 90days Fiber
257 Flax 105days Fiber
258 Flax 120days Fiber
259 Coffeearabica perennial Greenbeans
260 Coffeerobusta perennial Greenbeans
261 Teachinatea(cameliasinenses) perennial Leaves
262 Teahybrid(sinensisandassamica) perennial Leaves
263 Teaassamtea(cameliasinensesvar.assamica) perennial Leaves
264 Tobacco(tropicalcultivars) 105days Leaves
265 Tobacco(tropicalcultivars) 120days Leaves
266 Tobacco(tropicalcultivars) 135days Leaves
267 Tobacco(temperateandsubtropicalcultivars) 150days Leaves
268 Tobacco(temperateandsubtropicalcultivars) 165day) Leaves
269 Alfalfa(temperateandsubtropicalcultivars) perennial AGB
270 Alfalfa(tropicalcultivars) perennial AGB
135

Code Croptype Growthcycle Harvestedpart

271 Pasturelegumes(C3/Ispecies) perennial AGB
272 Pasturelegumes(C3/IIspecies) perennial AGB
273 Pasturegrasses(C3/Ispecies) perennial AGB
274 Pasturegrasses(C3/IIspecies) perennial AGB
275 Pasturegrasses(C4/IIspecies) perennial AGB
276 Pasturegrasses(C4/lspecies) perennial AGB
277 Miscanthus(C4/IItype) perennial AGB
278 Miscanthus(C4/Itype) perennial AGB
279 Switchgrass perennial AGB
280 Reedcanarygrass perennial AGB
ABG:Abovegroundbiomass.

TableA45Crops/commodities
Code Crop/commodity Crops
1 Wheat Wheat
2 Rice Rice
3 Maize Maize
4 Sorghum Sorghum
5 Millet Millet
6 Othercereals Barley,Rye,Oatandminorothercereals
7 Tubers Potato,Sweetpotato
8 Roots Cassava,Yams,otherRootsandPlantain
9 Sugarbeet Sugarbeet
10 Sugarcane Sugarcane
11 Pulses Pulses
12 Soybean Soybean
13 Rape Rapeseed
14 Sunflower Sunflower
15 Groundnut Groundnutsinshells
16 Oilpalm Oilpalm
17 Olive Olive
18 Cotton Cotton
19 Cashcrops1 Banana,Coconut
20 Vegetables Vegetables
21 Cashcrops2 Coffee,Tea,Cocoa
22 Fodder Fodder
23 Residual Othercropsnotlistedabove:mainlyfruit,nuts,
spices,tobacco,fibercrops,otheroilcrops
Color code indicates crops/commodities available in yield and
productiongapassessments

136

Appendix42 Parametersforcalculationofwaterlimitedyields

TableA46Waterlimitedyieldparameters

LengthofCropStage
(%ofgrowthcycle)
Cropwaterrequirementsrelativetoreference
evapotranspiration
Yieldlossfactors
NAME d1 d2 d3 d4 K1c K3c K5c KTc Ky1 Ky2 Ky3 Ky4 KyT
Wheat(winter) 10.00 30.00 35.00 25.00 0.40 1.10 0.30 0.85 0.20 0.60 0.75 0.50 1.05
Wheat(spring) 10.00 20.00 45.00 25.00 0.40 1.10 0.30 0.85 0.20 0.65 0.80 0.55 1.15
Rice(wetland) 10.00 30.00 30.00 30.00 1.10 1.20 0.90 1.10 1.00 2.00 2.50 1.00 2.00
Rice(dryland) 10.00 30.00 30.00 30.00 0.50 1.20 0.60 0.90 0.40 0.90 1.50 0.50 1.25
Maize/grain) 15.00 30.00 35.00 20.00 0.40 1.10 0.60 0.85 0.40 0.90 1.50 0.50 1.25
Barley(winter) 10.00 30.00 35.00 25.00 0.40 1.10 0.25 0.85 0.20 0.60 0.75 0.50 1.05
Barley(spring) 10.00 20.00 45.00 25.00 0.40 1.10 0.25 0.85 0.20 0.65 0.80 0.55 1.15
Sorghum 10.00 25.00 40.00 25.00 0.40 1.05 0.55 0.80 0.20 0.60 0.90 0.50 0.90
Sweetsorghum 10.00 25.00 40.00 25.00 0.40 1.05 1.00 0.95 0.20 0.60 0.90 0.50 0.90
Winterrye 10.00 30.00 35.00 25.00 0.40 1.10 0.25 0.85 0.20 0.60 0.75 0.50 1.05
Springrye 10.00 20.00 45.00 25.00 0.40 1.10 0.25 0.85 0.20 0.65 0.80 0.55 1.15
PearlMillet 10.00 25.00 40.00 25.00 0.35 1.05 0.30 0.80 0.20 0.60 0.80 0.50 0.90
FoxtailMillet 10.00 25.00 40.00 25.00 0.40 1.10 0.55 0.85 0.20 0.60 0.80 0.50 1.00
Springoat 10.00 20.00 45.00 25.00 0.40 1.10 0.25 0.85 0.20 0.65 0.80 0.55 1.15
Buckwheat 15.00 20.00 40.00 25.00 0.40 1.05 0.30 0.80 0.20 0.60 0.80 0.50 0.90
WhitePotato 20.00 25.00 35.00 20.00 0.50 1.10 0.75 0.85 0.50 0.80 0.80 0.70 1.10
SweetPotato 20.00 25.00 35.00 20.00 0.50 1.10 0.65 0.85 0.50 0.80 0.80 0.70 1.10
Sugarbeet 15.00 30.00 35.00 20.00 0.50 1.10 0.70 0.85 1.00 1.00 1.00 0.50 1.10
PhaseolousBean 20.00 33.00 33.00 14.00 0.40 1.10 0.90 0.85 0.20 0.60 1.10 0.75 1.15
Chickpea 20.00 33.00 33.00 14.00 0.40 1.00 0.60 0.85 0.20 0.60 1.10 0.75 1.15
Cowpea 20.00 33.00 33.00 14.00 0.40 1.05 0.60 0.85 0.20 0.60 1.10 0.75 1.15
Greengram 20.00 33.00 33.00 14.00 0.40 1.05 0.60 0.85 0.20 0.60 1.10 0.75 1.15
Pigeonpea 20.00 30.00 30.00 20.00 0.40 1.00 0.60 0.85 0.20 0.60 1.10 0.75 1.15
137

LengthofCropStage
(%ofgrowthcycle)
Cropwaterrequirementsrelativetoreference
evapotranspiration
Yieldlossfactors
NAME d1 d2 d3 d4 K1c K3c K5c KTc Ky1 Ky2 Ky3 Ky4 KyT
Groundnut 20.00 30.00 30.00 20.00 0.50 1.05 0.60 0.80 0.20 0.80 0.80 0.60 0.70
Soybean 15.00 20.00 45.00 20.00 0.40 1.10 0.50 0.85 0.20 0.80 1.00 0.80 0.85
Sunflower 17.00 28.00 35.00 20.00 0.40 1.10 0.40 0.80 0.25 0.60 1.00 0.80 0.95
Rape 15.00 25.00 40.00 20.00 0.50 1.10 0.40 0.80 0.20 0.80 1.00 0.80 0.85
Cotton 15.00 30.00 30.00 25.00 0.40 1.15 0.70 0.85 0.20 0.50 0.75 0.50 0.85
Flax 15.00 25.00 35.00 25.00 0.40 1.10 0.30 0.80 0.20 0.60 0.80 0.50 0.95
Whiteyam 20.00 25.00 35.00 20.00 0.50 1.05 0.65 0.85 0.50 0.80 0.80 0.70 1.10
Greateryam 20.00 30.00 30.00 20.00 0.50 1.10 0.65 0.85 0.50 0.85 0.85 0.70 1.10
Tombacco 10.00 25.00 30.00 35.00 0.35 1.10 0.80 0.90 0.20 1.00 0.80 0.50 0.90
Onion 15.00 25.00 30.00 30.00 0.60 1.05 0.80 0.80 0.45 0.45 0.80 0.30 1.10
Tomato 15.00 25.00 30.00 30.00 0.50 1.15 0.80 0.85 0.40 0.40 1.10 0.40 1.05
Cabbage 25.00 35.00 25.00 15.00 0.60 1.05 0.90 0.80 0.20 0.35 0.45 0.60 0.95
Carrot 20.00 25.00 40.00 15.00 0.70 1.05 0.95 0.90 0.40 0.40 0.80 0.30 1.10
Onion 15.00 25.00 30.00 30.00 0.60 1.05 0.80 0.80 0.45 0.45 0.80 0.30 1.10
Notes: The coefficients d1, , d4 relate to the characteristics of the crop growth cycle, denoting here the relative length (in percent) of four crop development stages, namely, initial stage, vegetative stage,
reproductivestage,andmaturationstage.Parametersk1
c
,k2
c
,andk3
c
definecropwaterrequirementsrespectivelyfortheinitialstage,thereproductivephase,andtheendofthematurationstage.Coefficient
k0
c
indicateswaterrequirementsrelativetoreferenceevapotranspirationovertheentiregrowthcycle.Finally,factorsK
y
quantifytheexpectedyieldlossinrelationtoacropevapotranspirationdeficit,bycrop
stageandfortheentiregrowthcycle,respectively.

138

Appendix43 TemperatureProfileRequirements
Thisdocumentisavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/Temperature_profile_requirements.pdf

139

Appendix44 Cropvernalizationrequirements
Somecropsrequireavernalizationperiod(i.e.dayswithcoldtemperatures)forperformingspecific
phenological development phases such as flowering. The production of flowers and grains, which
directly influences crop yield, is dependenton theextentand intensity of exposureto periods with
coldtemperature. This cold temperature requirement ismeasured in vernalization days (VD, days).
InGAEZ, there are four hibernating cropsthat need to fulfill vernalizationrequirements in orderto
produce:winterwheat,winterbarley,winterryeandwinterrape.

Therateofvernalization(fvn,VD/day)foradailyaveragetemperatureTaiscalculatedforeachLUT.

( ) ( ) ( )
( )

=
0
,
2
) (
2
2 2
else Tv Ta forTV
Tv T
Tv Ta Tv T Tv Ta
Ta fvn
x n
n opt
n n opt n

where:
Tv
n
, Tv
opt
, and Tv
x
are the cardinal temperatures for vernalization (minimum, optimum, and
maximum)
Thecoefficientiscalculatedas:

( ) ( )
n opt n x
Tv Tv Tv Tv
=
ln ln
2 ln

The accumulation of VD occurs during the dormancy period plus up to additional 60 days after
dormancy to account for cold temperature during early stages when temperatures increase above
5
o
C and vernalization processes continue. The parameters used for fvn calculation in GAEZ are
showninTableA4.

TableA47Parameterizationforthecalculationoftherateofvernalization

Crop Tv
opt
Tv
x
Tv
n
VD
0
VD
100

Winterwheat 5 15 1 10 45
Winterbarley 4 12 0 8 35
Winterrye 5 15 2 10 45
Winterrape 3 10 0 8 30
VD
100
isthenumberofvernalizationdaysrequiredforachievingfullvernalization
VD
o
istheminimumlevelofVDrequiredinGAEZforproceedingwithyieldcalculations

The number of vernalization days (VD) is then calculated by accumulating the rate of vernalization
(fvn,

VD/day) for the period between the start and the end of the dormancy period plus up to 60
days.
( )

= Ta fvn VD

140

YieldcalculationsforaLUTonlyproceedifVDisgreaterthanVD
0
,whichimpliesthatsomelevelof
vernalization occurred. If VD > VD
0,
a vernalization factor (fthz, fractional) is then calculated as a
functionofVD:

5 5
50
5
VD VD
VD
fthz
+
=

where:
VD
50
is50%ofvernalizationdaysrequiredforfullvernalization(VD
100
).

141

Appendix45 Biomassandyieldcalculation
The AEZ methodology for the calculation of potential net biomass and yields is based on eco
physiologicalprinciples,asoutlinedbelow:
To calculate the net biomass production (B
n
) of a crop, an estimation of the gross biomass
production(B
g
)andrespirationloss(R)isrequired:
B
n
=B
g
R (1)
The equation relating the rate of net biomass production (b
n
) to the rate of gross biomass
production(b
g
)andtherespirationrate(r)is:
b
n
=b
g
r (2)
The maximum rate of net biomass production (b
nm
) is reached when the crop fully covers the
groundsurface.Theperiodofmaximumnetcropgrowth,i.e.,thepointintimewhenmaximumnet
biomassincrementsoccur,isindicatedbytheinflectionpointofthecumulativegrowthcurve.When
the first derivative of net biomass growth is plotted against time the resulting graph resembles a
normaldistributioncurve.Themodelassumesthattheaveragerateofnetproduction(b
na
)overthe
entire growth cycle is half the maximum growth rate, i.e., b
na
= 0.5 b
nm
. The net biomass
productionforacropofNdays(B
n
)isthen:
B
n
=0.5b
nm
xN (3)
The maximum rate of gross biomass production (b
gm
) is related to the maximum net rate of CO
2

exchange of leaves (P
m
) which is dependent on temperature, the photosynthesis pathway of the
crop,andthelevelofatmosphericCO
2
concentration.
For a standard crop, i.e., a crop in adaptability group I with P
m
= 20 kg ha
1
hr
1
and a leaf area
indexofLAI=5,therateofgrossbiomassproductionb
gm
iscalculatedfromtheequation:
b
gm
=Fxb
o
+(1F)b
c
(4)
where:
F= thefractionofthedaytimetheskyisclouded,F=(A
c
0.5R
g
)/(0.8A
c
),whereA
c
(or
PAR) is the maximum active incoming shortwave radiation on clear days (de Wit,
1965),andR
g
isincomingshortwaveradiation(botharemeasuredincalcm
2
day
1
)
b
o
= grossdrymaterproductionrateofastandardcropforagivenlocationandtimeofthe
yearonacompletelyovercastday,(kgha
1
day
1
)(deWit,1965)
b
c
= grossdrymaterproductionrateofastandardcropforagivenlocationandtimeofthe
yearonaperfectlyclearday,(kgha
1
day
1
)(deWit,1965)
WhenP
m
isgreaterthan20kgha
1
hr
1
,b
gm
isgivenbytheequation:
b
gm
=F(0.8+0.01P
m
)b
o
+(1F)(0.5+0.025P
m
)b
c
(5)
WhenP
m
islessthan20kgha
1
hr
1
,b
gm
iscalculatedaccordingto:
b
gm
=F(0.5+0.025P
m
)b
o
+(1F)(0.05P
m
)b
c
(6)
To calculate the maximum rate of net biomass production (b
nm
), the maximum rate of gross
biomassproduction(b
gm
)andtherateofrespiration(r
m
)arerequired.Here,growthrespirationis
considered a linear function of the rate of gross biomass production (McCree, 1974), and
142

maintenance respiration a linear function of net biomass that has already been accumulated (B
m
)
Whentherateofgrossbiomassproductionisb
gm
,therespirationrater
m
is:
r
m
=kb
gm
+cB
m
(7)
wherekandcaretheproportionalityconstantsforgrowthrespirationandmaintenancerespiration
respectively, and B
m
is the net biomass accumulated at the time of maximum rate of net biomass
production. For both legume and non legume crops k equals 0.28. However, c is temperature
dependentanddiffersforthetwocropgroups.At30
o
C,factorc
30
foralegumecropequals0.0283
and for a nonlegume crop 0.0108. The temperature dependence of c
t
for both crop groups is
modelledwithaquadraticfunction:
c
t
=c
30
(0.0044+0.0019T+0.0010T
2
). (8)
ItisassumedthatthecumulativenetbiomassB
m
ofthecrop(i.e.,biomassattheinflectionpointof
thecumulativegrowthcurve)equalshalfthenetbiomassthatwouldbeaccumulatedattheendof
the crop's growth cycle. Therefore, we set B
m
= 0.5 B
n
, and using (3), B
m
for a crop of N days is
determinedaccordingto:
B
m
=0.25b
nm
xN (9)
By combining the respiration equation with the equation for the rate of gross photosynthesis, the
maximum rate of net biomass production (b
nm
) or the rate of net dry matter production at full
coverforacropofNdaysbecomes:
b
nm
=0.72b
gm
/(1+0.25c
t
N) (10)
Finally,thenetbiomassproduction(B
n
)foracropofNdays,where0.5b
nm
istheseasonalaverage
rateofnetbiomassproduction,canbederivedas:
B
n
=(0.36b
gm
xL)/(1/N+0.25c
t
) (11)
where:
b
gm
= maximumrateofgrossbiomassproductionatleafareaindex(LAI)of5
L= growthratio,equaltotheratioofb
gm
atactualLAItob
gm
atLAIof5
N= lengthofnormalgrowthcycle
c
t
= maintenance respiration, dependent on both crop and temperature according to
equation(8)
Potentialyield(Y
p
)isestimatedfromnetbiomass(B
n
)usingtheequation:
Y
p
=H
i
xB
n
(12)
where:
H
i
= harvestindex,i.e.,proportionofthenetbiomassofacropthatiseconomicallyuseful
Thus,climateandcropcharacteristicsthatapplyinthecomputationofnetbiomassandyieldare:(a)
heat and radiation regime over the crop cycle, (b) crop adaptability group to determine applicable
rateofphotosynthesisP
m
,(c)lengthofgrowthcycle(fromemergencetophysiologicalmaturity),(d)
lengthofyieldformationperiod,.(e)leafareaindexatmaximumgrowthrate,and(f)harvestindex.

143

Appendix46 Biomassandyieldparameters
Thisdocumentisavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/Biomass_yield_parameters.xls

144

Appendix47 OutputofModuleII
TableA48ContentoffixedoutputrecordsfromGAEZModuleII
Variable Parameter
Record
number
Type of
variable
Length of
variable
(inbytes)
btext Explanatorytextstring 1 Character 16
version Programversionstring 2 Character 24
datestr Datestringwhenfilewascreated 3 Character 9
Mrow Numberofrowsofgrid 4 Integer 2
Mcol Numberofcolumnsofgrid Integer 2
CR1SEL Indexoffirstcropinoutputfile Integer 2
CR2SEL Indexoflastcropinoutputfile Integer 2
irow0 Row number of upper left corner of subwindow (if
used,0else)
Integer 2
icol0 Columnnumberofupperleftcornerofsubwindow Integer 2
irow1 Rownumberoflowerrightcornerofsubwindow Integer 2
icol1 Columnnumberoflowerrightcornerofsubwindow Integer 2
Admsel Code of administrative unit selected for running (if
used,else0)
Integer 2
itech Inputlevel Integer 2
iflmst ControlparameterIFLMST Integer 2
iagclc ControlparameterIAGCLC Integer 2
irtawc ControlparameterIRTAWC Integer 2
daymin ControlparameterDAYMIN Integer 2
lenmin ControlparameterLENMIN Integer 2
itflg ControlparameterITFLG Integer 2
Rlps Lapserateapplied(degreeCper1m) 5 Real 4
Ppm AtmosphericCO2concentration(ppm) Real 4
dRl Parameter for change in water use efficiency under
elevatedCO2
Real 4
Sa0 AWClevel(mm/m) Real 4
Sdep0 Maximumapplicablesoildepth(m) Real 4
Rplim1 WaterbalancecontrolparameterRPLIM1 Real 4
Rplim2 WaterbalancecontrolparameterRPLIM2 Real 4
Samin WaterbalancecontrolparameterSAMIN Real 4
Tastr TemperaturethresholdTASTRT(usually5degC) Real 4
Kc1 WaterbalancecontrolparameterKc1 Real 4
Kc2 WaterbalancecontrolparameterKc2 Real 4
Kc3 WaterbalancecontrolparameterKc3 Real 4
KC4 WaterbalancecontrolparameterKc4 Real 4
KC5 WaterbalancecontrolparameterKc5 Real 4
Kc6 WaterbalancecontrolparameterKc6 Real 4
Kc7 WaterbalancecontrolparameterKc7 Real 4
dTx ClimatesensitivityruncontrolparameterdT1 Real 4
dTn ClimatesensitivityruncontrolparameterdT2 Real 4
dP ClimatesensitivityruncontrolparameterdP1 Real 4
dS ClimatesensitivityruncontrolparameterdS1 Real 4
dW ClimatesensitivityruncontrolparameterdW1 Real 4
Idxok IndicatorofLUTsusedinsimulation(0=off,1=on) 6 Integer 2*ncrp
Midx Referenceharvestindex(kgproduce/kgbiomass) 7 Real 4*ncrp
flnmap1 Inputfilenameofgridcelllandmaskused 8 Character 50
flninp Inputfilenameoflandpixelfile 9 Character 50
flncl1 Inputfilename:averagemonthlytemperature 10 Character 50
flncl2 Inputfilename:averagemonthlytemperaturerange 11 Character 50
flncl3 Inputfilename:monthlyprecipitation 12 Character 50
flncl4 Inputfilename:monthlywindrun 13 Character 50
145

Variable Parameter
Record
number
Type of
variable
Length of
variable
(inbytes)
flncl5 Inputfilename:averagemonthlysunshinefraction 14 Character 50
flncl6 Inputfilename:averagemonthlyrelativehumidity 15 Character 50
flncl7 Inputfilename:monthlywetdayfrequency 16 Character 50
flngem Inputfilename:climatechangefromGCM 17 Character 50
EoH EndofheaderstringEOH 18 Character 3
*ncrpnumberofcrops=indexlastcropindexfirstcrop+1

TableA49InformationcontainedineachpixeldatarecordofModuleII
Variable Description
Length of
variable
(inbytes)
Type of
variable
irow Pixelreference:rownumber 2 Integer
icol Pixelreference:columnnumber 2 Integer
alt Pixelreference:medianelevation[m] 2 Integer
lgpt2 LengthofLGPt=5 2 Integer
lgpt3 LengthofLGPt=10 2 Integer
lgptot Totalnumberofgrowingperioddays 2 Integer
ndwtot Number of days when estimated ETa of reference crop equals
referenceETo
2 Integer
ndhtot NumberofdayswhenprecipitationexceedsreferencetoEto 2 Integer
nlgp Numberofdistinctcomponentgrowingperiods 2 Integer
begdrm Beginningofdormancyperiod(dayofyear) 2 Integer
enddrm Endofdormancyperiod(dayofyear) 2 Integer
Ym0 Maximumradiation/temperaturelimitedyield(kgperhectare) 4*ncrp Real
fc1 Cropspecific yield reduction factor obtained by thermal profile
evaluation;indexranging010000.
2*ncrp Integer
fc2 Cropspecificyieldreduction factorduetowaterdeficit(CROPWAT
method);indexranging010000.
2*ncrp Integer
fc3 Cropspecificyieldreductionfactorduetoagroclimaticconstraints;
indexranging010000.
2*ncrp Integer
cdef CropwaterdeficitbyLUT(=cropspecificETa ETa,mm) 2*ncrp Integer
ceta Crop/LUTspecificETa(mm) 2*ncrp Integer
ctsum Crop/LUTspecific accumulated temperature during growth cycle
(degreedays)[Cd]
2*ncrp Integer
ccyl Crop/LUTspecificgrowthcyclelength[days] 2*ncrp Integer
ccbd Crop/LUTspecificbeginningofgrowthcycle[ dayofyear]
ncrpnumberofcrops=indexlastcropindexfirstcrop+1

146

Appendix48 SubroutinedescriptionsofModuleII
ModuleIIprocessesthegriddeddatabasesoftheAEZstudyareatoestimateLGP(s)andcalculate
maximumattainableyields.Thisincludes:

1. cropcyclethermalsuitabilityevaluation
2. pseudodailywaterbalancecalculation
3. biomassandyieldcalculation

Applicationoftheserulesyieldsanaveragesuitabilityfactorrelativetothemaximumtemperature
limitedattainableyieldofaparticularcrop/LUT.

Figure41isadiagramillustratingthelinkageslistedinTable410.Thesubroutinesandfunctions
arelistedinalphabeticalorderandwithashortdescriptionoftheirusewithinModuleIIandwhich
routinescallthemandwhicharecalledbythem.Table411liststhefortranfilesbelongingto
ModuleII,andliststheheaderfilesassociatedwitheachfortranfileandthesubroutinesand
functionsincludedineachfile.
147

A
n
n
u
a
l

p02
ctrlin
clrall
tbget0
readin
rdmpu
clr2bi
lgpfun
climate
thcs1
optbio
agrcst
biomass
higra
thcs2
tcc lgps gcmdat
rdymax rdwatq rdslu rdcrp2 rdcrp1 rdagcl loadt1
getbuf loadt0
Clr4br
et0 dayhr
f365s1
f365s2
daily
gaussj
amat5 thridx trp wdfrq
dayhr
scalefct
vrnfct
setdat
lendat
eta
psh
cythz cyprof cycval cycsum wstart
slim3 slim2 slim1
agcidx
watreq
cycyav interp laigra pmval tval
eta psh
daily higra cycval
clr2bi
vrnfct cythz cyprof cycval cycsum wstart
agrcst biomass thcs1 thcs2 watreq
Perennial
agcidx
interp
daily
higra
laigra
cycyav
pmval
tval cycval
cycsum
cycal
cyprof
cythz
vrnfct
wstart
vrnfct
cycval
cythz
slim1 wstart
cycsum
eta
psh
cypr
slim3 slim2
Figure41DiagramofthesubroutinesandfunctionsofGAEZModuleII
148

TableA410SubroutinesandfunctionsofModuleII
FileName Subroutine Description Calledfrom Callsto
BIO.F AGCIDX determineslgpindexandmultiplier
fractiondependingonlengthof
growingperiod
AGRCST
BIO.F AGRCST evaluatesagroclimaticconstraints P02(MAIN),OPTBIO AGCIDX
LGPSUB.F AMAT5 generatescoefficientsofasystemof
linearequationsforgeneratingdaily
valuesbysplineinterpolation
CLIMATE GAUSSJ
BIO.F BIOMASS calculatesunconstrainedcropbiomass
andyieldbasedonmethodssetoutin
FAOsWorldSoilResourcesReportNo
48/1oftheAgroEcologicalProject,
FAO1978,butincorporatingmany
enhancementstotheoriginal
procedure.
P02,OPTBIO CYCVAL,CYCYAV,
DAILY,HIGRA,
INTERP,LAIGRA,
PMVAL,TVAL
CLIM_GCM.F CLIMATE readsclimatedataforthecurrent
positionanddetermineLGPs
P02 AMAT5,DAILY,
DAYHR,ET0,
GCMDAT,LGPS,TCC,
THRIDX,TRP,WDFRQ
CLR.F CLR2BI utilityprocedurewhichinitializesn
elementsofa2byteintegerarrayxto
val
CLRALL,P02
CLR.F CLR4BI utilityprocedurewhichinitializesn
elementsofa4byteintegerarrayxto
val

CLR.F CLR4BR utilityprocedurewhichinitializesn

elementsofa4byterealarrayxtoval
CLRALL,CLRPIX,
CLRLNV

CLR.F CLR8BR utilityprocedurewhichinitializesn

elementsofa8byterealarrayxtoval

CLR.F CLRALL utilityprocedurewhichinitializes

variablesandarraysusedbyP02
P02 CLR2BI,CLR4BR
CLR.F CLRCH1 utilityprocedurewhichinitializesn
elementsofa1bytecharacterarrayx
toval
CLRPIX,CLRLNV
CLR.F CLRLNV clearsvariables,arraysbefore
processingnextlandinventoryentry
CLR4BR,CLRCH1
CLR.F CLRPIX clearsvariables,arraysbefore
processingnextrasterpoint
P02 CLR2BI,CLR4BR,
CLRCH1
P02.F CTRLIN readscontrolinformationandopens
I/Ofiles
P02 ERROR,GETBUF,
LOADT0
THCS1.F CYPROF calculatesdailytemperatureprofile THCS1,THCS2
LGPSUB.F DAILY convertsmonthlyordecadaldatato
dailyvaluesandcallsthenexttwo
functions.
BIOMASS,CLIMATE F365S1,F365S2
ERROR.F ERROR writesouterrormessages LOADT0,LOADT1,
CTRLIN,TBGET0

LGPSUB.F F365S1 fillsindailyvaluesbyspline

interpolation
DAILY
LGPSUB.F F365S2 fillsindailyvaluesbyspline
interpolationwhenthevaluemustbe
greaterorequalto0
DAILY
LGPSUB.F GAUSSJ performsgaussjordanelimination
withfullpivoting
AMAT5
CLIM_GCM.F GCMDAT readsGCMclimatechangedatafor
currentpixel
CLIMATE DAYHR,SCALEFCT
BIO.F INTERP producesweightedaverageoftwo
series
BIOMASS
LGP.F LGPS determinesthelengthofgrowing
period
CLIMATE ETA,LENDAT,PSH,
SETDAT
LOADTB.F LOADT0 readsdistrictorprovincecodes CTRLIN ERROR

149

FILENAME Subroutine Function CALLEDFROM CALLSTO

LOADTB.F LOADT1 readsadditionalcropspecific
information
TBGET0 ERROR
P02.F OPTBIO determinesoptimalcropcalendarfor
givenLUTtomaximizeattainable
agroclimaticyield
P02 AGRCST,BIOMASS,
THCS1,THCS2,
WATREQ
TBGET0.F RDCRP1 readscropparametersfromPAR.CN
file
TBGET0
TBGET0.F RDCRP2 readscropparametersfromPAR.BI
files
TBGET0
TBGET0.F RDMPU readstheoptimalsoilmappingunit
availablewatercontentdatafroman
inputfile(par.mpu)
P02
TBGET0.F RDSLU readssoilunitcodesfrompar.slu TBGET0
TBGET0.F RDWATQ readscropwaterrequirement
coefficientsfrompar.wq
TBGET0
TBGET0.F RDYMAX readsmaximumyieldsperinputlevel
frompar.ym
TBGET0
READIN.F READIN readsinthelandinformationfor
currentpixel
P02
CLIM_GCM.F SCALEFCT algorithmusedtocalculateascaling
fractionforscalingmonthlyclimatic
datavariablestomatchtheannual
changesfromGCM
GCMDAT
LGP.F SETDAT shiftsacalculateddayoftheyearbya
multipleof365tofitwithintherange
0365
LGPS
TBGET0.F TBGET0 setsupvarious'incore'tableswhich
areindependentofindividualland
inventoryentries
P02 ERROR,LOADT1,
RDAGCL,TDCRP1,
RDCRP2,RDSLU,
RDWATQ,RDYMAX
CLIM_GCM.F TCC calculatesthermalclimateclassand
thermalzoneclass
CLIMATE
CLIM_GCM.F TRP calculatestemperaturegrowing
periodsandthermalregime
parameters
CLIMATE THRIDX
BIO.F TVAL utilityroutinetolookupavalueina
table
BIOMASS,PMVAL
WATREQ.F WATREQ estimatestheimpactofwaterstress
onrainfedattainablecropyields
P02,OPTBIO ETA,PSH
CLIM_GCM.F WDFRQ distributesrainaccordingtowetday
frequency
CLIMATE
Functions
LGPSUB.F CYCSUM integratesanattributeovergrowth
cycle
THCS1,THCS2
LGPSUB.F CYCVAL averagesanattributeovergrowth
cycle
BIOMASS,THCS1,
THCS2

BIO.F CYCYAV averagesattributesovertheyearfor

grassandpermanentcrops
BIOMASS
THCS1.F CYTHZ calculatescyclecomponentinthermal
range
THCS1,THCS2
DAYHR.F DAYHR calculatesdaylengthforagiven
latitudeanddayoftheyear
CLIMATE,GCMDAT
ET0.F ET0 calculatespotential
evapotranspirationbythePenman
Monteithmethod
CLIMATE
ETAM.F ETA calculatesactualevapotranspiration
bysimulatingadailywaterbalancefor
aFAOreferencecrop(similartograss)
LPGS,WATREQ
BIO.F HIGRA calculatestheharvestindexforgrass
(i.e.theconsumablefraction)
BIOMASS,P02
P02.F ISRFED deprecatedfunction:checksifland
canbeusedforrainfedproduction

150

FILENAME Subroutine Function CALLEDFROM CALLSTO

BIO.F LAIGRA calculatestheleafareaindexofa
grass/legumemixture
BIOMASS
LGP.F LENDAT calculatesthenumberofdays
betweentwodatesincludingthestart
andenddate.
LGPS
BIO.F PMVAL calculatesphotosynthesisrateasa
functionoftemperatureandC02
concentration
BIOMASS TVAL
ETAM.F PSH calculatesthesoilwaterdepletion
fraction(p)foragivencroptypeand
levelofdailyETo
LGPS,WATREQ
THCS2.F SLIM interpolatesamultiplierforvalues
betweenrangeandoptimum
conditionsandappliedincaseofcycle
constraints
THCS2
THCS2.F SLIM2 Interpolatesamultiplierforvalues
betweenrangeandoptimum
conditions,appliedincaseof
temperaturesumandLGPT
constraints
THCS2
THCS2.F SLIM3 Interpolatesamultiplierforvalues
betweenrangeandoptimum
conditions,appliedincaseofwetland
ricestartup
THCS2
THCS1.F THCS1 evaluatesforagivenLUTandcrop
calendarthecriteriaforoptimum
growingcondition
P02,OPTBIO CYCSUM,CYCVAL,
CYPROF,CYTHZ,
VRNFCT,WSTART
THCS2.F THCS2 evaluatesforagivenforagivenLUT
andcropcalendarthecriteriafor
rangecondition
P02,OPTBIO CYCSUM,CYCVAL,
CYPROF,CYTHZ,
SLIM,SLIM2,SLIM3,
VRNFCT,WSTART
CLIM_GCM.F THRIDX determinestemperatureprofileclass
index
CLIMATE,TRP
THCS1.F VRNFCT calculatesvernalizationfactor THCS1,THCS2
THCS2.F WSTART calculatesaccumulatedwaterforrice
startup
THCS1,THCS2

151

TableA411HeaderandfortransourcefilessubroutinesandfunctionsforGAEZModuleII
Fortranfile AssociatedHeaderFiles Subroutines Functions
BIO.F aezdef.h,bio.h,clim.h,
control.h,iounit.h,tcc.h
AGCIDX,AGRCST,BIOMASS,
INTERP,TVAL
PMVAL,HIGRA,
LAIGRA,CYCYAV
BLKAEZ02.F aezdef.h,aez02.h,clim.h,
iounit.h,lnvrec.h,tabdef.h

CLIM_GCM.F aez02.h,aezdef.h,clim.h,
control.h,iounit.h,tcc.h
CLIMATE,LGPFUN,GCMDAT,
SCALEFCT,TCC,TRP,WDFRQ
THRIDX
CLR.F aezdef.h,aez02.h,bio.h,clim.h,
control.h,iounit.h,lnvrec.h,
tabdef.h
CLRALL,CLRPIX,CLRLNV,CLR8BR,
CLR4BR,CLR4BI,CLR2BI,CLRCH1

DAYHR.F DAYHR
ERROR.F ERROR
ET0.F ET0
ETAM.F ETA,PSH
LGP.F LGPS,SETDAT LENDAT
LGPSUB.F AMAT5,DAILY,F365S1,F365S2,
GAUSSJ
CYCVAL,CYCSUM,
TRCIDX
LOADTB.F aezdef.h,aez02.h,clim.h,
control.h,iounit.h
LOADT0,LOADT1
P02.F aezdef.h,aez02.h,bio.h,clim.h,
control.h,filter.h,iounit.h,
lnvrec.h,tabdef.h,tcc.h
CTRLIN,OPTBIO ISRFED,GETBUF
READIN.F aezdef.h,aez02.h,clim.h,
control.h,iounit.h,lnvrec.h
READIN
TBGET0.F aezdef.h,aez02.h,bio.h,clim.h,
control.h,iounit.h,lnvrec.h,
tabdef.h,usle.h
RDAGC1,RDCRP1,RDCRP2,
RDMPU,RDSLU,RDWATQ,
RDYMAX,TBGET0

THCS1.F aezdef.h,bio.h,clim.h,
control.h,tcc.h
CYPROF THCS1,CYTHZ,
VRNFCT
THCS2.F aezdef.h,bio.h,clim.h,
control.h,tcc.h
SLIM,SLIM2,SLIM3,
THCS2,WSTART
WATREQ.F aezdef.h,bio.h,clim.h WATREQ

152

Appendix49ExampleofModuleIIoutputatgridcelllevel
Example of information generated at grid cell level is given for Ilonga, Tanzania is available for
downloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/Example_grid_cell_output_Module_II.docx

Appendix51 Agroclimatic constraints for individual crop/LUTs

andinputlevelsforrainfedconditions
Thisdocumentisavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/Agroclimatic_constraints.xls

Appendix52 OutputsModuleIII
The output format of Module III is identical to output produced in Module II and described above.
Note, the main purpose of Module III is to compute and update the LUTspecific agroclimatic
constraintfactorsstoredintheresultfile.

153

Appendix53 SubroutinedescriptionsofModuleIII
In Module III, yield losses caused by agroclimatic constraints are quantified for each LUT and
location.TheresultisstoredasreductionfactortobeappliedtothisyieldcalculatedinModuleII.
Fivedifferentyieldconstraints(i.e.yieldreducingfactors)aretakenintoaccount:
a. Longterm limitation to crop performance due to yeartoyear soil moisture balance
variability
b. Pests,diseasesandweedsdamageonplantgrowth
c. Pests,diseasesandweedsdamageonqualityofproduce
d. Climaticfactorsaffectingtheefficiencyoffarmingoperations
e. Frosthazardandextremetemperatureevents

TableA51SubroutinesandfunctionsofModuleIII
Filename Subroutine Function Calledfrom Callsto
P03.F AGCIDX
determineslgpindexandweightfactorforlinear
interpolationofagroclimaticconstraintvalues
AGCIDX
P03.F AGRCST
calculatesreductionfactorforyieldlossescausedby
agroclimaticconstraints
P03 AGCIDX
P03.F RDAGCL
readsagroclimaticconstraintparametersfromdata
files
P03

Fortranfile AssociatedHeaderFiles Subroutines Functions
P03.F AEZDEF.H,CLIM.H,TCC.H,BIO.H,CONTROL.H,IOUNIT.H AGCIDX,AGRCST,RDAGCL

FigureA51DiagramofthesubroutinesandfunctionsofGAEZModuleIII
p03
agrcst
rdagcl
agcidx
154

Appendix61Soildrainageclasses
Soil drainage classes are based on the "Guidelines to estimation of drainage classes based on soil
type, texture, soil phase and terrain slope" (FAO, 1995). The estimation procedures have been
appliedtoallsoiltype,texture,soilphaseandbroadslopeclassesandresultshavebeendistributed
over GAEZ slope classes. The results of the soil drainage evaluation for the FAO 1974 and the FAO
1990soilclassificationareavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/GAEZ_Soil_Drainage_Characteristics.xls
Appendix62 Soilprofileattributesuitabilityratings
Relateddocumentsareavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/soil_evaluation.html
Appendix63 Soiltexturesuitabilityratings
Relateddocumentsareavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/soil_evaluation.html
Appendix64 Soildrainagesuitabilityratings
Relateddocumentsareavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/soil_evaluation.html
Appendix65 Soilphasesuitabilityratings
Relateddocumentsareavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/soil_evaluation.html
Appendix66 Terrainslopesuitabilityratings
Thisdocumentisavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/Terrain_slope_suitability_ratings.xls
Appendix67 Fallowperiodrequirements
Thisdocumentisavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/Fallow_requirements.xls

155

Appendix68 Suitabilityofwatercollectingsites
Inwatercollecting sites substantially more water can be available to plants as compared toupland
situations. Watercollecting sites are difficult to locate in a global study but can be approximately
determined on the basis of prevalence of specific soil types. Fluvisols
7
and to a lesser extent
Gleysols
8
aretypicallyrepresentingtheflatterrainofalluvialvalleysandotherwatercollectingsites.
The moisture suitability ratings devised for unprotected Fluvisols and Gleysols without artificial
drainage are organized in ten groups of crops with comparable growth cycle lengths and similar
tolerances to high groundwater levels, waterlogging and flooding. The rating tables are presented
below:

Shorttermdrylandcrops(I)
This group includes some short duration crops (wheat, barley, rye, oat, dryland rice, foxtail millet,
chickpea, rape, and alfalfa) which are somewhat tolerant toexcess moisture. ForLGPs lessthan 30
daysitisassumedthereisontheaverageinsufficientwatertobringthesecropstomaturationand
yield,especiallysincethecontributionfromrainfallisalsoalmostnonexistent.AtLGPslongerthan
120daysthesecropswillgrowirrespectiveadditionalwater.IthasbeenassumedthattheFluvisols
aretoowetinLGPsover300days.Mostofthesecropsaremarginaltonotsuitableinhumidareas.
AgroclimaticconstraintsalonewillrendertheselongLGPsalreadymarginaltonotsuitable.

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS 33 33 33 33 33 33
S 33
MS 33 33 33 33 33 33 33
mS 33 33
NS 100 100 67 67 34 34 34 34 34 34 34 100 100 100 100

Shorttermdrylandcrops(II)
The crops in this group (sorghum, pearl millet, buckwheat, sweet sorghum, cowpea) have either a
shorter duration than Group I (pearl millet and cowpea) or tolerance to both drought as well as to
excesswater(sorghum).ThereforeforsomepartsoftheFluvisolsin129daysgrowingperiodssome
modestyieldmaybeexpected(thoughnotinallyears).AtthewetendoftheLGPsthesecropsare
treatedsimilarlytoGroupI.

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS 33 33 33 33 33 33 33
S 33
MS 33 33 33 33 33 33 33 33
mS 33 33
NS 100 67 67 34 34 34 34 34 34 34 34 100 100 100 100

7
Fluvisols are by definition flooded by rivers. Fluvisols are young soils where sedimentary structures are
clearlyrecognizableinthesoilprofile.
8
Gleysolsaregenerallynotfloodedbyrivers.However,thesoilprofilesindicateregularoccurrenceofhigh
groundwatertablesthroughreduction(gley)features.LowlyingGleysolsmaybeponded/waterloggedby
highgroundwaterandrainfallduringtherainyseason.
156

Shorttermdrylandcrops(III)
The crops in Group III include maize, phaseolus bean, soybean, gram,dry pea,pigeon pea, tobacco
andsunflower.Theyaremoresensitivetoexcesswater(especiallywaterlogging)thanGroupIandII
crops.Therefore,theyarenotconsideredtobesuitableinareaswhereLGPexceeds270days.Their
waterrequirementsaresimilarorsomewhathigherthanGroupIcrops.

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS 33 33 33 33
S 33
MS 33 33 33 33 33 33
mS 33 33
NS 100 100 67 67 34 34 34 34 34 67 100 100 100 100 100
Shorttermdrylandcrops(IV)
Rootcrops(whitepotato,sweetpotato,sugarbeet)areallsensitivetohighgroundwaterlevelsand
waterlogging. Cotton and groundnut, cabbage, flax, onion and tomato are also very sensitive to
excess moisture. These crops can only be grown on the rarely flooded parts of the Fluvisols,
provided they are well drained. Apart from groundnut the growth cycles of the crops in this group
are slightly longer than the crops in Group IIII. This makes crops in Group IV slightly more
vulnerable.

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS
S
MS 33 33 33 33
mS 33 33 33 33 33 33
NS 100 100 100 100 67 34 34 34 34 67 100 100 100 100 100
Wetlandrice(V)
Wetland Rice is difficult to grow under rainfed conditions. In particular the water management is
problematic. Yields obtained from purely rainfed paddy is generally low. 23 t/ha is already good.
Floodwatersupplycomesinthesemiaridareasinanerraticfashion;toolittletoolateortoomuch
too soon. In the subhumid and humid areas the flood hazard makes management difficult
(submerging and flood damage by flowing water). LGPs less than 150 days have been considered
insufficient to obtain yield. Very long LGPs are assumed to be associated with high flood risks
(submerging,flowingwater,highwaterlevelsduringmaturingandharvest).

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS
S 33 33 33
MS 33 33 33 33
mS 33 33
NS 100 100 100 100 100 67 67 67 67 67 67 67 67 67 100
Cassava,citrus,coffee,jatropha,yamandcocoyam(VI)
157

Cassava,citrus,coffee,jatropha,yamandcocoyamarepreferablynotgrownonFluvisolsbecauseof
itssensitivityforexcessivewetnessinthesoil.OnthehigherpartsofFluvisolsshortdurationcassava
canbe found(e.g.,LGPof180270daysinGhana).Sincecassava isnotreallybenefitingfromextra
moisture,thebestLGPsarethosewherealsorainfedcassavawoulddoreasonablywell.Towardsthe
wetter end of the LGPs (more than 240270 days) cassava is not anymore to be considered on
Fluvisols.

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS
S
MS 33 33
mS 33 33
NS 100 100 100 100 100 100 67 67 67 67 100 100 100 100 100

Sugarcane,miscanthusandswitchgrass(VII)
Sugarcane, miscanthus and switch grass are fairly tolerant to flooding and waterlogging (e.g., see
FAO, 1988). The water from rainfall and whatever comes from the Fluvisols must meet full crop
waterrequirementsfor8to9months.Itisassumedthatthecontributionthroughadditionalwater
fromFluvisols sufficiently extends the growing period starting from LGP 180 210 days onwards. At
harvest presence of excess moisture is less favorable for both yield and management of the crop.
There need be a predictable period during which the Fluvisol environment provides at least 2
monthsofdryerconditions.

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS
S 33 33
MS 33 33
mS 33
NS 100 100 100 100 100 100 100 67 67 67 67 67 100 100 100

Banana/plantain,oilpalm,cocoa,coconutandtea(VIII)
Banana/plantain, oil palm, cocoa, coconut and tea prefer humid conditions. Banana is somewhat
toleranttowaterlogging,oilpalmsomewhatless.Highgroundwatertablesarenottolerated.Both
perennials require at least eight months during which full water requirements are met. Fluvisols
occurringinLGPsofmorethan300daysareassumedtobeassociatedwithlongerperiodswithhigh
groundwaterlevelsandarethereforeunsuitedforoilpalmandbanana/plantain.

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS
S
MS 33 33 33
mS 33 33
NS 100 100 100 100 100 100 67 67 67 67 67 100 100 100 100

158

Olives(IX)
Olives tolerate neither high groundwater tables nor waterlogging and flooding or inundation.
Therefore,olivesarenotconsideredforcultivationonFluvisols.

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365+
VS
S
MS
mS
NS 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

Naturalpasturesandreedcanarygrass(X)
Naturalpasturesandreedcanarygrassarewelladaptedtowetconditions.Normallythespeciesmix
is finetuned to the environmental conditions. Artificial (sown) pastures might grow unevenly on
Fluvisols depending on both local differences of soil fertility and water supply. The total period of
wateravailabilityonFluvisolscanbeconsideredanadequatemeasureoftheproductivityregarding
pastures(ofcourse,periodsofwaterlogging,floodingandinundationaretobesubtracted).

Suitability
class
PercentageofwatercollectingsitessuitableperLGPclass
0 1
29
30
59
60
89
90
119
120
149
150
179
180
209
210
239
240
269
270
299
300
329
330
364
365 365
+
VS 33 33 33 67 67 67 33
S 33 33 33 33 33 33 33 33
MS 33 33 33 34 33 33
mS 33 33 34 33 33 33
NS 100 67 67 34 34 34 34 34 34 67

159

Appendix69 OutputsofModuleIV
The main purpose of Module IV is to provide for each crop/LUT a comprehensive soil suitability
evaluation for all the soil units contained in the Harmonized World Soil Database (HWSD). This is
done by the use of individual soil quality ratings (SQ). Seven different SQs are calculated and are
combinedinasoilunitsuitabilityrating(SR,%).TheSRrepresentsthepercentageofpotentialyield
expectedforagivencrop/LUTwithrespecttothesoilcharacteristicspresentinasoilmapunitofthe
HWSDandisdependingoninput/managementlevel.
ModuleIVproducesaseparateoutputfilewithsoilevaluationresultsforeachcrop/LUT.Asubsetof
information contained in these files is later on used in Module V when agroecological potential
yields are estimated, accounting for yield reductions due to constraining soil and terrainslope
conditions.Theoutputfileisasalargematrix(inplainASCII),withrowsorganizedbysoilmapunit
and individual component soil type and columns representing relevant soil unit characteristics
followedby the estimated values of different soil qualities and the computedsoil suitability ratings
foreachinput/managementlevelandterrainslopeclass(seeTable61).
TableA61ContentofoutputfilefromGAEZModuleIV
Name Description
Typeof
variable
Fieldwidth
MU_GLOBAL HWSDglobalmappingunitidentifier Integer 8
COV Coveragecodeindicatingsourceofsoilpolygonanddatabase Integer 4
SEQ Componentsequencenumberwithinmappingunit Integer 4
SOIL_SYM FAOsoilunitsymbol Character 9
SOIL_NUM FAOsoilunitnumericcode Integer 9
TOPTEX Topsoiltexture Integer 7
PH1 Soilphasecode1 Integer 4
PH2 Soilphasecode2 Integer 4
ROO Classcodeforobstaclestoroots Integer 4
IL Classcodeforimpermeablelayer Integer 3
SWR Classcodeforsoilwaterregime Integer 4
SHARE Percentageofmappingunit Integer 6
AWC Availablesoilwaterstoragecapacity Integer 5
AC Availablesoilwaterstoragecapacityclass Integer 3
DEPTH Referencesoildepth Integer 6
Q1 SoilqualityratingforSQ1;allinputlevel Integer 5
Q2I SoilqualityratingforSQ2;low/intermediateinputlevel Integer 5
Q2H SoilqualityratingforSQ2;highinputlevel Integer 6
Q3 SoilqualityratingforSQ3;allinputlevel Integer 6
Q4L1 SoilqualityratingforSQ4;lowinputlevel;slopeclass1 Integer 5
Q4L2 SoilqualityratingforSQ4;lowinputlevel;slopeclass2 Integer 5
Q4L3 SoilqualityratingforSQ4;lowinputlevel;slopeclass3 Integer 5
Q4L4 SoilqualityratingforSQ4;lowinputlevel;slopeclass4 Integer 5
Q4L5 SoilqualityratingforSQ4;lowinputlevel;slopeclass5 Integer 5
Q4L6 SoilqualityratingforSQ4;lowinputlevel;slopeclass6 Integer 5
Q4L7 SoilqualityratingforSQ4;lowinputlevel;slopeclass7 Integer 5
Q4L8 SoilqualityratingforSQ4;lowinputlevel;slopeclass8 Integer 5
Q4I1 SoilqualityratingforSQ4;intermediateinputlevel;slopeclass1 Integer 5
Q4I2 SoilqualityratingforSQ4;intermediateinputlevel;slopeclass2 Integer 5
Q4I3 SoilqualityratingforSQ4;intermediateinputlevel;slopeclass3 Integer 5
Q4I4 SoilqualityratingforSQ4;intermediateinputlevel;slopeclass4 Integer 5
Q4I5 SoilqualityratingforSQ4;intermediateinputlevel;slopeclass5 Integer 5
Q4I6 SoilqualityratingforSQ4;intermediateinputlevel;slopeclass6 Integer 5
Q4I7 SoilqualityratingforSQ4;intermediateinputlevel;slopeclass7 Integer 5
Q4I8 SoilqualityratingforSQ4;intermediateinputlevel;slopeclass8 Integer 5
Q4H1 SoilqualityratingforSQ4;lowinputlevel;slopeclass1 Integer 5
Q4H2 SoilqualityratingforSQ4;lowinputlevel;slopeclass2 Integer 5
Q4H3 SoilqualityratingforSQ4;lowinputlevel;slopeclass3 Integer 5
Q4H4 SoilqualityratingforSQ4;lowinputlevel;slopeclass4 Integer 5
Q4H5 SoilqualityratingforSQ4;lowinputlevel;slopeclass5 Integer 5
Q4H6 SoilqualityratingforSQ4;lowinputlevel;slopeclass6 Integer 5
160

Name Description
Typeof
variable
Fieldwidth
Q4H7 SoilqualityratingforSQ4;lowinputlevel;slopeclass7 Integer 5
Q4H8 SoilqualityratingforSQ4;lowinputlevel;slopeclass8 Integer 5
Q5 SoilqualityratingforSQ5;allinputlevel Integer 6
Q6 SoilqualityratingforSQ6;allinputlevel Integer 6
Q7L SoilqualityratingforSQ7;lowinputlevel Integer 6
Q7I SoilqualityratingforSQ7;intermediateinputlevel Integer 6
Q7H SoilqualityratingforSQ7;highinputlevel Integer 6
SiL1 Soilsuitabilityrating;lowinputlevel;slopeclass1(00.5%) Integer 5
SiL2 Soilsuitabilityrating;lowinputlevel;slopeclass2(0.52%) Integer 5
SiL3 Soilsuitabilityrating;lowinputlevel;slopeclass3(25%) Integer 5
SiL4 Soilsuitabilityrating;lowinputlevel;slopeclass4(58%) Integer 5
SiL5 Soilsuitabilityrating;lowinputlevel;slopeclass5(816%) Integer 5
SiL6 Soilsuitabilityrating;lowinputlevel;slopeclass6(1630%) Integer 5
SiL7 Soilsuitabilityrating;lowinputlevel;slopeclass7(3045%) Integer 5
SiL8 Soilsuitabilityrating;lowinputlevel;slopeclass8(>45%) Integer 5
SiI1 Soilsuitabilityrating;intermediateinputlevel;slopeclass1(00.5%) Integer 5
SiI2 Soilsuitabilityrating;intermediateinputlevel;slopeclass2(0.52%) Integer 5
SiI3 Soilsuitabilityrating;intermediateinputlevel;slopeclass3(25%) Integer 5
SiI4 Soilsuitabilityrating;intermediateinputlevel;slopeclass4(58%) Integer 5
SiI5 Soilsuitabilityrating;intermediateinputlevel;slopeclass5(816%) Integer 5
SiI6 Soilsuitabilityrating;intermediateinputlevel;slopeclass6(1630%) Integer 5
SiI7 Soilsuitabilityrating;intermediateinputlevel;slopeclass7(3045%) Integer 5
SiI8 Soilsuitabilityrating;intermediateinputlevel;slopeclass8(>45%) Integer 5
SiH1 Soilsuitabilityrating;highinputlevel;slopeclass1(00.5%) Integer 5
SiH2 Soilsuitabilityrating;highinputlevel;slopeclass2(0.52%) Integer 5
SiH3 Soilsuitabilityrating;highinputlevel;slopeclass3(25%) Integer 5
SiH4 Soilsuitabilityrating;highinputlevel;slopeclass4(58%) Integer 5
SiH5 Soilsuitabilityrating;highinputlevel;slopeclass5(816%) Integer 5
SiH6 Soilsuitabilityrating;highinputlevel;slopeclass6(1630%) Integer 5
SiH7 Soilsuitabilityrating;highinputlevel;slopeclass7(3045%) Integer 5
SiH8 Soilsuitabilityrating;highinputlevel;slopeclass8(>45%) Integer 5

161

Launch SQ analysis
LoadCropRequirementsFromXLS
LoadAWC_FromXLS
ReadRecord
Implement Phase Split
Count Phase Split
Compute AWC
Analyse SQ record
ComputeAC ComputeAWC
AssessParameter
AssessTexture
AssessPhase
AverageExcludingLowest
AssessRainfall
AssessYesNoParameter
KillList
GetDrainageClass
ComputeSQ
CheckForNonSoil
KillListMembers
mSoilE
Appendix610 SubroutinedescriptionsofModuleIV
Intermsofcomputerimplementation,thesoilevaluationtoolisdifferentfromtheotherModules.It
iswrittenusingBorlandDelphi7.2SE,sotheinterfaceobjectisthemainprocedure.Soilattributesof
differentsoilmapunitsareretrieveddirectlyfromtheHWSDattributedatabasestoredinMSAccess
format. Figure 61 shows the structure and relationships of the main procedures and functions of
ModuleIVcodedinPascal.

FigureA61DiagramofthesubroutinesandfunctionsofGAEZModuleIV
162

TableA62SubroutinesandfunctionsofGAEZModuleIV
FileName Procedure Description Calledfrom Callsto
EVALUATION.pas Analyse_SQ_record For given crop,
soil record and
terrain slope
classes,
computes soil
qualities SQ1 to
SQ7 and
compiles
respective soil
suitabilityratings
Launch_SQ_analysis Assess_Parameter,
Assess_Texture,
Assess_Phase,
Assess_Drainage,
Assess_YesNoParameter,
AverageExcludingLowest,
Compute_SQ,
CheckForNonSoil, minimum,
frm,frmV,KillList
ROUTINES.pas Assess_Drainage For given crop
and input level,
rate drainage
class as part of
SQ4assessment
Analyse_SQ_record Get_Drainage_Class
ROUTINES.pas CheckForNonSoil Detects nonsoil
units
Analyse_SQ_record
ROUTINES.pas ComputeAC Computes soil
water class
number
Implement_Phase_S
plit
ROUTINES.pas EvalAWC Adjusts soil AWC
according to soil
phases
Implement_Phase_S
plit

ROUTINES.pas Implement_Phase_Split Applies splitting

rules to soil
recordduetothe
presence of
certain soil
phases
Launch_SQ_analysis ComputeAC,EvalAWC
FUNCTIONS.pas KillList Releasesmemory
not anymore
needed for
holding lists of
soil evaluation
data
Analyse_SQ_record KillListMembers
FUNCTIONS.pas KillListMembers Releasesmemory
for a specific list
ofdata
KillList
EVALUATION.pas Launch_SQ_analysis Main function
used to carry out
for all crops the
evaluation of all
soil units
contained in the
HWSD.
SoilEv Analyse_SQ_record,
ComputeAWC,
Count_Phase_Split,
LoadAWCFromXLS,
LoadCropRequirementsFromX
LS, Implement_Phase_Split,
ReadRecord
READ_SOIL_REC.pas ReadRecord Retrieves a data
record in MS
Access format
fromtheHWSD
Launch_SQ_analysis
FileName Function Description Calledfrom Callsto
ROUTINES.pas Assess_Parameter Evaluates rating
function for
given crop and
soil attribute
value
Analyse_SQ_record
ROUTINES.pas Assess_Phase Applies soil
phase
adjustmenttoSQ
rating
Analyse_SQ_record
ROUTINES.pas Assess_Texture Applies soil
texturerating
Analyse_SQ_record
ROUTINES.pas Assess_YesNoParameter Tests for
presence of
special soil
properties
Analyse_SQ_record
163

FileName Procedure Description Calledfrom Callsto

FUNCTIONS.pas AverageExcludingLowest Computes the
average over its
arguments
excluding the
lowestvalue.
Analyse_SQ_record
ROUTINES.pas ComputeAWC Retrieves and
assigns soil
specific water
holding capacity
value
Launch_SQ_analysis
ROUTINES.pas Compute_SQ Combines results
of topsoil and
subsoil
evaluation into
aggregate SQ
rating
Analyse_SQ_record
ROUTINES.pas Count_Phase_Split Checks if soil
record must be
split due to
presence of soil
phases
Launch_SQ_analysis
FUNCTIONS.pas frm Formatsoutput Analyse_SQ_record
FUNCTIONS.pas frmV Formatsoutput Analyse_SQ_record
ROUTINES.pas Get_Drainage_Class Determines FAO
drainageclassfor
given soil,
texture, soil
phase and slope
class
Assess_Drainage
READ_PARAMS.pas LoadAWCFromXLS Retrieves certain
soil data from
spreadsheet in
MSExcelformat
Launch_SQ_analysis
READ_PARAMS.pas LoadCropRequirementsFromXLS Retrieves various
soil evaluation
parameters from
spreadsheet in
MSExcelformat
Launch_SQ_analysis
FUNCTIONS.pas minimum Calculates
minimum value
of up to eight
inputparameters
Analyse_SQ_record

164

Appendix71 OutputsofModuleV
Each run of Module V typically executed for combinations of selected crops/crop groups, water
source (rainfed or irrigated), input level, and time period or future climate change scenario
generatesabinaryrandomaccessfileholdingcomputedresults.Theseoutputfilesareorganizedby
gridcell. Pixels are numbered consecutively, starting from upper left corner of the global 5 arc
minutelatitude/longituderasterandcountingalongpixelsinrowsdowntothelowerrightcorner.A
record is stored for each land pixel, i.e. gridcells not included in the GAEZ land mask are ignored.
TheinformationstoredforeachpixelincludesareferencetothespecificLUTselected,adistribution
of the gridcell area in terms of crop suitability classes, potential attainable production for each
suitabilityclass,agroclimaticpotentialproduction(i.e.,excludingsoil/terrainconstraints)forextents
ineachsuitabilityclass,andcalculatedcultivationfactors(=1fallowrequirementfactor).Twosets
of distribution parameters are stored: one for soils in a gridcell subjected to rules for water
collecting sites, and one summing up results for all other soils in the gridcell. Results are stored in
randomaccessdatarecordsasdescribedinTable71.
TableA71InformationcontainedineachpixeldatarecordofModuleV
Variable Description
Typeof
variable
Lengthof
variable
(inbytes)
af1 CropindicatortoidentifyLUTandinputleveldefiningresultsstoredin
gridcellrecordforsoilsnotsubjecttorulesforwatercollectingsites.
Integer 2
af2 CropindicatortoidentifyLUTandinputleveldefiningresultsstoredin
gridcellrecordforsoilswhicharesubjecttorulesforwatercollecting
sites (Fluvisols and Gleysols on flat terrain under low or intermediate
inputlevel).
Integer 2
acut1 Shares of gridcell by suitability class (VS, S, MS, mS, vmS, NS)
calculatedforsoilsnotsubjecttorulesforwatercollectingsites.(Note:
shares over suitability classes and all soils for total gridcell add to
10000).
Real 4*6
acut2 Shares of gridcell by suitability class (VS, S, MS, mS, vmS, NS)
calculatedforsoilswhicharesubjecttorulesforwatercollectingsites.
(FluvisolsandGleysolsonflatterrainunderlowandintermediateinput
level).
Real 4*6
aqu1 Attainable production by suitability class (VS, S, MS, mS, vmS, NS)
calculatedforsoilsnotsubjecttorulesforwatercollectingsites.
Real 4*6
aqu2 Attainable production by suitability class (VS, S, MS, mS, vmS, NS)
calculatedforsoilswhicharesubjecttorulesforwatercollectingsites
(FluvisolsandGleysolsonflatterrainunderlowandintermediateinput
level).
Real 4*6
aqx1 Agroclimatic potential production (i.e. without considering soil and
terrainconstraints)byextentindifferentsuitabilityclasses(VS,S,MS,
mS, vmS, NS) calculated for soils not subject to rules for water
collectingsites.
Real 4*6
aqx2 Agroclimatic potential production (i.e. without considering soil and
terrainconstraints)byextentindifferentsuitabilityclasses(VS,S,MS,
mS, vmS, NS) calculated for soils which are subject to rules for water
collecting sites (Fluvisols and Gleysols on flat terrain under low and
intermediateinputlevel).
Real 4*6
acf1 Cultivation factor by different suitability classes (VS, S, MS, mS, vmS,
NS) calculated for soils not subject to rules for watercollecting sites.
The calculation of cultivation factors depends on crop, climate
characteristicsandinputlevel.
Real 4*6
acf2 Cultivationfactorbydifferentsuitabilityclasses(VS,S,MS,mS,vmS,
NS)calculatedforsoilswhicharesubjecttorulesforwatercollecting
sites.
Real 4*6
165

RDFLV
RDSLP
RDFRQ
GETIMG3
GETIMG1
GETIMG2
CO2FUN
ISFLVS
ISGLYS
FALLOW
UPDATE
YCLASS
MODIFR
ISWETL
RVLE RVLT
RVLE RVLT
P05
Appendix72SubroutinedescriptionsofModuleV

ThismainprogramofModuleVhasasimplestructureandusesonlyasmallnumberofsubroutines
and functions. Calculations are essentially organized in a fourfold nested loop over blocks of 30
arcsec rows and columns being aggregated to 5 arcmin results. Within each grid cell, calculations
step throughrespective combinations ofrelevant soiltypes andslope classes.Results are stored in
randomaccessdatarecordsasdescribedinTable71.Relationshipsamongroutinesaresummarized
in Table 71. Relationships among routines are summarized in Table 72 and Table 73. Figure 71
providesasimplediagramofthesubroutinesandfunctionsinGAEZModuleV.

FigureA71DiagramofthesubroutinesandfunctionsofGAEZModuleV

166

TableA72SubroutinesandfunctionsofModuleV
Filename
Subroutines/
functions
Description
Called
from
Callsto
P05.F BESTCROP Determinebestcomponentyieldforarangesuitabilityclasses UPDATE
P05.F CO2FUN CalculateapplicableCO2fertilizationyieldincreasefactor P05.F
FALLOW.F FALLOW Calculate cultivation factor according to crop, input level,
temperatureandLGP
P05.F
P05.F GETIMG1 Read1bytethematicraster P05.F
P05.F GETIMG2 Read2bytethematicraster P05.F
P05.F GETIMG3 Read4bytethematicraster P05.F
P05.F ISFLVS ReturntrueforFluvisoils,elsefalse P05.F
P05.F ISGLYS ReturntrueforGleysols,elsefalse P05.F
FALLOW.F ISWETL Returntrueforwetlandrice,elsefalse FALLOW
RULE_G.F MODIFR Shifts extents and production among suitability classes
according to suitability rule (e.g. rules for water collecting
sites,sloperules,permafrostzones,etc.)
P05.F RULE,RULT
P05.F RDFLV Readsuitabilityrulesforwatercollectingsites(PAR.FLV) P05.F
P05.F RDFRQ Readfactorsforlowinputcondition(PAR.FRQ) P05.F
P05.F RDSLP Readsloperules(PAR.SLP) P05.F
RULE_G.F RULE Apply2waysuitabilityrule MODIFR
RULE_G.F RULE Apply3waysuitabilityrule MODIFR
P05.F UPDATE Update grid cell results of suitable land and potential
production
P05.F YIELD,
BESTCROP
P05.F YCLASS Determinesuitabilityclassforgivenyield P05.F
P05.F YIELD Calculateyieldfromproductionandharvestedarea UPDATE

TableA73FORTRANsourcefilesforModuleVandincludedheaderfiles,subroutinesandfunctions
Fortran
file
Associated
headingfiles
Subroutines Functions
FALLOW.F aezdef.h FALLOW,ISWETL
P05.F aezdef.h BESTCROP, GETIMG1, GETIMG2, GETIMG4,
RDFLV,RDFLQ,RDSLP,UPDATE
CO2FUN,ISFLVS,ISGLYS,YCLASS,YIELD
RULE_G.F MODIFR,RULE,RULT

167

Appendix73Cropsummarytabledescription

Crop summary tables provide standardized information ondistributions of crop suitability and crop
yield data. The data is based on aggregations of subgrid cells distributions and it provides data by
predefined land cover and protection classes. Crop summary tables provide detailed data by
predefinedlandcoverandprotectionclassesofcropareayieldandproductionpotentials.Thetables
are further organized by crop (49), water supply type (5), input level (4) and time period i.e.,
historical (19612000 individual years), baseline (19611990) and futureclimates (2020s, 2050s and
2080s).ThesummarytablesareavailableundertheSuitabilityandPotentialYieldthemeintheGAEZ
v3.0 data Portal. An example table for high input rainfed maize with detailed column heading
explanationsisavailablefordownloadat:
http://www.iiasa.ac.at/Research/LUC/GAEZv3.0/docs/Crop_summ_table_description.xlsx

168

Appendix81 Estimationofsharesofcultivatedlandbygridcell
The estimation of shares of rainfed cultivated land by 5 arcmin grid cell presents an approach to
formallyandconsistentlyintegrateuptodategeographicaldatasetsobtainedfromremotesensing
with statistical information compiled by FAO and/or national statistical bureaus, as a basis for
spatially detailed downscaling of agricultural production statistics to land units (grid cells) and
subsequentyieldgapanalysis,aswellasvariousenvironmentalassessmentsrequiringspatialdetail.
Theprocedureinvolvesasequenceofsteps,asfollows:
Collectionofnational(andpossiblysubnational)statisticsoncultivatedland;
Integrationofavailablehighresolutiongloballandcoverdatasets;
Aggregationofgeographicallandcoverdatasetstoobtaindistributionsoflandcoverclasses
fornationalandsubnationaladministrativeunits;
Crosssectional regressions of statistical cultivated land against land cover distributions
derived from geographical land cover data sets to obtain reference weights for each land
coverclassintermsofcultivatedlandcontained;
Estimation of urban/builtup land shares based on an empirical relationship of per capita
landrequirementsasafunctionofpopulationdensity,andapplicationtoaspatiallydetailed
populationdensitydatasetat30arcsec.Aggregationofresultsto5arcmingridcells;
Applicationofaniterativeprocedurefortheadjustmentoflandcoverclassweights,starting
fromestimatedreferencevalues,toachieveconsistencyofgeographicalandstatisticaldata,
i.e., such that weighted summation of land cover classes of an allocation unit (country or
subnational administrative unit) results in the total cultivated land as reported in the
statisticaldata.
The iterative algorithm for adjusting land cover weights is controlled by a parameter file specifying
three levels of increasingly wider intervals within which the respective class weights are adjusted.
The ranges of permissible class weights for each land cover category were defined by (i) where
possible,quantitativeinformationcontainedintheGLC2000legendclassdescription,and(ii)expert
judgmentontheplausibilityofthepresenceofcultivatedlandinalandcoverclass.
Thealgorithmnotonlyproducesformallyconsistentresultsforeachallocationunitbutalsoprovides
anindicationofthediscrepancybetweenmappedlandcoverdistributionsandstatisticalamountsof
cultivatedland.

169

Appendix82 Estimationofareayieldandproductionofcrops
The estimation of global processes consistent with local data and, conversely, local implications
emergingfromlongtermglobaltendencieschallengethetraditionalstatisticalestimationmethods.
These methods are based on the ability to obtain observations from unknown true probability
distributions.Infact,thejustificationofthesemethods,e.g.,theirconsistencyandefficiency,relyon
asymptotic analysis requiring an infinite number of observations. For the new estimation problems
referredtoabove,whichcanalsobetermedasdownscalingproblems,weoftenhaveonlylimited
or incomplete samples of real observations describing the phenomena and variables of interest.
Additionalexperimentstoachievemoreobservationsmaybeexpensive,timeconsuming,orsimply
impossible.
A main motivation for developing sequential downscaling methods initially was the spatial
estimation of agricultural production values. Agricultural production and land data are routinely
available at national scale from FAO and other sources, but these data give no indication as to the
spatialheterogeneityofagriculturalproductionwithincountryboundaries.Adownscalingmethod
in this case achieves plausible allocation of aggregate national land and production statistics to
individual spatial units, say pixels, by using all available evidence from observed or inferred geo
spatial information, such as remotely sensed land cover, soil, climate and vegetation distribution,
populationdensityanddistribution,transportationinfrastructure,etc.
ThedownscalingalgorithmappliedinGAEZv3.0proceedsiteratively.Itstartswithconstructingor
retrieving an initial prior allocation to individual crops based on the available geographical and
statistical information. Each iteration step then determines the discrepancy between statistical
totalsavailableatthelevelofspatialunits(countriesorsubnationalunits)andtherespectivetotals
calculated by summing harvested areas and production over gridcells. The magnitude of these
deviations is then used to revise the land and crop allocation and to recalculate discrepancies. The
processiscontinueduntilallaccountingconstraintsaremet(Fischeretal.,2006).
Inthefollowingwelisttheinputdatarequiredatthelevelofspatialunits(countriesorsubnational
administrativeunits),thegeographicallayersusedat5arcminspatialresolution,andtheequations
andaccountingconstraintsimposed.
Inputdatausedatadministrativeunitlevel:
Totalcultivatedland(annualandpermanentcrops) (TC) FAOSTAT
Totalcultivatedlandequippedwithirrigation (TC
I
) FAOSTAT
Harvestedarea,bycrops (TH
j
) FAOSTAT
Production,bycrops (TQ
j
) FAOSTAT
Producerprice,bycrops (P
j
) FAOSTAT
Shareofirrigatedharvestedareaintotalcropjharvestedarea (
j
) AQUASTAT
Shareofirrigatedproductionintotalcropjproduction (
j
) FAO
GISdata(5min):
Administrativeboundariesandcodes (adm) FAO
Gridcellareaextent (TA) IIASA
Gridcellshareofcultivatedland (c
T
) IIASA
Gridcellshareofcultivatedlandequippedwithirrigation (c
I
) AQUASTAT
Cultivationintensityclassfactor,rainfedcultivationofannualcrops (
R
m ) IIASAAEZ
Cultivationintensityclassfactor,irrigatedcultivationofannualcrops (
I
m ) IIASAAEZ
Farmingsystemzone (z) FAO
Potentialcropyield,rainfed,highinputlevel,bycrops ( ) GAEZv3.0
Potentialcropyield,rainfed,lowinputlevel,bycrops ( ) GAEZv3.0
, R high
ij
Y
, R low
ij
Y
170

Potentialcropyield,irrigated,highinputlevel,bycrops ( ) GAEZv3.0
Distancetomarket (d) FAO/IIASA
Populationdensity (pd) FAO
Ruminantlivestockdensity (rum) FAO
Locationcroppriorityfactorforrainfedcrops ( ) FAO/IIASA
Locationcroppriorityfactorforirrigatedcrops ( ) FAO/IIASA
Cropdistributionlayers,selectedcrops
9
(
j
) Monfredaetal.

Mainequationsandconstraints:
Totalirrigatedproductionofallocationunit,bycrops

j
I
j
I
j
TQ TQ = jcrops
Totalrainfedproductionofallocationunit,bycrops

j
I
j
R
j
TQ TQ ) 1 ( = jcrops
Totalirrigatedharvestedareaofallocationunit,bycrops

j
I
j
I
j
TH TH = jcrops
Totalrainfedharvestedareaofallocationunit,bycrops

j
I
j
R
j
TH TH ) 1 ( = jcrops
Gridcellcultivatedland

i
T
i i
TA c TC = igridcells
Gridcellirrigatedcultivatedland

i
I
i
I
i
TA c TC = igridcells
Gridcellshareofrainfedcultivatedland

I
i
T
i
R
i
c c c = igridcells
Gridcellrainfedcultivatedland

i
R
i
R
i
TA c TC = igridcells
Gridcellrainfedcroppingintensityapplicableforannualcrops
10

R
i
R R
i
m m = igridcells
Gridcellirrigatedcroppingintensityapplicableforannualcrops

I
i
I I
i
m m = igridcells
Gridcelltotalrainfedharvestedarea

R
i
R
i
R
i
TC m H = igridcells
Gridcelltotalirrigatedharvestedarea

9
Inthecurrentdownscalingapplicationforyear2000,informationfromthestudybyMonfredaetal.(2008)wasusedfor
selectedcropsincountrieswheremorethan50%wascoveredbysubnationalstatistics.
10
Note, this cropping intensity factor accounts for sequential multicropping of land within a year as well as for idle
cultivatedlandduetofallowrequirements.
, I high
ij
Y
R
jz

I
jz

171

I
i
I
i
I
i
TC m H = igridcells
Gridcellrainfedharvestedarea,bycrops
11

=
R
i
R
ij
P
R
i
R
ij
R
i R
ij
TC s m
TC s m
AH
crops perennial j
crops annual j

igridcells
Gridcellirrigatedharvestedarea,bycrops

=
I
i
I
ij
P
I
i
I
ij
I
i I
ij
TC s m
TC s m
AH
crops perennial j
crops annual j

igridcells
Totalrainfedharvestedareaofallocationunit,bycrops

=
cells grid i
R
ij
R
j
AH TH jcrops
Totalirrigatedharvestedareaofallocationunit,bycrops

=
cells grid i
I
ij
I
j
AH TH jcrops
Gridcellrainfedyield,bycrops
) ) 1 ((
, , high R
ij
R
ij
low R
ij
R
ij
R
j
R
ij
Y Y Y + = jcrops,igridcells
The spatial layer of location factors
ij
is used to reflect differences in farm management
intensityandinputuse.Observationstoportrayrelativespatialinputintensitiesmaybeobtained
from remote sensing products or be based on georeferenced household survey data providing,
for instance, information on farm size, input use and market orientation of households.
Alternatively, factors such as population density, type of suitable crops, and distance to market
canbeusedtodifferentiateamonglandunits.
Gridcellirrigatedyield,bycrops

high I
ij
I
j
I
ij
Y Y
,
= jcrops,igridcells
Totalrainfedproductionofallocationunit,bycrops

=
cells grid i
R
ij
R
ij
R
j
Y AH TQ jcrops
Totalirrigatedproductionofallocationunit,bycrops

=
cells grid i
I
ij
I
ij
I
j
Y AH TQ jcrops
Gridcellrelativeyieldfactor,byrainfedcrops
) ( max /
, , high R
kj
cells grid k
high R
ij
R
ij
Y Y

= jcrops,igridcells
Gridcellrelativeyieldfactor,byirrigatedcrops
) ( max /
, , high I
kj
cells grid k
high I
ij
I
ij
Y Y

= jcrops,igridcells

11
Thecroppingintensityofperennialcropsinbothrainfedandirrigatedcultivatedlandiskeptconstantata
valueof0.95.
172

Gridcellcropshareallocation:
Allocation of land to cropping at grid cell level is computed in a 2stage nested way. First, land is
allocatedtotwobroadsetsofcrops,describedbyindexset
1
I (cropsforwhichaspatialdistribution
layerwithshares
ij
isavailable)andindexset
2
I (cropsforwhichaspatiallayerislacking).
Theshareoftotalrainfedcultivatedlandallocationtocropsinindexset
1
I

=
R R
R
I I j
R
ij
R
j j
R
ij
R
i
I j
R
ij
R
j j
R
ij
R
i
R
i
P Y m
P Y m
S
2 1
1
1


igridcells
whereindexset
R
I
1
ofrelevantrainfedcropsin
1
I isdefinedas
} 0 {
1 1
R
j
R
ij ij
R
I j I > =
andindexset
R
I
2
ofrelevantrainfedcropsin
2
I isdefinedas
} {
2 2
R
j
R
ij
R
I j I =
Similarly,theshareoftotalirrigatedcultivatedlandallocationtocropsinindexset
1
I is

=
I I
I
I I j
I
ij
I
j j
I
ij
I
i
I j
I
ij
I
j j
I
ij
I
i
I
i
P Y m
P Y m
S
2 1
1
1


igridcells
withindexset
I
I
1
ofrelevantirrigatedcropsin
1
I definedas
} 0 {
1 1
I
j
I
ij ij
I
I j I > =
andindexset
I
I
2
ofrelevantirrigatedcropsin
2
I definedas
} {
2 2
I
j
I
ij
I
I j I =
Sharesoftotalcultivatedlandallocatedtocropswithinindexset
2
I arethencomputed
respectivelyforrainfedandirrigatedconditionsas

R
i
R
i
S S
1 2
1 = and
I
i
I
i
S S
1 2
1 =

igridcells
Inasecondstep,thecroplevelareashares
R
ij
s and
I
ij
s forrespectivelyrainfedandirrigation
conditionsarecalculatedforthetwosetsofcrops:
173

R
R
I k
R
ik
R
k k
R
ik
R
ik
R
ij
R
j j
R
ij
R
ij R
i
I k
R
k
R
ik
R
j
R
ij R
i
R
ij
P Y m
P Y m
S
S
s
2
1
2
1
0
0

R
R
R
R
I j I j
I j I j
I j I j
I j I j
2 2
2 2
1 1
1 1




igridcells
andforirrigatedland

I
I
I k
I
ik
I
k k
I
ik
I
ik
I
ij
I
j j
I
ij
I
ij I
i
I k
I
k
I
ik
I
j
I
ij I
i
I
ij
P Y m
P Y m
S
S
s
2
1
2
1
0
0

I
I
I
I
I j I j
I j I j
I j I j
I j I j
2 2
2 2
1 1
1 1




igridcells
Withcultivatedlandallocatedaccordingtothesecomputedlandshares,thecropspecific
harvestedareasingridcellicanbewrittenas:

R
ij
crops k
R
ij
crops k
R
ij
R
ij
R
i
R
i
R
ij
s
s
m s
TA c AH

jcrops,igridcells
and

I
ij
crops k
I
ij
crops k
I
ij
I
ij
I
i
I
i
I
ij
s
s
m s
TA c AH

jcrops,igridcells

174

Solutionalgorithm:
After initialization of all variables, the solution algorithm of the iterative rebalancing method
updates the various multipliers
R
j
and
I
j
for area,
R
and
I
for cropping intensity, and
R
j
and
I
j
foryieldandproductionsuchthatallconditionsandaccountingconstraintsaremet.Asaresultit
produces a gridcell specific allocation of crop harvested area and production for rainfed and
irrigatedcultivatedland(i.e.thephysicalland).Intheprocess,respectivecroppingintensityfactors
R
i
m and
I
i
m are estimated. The multipliers
R
and
I
provide a measure of actual cropping
intensity compared to potential multicropping. The multipliers
R
j
and
I
j
represent the ratios of
actualachievedtoapplicablepotentialcropyields, i.e.an indicationofyieldgapsfortheestimated
croppingpatternandhistoricalobservedproduction.

175

Appendix9 Globalterrainslopeandaspectdatadocumentation
TheNASAShuttleRadarTopographicMission(SRTM)hasprovideddigitalelevationdata(DEMs)for
over80%oftheglobe.TheSRTMdataispubliclyavailableas3arcsecond(approximately90meters
resolutionattheequator)DEMs(CGIARCSI,2006).
Forlatitudesover60degreesnorthelevationdatafromGTOPO30(USGS,2002)witharesolutionof
30arcseconds(dependingonlatitudethisisapproximatelya1by1kmcellsize)wereused.
Datacreationdateandversion
Creationdate:December2006(Version1.0)
ProcessingSteps
Under an agreement with the National Aeronautics and Space Administration (NASA) and the
Department ofDefense's NationalGeospatialIntelligenceAgency(NGA),theU.S.Geological Survey
(USGS) is now distributing elevation data from the Shuttle Radar Topography Mission (SRTM). The
SRTMisajointprojectbetweenNASAandNGAtomaptheEarthslandsurfaceinthreedimensions
at a level of detail unprecedented for such a large area. Flown aboard the NASA Space Shuttle
EndeavourFebruary1122,2000,theSRTMsuccessfullycollecteddatafromover80percentofthe
Earthslandsurface,formostoftheareabetween60oN.and56oS.latitude.
The data currently being distributed by NASA/USGS (finished product) contains nodata holes
where water or heavy shadow prevented the quantification of elevation. These are generally small
holes, which nevertheless render the data less useful, especially in fields of hydrological modelling.
Dr. Andrew Jarvis of the CIAT Land Use project, in collaboration with Dr. Robert Hijmans and Dr.
Andy Nelson, have further processed the original DEMs to fill in these nodata voids. This involved
the production of vector contours, and the reinterpolation of these derived contours back into a
rasterDEM.TheseinterpolatedDEMvalueswerethenusedtofillintheoriginalnodataholeswithin
theSRTMdata.
TheDEMfileshavebeenmosaicedintoaseamlessglobalcoverage,andareavailablefordownload
as5x5tiles,ingeographiccoordinatesystemWGS84datum.Theavailabledatacoverarasterof
24rowsby72columnsof5x5latitude/longitudetiles,fromnorth60degreelatitudeto56degree
south.
TheseprocessedSRTMdata,witharesolutionof3arcsecond(approximately90mattheequator),
i.e.6000rowsby6000columnsforeach5x5tile,havebeenusedforcalculating:(i)terrainslope1
gradients for each 3 arcsec grid cell; (ii) aspect of terrain slopes for each 3 arcsec grid cell; (iii)
terrain slope class by 3 arcsec grid cell; and (iv) aspect class of terrain slope by 3 arcsec grid cell.
Products (iii) and (iv) were then aggregated to provide distributions of slope gradient and slope
aspectclassesby30arcsecgridcellandfora5x5latitude/longitudegridusedinglobalAEZ.
Thecomputeralgorithmusedtocalculateslopegradientandslopeaspectoperatesonsubgridsof3
by3gridcells,saygridcellsAtoI:
A B C
D E F
G H I

SRTMdataarestoredin5x5tiles
12
.WhenEfallsonaborderroworcolumn(i.e.,rowsorcolumns1
or6000ofatile)therequiredvaluesfallingoutsidethecurrenttilearefilledinfromtheneighboring
tiles.

12
For the globe the computer program processes 36 million subgrids, in total 32.4 billion subgrids are
considered.
176

TocalculateterrainslopeforgridcellE,thealgorithmproceedsasfollows:
1)IfthealtitudevalueatEisnodatathenbothslopegradientandslopeaspectaresettonodata.
2)ReplaceanynodatavaluesinAtoDandFtoIbythealtitudevalueatE.
LetPx,PyandPzdenoterespectivelycoordinatesofgridpointPinxdirection(i.e.longitudeinour
case), y direction (i.e. latitude in our application), and z in vertical direction (i.e., altitude), then
calculatepartialderivatives(dz/dx)and(dz/dy)from:
(dz/dx)=((AzCz)+2(DzFz)+(GzIz))/(8size_x)
(dz/dy)=((AzGz)+2(BzHz)+(CzIz))/(8size_y)
When working with a grid in latitude and longitude, then size_y is constant for all grid cells.
However,size_xdependsonlatitudeandiscalculatedseparatelyforeachrowofatile.
Theslopegradient(indegrees)atEis:
slgE=arctan(Jz Jx )
2
+ (Jz Jy )
2

andinpercentisgivenby
slpE=100(Jz Jx )
2
+(Jz Jy )
2

Theslopeaspect,i.e.theorientationoftheslopegradient,startingfromnorth(0degrees)andgoing
clockwise,iscalculatedusingthevariablesfromabove,asfollows:
aspE=arctan(Jz Jx ) + (Jz Jy )
Theaboveexpressioncanbeevaluatedfor(dz/dy)0.OtherwiseaspE=45(for(dz/dx)<0)oraspE
=270(for(dz/dx)>0)
3) To produce distributions of slope gradients and aspects for grids at 30 arcsec or 5 min
latitude/longitude,slopegradientsaregroupsinto9classes:
C1:0%slope0.5%
C2:0.5%slope2%
C3:2%slope5%
C4:5%slope10%
C5:10%slope15%
C6:15%slope30%
C7:30%slope45%
C8:Slope>45%
C9:Slopegradientundefined(i.e.,outsidelandmask)
Slopeaspectsareclassifiedin5classes:
N:0<aspect45or315<aspect360
E:45<aspect135
S:135<aspect225
W:225<aspect315
U:Slope aspectundefined;thisvalue isused forgridswhereslopegradientisundefined or
slopegradientislessthan2%.

Detaileddatadescription
DataFormat:
ThedataareprovidedasASCIIfilesinagridformat.Theyconsistofheaderinformationcontaininga
setofkeywords,followedbycellvaluesinrowmajororder.Thefileformatis:

177

NCOLSxxx
NROWSxxxXLLCENTERxxx|xllcornerxxx>
YLLCENTERxxx|yllcornerxxx>
CELLSIZExxx
NODATA_VALUExxx
row1
row2

rown
wherexxxisanumber.Row1ofthedataisatthetopofthegrid,row2isjustunderrow1andso
on. The end of each row of data from the grid is terminated with a carriage return in the file. The
gridisdefinedintheheaderinformationwiththefollowingkeywords:
NCOLS:numberofcolumns
NROWS:numberofrows
XLLCENTER:xcoordinateoflowerleftcentre
YLLCENTER:ycoordinateoflowerleftcentre
CELLSIZE:gridcellsize
NODATA_VALUE:Thevalueassignedtonodatainformation

Geographicaldetails
Spatialcoverage: Global
Gridcellsize: 5minutesand30arcseconds
Projection: Geographiccoordinatesystem(Longitude,latitude)
Units:Decimaldegrees,
Datum:WGS84

Datacontent
Thedatacompriseoneelevationmapdescribingmedianelevationineachgridcell,eightslopeand
four aspect maps describing percentage distributions of the respective slope or aspect classes. The
sumofallclassesforslopesandaspectsrespectivelyis100percentages.
Units:
Elevationdata: meters
Slopeandaspectclasses: percentage*1000
Landmask:
In addition a land mask has been provided. The land mask shows the number of 3 arc second grid
cellsintheSRTMdatathatfallintoa5minutesor30arcsecondgridcell.Alongcoastlines5minutes
or 30 arcsecond grid cells usually only contain a fraction of the higher resolution 3 arc second grid
cells, which were used for the slope and aspect calculations. In the 5 minutes and 30 arc second
gridstheslopesandaspectdistributionsalwayssumupto100percent.Thusiftherealpercentage
distributionofaparticular5minutesor30arcsecondisrequireditcanbecalculatedusingtheland
mask.

178

TableA111DescriptionoffilenamesoftheIIASALUCGlobalTerrainSlopesandAspectDatabase
FILENAMES Description
gridcellsize:
5x5minutes
gridcellsize:
30arcseconds

LANDMASK
GloLand_5min GloLand_30as Numberof3arcsecondgridcellsthatbelongtothelandmaskand
fallintorespective5minutesor30arcsecondgridcells
ELEVATION
GloElev_5min GloElev_30as Medianelevation(meters)
SLOPES Slopeclass
GloSlopesCl1_5min GloSlopesCl1_30as 0%slope0.5%
GloSlopesCl2_5min GloSlopesCl2_30as 0.5%slope2%
GloSlopesCl3_5min GloSlopesCl3_30as 2%slope5%
GloSlopesCl4_5min GloSlopesCl4_30as 5%slope10%
GloSlopesCl5_5min GloSlopesCl5_30as 10%slope15%
GloSlopesCl6_5min GloSlopesCl6_30as 15%slope30%
GloSlopesCl7_5min GloSlopesCl7_30as 30%slope45%
GloSlopesCl8_5min GloSlopesCl8_30as Slope>45%
ASPECT Aspectclass
GloAspectClN_5min GloAspectClN_30as North:0<aspect45or315<aspect360
GloAspectClE_5min GloAspectClE_30as East:45<aspect135
GloAspectClS_5min GloAspectClS_30as South:135<aspect225
GloAspectClW_5min GloAspectClW_30as West:225<aspect315
GloAspectClU_5min GloAspectClU_30as Undefined: Slope aspect undefined; this value is used for grids
whereslopegradientisundefinedorslopegradientislessthan2%.


179

GlobalAgroEcologicalZones(GAEZv3.0)
ModelDocumentation

Food and Agriculture

Organization of the
United Nations
I I A S A
International Institute
for Applied
Systems Analysis
GAEZ
Global Agro-ecological Zones