简介：Theobjectivesofthisstudyaretoassesslandsuitabilityandtopredictthespatialandtemporalchangesinlandusetypes(LUTs)byusingGIS-basedlandusemanagementdecisionsupportsystem.AGISdatabasewithdataonclimate,topography,soilcharacteristic,irrigationcondition,fertilizerapplication,andspecialsocioeconomicactivitieshasbeendevelopedandusedfortheevaluationoflandproductivityfordifferentcropsbyintegratingwithacropgrowthmodel--theerosionproductivityimpactcalculator(EPIC).Internationalfoodpolicysimulationmodel(IFPSIM)isalsoembeddedintoGISforthepredictionsofhowcropdemandsandcropmarketpriceswillchangeunderalternativepolicyscenarios.Aninferenceengine（IE)includinglandusechoicemodelisdevelopedtoillustratelandusechoicebehaviorbasedonlogitmodels,whichallowstoanalyzehowdiversifiedfactorsrangingfromclimatechanges,croppricechangestolandmanagementchangescanaffectthedistributionofagriculturallanduse.Atestforintegratedsimulationistakenineach0.1°by0.1°gridcelltopredictthechangeofagriculturallandusetypesatgloballevel.Globallandusechangesaresimulatedfrom1992to2050.