黃淮海平原農(nóng)田土壤水氮行為模擬與管理分析.pdf

1、Y773489分類號：密級：單位代碼：10019學號：B01127哼目差≮夭等’學位論文黃淮海平原農(nóng)田土壤水氮行為模擬與管理分析SimulationofSoilWaterandNitrogenBehaviorandAnalysisofAgriculturalManagementinHuangHuai—HaiPlain研究生：高翅蠢指導教師：睦縫掛熱援奎握國熬援蓋亙值茲援申請學位門類級別：盔堂擅專業(yè)名稱：攮堂研冤方向：地型用皇信息拉盔所在

2、學院：童遂生玨縫堂院2005年6月AbstractSpatialvariationcharacteristicsofsoilparticlecomposition(sPc)，bulkdensity(BD)’soilnutrientsandCECintheHuangHuaiHaiPlain(HHHP)werestudiedbycouplingGeographicInformationSystem(GIS)withgeostafisties

3、inthispaperAconjunctivesimulationmodelincoporatingsoilwaterheat，nitrogenbehaviorsandcropgrowthwasestablishedandvalidatedusingfieldexperimentdataandthencoupledwithGIStoestimatesoilwaterandnitrogendynamicsatregionalscaleTh

4、eregionaldistributionofwaterandnitrogenaseefficiency(NUE)，nitrogenlossundersummermaizewinterwheatrotationsystemwerestudiedunderthemeteorologicalandruralsocio—economicalconditionofHHHP，andtheoptimalwaterandnitrogentreatme

5、ntsweredeterminedbasedonthesimulationresultsThemainresultswere船follows：SPC，BDofeachsoillayersandsoilnutrients，CECof020cmsoillayerwerevariedspatiallywithregionalfeaturesTheirtrendeffectsandanisotropywereobviousTheresultsi

6、ndicatedthattheanalysisoftrendandanisotropywasnecessaryandsuitabletobeinvolvedinthestudyonsemivariogramofsoilpropertiesatregionalscaleStochasticvariabilityforsoilpropertiesisbeyond05Spatialdistributionofsoilpropertiesdis

7、tributedass仃ipeorlumppatternsPedoTransferFunctions(PTFs)modelsusingBackPropagation∞P)neuralnetworkwereestablishedtopredictsoilhydrauliccharacteristicparametersThepredictedvaluesagreedwiththeobservedonesfairlywell，andtheB

8、PneuralnetworkmodelhadratherhighpredictionprecisionTheTtestresultshowedthatwithinputofsoiltexture，itcallsatisfythesimulationrequirementsofsoilhydrauliccharacteristicparametersatregionalscaleAcombinedsoilwaterheat—nitroge

9、nandcropgrowthsimulationmodelwasestablishedandvalidatedbasedonthefielddataIt’sfoundthatthesimulatedvaluesonsoilwaterandnitrogencontentfitwellwiththeobservedonesCouplingwithGISbyusingVBAenvironmentembeddedwithinArcGIS81an

10、ddevelopingtherelevantsharingdatabase，thecombinedmodelcouldbeusedforwaterheatnitrogensimulationatregionallevelTheGIS—basedmodelwasusedforregionalsimulationonsummermaizewinterwheatrotationsystemin19992000Themultivariatest

11、epwiseregressionresultsshowthatinsummermaize‘growingSeasonnitrogenleachingwassignificantlypositivecorrelationwithnitrogenfertilizeramount(NFA)，saturatedhydraulicconductivity(曲)，Activeaccumulatedtemperature(ACT)，rainfalla

12、ndirrigationamount(RIA)，wateruseefficiency(WUE)wassignificantlypositivecorrelatedwithNFAandsignificantlynegativecorrelatedwithACT，nitrogenaseefficiency(NUE)WassignificantlypositivecorrelatedwithRIAandsignificantlynegativ

13、erelativewithNFAandsoilorganicmatter(SOM)Inwinterwheatgrowingseason，NitrogenleachingwassignificantlypositivecorrelatedwithNFA，KsandRIA，WUEwassignificantlypositivecorrelatedwithNFAandsunshinehours(SH)andsignificantlynegat