基于生化参数优化的典型农作物叶面积指数反演研究
II
北华航天工业学院硕士学位论文
Abstract
TheLeafAreaIndex(LAI)isacrucialindicatordescribingthevegetationgrowth
condition.Itisnotonlyusedtoassessprocessessuchasphotosynthesis,evaporation,and
transpiration,aswellasestimatethenetproductivityofterrestrialecosystems,butalsoserves
asaparameterinputformodelsrelatedtowaterbalance,globalcarboncycling,andmore.
LAIcanbeacquiredthroughgroundmeasurementsandremotesensing.Whileground
measurementsprovideauthenticvaluesforremotesensingvalidation,theyaregenerally
time-consumingandlabor-intensive,makingthemimpracticalforlarge-scalemonitoring.To
overcomethislimitationandobtainextensiveLAIdata,researchershavedevelopedvarious
remotesensinginversionalgorithms,includingempiricalmodels,physicalmodels,and
integratedmodels.Amongthem,thePROSAILmodelhasawidespreadapplication
foundationinthefieldsofspectralremotesensingandvegetationparameterestimation,
representingaclassicalexampleofaphysically-basedmodelforLAIinversion.
Toenhancetheaccuracyofthephysically-basedmodelforLAIinversion,thisstudy
focusedonwinterwheatandsummermaizeattheYuchengcomprehensiveexperimentalsite
inShandong.UtilizingGF1remotesensingimagery,weconstructedanLAIinversionmodel
basedonbiochemicalparameteroptimizationusingthePROSAILmodel.Theresearch
methodologyandfindingsareoutlinedasfollows:
(1)ThisstudyconductedasensitivityanalysisofinputparametersinthePROSAIL
model,analyzingtherelationshipsbetweenmultipleinputparameters(leafchlorophyll
content,leafdrymattercontent,structuralcoefficient,carotenoidcontent,equivalentwater
thickness)andLeafAreaIndex(LAI)duringthreegrowthstagesofwinterwheat(jointing
stage,headingstage,g