Retrieving Soil Moisture Based On MODIS Data

Soil moisture is the control factor in crop growth ,the important indictor for monitoring soil degradation .It has devastating impacts on agriculture ,water resources ,economy and environment .It plays an important role on the water and energy exchange at the land surface/atmosphere interface .So it is significant to do research on how to monitor soil moisture by remote sensing ,which can provide information on large area quickly and easily .MODIS data have high spectral resolution and we can get them on free of charge ,hence it has practical value to extract land surface information using MODIS data .After analyzing all kinds of information about soil moisture and comparing different quantitative remote sensing models ,including the principium of the models ,the using scope of the models and the advantage/disadvantage of the models ,then we think it is reasonable to choose thermal inertia for retrieving soil moisture where the land is bare or has low vegetation cover .However ,when the vegetation cover is high ,there are many models can be used . Up to now ,it is a moot to choose which one to retrieve soil moisture .On most conditions ,we just choose the models according the real situation .This paper focuses on the soil moisture distribution of Jilin province in April .And in Jilin province ,April is the important time for planting .It has significant referencing value for agriculture production to monitor the soil moisture .In this way the relevant departments can forecast the happenstance of drought timely and make a decision to overcome it .This paper use thermal inertia model to retrieve the soil moisture and retrieve the parameters which are needed in this model based on the band 1,2,3,4,5,7,19,31,32 of MODIS data .The parameters include NDVI , the ratio of vegetation cover , the land surface ratio radiance ,the atmospheric transmittance , land surface temperature ,albedo .The land surface temperature retrieving use 2-factor model of split-window algorithm with band 31,32 MODIS data .Then construct the linear relationship between P_ATI and observing data .Finally ,we get the soil moisture distribution of western region of Jilin province with the linear model .