Study On Soil Moisture Retrieval Method Based On ASAR Data And TM Images

Soil moisture is one of the most important parameters in land surface processes,which is of great significance to the global water cycle,climate change monitoring and energy balance.Not only that,in the production of human life,the soil moisture is not only the basic conditions for crop growth and development,but also affects the conversion of energy and consumption,and plays an important role in surface runoff process.Therefore,to establish a set of effective,accurate and real-time dynamic monitoring system of soil moisture has important scientific significance for agricultural production,drought and ecological environment evaluation.In this paper,a method is developed to study the soil moisture content in the low vegetation coverage area,such as agricultural land.Based on the soil scattering model and vegetation scattering model,simulation of vegetation using cloud model of backscattering and scattering attenuation of vegetation on two-way soil,and the leaf area index is simulated by passive optical remote sensing data as the input parameters of the vegetation canopy water cloud model,so the model can be applied to large scale vegetation backscattering simulation.Secondly,the Oh model is used to simulate the backscattering of soil,and a new method of active and passive remote sensing for soil moisture retrieval is established.Because the water cloud model original soil backscattering simulation for simple linear equation,replace the original cloud model in soil by using Oh model of backscattering simulation,considering the soil roughness on the back scattering effect.Therefore,the model is more accurate in describing the backscattering of soil,so as to improve the simulation precision of the model.At last,this paper selects the oasis irrigation area of Zhangye oasis in the middle reaches of Heihe River Basin as the research area in the middle reaches of Gansu River Basinon the basis of active microwave remote sensing Envisat ASAR data,passive optical remote sensing Landsat-5 TM data and field observation data in the study area,the feasibility and accuracy of the model are compared and verified.The results show that the tested by field observation data of the study area using simulation results of the model,the improved cloud model simulation of soil moisture values obtained result has higher accuracy compared with the traditional water cloud model simulation.At the same time,we also found that the improved model in soil on soil moisture is more humid conditions compared to the simulation results under dry conditions of lower accuracy,but compared to the traditional cloud model has improved.Therefore,further research work should be carried out in the area where the vegetation cover is relatively dense and the soil moisture changes greatly.