Study On Multi-source Remote Sensing Collaborative Inversion Of Farmland Soil Moisture

Soil moisture measurement plays an important role in drought prediction,weather simulation,crop yield estimation and water resources management.Although the traditional soil moisture measurement methods can provide accurate results,they need huge manpower and financial resources,and can not obtain a large range of soil moisture information in a short time.With the development of remote sensing technology,Synthetic Aperture Radar(SAR)with its advantages of all day and all-weather work,stands out in many soil moisture estimation methods.However,due to the complexity of the actual situation of the surface,the interaction between electromagnetic waves and ground objects is more complex.Radar waves are not only affected by soil moisture,but also affected by roughness,vegetation coverage,radar system parameters and other factors,which brings challenges to soil moisture estimation.Based on Sentinel-1 SAR microwave data and Sentinel-2 Multispectral Imager(MSI)optical data,this paper takes the winter wheat farmland in Xiangfu District of Kaifeng City,Henan Province as the study area,aiming at the problem of reducing the sensitivity of radar signal caused by crop coverage and surface roughness in soil moisture retrieval,and carries out the research on farmland soil moisture inversion by remote sensing.The main contents and innovative work of this paper can be summarized as follows:(1)A new vegetation index called as Fusion Vegetation Index(FVI)is defined.Combined with the modified Water Cloud model by Mc Laren series,a semi empirical model of soil moisture inversion based on active and passive remote sensing data is developed to reduce the influence of agricultural crop cover on radar backscattering coefficient in the process of soil moisture inversion.The experimental results show that when FVI and VV/VH polarization data are used in combination,the semi-empirical model developed in this paper has the best accuracy of soil moisture inversion,and can effectively reduce the influence of crop mulch on backscattering coefficient.(2)The optimization theory is used to determine the value of vegetation related empirical parameters in the Water Cloud model,so as to obtain the optimal parameter value combination suitable for the study area.Based on the sensitivity analysis of the parameters in the Water Cloud model,the empirical parameters which have little influence on the radar backscatter coefficient but great influence on the vegetation attenuation factor are set as fixed values,and the optimization theory is used to determine other empirical parameters.Different optimal parameter combinations are obtained for different research areas.(3)Based on the active and passive remote sensing data,a Look-Up Table is constructed,combined with the Water Cloud model and the Oh model to perform farmland soil moisture inversion to reduce the influence of surface roughness on the radar backscattering coefficient during the soil moisture inversion process.The experimental results show that the improved water cloud model by introducing vegetation coverage is used to modify the backscattering coefficient,and then combined with the Oh model to construct the Look-Up Table to inversion the soil moisture can improve the accuracy of the inversion,and effectively reduce the influence of surface roughness on the radar backscattering coefficient in the process of soil moisture inversion.