Collaborative soil moisture inversion with multi-source remote sensing data

Abstract. Soil moisture is a key variable in the global water cycle, carbon balance and energy conversion, and is crucial for hydrological control, meteorological forecasting and crop growth. Pengyang County in Ningxia is a typical region with fragile ecology. In this paper, we utilize Sentinel 1 SAR data and Landsat 8 optical imagery to synergistically invert soil moisture in Pengyang County by combining the advantages of optical and microwave remote sensing. The study calculates the vegetation water content through the VWC model, and uses the water cloud model to eliminate the influence of vegetation on the radar signal to obtain the soil backscattering coefficient with the removal of the influence of vegetation. Finally, the BP neural network model was utilized to invert the soil moisture in Pengyang County. The results show that the VH-polarized SAR data are more sensitive to the vegetation structure and moisture content, which is more suitable for soil moisture inversion in this region, and the NDMI has the highest sensitivity to the vegetation moisture content, which contributes more to the soil moisture estimation. The inversion results of the BP neural network model have a high correlation with the measured values, which indicates that the method can effectively invert the soil moisture in Pangyang County. The results of the study can provide a reference for soil moisture monitoring in the region, and provide a basis for decision-making in ecological protection, water conservation and comprehensive regional management.