A synergic method of Sentinel-1 and Sentinel-2 images for retrieving soil moisture content in agricultural regions

Abstract

Remote sensing techniques, which provide timely and accurate soil moisture information to improve food production, are becoming increasingly popular among agricultural production in China and many other regions. It also shows its importance in monitoring agricultural basins where the meteorological and topographic conditions vary greatly. In this case, the reliability of soil moisture content (SMC) monitoring for theses basins could be promised. The main objective of this paper is to develop a synergic method of radar and optical data, as well as an on-site sampling scheme, to accurately estimate SMC in the basin. The reason is that in the winter wheat production area of Dawen River basin in China, there are few conventional SMC testing equipment. Considering the characteristics of different crops and the conditions of crops in different growth periods, the VV polarization mode of Sentinel-1 C-band Synthetic Aperture Radar (SAR) images was used to retrieve SMC. Combined with the Normalized difference vegetation index (NDVI) extracted from Sentinel-2 MSI optical images and the optimized Water Cloud Model (WCM), the surface backscattering coefficient was obtained by eliminating vegetation moisture content (VMC), and the SMC model was established. The results showed that from March to May in 2020, the coefficient of determination (R2) between the estimated VMC and the field observation was 0.655, and the Root Mean Square Error (RMSE) between the two was 0.334. In the inversion algorithm model, the R2 between the retrieved SMC and the field measurement was 0.506, and the RMSE was 3.725.