Simultaneous retrieval of soil moisture and salinity in arid and semiarid regions using Sentinel-1 data and a revised dielectric model for salty soil

Abstract

Soil moisture and salinity (SMS) are two critical factors in crop growth, and monitoring their dynamic has important scientific value and social benefits for preventing land degradation and improving land productivity. However, the current methodology treats soil moisture and salinity as two independent variables to be estimated separately, completely ignoring their joint influence on the microwave signal. In this paper, the Jingdian irrigated region, which is located in northwestern China, is selected as an example, the contents of soil volumetric water and soil salt are measured separately for different seasons in the research area, and they are also retrieved simultaneously by combining Sentinel-1 data and a revised dielectric model of salty soil. The results demonstrate that the Sentinel-1 data can achieve satisfactory results in the simultaneous retrieval of SMS, with R² values higher than 0.53. The RMSE values in the upward track are less than 0.042 m³/m³ and 3.132 mS/cm, respectively, which are smaller than in the downward track, with the RMSE values less than 0.051 m³/m³ and 3.84 mS/cm, respectively. The average value of soil moisture content in winter is 0.17 m³/m³, which is higher than in spring, with a value of 0.21 m³/m³. The soil salt content increases gradually over the study period, with average values of 1.88 mS/cm in spring and 3.58 mS/cm in winter, respectively. In addition, vegetation, surface roughness, precipitation, and agricultural activities are the main factors affecting the simultaneous retrieval of SMS.