Assimilation of satellite and in-situ data for soil moisture retrieval: evaluation of AIEM over grasslands

This research explores the application of the Advanced Integral Equation Model (AIEM) for retrieving surface soil moisture in grassland ecosystems. The AIEM was utilized to analyze VV-polarized backscatter coefficients obtained from Sentinel-1, incorporating Sentinel-2-derived NDVI to address the challenge of separating soil and vegetation scattering contributions in grasslands. Calibration of the model was performed using field measurements from the Texas Soil Observation Network (TxSON), spanning from 2019 to 2021. The soil moisture estimates derived from the model were rigorously validated against ground-truth data from TxSON for the year 2023. To evaluate the accuracy and reliability of the retrieved soil moisture data, statistical analyses, including trend analysis, Sen’s slope estimation, and the Mann-Kendall test, were conducted. The results demonstrated that the model successfully captured soil moisture dynamics throughout the entire vegetation period. Although the modeled soil moisture values were marginally higher than the observed values, the overall trend between the modeled and observed data was consistently aligned. This study underscores the potential of AIEM, in conjunction with multi-sensor satellite data, for accurate soil moisture retrieval in grassland ecosystems. The findings provide valuable insights into soil-vegetation-atmosphere interactions and have significant implications for rangeland management, precision agriculture, and hydrological modeling in grassland environments.