Soil moisture disaggregation via coupling geographically weighted regression and radiative transfer model

•36 km SMAP brightness temperature (TB) are disaggregated via GWR model.•Disaggregated TB are converted into 1 km soil moisture based on RTM model.•1 km SM is validated and compared with in-situ measurements and other SM products. Passive microwave soil moisture (SM) products hold tremendous potential for both scientific research and practical applications. However, due to the limitations imposed by brightness temperature (TB), passive microwave SM suffer from lower spatial resolution, which significantly restricts their application in regional scale like crop growth monitoring, water resource management, runoff forecasting, and others. To address this challenge, this study proposed a feasible methodology to disaggregate 36 km Soil Moisture Active Passive (SMAP) SM in the ShanDian River Basin via coupling geographically weighted regression (GWR) and radiative transfer model (RTM). Firstly, we employed GWR to disaggregate 36 km SMAP TB to 1 km via MODIS LST, NDVI and DEM data. Subsequently, in accordance with the microwave radiative transfer model, the disaggregated TB was converted to 1 km SM using a single-channel algorithm (SCA) in V polarization. The disaggregated SM, derived from the disaggregated TB, was validated against in-situ SM from the SMN-SDR network and compared with various SMAP SM products at different spatial resolutions. The results show the disaggregated SM effectively inherits the performance of SMAP official products, while offering enhanced spatial resolution and enabling more accurate applications. This study effectively enhances the performance of coarse-resolution passive microwave brightness temperature data at fine scales.