Improving Surface Soil Moisture Estimates in Humid Regions by an Enhanced Remote Sensing Technique
Peilin Song, Yongqiang Zhang, Jing Tian
Abstract
Abstract Using optical remote sensing data to downscale microwave soil moisture is currently a crucial technique for producing fine‐resolution soil moisture data set at large scales. Conventional soil moisture downscaling models are mainly developed and evaluated in arid or semi‐arid regions, but have serious limitation in humid regions. To improve soil moisture estimates under a wide range of climate regimes, this study developed a nonlinear enhanced soil moisture downscaling model and evaluated it at a typical microwave resolution over the contiguous China. The enhanced model clearly outperforms the conventional approach in estimating surface soil moisture by reducing the Root Mean Square Error (RMSE) from 0.175 vol/vol to 0.075 vol/vol under wet climate. Finally, soil moisture downscaling is conducted by applying such models (built at the 25‐km scale) at a finer (1‐km) resolution. Validation result of downscaled soil moisture against in situ data set demonstrates the advantage of the enhanced model.