A New Operational Northern Hemisphere Snow Water Equivalent Retrieval Algorithm for FY-3F/MWRI-II

Snow water equivalent (SWE) quantitatively describes water storage in snowpack. Satellite-based passive microwave (PMW) remote sensing is a valid tool for monitoring SWE in the Northern Hemisphere. However, the current operational SWE retrieval methods, especially those without assimilating near real-time station snow depth, still utilize globally-constant coefficients to build regression-based retrieval algorithms. In the context of the successful launch of the FY-3F satellites in 2023, we are attempting to work out a better Northern Hemisphere algorithm for the Micro-Wave Radiation Imager-II (FY-3F/MWRI-II), using a set of pixel-sensitive dynamic coefficients regressed based on a spatiotemporally continuous reference SWE dataset. We first utilize a method that couples random forest with the HUT snow emission model to calculate a high-accuracy SWE reference dataset. Then the linear-regression equations are used to fit the reference SWE data and satellite brightness temperature observations at each pixel to build the new operational FY-3F algorithm. Finally, the proposed FY-3F algorithm is validated extensively using four spatially independent datasets. The proposed FY-3F algorithm will improve the global monitoring capabilities for snow cover and enhance a complete and timely understanding of changes in SWE.