Evaluation and Blending of AMSR2 and ATMS Snow Water Equivalent Retrievals over the Conterminous United States

Satellite-sensed snowpack has seen increasing use in the prediction of water supply and flooding in the US. In this study, we explore the complementary skills of two passive microwave snow water equivalent (SWE) retrievals, namely that provided by the Advanced Microwave Scanning Radiometer 2 (AMSR2) onboard the Global Change Observation Mission (GCOM) and the Advanced Technology Microwave Sounder (ATMS) onboard the Joint Polar Satellite System (JPSS). We further experiment with blending these products with in situ observations. In our experiments, the AMSR2 and ATMS SWE retrievals are first screened using snow cover product from the National Ice Center's Interactive Multisensor Snow and Ice Mapping System (IMS); the masked SWE data are then validated and corrected for bias using in situ observations from the Natural Resources Conservation Service (NRCS) Snow Telemetry (SNOTEL) and the National Weather Service (NWS) Cooperative Observer Program (COOP) networks. Then, the bias-corrected SWE are blended with in situ observations using optimal interpolation. The cross-validation results demonstrate that bias correction is effective in reducing systematic bias of the two products. The blended SWE product is shown to improve upon each individual product as shown by lower bias and higher correlation with in situ observations.