A decade of GCOM-W1 AMSR2 rainfall record at NOAA

Rain rate estimates from the GCOM-W1 Advanced Scanning Microwave Radiometer 2 (AMSR2) are operationally produced by NOAA to meet the requirements of the Joint Polar Satellite System (JPSS) program. Following the evaluation of the most recent passive microwave precipitation algorithm from NASAs Precipitation Measurement Mission (PMM), it has been determined that the inclusion of environmental information (such as surface cover type, total precipitable water, and near-surface temperature) to the Bayesian retrieval scheme over land greatly improves precipitation detection seen in the currently operational retrieval version (GPROF2010v3). Based on these analyses, a new PMM GPROF2017 package has been adapted, tested, and set for transition to the NOAA operations. The new algorithm version fully replaces the existing retrieval relying exclusively on NOAA-unique products and employing a fully parametric Bayesian scheme over all surface types. To achieve this, a new frontend and preprocessing modules are developed to complement a unified algorithm core script. To evaluate performance and estimate regional and seasonal biases, instantaneous and long-term validation of rainfall rate is performed against the MRMS and GPM DPR products. Key performance metrics (including but not limited to probability of detection, false alarm rate, root mean squared difference, and correlation), indicating stability of the new retrieval version and significant improvements relative to the current AMSR2 product, will be presented.