Rain-to-Snow Transition: Evaluation of Precipitation Phase Algorithms in Southern New England

Abstract

As part of NASA’s Global Precipitation Measurement (GPM) ground validation program, a multi-year winter field study in Storrs, Connecticut brought a unique opportunity to study the long-lasting problem of precipitation phase determination within retrieval algorithms. This study evaluates the phase algorithms of NASA’s Integrated Multi-satellitE Retrievals for GPM (IMERG) dataset, the flagship GPM global precipitation product, and NOAA’s Multi-Radar Multi-Sensor (MRMS) dataset, which is frequently used to validate IMERG precipitation rates over the continental United States. The study selected ten long-lasting phase transition events where rain and snow were recorded for more than one hour with the presence of mixed precipitation and focused on two diverse events. Using the PARSIVEL disdrometer phase as a reference, phase transition occurs at around wet-bulb temperature of 0.5˚C, which agrees with prior studies. IMERG and MRMS phase algorithms performed better when using High Resolution Rapid Refresh (HRRR) and European Center for Medium-Range Weather Forecast Reanalysis (ERA5) temperatures than Modern-Era Retrospective Analysis for Research and Applications reanalysis (MERRA2) temperatures. Overall, IMERG Version 07 performed better than its predecessor Version 06. This study also evaluated HRRR deterministic and probabilistic phase algorithms, which performed better than HRRR temperature-based MRMS and IMERG algorithms, respectively.