Utilizing the Vertical Variability of Precipitation to Improve Radar QPE

Radar-based measurements of rainfall are vital in understanding the distribution of precipitation over large watersheds. However, radar measurements at this scale are rarely obtained very close to the ground and hence the vertical distribution of precipitation remains an outstanding problem for radar-rainfall mapping. Characteristics of the melting layer and raindrop size distribution can be exploited to further improve radar quantitative precipitation estimation. Using dual-polarimetric radar and disdrometers, we found that the characteristic size of raindrops reaching the ground in stratiform precipitation often varies linearly with the depth of the melting layer. As a result, a radar rainfall estimator was formulated using Dm that can be used by polarimetric as well as dual-frequency radars (e.g., space-based radars such as the GPM Dual-Frequency Precipitation Radar), to lower the bias and uncertainty of conventional single radar parameter rainfall estimates by as much as 20%. Polarimetric radar also suffers from issues associated with sampling the vertical distribution of precipitation. Hence, we characterized the vertical profile of polarimetric parameters (VP3)—a radar manifestation of the evolving size and shape of hydrometeors as they fall to the ground—on dual-polarimetric rainfall estimation. The VP3 revealed that the profile of ZDR in stratiform rainfall can bias dual-polarimetric rainfall estimators by as much as 50%, even after correction for the vertical profile of reflectivity (VPR). The VP3 correction technique that we developed can improve operational dual-polarimetric rainfall estimates by 13% beyond that offered by a VPR correction alone.