Discrimination between winter precipitation types using radar observations and spectral bin microphysics

In this presentation, a novel approach for discriminating the surface precipitation type, the Spectral Bin Classifier (SBC) is presented. The SBC diagnoses six categories of precipitation: rain (RA), snow (SN), a RA/SN mix, freezing rain (FZRA), ice pellets (PL), and a FZRA/PL mix. This algorithm combines input from radars and numerical models. It works by calculating the mass fraction of water for a spectrum of hydrometeors for a given temperature (T) and relative humidity (RH) at each model level from the cloud top to the surface. Radar observations of reflectivity and riming are used to determine the riming factor and drop-size distribution. Demonstrations of the SBC output for individual T and RH profiles shows that it provides reasonable estimates of the mass water fraction of various-sized hydrometeors for the different categories of precipitation. Consideration of the plan views of precipitation type for select events shows the SBC faithfully represents the horizontal distribution of precipitation type in as much as the model analysis is accurate. When applied to a collection of observed soundings associated with surface observations of RA, SN, FZRA, and PL, the classifier has probabilities of detection (PODs) that range from 61.6 to 98.3\%. As one might expect, the PODs decrease when the effects of model uncertainty are accounted for. This decrease is quite modest for RA, SN, and PL, but is rather large for FZRA due to the fact that this form of precipitation is quite sensitive to small changes in the thermal profile. This points to a need for probabilistic analyses and forecasts of precipitation type. A demonstration of how this algorithm can be used toward that end is included.