Combining Polarimetric Radar Data and HRRR Model Output to Determine Melting Layer Coverage and Surface-Based Precipitation Types in Winter Storms: New Algorithms for the WSR-88D

The classification of cold-season precipitation type at the surface is complicated by the broad range of precipitation types that might result from processes that occur below the height of the radar’s lowest elevation sweep. For example, a shallow layer of subfreezing air near the surface might lead to either a complete refreezing of drops (ice pellets) or refreezing upon contact with the surface (freezing rain). Both of these precipitation types are difficult to determine using radar data alone, and may not be observed at all at distances > 50 km from the radar. Because of this, the fuzzy-logic-based Hydrometeor Classification Algorithm (HCA), which gives classifications on conical surfaces, often provides results in transitional winter weather events that are not at all representative of the precipitation type observed at ground level.

In this paper, we describe a new surface-based Hydrometeor Classification Algorithm (sHCA) that combines polarimetric radar data with thermodynamic output from the High Resolution Rapid Refresh (HRRR) model. The algorithm allows fuzzy-logic-based classifications from the lowest elevation sweep to be projected to the surface as snow or ice crystals for cold season events where the entire atmospheric column above a location has T < -5ºC and as rain, big drops, or hail for warm season events where the surface temperature at a location has a T > 5ºC. For intermediate conditions typical of transitional winter weather events, the algorithm uses output from a 1-D spectral bin model, which is initialized at all grid points with vertical profiles of wet bulb temperature profiles obtained from the High Resolution Rapid Refresh (HRRR) forecast model, to provide a background precipitation classification type. Polarimetric radar observations, when available, are then used to either confirm or reject the background classification. In short, the introduction of thermodynamic output from the HRRR provides an opportunity to not only enhance classification in regions where radar data are available, but also to extend classification capabilities to more distant ranges where low-level radar data are not available.

Since the accurate detection of the presence/absence of an elevated warm layer is a fundamental requirement of the development of a sHCA for transitional winter events, the paper also highlights a new “hybrid” Melting Layer Detection Algorithm (HMLDA) that blends radar-based ML detections at all elevation angles (for high SNR regions that are typically close to the radar) with model-based estimates (for low SNR regions that are typically more distant from the radar) to provide a comprehensive ML detection coverage map. This differs from existing ML detection techniques that use only high elevation angle scans to detect ML coverage near the radar and then assumes horizontal homogeneity and projects that coverage to more distant ranges.