A Warm Season Radar QPE Algorithm Using Adaptive Z-R Relationships

Quantitative precipitation estimation (QPE) from radar is subject to errors including uncertainties in Z-R relationships, contaminations from non-precipitation targets, partial beam filling due to blockages, beam spreading, overshooting, etc. Among them inaccurate Z-R relationship is one of the most significant errors. An accurate Z-R relationship should correctly represent spatial and temporal variations of the microphysical processes and associated drop size distributions in the observed precipitation system. However, most of existing operational radar QPEs use single Z-R relationship for the whole radar coverage area. This paper presents an algorithm that identifies different microphysical processes in warm season precipitation systems using three-dimensional radar reflectivity structure and environmental thermal field. Precipitation echoes are segregated into convective, stratiform, hail, and warm rain regimes and corresponding Z-R relationships are applied. The scheme was tested for events from several regions in the United States and was shown to provide improvements over single Z-R approaches. The scheme is fully automated and can be potentially used for operations.