The Joint Polarization Experiment: Polarimetric Rainfall Measurements and Hydrometeor Classification

As part of the evolution and future enhancement of the Next Generation Weather Radars (NEXRAD), the National Severe Storms Laboratory recently upgraded the KOUN Weather Surveillance Radar-1988 Doppler (WSR-88D) to include a polarimetric capability. The proof of concept was tested in central Oklahoma during a 1-yr demonstration project referred to as the Joint Polarization Experiment (JPOLE). This paper presents an overview of polarimetric algorithms for rainfall estimation and hydrometeor classification and their performance during JPOLE. The quality of rainfall measurements is validated on a large dataset from the Oklahoma Mesonet and Agricultural Research Service Micronet rain gauge networks. The comparison demonstrates that polarimetric rainfall estimates are often dramatically superior to those provided by conventional rainfall algorithms. Using a synthetic R(Z, KDP, ZDR) polarimetric rainfall relation, rms errors are reduced by a factor of 1.7 for point measurements and 3.7 for areal estimates [when compared to results from a conventional R(Z) relation]. Radar data quality improvement, hail identification, rain/snow discrimination, and polarimetric tornado detection are also illustrated for selected events.