With few exceptions, the majority of the dual-polarization radar-gage comparisons were made for warm season precipitation and at distances less than 100 km from the radar. The National Weather Service requires estimating rainfall at ranges up to 230 km from the radar. As part of the evolution and future enhancement of the WSR-88D, the National Severe Storms Laboratory recently upgraded the KOUN WSR-88D radar to include polarimetric capability. In this paper, we assess the quality of rainfall estimation at far distances from this radar. The study incorporates gage information from a regular network of 108 Oklahoma Mesonet gages located in central Oklahoma. Dataset includes 12 rain events with 35 hours of observations during the period from August to December 2002.

One-hour rain accumulations were estimated using a conventional method based on the radar reflectivity factor Z, a polarimetric algorithm based solely on specific differential phase KDP, and a new polarimetric algorithm utilizing Z, KDP, and differential reflectivity ZDR. The new polarimetric method combines Z and ZDR for light rain, KDP and ZDR for moderate rain, and relies exclusively on KDP for heavy rain.

At distances up to 140-150 km from the radar, where both conventional and polarimetric measurements of rain are not affected by the bright band, the new algorithm demonstrates superior performance compared to the conventional algorithm and the R(KDP) algorithm. The RMS error of the one-hour total estimate is reduced by a factor of 1.5 for point estimates when compared to the conventional one, and is reduced by a factor of 1.25 for point estimates when compared to the R(KDP) estimate. In the range interval between 150-200 km, where bright band contamination was maximized for the observed rain events, the R(KDP) algorithm outperforms the others. The RMS error of the one-hour total estimate is reduced by a factor of 1.3 for point estimates when compared to the conventional algorithm, and is reduced by a factor of 1.12 compared to the synthetic polarimetric algorithm. All algorithms demonstrated progressively degraded performance at distances beyond 200 km from the radar.

The results indicate the ability to improve rainfall estimates at distant ranges from the radar using a combination of polarimetric techniques. Further, the results suggest that rainfall estimates can be improved up to 230 km from the radar, even in the presence of bright band contamination.