Extended Abstract (616K)P2.11
A multi-site evaluation of the range correction and convective-stratiform separation algorithms for improving WSR-88D rainfall estimates
Feng Ding, RS Information System and NOAA/NWS, Silver Spring, MD; and D. Kitzmiller, D. J. Seo, D. Riley, C. Dietz, C. Pham, and D. Miller
The Range Correction Algorithm (RCA), which has been developed by the Office of Hydrologic Development (OHD), National Weather Service (NWS), NOAA, functions to reduce range-dependent biases in radar precipitation estimates due to nonuniform vertical profiles of reflectivity (VPR). It improves estimates of stratiform precipitation by approximating the mean VPR over the entire radar umbrella, then using the VPR to calculate and apply range-dependent multiplicative correction factors to radar reflectivity and/or Z R precipitation estimates. RCA reduces bright band contamination at near-to-mid ranges from the radar, and enhances rainfall estimates at farther ranges, where the radar beam typically intercepts ice particles above the melting layer. To deal with situations of embedded convection, a companion algorithm, the Convective Stratiform Separation Algorithm (CSSA), has been implemented in conjunction with RCA. If CSSA classifies precipitation as convective, no adjustment is made in that portion of the radar umbrella.
The OHD and several NWS field offices conducted a field evaluation of the RCA and CSSA during the period March-June 2004, using real-time data from 6 WSR-88D sites: KRTX (Portland OR), KEAX (Pleasant Hill MO), KMPX (Minneapolis MN), KTLX (Twin Lakes OK), KPBZ (Pittsburgh PA), and KRLX (Charleston WV). Real-time data from KLWX (Sterling VA) were also collected for analysis in OHD. The purpose of the field evaluation was to obtain feedback from forecasters and hydrologists on the utility of these corrections to basic radar estimates, and to obtain a geographically diverse set of precipitation estimates for objective verification. During the field evaluation, forecasters at the field offices were given access to a web page featuring rainfall accumulations with and without range correction. Their evaluations of the performance of RCA and CSSA over discrete precipitation events were collected through standard questionnaires.
During the experiment, RCA performed as intended in a clear majority of cases. Verification statistics based on rain gauge observations showed that the rainfall estimates with RCA/CSSA were significantly improved over the original uncorrected estimates, and that applying both range correction and mean field bias correction resulted in more improvement than could be realized than by applying either correction alone.
Poster Session 2, CURRENT AND FUTURE PRECIPITATION MEASUREMENTS POSTERS
Tuesday, 11 January 2005, 9:45 AM-9:45 AM
Previous paper Next paper