An intercomparison of polarimetrically-derived and VPR-corrected rainfall with conventional estimates

Estimating surface rainfall from volume scan radar data continues to be an important and challenging research problem in radar meteorology. Many studies have indicated that rainfall estimates can be improved by incorporating vertical profile of reflectivity (VPR) corrections. Another significant improvement in radar rainfall estimation has come from the emergence and subsequent application of dual-polarized scanning radar to operational use, utilizing polarimetrically-derived rainfall estimates. While both solutions are different in application, each promises (in theory) to yield better estimates of precipitation over the entire radar sampling area than traditional reflectivity-only radar rainfall estimation techniques. It is the goal of this study to evaluate and characterize the performance of each of these methods by comparing to gauge-bias corrected rainfall amounts derived from standard reflectivity measurements. Volume scan radar data collected over 27 days in July 2002 from the NASA S-Band polarimetric Doppler Radar (NPOL) in Southwestern Florida provides an excellent dataset for this study.