How good are D0 estimates using polarimetric variables?

One of the major applications of polarimetric radar is improvement in rainfall estimation by estimating raindrop size distributions. However, these are difficult to verify at the radar resolution volume. In this study, we use profiler DSD estimates, which have reasonably well understood error characteristics, to evaluate the accuracies of the polarimetric estimations. Data from forty-five rain events from both monsoon and continental regimes near Darwin, Australia, have been classified into convective and stratiform rain. Preliminary results show that during stratiform rain, the polarimetric estimate of median raindrop diameter is larger than the profiler estimate. This indicates that actual raindrops are more oblate than that assumed in the polarimetric estimates.

In this study, the C-band polarimetric scanning radar (C-Pol) operating near Darwin, Australia, is used to estimate the median raindrop diameter over two vertically-pointing profiling radars located 24 km away. Several different polarimetric relationships are used to estimate the median raindrop diameter. The two profiling radars estimate the vertical air motion and the raindrop size distribution from about 1.7 km to just below the melting layer at 4 km. The uncertainties of the polarimetric and profiler DSD estimates and the statistical comparison between the two instruments for the forty-five rain events including dependencies on the height, regime, and rain type will be presented at the conference.