Revisiting Polarimetric Signatures of Hail and Rainfall Estimation in the Presence of Hail

Historically, the original concept of radar estimation of rain in the presence of hail rests on the basic assumption that, as opposed to rain, hail is “polarimetrically isotropic” and the intrinsic values of differential reflectivity (ZDR) and specific differential phase (KDP) of hail are close to 0 dB and 0 deg/km, respectively. This means that ZDR and KDP of the rain / hail mixture should be lower than their maximal values in pure rain. However, anomalously high magnitudes of ZDR and KDP exceeding their maximal possible values in rain are persistently reported in precipitation from hail-bearing storms. While ZDR of very big hailstones with sizes larger than 5 – 6 cm tends to be negative at S band, the precipitation containing small melting hailstones may have ZDR as high as 7 dB which is much higher than the highest possible ZDR of raindrops (about 5 dB at S band). Very high values of KDP (well over 6 - 7 deg/km at S band) are quite common in hailstorms. As a result, the use of a standard R(KDP) relation may cause strong overestimation of rain in the presence of hail. On the other hand, extreme rain rates (exceeding 100 mm/h) reported at the surface are often underestimated in deep convective storms containing hail.
The physical reasons for anomalous polarimetric signatures and polarimetric radar QPE biases in hail-bearing storms are discussed and examined in this talk using a 1D spectral bin model of melting hail and a large dataset of polarimetric observations of hail. We hypothesize that very high ZDR and KDP are associated with melting hail of relatively small size whereas underestimation of extreme rainfall is commonly attributed to the fact that none of the existing radar QPE techniques takes into account increased fall velocities of hydrometeors in the strong convective downdrafts.