Considerations in rainfall estimation using X-band polarimetric radar data

An issue of radar rainfall estimation using X-band polarimetric radar data is discussed in this communication. First of all, the removal of noises and clutters using polarimetric parameters is considered. A method based on fuzzy logic works well for the removal, and spatial deviations of the differential reflectivity Zdr and the differential propagation phase are nice parameters. The additional use of the copular correlation coefficient makes better removal for lower elevation angles. Second, horizontal reflectivity Zh and Zdr are calibrated using the vertical incidence radar data and 2-D video disdrometer data. Especially, the bias correction of Zdr and the attenuation correction of Zh have to be performed accurately to use their values for rainfall estimation and hydrometeor classification. Lastly, several types of algorithm are tested and compared for radar rainfall estimation. The algorithms include simple estimators such as R(Zh), R(Zh,Zdr), R(Kdp), and R(Kdp, Zdr) as well as hybrid methods to use the two of their estimators. Radar data collected during 2 years are used. Results suggest that a hybrid algorithm using R(Zh,Zdr) and R(Kdp, Zdr) shows the best performance from the comparison with surface measurements. Basically, a close relation of the specific differential phase Kdp with rainfall amount can be found for convective rainfall, and the use of Kdp is crucial for estimation from X-band polarimetric radar data. This is the same as the previous works. The effect of the additional use of Zdr is small, but it is useful because a slightly underestimation of heavy rainfall from an estimator R(Kdp) is improved.