2007: Supervised fuzzy-logic classification of hydrometeors using C-band weather radars

Abstract—A model-based fuzzy-logic method for hydrometeor classification using C-band polarimetric radar data is presented and discussed. Membership functions of the fuzzy-logic algo-rithm are designed for best fitting simulated radar signatures at C-band. Such signatures are derived for ten supervised hydrom-eteor classes by means of a fully polarimetric radar scattering model. The Fuzzy-logic Radar Algorithm for Hydrometeor Clas-sification at C-band (FRAHCC) is designed to use a relatively small set of polarimetric observables, i.e., copolar reflectivity and differential reflectivity, but a version of the algorithm based on the use of specific differential phase is also numerically tested and documented. The classification methodology is applied to volume data coming from a C-band two-radar network that is located in north Italy within the Po valley. Numerical and experimental re-sults clearly show the improvements of hydrometeor classification, which were obtained by using FRAHCC with respect to the direct use of fuzzy-logic-based algorithms that are specifically tuned for S-band radar data. Moreover, the availability of two C-band rainfall observations of the same event allowed us to implement a path-integrated attenuation correction procedure, based on either a composite radar field approach or a network-constrained variational algorithm. The impact of these correction procedures on hydrometeor classification is qualitatively discussed within the considered case study. Index Terms—C-band weather radar, fuzzy-logic method, hydrometeor classification, path-attenuation correction, radar meteorology, radar polarimetry. I