Test and validation of particle classification based on meteorological model and weather radar simulator

Dual-polarization weather radars provide a number of measurements from which considerable meteorological information is obtained. This allows to identify procedures to classify hydrometeors and to estimate rainfall and ice mass. An important challenge is to test and validate such procedures. In this work, an approach is proposed to validate a hydrometeor classification based on a Support Vector Machine algorithm using a polarimetric radar simulator based on microphysics provided by the WRF meteorological model. Outputs of such model are used as truth reference. The compared analysis of the classifications obtained on radar simulations and those based on real radar measurements shows good performances of the hydrometeor classification method under test.