摘要：Weather radar calibration is a topic of great current interest because it is useful for various hydrological applications. Nevertheless, several methods have been developed for adjusting the relation between reflectivity data Z and rainfall intensity R, Z/R, because droplet size distributions in different storm events are unknown and highly variable in time and space. The present study developed and tested a new space and time window-based procedure for optimal local calibration of weather radar using Z/R relations and applied to convective and stratiform storms in the Lower Grijalva River Basin in Mexico. Improving rain estimates from the Sabancuy, Campeche, radar is essential because it monitors this basin, which is prone to floods. The resulting estimates of the optimal power-law (Z = AR b ) window-based procedure (OP) are compared with those of the default Marshal and Palmer (MP) relation using the observed rain gauge records. The appropriate window was selected using a criterion that considers factors that affect raindrop free fall. For most of the storms tested, metrics for the OP models showed better values than those calculated for the MP ones. The best MP performance is when using smooth calibration data, but it only achieves similar metric results to that of the OP. We conclude that the proposed calibration method is useful to improve the default MP model estimates at any weather radar with similar characteristics to the ones analyzed in this work. The resulting Z/R relations could improve precipitation radar estimates for hydrologic model inputs.
关键词：Optimization; Flash-flood; Power-law; Grijalva River; Space-time windows
