Multiparameter radar snowfall estimation using neural network techniques

Abstract

WISP94 CSU-CHILL radar data and ground snowgage measurements at Stapleton International Airport (SIA) and Denver International Airport (DIA), Denver, Colorado, are analyzed in this paper. Traditionally, the radar estimation of ground snowfall is estimated by Z-S relations. The performance of such parametric relations are not satisfactory due to the complexities of the snow process. In this paper a radial-basis function neural network based algorithm is applied to map the relationship between the radar observations and ground snowfall measurements. The development of the neural network based technique, and the snowfall estimation results are presented.