Identification of raining clouds using a method based on optical and microphysical cloud properties from Meteosat second generation daytime and nighttime data

A new scheme for the delineation of raining and non-raining cloud areas applicable to mid-latitudes from daytime and nighttime multispectral satellite data is developed. The technique is based on optical and microphysical cloud properties using an artificial neural network. The tests have been conducted during the rainy season of 2006/2007. The proposed algorithm uses the spectral parameters of SEVIRI (Spinning Enhanced Visible and Infrared): brightness temperature T IR10.8 and brightness temperature differences ΔT IR10.8–IR12.1, ΔT IR8.7–IR10.8, ΔT IR3.9–IR10.8 and ΔT IR3.9–WV7.3 during the nighttime and reflectances R VIS0.6, R NIR1.6, brightness temperature T IR10.8, brightness temperature difference ΔT IR8.7–IR10.8 and ΔT IR10.8–IR12.0 during the daytime. The algorithm is calibrated by instantaneous meteorological radar using multilayer perceptron. Radar provided the ‘‘ground precipitation truth’’ for training and validation. The application shows interesting and encouraging results.