RAINFALL MONITORING BASED ON METEOSAT DATA - A COMPARISON OF TECHNIQUES APPLIED TO THE WESTERN SAHEL

This study was undertaken to evaluate a number of techniques to monitor rainfall over West Africa using automated processing of full-resolution digital data from the Meteosat satellite, using techniques developed by the Universities of Reading and Bristol in the U.K. and the consulting firm EARS bv in the Netherlands. This study was intended to determine an appropriate technique for the monitoring of rainfall over Africa and the Near-East for application to operational rangeland and crop production monitoring. The results indicate that all the estimates are a considerable improvement compared to estimates made by interactive man-machine processing of data from a Meteosat Secondary Data Users Station. Overall, none of the techniques yielded better results than the others, but there were distinct differences in performance. The techniques developed by the University of Reading and EARS bv performed best at the higher latitudes of the study area, while the Bristol methods performed best at the lower latitudes. The use of Meteosa visible data to complement the thermal-infrared data, as performed by one of the Bristol methods, did not improve the estimates very much.