A comparison of Remotely-Sensed Precipitation Estimates with observed data from rain Gauges in the Western Cape, South Africa

Abstract

Precipitation data are critical for management of water resources. Precipitation can be measured through ground measurement by means of rain gauges and by remote sensing techniques. Rain gauges often give accurate measurements, however networks of rain gauges are often sparse as they are costly and cannot be placed at all the desirable locations. The alternative is remotely-sensed data which have large spatial coverage and high temporal resolution. Remotely-sensed estimates need to generate realistic and reliable data in order to be used in water resource assessments. Therefore there is need to evaluate the accuracy of remote sensing techniques. This study investigated the reliability of the following satellite derived rainfall estimates; Multi-Sensor Precipitation Estimator (MPE), Tropical Applications of Meteorology using SATellite (TAMSAT) and Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) in areas with contrasting topographical and climatic characteristics in the Western Cape Province of South Africa.