A simple self-calibrating cold cloud duration technique applied in West Africa and Bangladesh

Abstract

Two satellite-based rainfall estimation routines (RI1 and RI2) have been applied in two different climatic environments: the semi-arid to sub-humid conditions of the Senegal River basin in West Africa and the humid conditions of Bangladesh. Both routines are based on bi-hourly thermal infra-red data from METEOSAT satellites. The RI1 and RI2 models are both variants of Cold Cloud Duration (CCD) techniques and employ two parameters for rainfall estimation: a rain rate and a rain threshold temperature, where the latter is used to identify rain areas. The RI1 model, like ordinary CCD algorithms, assumes an equal amount of rain falls from all clouds colder than the temperature threshold, whereas the RI2 model assumes that clouds precipitate more with decreasing temperatures. Continuous calibration of rain rates and temperature thresholds is suggested to capture short term variations in rain cloud properties. Results are compared to a well-established CCD routine, the GOES Precipitation Index (GPI) that uses constant values of rain rate and temperature threshold. Tests of RI1. RI2 and GPI shows that neither RI1 nor RI2 produces better results than the GPI in Bangladesh, while in the Senegal River basin the continuously calibrated RI1 model performs best.