Validation of Satellite Rainfall Estimates over Equatorial East Africa

Abstract Rain gauge data sparsity over Africa is known to impede the assessments of hydrometeorological risks and skill numerical weather prediction models. Satellite rainfall estimates (SREs) have been used as surrogate fields for a long time are continuously replaced by more advanced algorithms new sensors. Using unique daily dataset from 36 stations across equatorial East period 2001–18, this study performs multiscale evaluation gauge-calibrated SREs, namely, IMERG, TMPA, CHIRPS, MSWEP (v2.2 v2.8). Skills were assessed annual scales, extreme precipitation, TMPA IMERG near-real-time (NRT) products. Results show that 1) SREs reproduce pattern seasonal cycle well, despite exhibiting biases up 9%; 2) best shorter temporal scales while MSWEPv2.2 CHIRPS perform at monthly steps, respectively; 3) performance all varies spatially, likely due an inhomogeneous degree calibration, with largest variation seen in MSWEPv2.2; 4) miss between 79% (IMERG-NRT) 98% (CHIRPS) events recorded rain gauges; 5) IMERG-NRT regarding event detection accuracy; 6) return values rainfall, least errors MSWEPv2.8 cannot be recommended. The also highlights improvements decline compared MSWEPv2.2, potential flood risk assessment Africa.