Performance evaluation of integrated multi-satellite retrieval for global precipitation measurement products over Gilgel Abay watershed, Upper Blue Nile Basin, Ethiopia

Abstract

Precipitation incorporates a substantial consequence on hydrology, environment, irrigation scheduling, vegetation, wildlife and ecology of the world. The recording of accurate and reliable rainfall data on good spatial and temporal resolutions is significant for environmental and water resources applications. The Integrated Multi-Satellite Retrievals (IMERG) of Global Precipitation Measurement (GPM) products are accessible within the variety of IMERG early, late, and final. This study aimed to assess the performance of IMERG on GPM of rainfall generation by all the three IMERG runs using gauge-based gridded data within the Gilgel Abay watershed, Upper Blue Nile Basin, Ethiopia. The assessment of satellite rainfall products was done by the areal-averaged rainfall computed using Thiessen polygon method. The satellite rainfall products were quantitatively compared with ground observations using three statistical indices (relative percent of bias (PBias), Pearson parametric statistic (R2), and root-mean-square error (RMSE)). The findings of this study revealed that IMERG early performs better than the other two satellite rainfall products in annual, monthly, and daily timescales. The IMERG late satellite rainfall overestimated and, on the contrary, IMERG final underestimated as compared to the ground-based gauged rainfall. The IMERG early rainfall product showed a worth of 1.47, 15.08, and 0.94, respectively, for RMSE, PBias, and R2. From the performance indicators of satellite rainfall products, it’s better to recommend the use of satellite IMERG early GPM to be employed in data-scarce areas for various water resources development applications by engaging bias correction.