Evaluation and bias correction of the NASA POWER precipitation for representing the spatiotemporal variability of rainfall over the upper Blue Nile basin, Ethiopia

This study evaluates the performance of NASA POWER precipitation data in representing the spatiotemporal variability of rainfall in the upper Blue Nile basin, Ethiopia. To address identified deficiencies, bias correction was implemented using Empirical Quantile Mapping (EQM). Daily observed rainfall from local gauges served as the reference for evaluating performance using standard statistical metrics, including Correlation Coefficient (CC), Nash–Sutcliffe Efficiency (NSE), Root Mean Square Error (RMSE), Percent Bias (PBIAS), and Kling–Gupta Efficiency (KGE). Results indicate that the raw NASA POWER product exhibits substantial systematic errors, characterized by weak linear agreement (mean CC = 0.23) and consistent underestimation of magnitude (mean PBIAS = −23.4%). These limitations highlight the risks of using uncorrected satellite–reanalysis data in high-relief terrain. The application of EQM significantly improved performance by correcting distributional biases, increasing mean CC to 0.56 and neutralizing the PBIAS to 1.31%. While event-level timing remains a challenge, the correction notably enhanced distributional consistency and volumetric accuracy. These findings underscore that while raw satellite products struggle in topographically complex regions, robust bias correction substantially increases their reliability for hydrological modeling, drought monitoring, and climate impact assessments. Ultimately, the corrected dataset provides a viable and accessible alternative for water resource management in data-scarce mountainous watersheds.