PERFORMANCE EVALUATION OF MULTIPLE STATE-OF-THE-ART SATELLITE PRODUCTS FOR PRECIPITATION ESTIMATION AND FLOOD-CAUSING POTENTIAL OVER NEPAL

Abstract

Understanding the hydrological cycle, particularly in regions with complex terrain like Nepal, is essential for effective water resource management, disaster preparedness, and climate change adaptation. However, the absence of reliable precipitation data poses a significant challenge to this understanding. In mountainous and rugged terrains, such as those found in Nepal, this challenge is particularly pronounced. To address this issue, satellite precipitation products (SPPs) offer a promising solution by providing valuable insights into precipitation patterns over remote and inaccessible areas. Despite their potential utility, the efficiency of SPPs over Nepal has not been comprehensively assessed until now. This study represents the first comprehensive evaluation of SPPs in Nepal, aiming to fill the gap in understanding the performance of these products in the region. Four SPPs, CMORPH, CHIRPS v2.0, PERSIANN-CDR, and PDIR-Now, were subjected to rigorous analysis in two stages. Firstly, they were compared to the observation-based APHRODITE dataset, followed by a comparison with observed gauge datasets from 28 rain gauges in Narayani River Basin of Nepal. The findings of the study reveal that CHIRPS v2.0 emerged as the top performer across all analyses. It exhibited superior performance in estimating mean rainfall amounts and detecting extreme rainfall events compared to the other SPPs evaluated. In contrast, the other SPPs showed varying degrees of underestimation or overestimation, highlighting the importance of selecting the most appropriate SPP for specific applications. Moreover, CHIRPS v2.0 demonstrated better correlation with observed rainfall at a monthly scale, indicating its reliability in capturing temporal rainfall patterns. Additionally, it exhibited a robust rainfall detection capacity, characterized by higher probability of detection (POD) and lower false alarm rate (FAR), further affirming its suitability for precipitation monitoring in Nepal's challenging terrain.