A comparative analysis of rainfall data from IMERG early run and ground-based rain gauges on Bali Island

Abstract

Ground-based precipitation measurements encounter challenges in various parts of Bali due to the limited number of gauge stations. Therefore, it is essential to identify dependable alternatives like satellite-derived precipitation data, which offer continuous precipitation time series with high spatial resolution. This study assessed the effectiveness of near real-time global satellite precipitation products, specifically the Integrated Multi-satellite Retrievals for Global Precipitation Measurement-Early Run (IMERG-E) compared to gauge data from 43 stations across Bali Province. To evaluate IMERG-E datasets, traditional point-to-pixel comparison techniques were employed, alongside statistical metrics including correlation coefficient (CC), mean absolute error (MAE), relative bias (RB), and Nash-Sutcliffe efficiency (NSE). The comparative analysis showed that the daily IMERG-E dataset performs moderately well, as evidenced by weak to moderate correlation and low MAE. IMERG-E showed evidence of underestimating rainfall, as indicated by the RB value. Conversely, IMERG-E demonstrates poor accuracy according to the NSE value. It is necessary to explore effective correction methods for IMERGE-E to establish it as a viable alternative data source.