The Quantitative Comparison of Grid Re-analysis Rainfall Products, Satellite Rainfall Products, and Hourly Rainfall Gauge Observation over Bali Province

Abstract

Grid re-analysis and satellite-based rainfall products offers a means of estimating rainfall information at both regional and worldwide level. Numerous rainfall estimates are accessible, each varying in terms of the retrieval algorithm, sensor instrumentation, spatial-temporal resolution, and geographical coverage. The objective of this research was to assess the accuracy of grid re-analysis precipitation dataset (ERA-5) and the satellite-derived rainfall dataset (IMERG) when compared to hourly rain gauge observations in Bali Province during the period spanning 2017 to 2020. To assess grid-based datasets, conventional point-to-pixel comparison methods, along with statistical metrics in the forms of continuous, categorical, and volumetric measurements. The comparative findings illustrate that IMERG exhibits superior performance at sub-daily scales in accurately detecting volume, whereas ERA-5 demonstrates greater capability in identifying rainfall events. Both products display a tendency to overestimate the capture of low to moderate rainfall occurrences and for underestimating intense to exceptionally heavy rainfall occurrences. The IMERG product excels across various elevations.