Performance evaluation of IMERG and TMPA daily precipitation products over CONUS (2000–2019)

Abstract

Ground validation of satellite-based precipitation products is necessary to evaluate the performance of satellite observation at local, regional, and global scales. The evaluation of the rainfall data accuracy will reveal both its strengths and weaknesses, thereby determining its suitability for hydrometeorological applications. To reach this goal, this study comprehensively evaluated and characterized the systematic bias in the Global Precipitation Measurement (GPM) Integrated Multi-SatellitE Retrievals for GPM (IMERG) precipitation product, to be compared with its predecessor Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA), based on the wide-spread ground-based observation data, PRISM (Parameter-Elevation Regression on Independent Slopes Model) over the Conterminous United States. The evaluation was conducted for daily datasets from June 2000 to the end of December 2019, in which IMERG and TMPA overlap using statistical error metrics both in space and time. While some statistical measures do not show significant improvement in daily IMERG compared to TMPA, systematic bias is substantially diminished. We detected systematic bias related to the vertical relief for both IMERG and TMPA but with a substantial decrease in the IMERG. The systematic bias in both satellite products distinctively differs in the summer and winter seasons, but, both products showed high performance in summer. According to our findings, the spatiotemporal performance of the IMERG precipitation product varies in different seasons, regions, time scales, and topographic conditions but with superiority over the TMPA daily product.