QUANTIFYING THE RELIABILITY OF VARIOUS REANALYSIS PRECIPITATION PRODUCTS WITH IMD DATASET

Abstract

Precise and consistent time series of precipitation is crucial for numerous purposes, including climate and land use change studies, environmental and agricultural research, the management of natural hazards, and the planning of water resources and hydrological systems. Reanalysis products have been released worldwide to investigate the validity of reanalysis on a global and regional basis. The precipitation data obtained from reanalysis products is considered as an important resource for various research applications. Reanalysis products have been made available worldwide to assess their validity on both global and regional scales. However, the applicability of these products can vary depending on the geographic region and the spatial scale being considered. In this work, we compared different precipitation reanalysis products viz the fifth generation of ECMWF global atmospheric reanalysis (ERA5), Modern era retrospective analysis for research and application, Land version 2 (MERRA-2 Land), climate forecast system reanalysis (CFSR), Japanese global atmospheric reanalysis project (JRA55), and Multi-Source Weather (MSWX) with gridded Indian Meteorological Department dataset using various descriptive indices and found that MSWX was a better representative of IMD data than other reanalysis products. The products were evaluated both at national scale and basin scale. On a pan-India scale, MSWX and MERRA-2 Land performed better compared to others. Further we also compared the reanalysis products for each of the 23 major river basins of India and validated our results by setting up the SWAT model for Baitarani river basin. Results obtained from descriptive indices and hydrological modelling indicated that MERRA2 Land is a better representative of IMD data than other reanalysis products for Baitarani river basin, thus validating our analysis.