Study of Downscaling Techniques and Standings of Bias Corrected Global Climate Models for Brahmani Basin at Odisha, India

Climate change is emerging as one of the most pressing issues facing our environment since it will have severe consequences for both natural and human systems. The ability to estimate future climate is required to investigate the influence of climate change on a river basin. The most reliable instruments for simulating climate change are Global Climate Models (GCMs), also known as General Circulation Models. The performance of a precipitation simulation for the Brahmani river basin spanning 94 locations (with a grid resolution of 0.25° X 0.25°) is evaluated in the present study. The observed and model historical temperature datasets cover the period from 2000-2019. Twelve Coupled Model Intercomparison Project – Phase 6 (CMIP6) GCMs (ACCESS- CM2, CESM2, CIESM, FGOALS- g3, HadGEM3, GFDL- ESM4, INM- CM5-0, MIROC- ES2L, NESM3, UKESM1, MPI- ESM1, NorESM2) are used for the climate variable (Pr) using five indicators of performance. Indicators used are Average Absolute Relative Deviation (AARD), Skill Score (SS), Absolute Normalized Mean Bias Deviation (ANMBD), Correlation Coefficient (CC), Normalized Root Mean Square Deviation (NRMSD). GCMs are downscaled to finer spatial resolution before ranking them. The statistical downscaling technique is applied to eliminate the systematic biases in GCM simulations. Weights are determined using the Entropy technique for each performance metric. Cooperative Game Theory (CGT), Compromise programming (CP), Weighted Average Technique, Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS), and Preference Ranking Organization Method of Enrichment Evaluation (PROMETHEE-2) methods are utilized to rank the GCMs for the study area. GDM is an approach utilized to integrate the ranking techniques of GCMs to get a collective single rank. The results obtained for precipitation suggest that MIROC-ES2L, HadGEM3, GFDL-ESM4, UKESM1, FGOALS-g3 are the top five models that are preferred for the prediction of precipitation in the Brahmani River Basin.