Identification of Best CMIP6 Global Climate Model for Rainfall by Ensemble Implementation of MCDM Methods and Statistical Inference

Global climate models (GCMs) are becoming more and more important due to their ability to accurately identify climatic factors, which will be highly helpful in establishing planning and strategizing for water resources engineers. Therefore, studying the performance of GCMs is essential, as it can simulate and predict climatic scenarios. This study examines the best performance of 24 GCMs using NASA NEX-GDDP dataset reproduced rainfall in four catchments: Silabati, Keliaghai, Kangsabati and Dwarkeswer from West Bengal, India. This study used historical period (1950–2014) rainfall data to evaluate the performance of 24 CMIP6 GCMs. The GCMs output was compared with observed data using 17 performance indicators (PIs). Five distinct techniques for Multicriteria Decision Making (MCDM) were applied to assess rankings, aiming to achieve a consistent evaluation of GCMs, despite encountering conflicting outcomes from both the parameters and the MCDM methods. A combined evaluation of PIs and MCDM was used to rank GCMs. The research uncovered that out of all the GCMs examined, five were identified as the top-performing ones in terms of accurately estimating rainfall based on comparisons with observed data. The top-ranked GCMs observed were canESM5, MIROC-ES2L, IITM-ESM, BCC-CSM2-MR and EC-Earth3-Veg-LR. The assessed GCMs, despite their advancements, struggle with accurately capturing regional intricacies, potentially affecting the precision of forecasts. The findings from this research will provide valuable insights for both climate researchers and decision-makers in selecting the most suitable CMIP6 GCMs. It can be asserted that the proposed methodology is readily adaptable to any situation, offering practical applicability.