Abstract:
Climate change refers to long-term variations in climate parameters. Significant
long-term variation of climate parameters is referred to as climate change. Future climate information can be projected using General Circulation Models (GCMs). Projections are perplexed from GCM model to model leading to a certain degree of uncertainty. The selection of best-fitted GCMs with recent updates can significantly reduce the uncertainty in climate projections. This study aimed to rank and select a subset of best-fitted models for Sri Lanka by evaluating the performance of GCMs. Precipitation data of 45 years from 51 stations were compared with the outputs of historical runs for 30 CMIP6 GCMs to evaluate the performance of GCMs. The maximum and minimum temperatures of 33 years
from 18 stations were also compared with the outputs of historical runs for the same period for 17 CMIP6 GCMs. Three indicators, namely, Percentage bias (PBIAS), Root Mean Square Error-observations standard deviation ratio (RSR), and Pearson correlation coefficient (r), were used to assess the performance of the GCMs, and the weights for each indicator were determined using the entropy method. For performance ranking of GCMs, compromise programming (CP) approach and Multi-Criteria Decision Making (MCDM) technique, were employed. The top-ranked models for simulating annual precipitation were CESM2, FGOALS-f3-L, FIO-ESM-2-0, and GFDL-ESM4. For simulating Yala season, GCM model CanESM5-CanOE, CMCC-CM2-SR5, CMCC-ESM2, and
CNRM-CM6-1-HR were accurate, while CNRM-CM6-1, CNRM-CM6-1-HR, CNRM-ESM2-1, and EC-Earth3-Veg-LR resulted higher accuracy for Maha season. The maximum temperature was accurately simulated by INM-CM4-8, INM-CM5-0, KIOST-ESM, and MIROC6, while AWI-CM-1-1-MR, CNRM CM6-1, CNRM-ESM2-1, and INM-CM4-8 delivered the highest accuracy for the minimum temperature. Identified high-accuracy GCMs can be used for
simulating future climate change impacts on water resources and crop production,
as well as to identify vulnerable regions for proactive adaptation and mitigation
strategies.
Keyword
