Performance evaluations of CMIP6 model simulations and future projections of rainfall and temperature in the Bale Eco-Region, Southern Ethiopia

Abstract

Identifying best performing climate models is indispensable for better understanding of the future climate and its impact as well as for planning effective climate change adaptation and mitigation measures. This research aims to identify the best performing Global Climate Models (GCMs) products from the Coupled Model Inter-comparison Project phase 6 (CMIP6) in simulating rainfall and temperature in the Bale Eco-Region (BER), Southern Ethiopia. In this study, evaluations were performed for ten CMIP6 GCMs against observed and reanalysis rainfall and temperature products in terms of how well the GCMs reproduce rainfall, maximum temperature (Tmax) and minimum temperature (Tmin) from daily to annual temporal scales during 1995–2014 period. Performance evaluations were performed using the Comprehensive Rating Index (CRI), which is based on four statistical metrics. The best performing CMIP6 model(s) were bias-corrected by Distribution Mapping (DM) for future climate analysis at different agro-ecological zones (AEZs) and at the eco-region level. The study used projections of climate variables in the near future (2021–2040), mid-century (2041–2060) and late century (2081–2100) periods. Three shared socioeconomic pathways (SSP2-4.5, SSP3-7.0, and SSP5-8.5) were considered as future climate scenarios. The result indicated that BCC-CSM2-MR, CNRM-CM6-1 and MRI-ESM2-0 are relatively better for simulating the rainfall climatology of the BER from the daily to annual temporal scales. EC-Earth3, Ec-Earth3-Veg and MPI-ESM1-2-LR are also comparatively better for simulating Tmax while CNRM-CM6-1, EC-Earth3-Veg and EC-Earth3 outperformed for simulating Tmin in the studied temporal scales. After careful evaluations, climate change analysis was performed using the ensemble mean of BCC-CSM2-MR, CNRM-CM6-1 and MRI-ESM2-0 for rainfall, EC-Earth3 for Tmax and the ensemble mean of CNRM-CM6-1 and EC-Earth3-Veg for Tmin. Accordingly, the annual rainfall is expected to decrease in the near future in the three scenarios in the alpine (2–5%), temperate (11–14%) and sub-tropical (7–9%) AEZs as well as the BER spatial scales (2–5%), but rainfall is expected to increase in the late century period. In contrast, rainfall is expected to increase in the tropical AEZ in both the near future (3–11%) and late century (25–45%) periods. In the mid-century period, rainfall is expected to increase in the tropical AEZ in all the three scenarios, but it exhibits different directions of changes in the remaining AEZs and BER scale at different scenarios. The finding also revealed an expected increase in both Tmax and Tmin in the different AEZs as well as the BER scale, but the projected temperature increase is high in temperate AEZ. The projected increase of rainfall in the near future in tropical AEZ may reduce the frequently occurring droughts mainly in the lowland parts of the BER. Conversely, the reductions of rainfall in the remaining AEZs may introduce challenges for agriculture, water resources as well as endemic animals. The findings of this study justifies the need for climate models evaluation for each climate variables in order to choose the most appropriate models for localized climate change impact and adaptation studies. In addition, this study also suggested projections of future climate for different AEZs for better decision-making process for the specific AEZs.