Evaluation of the performance of bias-corrected CORDEX regional climate models in reproducing Baro–Akobo basin climate

Abstract

The applicability of the regional climate model (RCMs) for catchment hydroclimate is obscured due to their systematic bias. As a result, bias correction has become an essential precondition for the study of climate change. This study aimed to evaluate the skill of seven rainfall and five maximum and minimum temperature RCM outputs against observed data in simulating the characteristics of climate at several locations over the Baro–Akobo basin in Ethiopia. The evaluation was performed based on raw and bias-corrected RCMs against observed for a long-term basis. Several statistical metrics were used to compare RCMs against observed using a pixel-to-point approach. In this finding, raw RCMs showed pronounced biases such as lower correlation and higher PBIAS in estimating rainfall and minimum temperature than maximum temperature. However, most RCMs after bias correction showed better performance in reproducing the magnitude and distribution of the mean monthly rainfall and temperature and improve all the statistical metrics. The Mann–Kendall trend test for observed and bias-corrected RCMs indicated a decreasing annual rainfall trend while the maximum and minimum temperature showed an increasing trend in most stations. In most statistical metrics, the ensemble mean resulted in better agreement with observation than individual models in most stations. In general, after bias correction, the ensemble adequately simulates the Baro–Akobo basin climate and can be used for evaluation of future climate projections in the region.