Performance assessment of bias correction methods using observed and regional climate model data in different watersheds, Ethiopia

Abstract

Abstract Bias correction methods are used to compensate for any tendency to overestimate or underestimate the downscaled variables. Rainfall, maximum, and minimum temperatures are the key climate variables where the socioeconomic activities of the regions are principally based on rain-fed agriculture. This paper compares the performance of regional climate models (RCMs) and bias correction methods in Gelana and Deme watersheds in Ethiopia during the base period of 1988–2019. Observed data obtained from the Ethiopian National Meteorological Agency were used for performance evaluation of the RCM outputs. The performance of the three selected RCMs and four bias correction methods were evaluated by using four statistical indicators: Pearson correlation coefficient (R), root mean square error, Nash–Sutcliffe efficiency, and percent bias. The results show that the RACMO22T and HIRHAM5 models performed better than the RCA4 model in reproducing daily precipitation, and maximum and minimum temperatures in the Deme and Gelana watersheds. Similarly, the empirical quantile mapping method for precipitation and maximum temperature bias correction, and the distribution mapping method for minimum temperature bias correction, were well performed and preferable to adjust the climate variables of the future periods in these watersheds. Moreover, all RCMs performed better in the Deme watershed than in the Gelana watershed.