Performance Evaluation of the HadGEM2-ES and Eta Models from Climatic Extremes of Precipitation for the Amazon Basin

Abstract The calculation of climate index from observed data and climate models is an efficient way of identifying possible climatic variations on the Amazon basin. Although it is frequent to evaluate the performance of climate models from precipitation and temperature data, the evaluation of their performances using climate index is still little explored, mainly for the Amazon basin. Thus, the present work main objective was to quantify and evaluate the dexterity of the Had- GEM2-ES and Eta models in the representation of eight climate index. The climate index was determined from the daily precipitation from the Climate Prediction Center. To quantify the performance of both models, the bias and mean square error were calculated. These methods were applied on the basis of the annual average of the climate index for the period 1981 to 1990. The results show that the bias of HadGEM2-ES and Eta models is similar in the most climate index. Regarding the mean square error, it was observed that not all climatic extremes indicators based on precipitation data are well represented by both models, but Eta presented better performance in relation to HadGEM2-ES.