Evaluation of the EC-EARTH3-VEG climate model in reproducing the evolution of maximum and minimum air temperatures in Brazil

Abstract

According to the Sixth Report of the Intergovernmental Panel on Climate Change (IPCC), published in 2023, between 2011 and 2020, global surface temperatures increased by 1.1°C compared to the period 1850-1900. This upward trend in temperature is related to changes in the observed climate patterns, which will potentially lead to a greater incidence of extreme weather events and ecosystem changes, as well as impacting the health of human populations. In Brazil, this scenario could also result in biodiversity losses, reduced agricultural productivity and changes in the availability of water resources, with consequences for the country's economy and energy security. Considering this context, this study aims to evaluate the ability of the EC-Earth3-Veg model, part of the Coupled Model Intercomparison Project Phase 6 (CMIP6), to reproduce the evolution of maximum and minimum air temperatures at 2 meters in Brazil during the historical period, at intervals between 1961 and 2014, comparing it with ERA5 reanalysis data. The CMIP6 data was interpolated to the ERA5 grid to carry out the desired analysis. Based on this, it was observed that EC-Earth3-Veg was able to reproduce the historical climatology for Brazil but showed climatological differences when compared to ERA5 in the four periods observed. It is common among the periods analyzed that the further north of the country is warmer, with maximum temperatures in summer and autumn. In winter and spring, the same happens in the eastern part of the North, the Center-West, and the northern part of the Southeast of Brazil. In the fall, the Northeast and Midwest show cooler highs. The period in which the model's results were closest to ERA5 was from 1961 to 1990, especially for minimum temperatures in summer and winter. Even so, in the fall and spring of this period, the model showed warming in relation to the minimums in the South, and, in all the quarterly cut-outs, it showed cooling in the minimums near the Northeast. In general, certain regional and seasonal patterns were observed in the results, which may indicate a limitation of the model in terms of horizontal resolution in considering a more characteristic atmosphere for Brazil. In the southern region, for example, the maximum and minimum temperatures in the model showed warming. This may indicate that the model is limited in its horizontal resolution and does not consider an atmosphere that is more characteristic of Brazil. Therefore, to improve the model's performance in simulating the climate in Brazil, it is necessary to correct the bias and use EC-Earth3-Veg in conjunction with other models to reduce systematic errors. This study aims to replicate this work for other models.