A precipitação é uma variável de grande importância para a compreensão do clima em diferentes escalas espaciais como, por exemplo, a regional e global. Se os modelos climáticos conseguem reproduzir a variabilidade espacial e temporal da precipitação, se tem confiabilidade para o uso deles em estudos do clima futuro. Portanto, este estudo tem como objetivo avaliar a performance de 50 modelos climáticos globais do Coupled Model Intercomparison Project Phase 6 (CMIP6) a fim de selecionar aqueles que melhor simulam a climatologia da precipitação sobre dois subdomínios da América do Sul (sul da Amazônia – AMZ e sudeste do Brasil – SDE), no período histórico (1995-2014). Para isso, são realizadas análises estatísticas entre cada modelo e dados considerados como referência (Global Precipitation Climatology Project Version 1.2 - GPCP e Climate Prediction Center Merged Analysis of Precipitation - CMAP). Os resultados indicam que o modelo INM-CM5-0 é o que melhor simula a precipitação na AMZ e os modelos MPI-ESM1-2-HR e NESM3 na região SDE. Como estudos com modelos globais necessitam de muitos recursos computacionais, é mais conveniente selecionar aqueles que mostram boa performance para ambas as regiões, assim, são indicados os modelos EC-Earth3-Veg-LR, INM-CM4-8, INM-CM5-0 e MPI-ESM1-2-HR.Palavras-chave: Modelos de Circulação Geral, CMIP, IPCC, América do Sul.Performance of CMIP6 climate models in simulating precipitation in subdomains of South America in the historical period A B S T R A C TPrecipitation is a variable of great importance in understanding the climate at different spatial scales, such as regional and global. Whether climate models can reproduce the spatial and temporal variability of precipitation, they are reliable for their use in future climate studies. Therefore, this study aims to evaluate the performance of 50 global climate models from the Coupled Model Intercomparison Project Phase 6 (CMIP6) in order to select those that best simulate the climatology of precipitation over two subdomains of South America in the historical period (1995- 2014): southern Amazon (AMZ) and southeastern Brazil (SDE). In this sense, statistical analyses are performed between each model and data considered as reference (Global Precipitation Climatology Project Version 1.2, GPCP and Climate Prediction Center Merged Analysis of Precipitation - CMAP). The results indicate that the INM-CM5-0 model is the one with best performance in simulating precipitation in AMZ and MPI-ESM1-2-HR and NESM3 models in SDE. As studies with global models need a lot of computational resources, it is more convenient to select those models with good performance in both regions. In this way, EC-Earth3-Veg-LR, INM-CM4-8, INM-CM5-0 and MPI-ESM1-2-HR are the models indicated for both regions.Keywords: General Circulation Models, CMIP, IPCC South America

The Paraiba do Sul River Basin (PSRB) is a vital source of water resources in Brazil, providing water for human consumption, industry, agriculture, and hydroelectric energy generation. As part of one of the most developed areas of the country, in the Southeast of Brazil, the region is vulnerable to the impacts of climate change, with evidence of extreme events such as droughts and floods affecting the availability and quality of water. Hence, this study analyzes precipitation and streamflow rates data from the PSRB between 1939–2020 to investigate the spatial variability of average patterns and extreme events, trends, and their relationship with urban growth and socioeconomic development. The analysis reveals significant spatial variations in precipitation and runoff rates, with higher altitude areas, such as the Serra da Mantiqueira, exhibiting higher average values. Moreover, the Mann–Kendall trend results showed in most of the sites no significant trend regarding precipitation data; however, about 50% of the sites in the PSRB presented a decreasing trend of runoff rates. Since the precipitation does not explain identified changes in the hydrological patterns, the evaluation of the area’s urban growth and socioeconomic development throughout the decades suggested that human activities, such as those associated with urbanization, have played a significant role in altering the runoff patterns in the basin. These findings highlight the importance of sustainable land-use planning and water resource-management practices in the PSRB to mitigate the negative impacts of urbanization on the hydrological cycle and to enhance the resilience of the region’s water resources.