This study provides a comparative analysis of the ERA5 reanalysis and Integrated Surface Database (ISD) datasets to evaluate their effectiveness in identifying extreme heat and cold events across Brazil from 1996 to 2022. With Brazil’s vast area and uneven distribution of weather stations, the research emphasizes the suitability of ERA5 in capturing extreme temperature events (ETEs) in data-sparse regions. Key metrics—Pearson correlation, Mean Absolute Error (MAE), Root Mean Square Error (RMSE), hit rates for extreme events, and matching days at the 5th and 95th percentiles—were used to compare datasets. Results revealed a high overall correlation (r = 0.89) between ERA5 and ISD, with notable regional variability, particularly in the North and Northeast. ERA5 data indicated a significant increase in heatwave frequency, especially in the North, where heatwave events rose from an average of 0.96 per year (1996–2004) to 3.10 per year (2014–2022). Conversely, ISD data showed declining heatwave trends in some regions, notably in the Northeast, where heatwave frequency decreased from 2.38 to 1.05 events per year over the same periods. Coldwave trends showed a general decline in both datasets, consistent with global warming. ERA5, for instance, recorded a decrease in coldwave events in the North from 2.38 per year to 0.53 per year. However, ISD data showed less consistency, with minimal declines in the North, Northeast, and Midwest. These discrepancies highlight challenges due to data sparsity and methodological differences but affirm ERA5's utility for ETE detection in regions lacking dense weather station coverage. This study provides critical insights into the strengths and limitations of reanalysis versus station-based data for climate monitoring in Brazil, with implications for future climate research and policy. Keywords: Extreme Heat, Extreme Cold, Reanalysis, Station Data, Brazil.