This analysis focuses on modeling temperature change in South America, a region particularly vulnerable to the impacts of climate change due to its diverse ecosystems, from 1917 to 2020. We hypothesize that ARIMA-based predictive algorithms can help us understand future global temperatures and find preventive measures for environmental disasters. The proposed ARIMA model was applied to test data obtained from Kaggle, which includes temperature changes in the South American region over the last century. ARIMA is a time series algorithm that employs an autoregressive integrated moving average technique to model data for future predictions. The model performed moderately across all regions, with root mean square error (RMSE) values averaging 0.42 in South America during the months of Summer. Significant temperature changes over select months in South America contribute to extreme environmental events, such as the landslides in Colombia (2017) and the volcanic eruption in Chile (2015). For summer data in South America, temperatures transitioned from a decrease of 0.4°C in 1960 to a projected increase of 2°C by 2030, indicating a sharp rise even without accounting for external factors. Including external influences like carbon emissions and energy consumption may improve the model’s accuracy, offering deeper insights. This enhancement would also enable the identification of key factors that cause temperature anomalies, allowing for targeted strategies to fight global warming. Using external factors with natural variability will also help predict region-specific solutions and policies. Similar analyses conducted for other regions revealed similar trends, suggesting that temperature fluctuations will significantly increase compared to current temperature value. This research aims to explore time series-based algorithms, their applications, and their performance limitations. A detailed analysis will be presented at the conference.