This study investigated the relationship between surface meteorological variables and the levels of surface air pollutants (O₃, PM₁₀, and PM_2.5) in scenarios of simulated temperature increases in Rio Grande, a medium-sized Brazilian city with strong industrial influence. This study utilized five years of daily meteorological data (from 1 January 2019 to 31 December 2023) to model atmospheric conditions and two years of daily air pollutant data (from 21 December 2021 to 20 December 2023) to simulate how pollutant levels would respond to annual temperature increases of 1 °C and 2 °C, employing a Support Vector Machine, a supervised machine learning algorithm. Predictive models were developed for both annual averages and seasonal variations. The predictive analysis results indicated that, when considering annual averages, pollutant concentrations showed a decreasing trend as temperatures increased. This same pattern was observed in seasonal scenarios, except during summer, when O₃ levels increased with the simulated temperature rise. The greatest seasonal reduction in O₃ occurred in winter (decreasing by 10.33% and 12.32% under 1 °C and 2 °C warming scenarios, respectively), while for PM₁₀ and PM_2.5, the most significant reductions were observed in spring. The lack of a correlation between temperature and pollutant levels, along with their relationship with other meteorological variables, explains the observed pattern in Rio Grande. This research provides important contributions to the understanding of the interactions between climate change, air pollution, and meteorological factors in similar contexts.