Abstract: A large body of evidence has pointed to the effects of particle pollution on human health, particularly in urban centers. Fine particles are responsible for approximately 6.4 million deaths each year, and of the total number of deaths, approximately 95% occur in countries in the Global South. The concentration of pollutants in the troposphere is influenced by dynamic atmospheric processes, with precipitation and relative humidity being the main mechanisms for removing particles from the air. Thus, statistics and geostatistics allow, in a temporal and regionalized way, to carry out mapping, quantification, and modeling of the climatic aspects of a given region. Therefore, the present work aims to map atmospheric dynamics in the city of Belo Horizonte, with a simultaneous focus on climate processes and air pollution, to contribute to debates on air quality control in the city. Hourly particle concentration data from the Minas Gerais automatic air quality monitoring network were used, for the four stations located in the Belo Horizonte, capital of the state. The analysis of the annual intensity classification based on the Standard Year classification and annual anomaly verification was based on daily meteorological data from the local station of the National Institute of Meteorology and the groupings were carried out using the Standard Deviation. Regarding mapping, this step was carried out using spatial interpolation methods, the Voronoi Diagram, when data from automatic stations across the entire Metropolitan Area of Belo Horizonte (MABH) were included in the analysis, to minimize problems of edge effects in the Belo Horizonte. Statistical techniques were used, such as the Mann-Kendall trend test (α = 0.05) to investigate climate and pollution trends. A bottleneck analysis technique was carried out to evaluate the influence of influential factors on pollution, starting with an annual analysis and moving on to monthly and daily analyses, filtering the best and worst situations for comparison. Regarding extreme years, the evaluation of positive and negative anomalies did not demonstrate a pattern of intensification of dry or rainy period events, demonstrating the occurrence of both situations in oscillation. For particle pollution, the results obtained for extreme years suggest the existence of factors other than climatic conditions for the increase or decrease in pollution over the years, since there was no frequent agreement between dry years and intensification of particle pollution. With temporal funneling, the results of seasonal and monthly analyzes allowed for a greater understanding of the climate influence on particle pollution. In general, September was the month with the highest PM2.5 pollution, which corroborates the hypothesis of the strong influence of emissions from biomass burning on fine particle pollution in all stations in the city of Belo Horizonte. Stations predominantly influenced by emissions from vehicular traffic had the best situation recorded in 2020, which despite having lower accumulated precipitation and lower average relative humidity, had the best air quality situation. It is important to highlight that 2020 was marked by the beginning of the COVID-19 pandemic, with actions to restrict urban mobility to contain the transmission of the virus, which suggests that the decrease in traffic was more important for air quality than weather conditions unfavorable. The stations where meteorological factors were most prevalent in improving air quality were those that, in addition to the influence of emissions from vehicular traffic and biomass burning, are located close to industrial installations, being influenced by their emissions. Spatially, PM2.5 concentrations are higher in stations located close to industrial areas. Understanding the factors that influence the dynamics of particle pollution is an important step towards establishing strategies to control and improve air quality.

Keywords: Fine particle pollution, Meteorological influence, Air pollution in a Brazilian urban center, Geostatistical analysis of air quality