Adjustment of statistical distributions to O₃ and NO₂ concentrations in the pre-COVID-19 periods and during the pandemic in Campo Grande, MS, Brazil

COVID-19 has triggered a series of studies based on environmental factors that may influence its dissemination. Among them, the role of air pollution and climate variables in the dissemination of SARS-CoV-2 stands out. Campo Grande (MS), located in the Central-West region of Brazil, offers a scenario to investigate such relationships. This study evaluated the relationship between ozone (O₃) and nitrogen dioxide (NO₂) concentrations, meteorological variables, and confirmed cases of COVID-19, focusing on the statistical modeling of the distributions of these pollutants in the periods before and after COVID-19. Daily time series of O₃, NO₂, temperature, relative humidity, precipitation, and COVID-19 data were analyzed, referring to the period from January to March 2020. Probabilistic models (log-normal, gamma, Weibull, Gumbel, and log-logistic) were applied and evaluated based on the Akaike and Bayesian information criteria to identify the best-fit statistical distribution for the pollutants. The results showed a negative correlation of COVID-19 with humidity and precipitation and a positive correlation with temperature, O₃ and NO₂. An increase in the mean concentrations of O₃ (+2.7 DU) and NO₂ (+3.731015 molecules/cm²) was observed during COVID-19. The log-normal distribution was best suited to O₃, while the Gumbel distribution performed best for NO₂.