Variance analysis applied to ground-level ozone concentrations in the state of São Paulo, Brazil

Abstract

This study applied the analysis of variance (ANOVA) technique and Scott–Knott test to investigate the patterns associated with the variability of daily maximum ozone concentrations registered between 1996 and 2013 at 32 sites in the State of São Paulo, Brazil. The dataset was analyzed using the ANOVA three way test, arranged in a factorial design (4 × 2 × 32), with four replicates per treatment, followed by the Scott–Knott test. In the ANOVA and Scott–Knott test, the first factor (a) represented the seasons, the second factor (b) distinguished measurements taken on weekdays and weekends, and the third factor (c) distinguished each of the 32 sites. The ANOVA technique and Scott–Knott test produced significant results, showing a three-way interaction between the factors mentioned. We also identified that maximum ozone concentrations occurred generally during the spring, followed by the summer; also, greater values of ground level ozone were registered on weekends in most sites, especially in those influenced mainly by vehicular emissions. Furthermore, results suggest that vehicular emissions play an important role on triggering the weekend effect on ozone levels.