Spatiotemporal analysis of particulate matter, sulfur dioxide and carbon monoxide concentrations over the city of Rio de Janeiro, Brazil

Time series of pollutants and weather variables measured at four sites in the city of Rio de Janeiro, Brazil, between 2002 and 2004, were used to characterize temporal and spatial relationships of air pollution. Concentrations of particulate matter (PM10), sulfur dioxide (SO2) and carbon monoxide (CO) were compared to national and international standards. The annual median concentration of PM10 was higher than the standard set by the World Health Organization (WHO) on all sites and the 24 h means exceeded the standards on several occasions on two sites. SO2 and CO did not exceed the limits, but the daily maximum of CO in one of the stations was 27% higher on weekends compared to weekdays, due to increased activity in a nearby Convention Center. Air temperature and vapor pressure deficit have both presented the highest correlations with pollutant's concentrations. The concentrations of SO2 and CO were not correlated between sites, suggesting that local sources are more important to those pollutants compared to PM10. The time series of pollutants and air temperature were decomposed in time and frequency by wavelet analysis. The results revealed that the common variability of air temperature and PM10 is dominated by temporal scales of 1-8 days, time scales that are associated with the passage of weather events, such as cold fronts.