Impact of Meteorological Factors on Seasonal and Diurnal Variation of PM2.5 at a Site in Mbarara, Uganda

Abstract

Because PM2.5 concentrations are not regularly monitored in Mbarara, Uganda, this study was implemented to test whether correlations exist between weather parameters and PM2.5 concentration, which could then be used to estimate PM2.5 concentrations. PM2.5 was monitored for 24 h periods once every week for eight months, while weather parameters were monitored every day. The mean dry and wet season PM2.5 concentrations were 70.1 and 39.4 µg/m3 , respectively. Diurnal trends for PM2.5 levels show bimodal peaks in the morning and evening. The univariate regression analysis between PM2.5 and meteorological factors for the 24 h averages yields a significant correlation with air pressure when all data are considered, and when the data are separated by season, there is a significant correlation between PM2.5 concentration and wind speed in the dry season. A strong correlation is seen between diurnal variations in PM2.5 concentration and most weather parameters, but our analysis suggests that in modeling PM2.5 concentrations, the importance of these meteorological factors is mainly due to their correlation with underlying causes including diurnal changes in the atmospheric boundary layer height and changes in sources both hourly and seasonally. While additional measurements are needed to confirm the results, this study contributes to the knowledge of short-term and seasonal variation in PM2.5 concentration in Mbarara and forms a basis for modeling short-term variation in PM2.5 concentration and determining the effect of seasonal and diurnal sources on PM2.5 concentration.