Large spatio-temporal variations of size-resolved particulate matter and volatile organic compounds in urban area with heavy traffic

A monitoring campaign, the first of this kind for a heavy traffic urban area of Vietnam, was conducted which generated nearly 200 daily filter samples of PM2.5, PM10, and black carbon (BC), 1300 online hourly PMx (PM10, PM2.5, and PM1), 900 hourly/bi-hourly BTEX data, 700 h of traffic counts, and online meteorology records. PMx and BTEX levels show large horizontal gradients across this small urban area of 300 m width suggesting that the pollution data should be generated with sufficient spatial resolutions for assessment of the exposure and health effects. This paper focuses on analyzing PMx with reference to the previously published BTEX to provide a more complete picture of the traffic-related pollution in the area. Spatio-temporal variations of pollutants are analyzed in relation to traffic flows and fleet compositions, weekday-weekend effects, local and regional meteorology. PM10 and BTEX levels had larger variations between the sites indicating their stronger associations with the traffic activities than the finer particles. Twenty-four-hour (24 h) PM2.5 levels ranged between 19 and 191 mu g/m(3) with high PM1/PM2.5 ratios of above 0.8 at ambient site (AA) and above 0.7 at roadsides. Multivariate relationship analysis (PCA) for the bi-hourly datasets of meteorology, traffic flows, and pollutant levels indicated overwhelming influence of on-road traffic fleet compositions on the roadside pollutants levels. At AA, PCA results showed a complex interaction between local emissions, meteorological conditions, and regional/long-range transport. Higher pollution levels were associated with the airmass types having the continental origin and pathways.