Abstract
Airborne particulate matter in the PM10, PM2.5 and PM1.0 size ranges has been sampled at three sites within 20km of one another, representing urban background, urban roadside and rural locations. The samples have been subject to chemical analysis for major constituents and the gravimetrically measured mass reconstructed using the pragmatic mass closure model of Harrison et al. [2003. A pragmatic mass closure model for airborne particulate matter at urban background and roadside sites. Atmospheric Environment 37, 4927-49331. Despite the separation in both time and space and the inclusion of a rural site, the coefficients determined in the earlier mass closure study provide an equally good mass closure on the current dataset. This extends also to the PM1.0 fraction when the coefficients deter-mined for PM2.5 are applied. The mass and composition data for PM2.5 and PM1.0 are intercompared and perhaps surprisingly the differences are accounted for more by components typical of fine fraction particles such as ammonium sulphate and ammonium nitrate than those residing primarily in the coarse fraction such as sea salt, calcium- and iron-rich dusts. A comparison of the composition of 24-h samples collected on days when average PM10 exceeded 50 mu g m(-3) with data for all days demonstrates the immense importance of nitrates, which together with their strongly bound water, account for on average 39% of PM10 and 46% of PM2.5 during episode conditions, which is more than double their contribution to the overall dataset. (c) 2007 Elsevier Ltd. All rights reserved.