Use of atmospheric elemental size distributions in estimating aerosol sources in the Helsinki area

Citation
Ta. Pakkanen et al., Use of atmospheric elemental size distributions in estimating aerosol sources in the Helsinki area, ATMOS ENVIR, 35(32), 2001, pp. 5537-5551
Citations number
43
Categorie Soggetti
Environment/Ecology,"Earth Sciences
Journal title
ATMOSPHERIC ENVIRONMENT
ISSN journal
13522310 → ACNP
Volume
35
Issue
32
Year of publication
2001
Pages
5537 - 5551
Database
ISI
SICI code
1352-2310(200111)35:32<5537:UOAESD>2.0.ZU;2-0
Abstract
In June 1996-June 1997 Berner impactors were used in the Helsinki area to m easure size distributions of atmospheric aerosols simultaneously at an urba n and at a rural site. Ten sample pairs were collected in the size range of 0.03-15.7 mum of equivalent aerodynamic diameter (EAD). Average size distr ibutions at the two sites were calculated for 29 elements, particulate mass , and sulphate. At both sites especially sulphate, As, B, Bi, Cd, Ni, Tl, a nd V were enriched in fine particles (EAD <2.3 mum). In order to estimate l ocal fine-particle sources of the various chemical components, the similari ties and dissimilarities in the accumulation-mode parameters were studied s eparately for both sites. It was observed that often in different samples, different components had similar accumulation modes. At both sites, particu late mass, As, and Ph had similar accumulation modes to sulphate which sugg ests that long-range transport (LRT) is important for these components. V. Ni. Mo, and Co formed mother group of similar accumulation modes at both si tes suggesting that these elements largely originated from local and region al oil combustion. In addition, other groups of similar accumulation modes were observed but these groups were different between the sites. The meteor ological parameters indicated that seven sample pairs formed a subset of th e data in which the local emissions of the Helsinki area were transported t o the urban site but not to the rural site. For this subset the rural fine- particle concentrations were considered to represent an upper limit estimat e for the LRT. These upper limit LRT estimations were further improved by u tilising the quantitative relative size distributions (QRSD) method at the rural site. The QRSD method supposes that in the fine-particle size range t he LRT fractions of all chemical components have a similar shape in their s ize distributions. Fine-particle sulphate is typically long-range transport ed, and was therefore selected as the model component that represents the s hape of LRT material. Sulphate size distribution was then scaled to give an estimation of the LRT contribution of each component at the rural site. Th ese rural "sulphate scaled" LRT estimates were subtracted from the correspo nding urban concentrations to give the local contributions (ng/m(3)) downwi nd of the Helsinki area. In particles with EAD below 2.3 mum. the highest a bsolute and relative downwind local contributions were observed for several common sea-salt and road-dust components. Also the combustion-related elem ents Ni and V showed fairly high downwind local contributions. Because of t he limited number of samples, the local and LRT contributions were not esti mated for different wind directions. (C) 2001 Elsevier Science Ltd. All rig hts reserved.