RESUMO
Ambient aerosols were sampled at three selected sites in the coastal region of central Taiwan to obtain composition data for use in receptor modeling. All the samples were analyzed for 20 elements with an x-ray fluorescence spectrometer. The mass percentage of sulfates in particle samples was determined by ion chromatography, and mass percentages of elemental carbon (EC) and organic carbon (OC) were determined by an elemental analyzer. Because the three sampling sites were located within 25 km of each other, the average chemical compositions were similar for particle samples taken at the three sites on the same day. However, the variation in composition from day to day was significantly influenced by wind direction and change in local sources, such as the burning of agricultural wastes. The abundant species in the coarse fraction (2.5-10 µm) were Al (0.5-4.0 µg/m3), Cl (0.1-4.8 µg/m3), Ca (0.2-3.4 µg/m3), Fe (0.2-2.8 µg/ m3), and K (0.1-1.4 µg/m3), while the abundant species in the fine fraction (<2.5 µm) were S (0.3-3.5 µg/m3), Cl (0.01-1.9 µg/ m3), K (0.04-0.98 µg/m3), organic carbon (0.01-10.5 µg/m3), elemental carbon (0-10.7 µg/m3), and sulfates (1.2-15.7 µg/m3). Calculations for source apportionment were carried out using the CMB7 software developed by the U.S. Environmental Protection Agency (EPA). The main sources for the coarse fraction of ambient aerosols in the region were found to be marine aerosol, coal and fuel oil combustion, burning of agricultural wastes, and paved road dust. The main sources for the fine fraction were burning of agricultural wastes, diesel exhaust, coal and oil combustion, and sulfates. Source apportionment for the fine fraction was relatively sensitive to the types of sources selected for calculations and the compositions of the sources. The problem can be ameliorated by careful examination of possible sources and by use of local source profiles.
