The objective of the project was to develop a method or measuring process to determine the composition of airborne particulate matter.
For this purpose, both physical and chemical analysis were performed and multivariate calculations used to identify or simulate the mass fractions of each source material in real samples of particulate matter.
Several source materials were selected, soil, asphalt, soot, brake lining and salt. The method successfully produced estimates for the composition of PM10 winter samples collected in Reykjavik city. Traffic related material is dominating the winter samples but soil, originating from soil erosion, does have a strong impact on the airborne pollution.
The average combination for the ambient PM10 samples was found to constitute of asphalt 55%, soil 25%, soot 7%, salt 11% and brake lining around 2%. On the most problematic days when the PM10 concentrations in ambient air are above limit values set in regulations, asphalt is almost 60% of the total PM. Soil and asphalt are seen in higher concentrations in coarse fraction, PM2,5-10 while soot is detected in larger amount in the fine fraction, PM2,5. A comparison between wet and dry days indicates that asphalt is strongly dominating dry days while soot and salt are seen in fairly high concentrations on wet days.
There is an indication for a source of PM in summers that is not detectable by the source samples used in this modelling. It is suggested here that pollen and spores might be relevant sources not taken into account in the modelling at hand.