Source apportionment of PM10, organic carbon and elemental carbon at Swiss sites: An intercomparison of different approaches

摘要

In this study, the results of source apportionment of particulate matter (PM10), organic carbon (OC), and elemental carbon (EC) - as obtained through different approaches at different types of sites (urban background, urban roadside, and two rural sites in Switzerland) - are compared. The methods included in this intercomparison are positive matrix factorisation modelling (PMF, applied to chemical composition data including trace elements, inorganic ions, OC, and EC), molecular marker chemical mass balance modelling (MM-CMB), and the aethalometer model (AeM). At all sites, the agreement of the obtained source contributions was reasonable for OC, EC, and PM10. Based on an annual average, and at most of the considered sites, secondary organic carbon (SOC) is the component with the largest contribution to total OC; the most important primary source of OC is wood combustion, followed by road traffic. Secondary aerosols predominate in PM10. All considered techniques identified road traffic as the dominant source of EC, while wood combustion emissions are of minor importance for this constituent. The intercomparison of different source apportionment approaches is helpful to identify the strengths and the weaknesses of the different methods. Application of PMF has limitations when source emissions have a strong temporal correlation, or when meteorology has a strong impact on PM variability. In these cases, the use of PMF can result in mixed source profiles and consequently in the under- or overestimation of the real-world sources. The application of CMB models can be hampered by the unavailability of source profiles and the non-representativeness of the available profiles for local source emissions. This study also underlines that chemical transformations of molecular markers in the atmosphere can lead to the underestimation of contributions from primary sources, in particular during the summer period or when emission sources are far away from the receptor sites.