|Title||Benzo[a]pyrene modelling over Italy: Comparison with experimental data and source apportionment|
|Publication Type||Articolo su Rivista peer-reviewed|
|Year of Publication||2012|
|Authors||Silibello, C., Calori G., Costa M.P., Dirodi M.G., Mircea Mihaela, Radice P., Vitali Lina, and Zanini Gabriele|
|Journal||Atmospheric Pollution Research|
This work describes the extension of the Flexible Air quality Regional Model (FARM) to polycyclic aromatic hydrocarbons (PAHs). Modules accounting for the partitioning of these species between gaseous and particulate phases were inserted in a simplified version of the model and in a more state-of-the-art configuration implementing the SAPRC99 gas-phase chemical mechanism coupled with the aero3 aerosol module. Both versions of FARM were applied over Italy for the year 2005. The analysis of model results was focused on benzo[a]pyrene (B[a]P), which is considered a marker substance for the carcinogenic risk of PAHs. Simulated B[a]P concentrations were compared with observed data, collected at background sites mainly located in Po Valley, and with concentrations produced at continental scale by EMEP/MSC-E model. Higher B[a]P yearly average concentrations were simulated by the national modelling system as a result of different factors: the higher resolution adopted by the national modelling system, the greater Italian emissions estimated by the national inventory and the effects induced by the use of a high resolution topography on meteorological fields and thus on the dispersion of pollutants. The comparison between observed and predicted monthly averaged concentrations evidenced the capability of the two versions of FARM model to capture the seasonal behaviour of B[a]P, characterised by higher values during the winter season due to the large use of wood for residential heating, enhanced by lower dispersion atmospheric conditions. The statistical analysis evidenced, for both versions of the model, a good performance and better indicators than those associated to EMEP/MSC-E simulations. A source apportionment was then carried out using the simplified version of the model, which proved to perform similarly to the full chemistry version but with the advantage to be computationally less expensive. The analysis revealed a significant influence of national sources on B[a]P concentrations, with non-industrial combustion employing wood burning devices being the most important sector. The contribution of the industrial sectors is relevant around major industrial facilities, with the largest absolute contribution in Taranto (above 1 ng m-3), where steel industries are the largest individual source of PAHs in the country. © Author(s) 2012.
cited By 10