The results of a source contribution analysis study performed with the CAMx / PSAT v6.3 model and related to an urban receptor located in downtown Milan are presented. CAMx was applied to a domain covering the Po Valley for the 2010 calendar year. The simulations covered nitrogen dioxide (NO2), fine particulate matter (PM2.5) and primary and secondary components, namely: elemental carbon (EC) and nitrate (NO3-). The results of the breakdown of emission contributions are reported by emission areas (e.g., local areas, urban, metropolitan, counties) and by categories (e.g., transportation, heating, industrial activities) and as a combination of areas and categories. Five emission areas were considered: the centre of Milan, the municipality of Milan, the metropolitan area of Milan, the Lombardy region and the entire Po Valley. In terms of the contributions by emission area, the municipality of Milan, its metropolitan area and the Lombardy region account for about 60% of the total mass of PM2.5 at the selected receptor. In contrast, local-scale emissions contribute more than 50 percent to ambient levels of elemental carbon, again assessed at the Milan receptor. Conversely, sources located in the most distant emission areas (Lombardy and the Po Valley) and long-range transport determine the largest contribution (80%) to NO3 concentrations. For NO2, local-scale emissions are responsible for more than 60% of the concentration levels in the centre of Milan. In terms of source categories, traffic is the main contributor to NO2 and NO3, while biomass burning and traffic are the main sources of elemental carbon and PM2.5. The contributions of emission categories to PM2.5 estimated by CAMx / PSAT for the selected receptor show fairly good agreement with the partitioning results obtained from the Positive Matrix Factorization (PMF) application available for the Milan area. However, the two approaches provided similar estimates only for biomass combustion and traffic contribution (24% and 20%, respectively), while CAMx provided significantly lower estimates for the portion of secondary organic aerosol (SOA), probably due to the lack of adequate algorithms for the correct reconstruction of particulate organic matter.