Urban air quality represents a major public health burden and is a long-standing concern to European citizens. Combustion processes and traffic-related emissions represent the main primary particulate matter (PM) sources in urban areas. Other sources can also affect air quality (e.g., secondary aerosol, industrial) depending on the characteristics of the study area. Thus, the identification and the apportionment of all sources is of crucial importance to make effective corrective decisions within environmental policies. The aim of this study is to evaluate the impacts of different emissions sources on PM2.5 concentrations and compositions in a mid-size city in the Po Valley (Treviso, Italy). Data have been analyzed to highlight compositional differences (elements and major inorganic ions), to determine PM2.5 sources and their contributions, and to evaluate the influence of air mass movements. Non-parametric tests, positive matrix factorization (PMF), conditional bivariate probability function (CBPF), and concentration weighted trajectory (CWT) have been used in a multi-chemometrics approach to understand the areal-scale (proximate, local, long-range) where different sources act on PM2.5 levels and composition. Results identified three levels of scale from which the pollution arose: (i) a proximate local scale (close to the sampling site) for traffic non-exhaust and resuspended dust sources; (ii) a local urban scale (including both sampling site and areas close to them) for combustion and industrial; and (iii) a regional scale characterized by ammonium nitrate and ammonium sulfate. This approach and results can help to develop and adopt better air quality policy action