Abstract:Vehicle-emitted fine particulate matter (PM 2.5 ) has been associated with significant health outcomes and environmental risks. This study estimates the contribution of traffic-related exhaust emissions (TREE) to observed PM 2.5 using a novel factorization framework. Specifically, co-measured nitrogen oxides (NO x ) concentrations served as a marker of vehicle-tailpipe emissions and were integrated into the optimization of a Non-negative Matrix Factorization (NMF) analysis to guide the factor extraction. The novel TREE-NMF approach was applied to long-term (2012–2019) PM 2.5 observations from air quality monitoring (AQM) stations in two urban areas. The extracted TREE factor was evaluated against co-measured black carbon (BC) and PM 2.5 species to which the TREE-NMF optimization was blind. The contribution of the TREE factor to the observed PM 2.5 concentrations at an AQM station from the first location showed close agreement ( R2=0.79 ) with monitored BC data. In the second location, a comparison of the extracted TREE factor with measurements at a nearby Surface PARTiculate mAtter Network (SPARTAN) station revealed moderate correlations with PM 2.5 species commonly associated with fuel combustion, and a good linear regression fit with measured equivalent BC concentrations. The estimated concentrations of the TREE factor at the second location accounted for 7–11 % of the observed PM 2.5 in the AQM stations. Moreover, analysis of specific days known to be characterized by little traffic emissions suggested that approximately 60–78 % of the traffic-related PM 2.5 concentrations could be attributed to particulate traffic-exhaust emissions. The methodology applied in this study holds great potential in areas with limited monitoring of PM 2.5 speciation, in particular BC, and its results could be valuable for both future environmental health research, regional radiative forcing estimates, and promulgation of tailored regulations for traffic-related air pollution abatement.

Decomposing PM 2.5 concentrations in urban environments into meaningful factors 2. Extracting the contribution of traffic-related exhaust emissions