High resolution source-resolved PM2.5 spatial distribution and human exposure in a large urban area

Abstract

Chemical transport models often use moderate spatial resolution to simulate atmospheric pollution, thereby limiting the model's ability to represent variations in urban areas. Additionally, the contributions of individual sources of pollution transported to the urban areas of interest from elsewhere are rarely quantified. In this study, we developed an approach to simulate air quality, focusing on PM_2.5 (particulate matter with a diameter lower than 2.5 μm), and its local and regional sources at high spatial resolution of 1 × 1 km². The approach is applied in the largest city of Greece, Athens. The PMCAMx chemical transport model is employed in combination with the source apportionment algorithm, PSAT (Particle Source Apportionment Technology), to quantify the concentrations and sources of PM_2.5, organic aerosol (OA) and elemental carbon (EC) for a typical summer and winter month. A novel approach is developed, allowing the quantification of the contributions of sources not only inside the simulated urban area but also of the regional sources located outside. Model predictions were combined with population distribution data to provide estimations for human exposure not only to total PM_2.5 concentrations but also to specific sources within the city. Residential biomass burning and transportation were found to be the dominant local sources of PM_2.5 exposure. The higher resolution (1 × 1 km²) offered a more detailed representation of PM_2.5 spatial variability than a coarser one (36 × 36 km²). This underscores the importance of capturing local sources in specific areas of the domain. The proposed approach can be used to provide estimates of human exposure to specific local and regional sources of primary and secondary PM_2.5 in an urban area.