Background Influence of PM2.5 in Dallas–Fort Worth Area and Recommendations for Source Apportionment

Abstract

Source apportionment of observed PM2.5 concentrations is of growing interest as communities seek ways to improve their air quality. We evaluated publicly available PM2.5 data from the USEPA in the Dallas–Fort Worth metropolitan area to determine the contributions from various PM2.5 sources to the total PM2.5 observed. The approach combines interpolation and fixed effect regression models to disentangle background from local PM2.5 contributions. These models found that January had the lowest total PM2.5 mean concentrations, ranging from 5.0 µg/m3 to 6.4 µg/m3, depending on monitoring location. July had the highest total PM2.5 mean concentrations, ranging from 8.7 µg/m3 to 11.1 µg/m3, depending on the location. January also had the lowest mean local PM2.5 concentrations, ranging from 2.6 µg/m3 to 3.6 µg/m3, depending on the location. Despite having the lowest local PM2.5 concentrations, January had the highest local attributions [51–57%]. July had the highest mean local PM2.5 concentrations, ranging from 2.9 µg/m3 to 4.1 µg/m3, depending on the location. Despite having the highest local PM2.5 concentrations, July had the lowest local attributions [33–37%]. These results suggest that local contributions have a limited effect on total PM2.5 concentrations and that the observed seasonal changes are likely the result of background influence, as opposed to modest changes in local contributions. Overall, the results demonstrate that in the Dallas–Fort Worth metropolitan area, approximately half of the observed total PM2.5 is from background PM2.5 sources and half is from local PM2.5 sources. Among the local PM2.5 source contributions in the Dallas–Fort Worth metropolitan area, our analysis shows that the vast majority is from non-point sources, such as from the transportation sector. While local point sources may have some incremental site-specific local contribution, such contributions are not clearly distinguishable in the data evaluated. We present this approach as a roadmap for disentangling PM2.5 concentrations at different spatial levels (i.e., the local, regional, or state level) and from various sectors (i.e., residential, industrial, transport, etc.). This roadmap can help decision-makers to optimize mitigatory, regulatory, and/or community efforts towards reducing total community PM2.5 exposure.