Traffic contribution to PM2.5 increment in the near-road environment

A growing number of studies have reported the adverse health effects of long-term exposure to air pollutants, especially fine particulate matter (PM_2.5). Vehicular emission sources have been shown to contribute to elevated air pollution concentrations in the near-road environment, including PM_2.5, based on monitoring data collected mainly during short-term campaigns. The United States Environmental Protection Agency (EPA) added near-road monitors to its national network to collect long-term National Ambient Air Quality Standard (NAAQS)-comparable data in the near-road environment. The EPA also mandated inclusion of near-road monitoring data in the Air Quality Index to reflect the elevated level of near-road PM_2.5 concentrations to which millions of people in major urban areas are exposed to on a daily basis. For the first time, PM_2.5 data collected at one of these near-road monitors were compared with those of other NAAQS monitors during 2016 in Houston, Texas. One of these NAAQS monitors was selected based on EPA guidance for quantitative hotspot analyses of particulate matter to represent background concentrations. The near-road PM_2.5 increment was statistically significant. The traffic contribution to 24-h PM_2.5 increment in the near-road environment was estimated to be about 23% of background concentration, which is close to estimates given by previous studies (22%) and is greater than a recent estimate based on a national-scale data analysis (17%), emphasizing the importance of background monitor selection criteria. Wind speed and direction were shown to have a considerable effect on PM_2.5 increment in the near-road environment. A multiple linear regression model was developed to predict 24-h near-road PM_2.5 concentrations using background PM_2.5 concentration, wind speed, and wind direction. This model explained 83% of the variability of 24-h PM_2.5 concentrations in the near-road environment and showed improvement in near-road concentration predictions when accounting for wind speed and wind direction.