Prioritizing the control of emission sources to mitigate PM2.5 disparity in California

Abstract

Previous research has investigated the spatial disparity of ambient PM2.5 concentrations in the context of environmental justice (EJ). However, source emissions associated with the PM2.5 disparity have not been well understood. In this study, we found 2.54 mu g/m(3) (40.9%, p < 0.0001) higher PM2.5 concentration, on average, in more vulnerable (MV) communities than in less vulnerable (LV) communities in California for the period 2012-2014. Multiple linear regression models were employed to quantify the contributions of on- and off-road vehicles and point source emissions to the PM2.5 disparity between MV and LV communities while adjusting for local meteorology and site-specific characteristics. Controlling for on-road vehicular emissions associated with PM2.5 reduced the spatial PM2.5 disparity the most between MV and LV communities down to 1.05 mu g/m(3) (p = 0.0105), followed by off-road vehicular emissions (1.75 mu g/m(3), p < 0.0001) and PM2.5 point source emissions (2.17 mu g/m(3), p < 0.0001). The comparison of the full (including 3 emission predictors together) and reduced (including 2 emission predictors) models also demonstrated the strongest association of on-road vehicular emissions with the observed PM2.5 disparity. The largest contribution of on-road vehicular emissions to PM2.5 disparity seems to be attributable to disproportionately higher road density, especially for limited access roads such as Interstate highways (a factor of 1.76-2.39) and higher traffic volume (a factor of 1.62) in MV communities. These findings suggest that continuing efforts to reduce on-road traffic emissions and consider the spatial relation between MV communities and high-traffic roadways may be beneficial to alleviate potential PM2.5 health risks, particularly in MV communities.