The Environmental and Population Health Impacts of Road Urban Transportation in the Puget Sound Region, WA

Abstract

The preference for private vehicles over public transportation and physically active modes of transportation in urban areas poses important environmental and public health challenges. On-road motor vehicles are a major source of urban air pollution such as particulate matter less than 2.5 micrometers (PM2.5), oxides of nitrogen (NOx), carbon monoxide (CO), volatile organic compounds (VOC) and black carbon (BC), with important impacts on population health due to increased risks of mortality and adverse cardiovascular and respiratory endpoints. A reliance on on-road motor vehicles also impacts population health through noise, greenhouse gas emissions and reduced physical activity. Thus, there is a need for further epidemiologic research to more accurately characterize the links between on-road urban transportation and public health and to evaluate the public health benefits of transportation policies to support decision-making. To this end, this dissertation proposed an integrated approach. First, we characterized the influence of highway traffic emissions on primary PM2.5 and NOx concentrations generated by highway traffic in communities near highways using a dispersion model with fine-scale spatiotemporal variations of traffic volume and flow and compared these results to those of a model with more aggregated traffic data. Second, we evaluated the association between primary PM2.5, NOx and BC generated by highway traffic and daily mortality in near-road communities using a time-stratified case-crossover design. Third, we conducted a health impact assessment to quantify the air pollution and health benefits of urban transportation policies promoting electric vehicle use and replacement of short car trips with walking and bicycling. In the first aim, we found that spatiotemporal variations between and within highways in traffic volume and flow are complex and not completely captured by traditional aggregated traffic metrics. For both sets of data, we observed highly variable concentrations over space and time of primary PM2.5 and NOx generated by highway traffic in near-road communities. While modeled concentrations of these pollutants from the fine-scale data had largely similar spatial and temporal distributions as the aggregated data, we observed some areas with larger differences between the two sources of traffic data, especially in communities closest to highly congested highways. In our second aim, we found no conclusive evidence of increased mortality with higher daily concentrations of PM2.5, NOx and BC generated by highway traffic. However, there was suggestive evidence that greater short-term exposures to these air pollutants were associated with greater odds of respiratory mortality. Contrary to our hypothesis, we found reduced odds of cardiovascular and cerebrovascular mortality with higher levels of these air pollutants. For both outcomes, observed associations were stronger among those living closest to the highways. In the third aim, we found that transportation policy scenarios promoting cleaner vehicles and replacing car trips with walking and bicycling reduce NOx and PM2.5 concentrations as well as CO2 emissions as compared with a business as usual scenario. These policy interventions also reduce burden of mortality with fewer premature deaths in adults of the general population. From an exposure assessment and environmental epidemiology standpoint, this dissertation provides evidence that improvements are needed over standard exposure assessment approaches when characterizing near-road exposures to air pollution. Furthermore, from a decision-making perspective, this dissertation provides evidence that transportation policies may mitigate the population health burdens of motorized transportation.