Urban development patterns and exposure to hazardous and protective traffic environments

Estimates of the proportion of the U.S. population living close to high-traffic roads range from four to 19%. These proportions are higher for minority and low-income populations. Although the relationships among traffic exposure, race, and socioeconomic status have been consistent and reproducible, they are spatially heterogeneous and there has been limited investigation into the patterns heterogeneity. Using a spatially-explicit global regression and an exploratory geographically weighted regression (GWR), we examined variation in residential exposure to traffic at regional and neighborhood scales. Race/ethnicity, income, and college attainment are our variables of interest. Consistent with prior studies, we found that minority and lower socioeconomic status are systematically linked to higher exposure to traffic. Furthermore, the GWR approach has the potential to uncover patterns of disparities at a more localized level. However, a richer set of land use variables needs to be evaluated within this framework.