Patterns of distributive environmental inequity under different PM2.5 air pollution scenarios for Salt Lake County public schools

Previous studies have cataloged social disparities in air pollution exposure in US public schools with respect to race/ethnicity and socioeconomic status. These studies rely upon chronic, averaged measures of air pollution, which fosters a static conception of exposure disparities. This paper examines PM2.5 exposure disparities in Salt Lake County (SLC), Utah public schools under three different PM2.5 scenarios—relatively clean air, a moderate winter persistent cold air pool (PCAP), and a major winter PCAP—with respect to race/ethnicity, economic deprivation, student age, and school type. We pair demographic data for SLC schools (n = 174) with modelled PM2.5 values, obtained from a distributed network of sensors placed through a community-university partnership. Results from generalized estimating equations controlling for school district clustering and other covariates reveal that patterns of social inequality vary under different PM2.5 pollution scenarios. Charter schools and schools serving economically deprived students experienced disproportionate exposure during relatively clean air and moderate PM2.5 PCAP conditions, but those inequalities attenuated under major PCAP conditions. Schools with higher proportions of racial/ethnic minority students were unequally exposed under all PM2.5 pollution scenarios, reflecting the robustness of racial/ethnic disparities in exposure. The findings speak to the need for policy changes to protect school-aged children from environmental harm in SLC and elsewhere. •We examined social disparities in PM2.5 levels at Salt Lake County schools.•A community-university partnership provided PM2.5 measures from ~190 sensors.•Social disparities in school-based PM2.5 exposure varied under 3 different scenarios.•Charter schools were disproportionately exposed when PM2.5 was low and moderate.•Predominantly minority schools were disproportionately exposed under all scenarios.