Satellite-based detection of heavy-duty trucks: distribution and health impacts in southern California

Shahir Masri, Yiqiao Li, Luyi Gui, Shupeng Zhu, Xiaohui Xie, Jun Wu

Heavy-duty trucks contribute to ambient noise, greenhouse gas emissions and hazardous air pollution, which often place disproportionate burdens on disadvantaged communities. In this study, we applied our newly developed satellite-based truck-counting method, which fuses satellite images with publicly available GIS data, to map the spatial distribution of heavy-duty trucks and containers throughout 746 census tracts in southern California. We examined the relationship between heavy-duty truck and container density and census-tract level characteristics such as warehouse density, PM 2.5 concentrations, multiple environmental burden indicators, as well as sociodemographic, demographic, and population vulnerability factors. Results showed warehouse density to be an important predictor of truck density (p < 0.0001) for both trucks with and without containers, while population density and the proportion of White residents were both inversely associated, and PM 2.5 concentrations and the number hazardous waste facilities positively associated, with truck density (<0.0001). These results suggest the existence of environmental inequities in which truck-related exposures (e.g., air pollution, noise) are disproportionately realized among minority groups, as is the compound burden of increased hazardous waste facilities. Such findings are important for city planners when considering zoning and the siting of new warehouses and other industrial facilities, particularly as it relates to strengthening public health and sustainability. Moreover, our demonstration of the warehouse-truck density relationship is useful for future researchers who aim to develop landuse regression models to estimate regional air pollution. Finally, our analysis serves as proof-of-concept, demonstrating the utility and application of a new satellite-based data source for future environmental justice analyses and other research.