Plant and soil contaminant modeling paired with risk-based soil screening levels and ingestion rates for home gardening in environmental justice communities.

Pollution is ubiquitous and poses substantial health risks for environmental justice (EJ) communities via growing and eating contaminated plants. Using community and citizen science frameworks, we collected and analyzed 435 soils and 997 plants for metal(loid) concentrations in 10 communities facing contamination in the United States. Linear mixed models showed that community was a major driver of soil contamination, plant contamination, and bioconcentration factors. Plant category contamination trends were not consistent across all contaminants, but generally, herbs, tree leaves, brassica vegetables, and leafy vegetables had the highest contamination while fruits and fruiting vegetables had the lowest. Leafy vegetables, herbs, flowers, and tree leaves had high bioconcentration factors, while fruits and fruiting vegetables were typically lower. Estimated risk-based screening levels indicated that existing soil screening levels can generally protect against negative health effects from growing food in contaminated soil but may not be strict enough to protect against risks from arsenic, nickel, barium, and copper exposures. Based on EJ community metal(loid) concentrations observed in this study, estimated risk-based ingestion rates indicated that generally, tree leaves, herbs, flowers, and brassica vegetables should be consumed at the lowest rates due to their high potential arsenic contamination. For a target arsenic cancer risk of 10