A review of background dioxin concentrations in urban/suburban and rural soils across the United States: implications for site assessments and the establishment of soil cleanup levels.

Over the last several decades, dioxin releases have decreased >90%, leading to a corresponding decrease in human body burdens. In addition, the weight-of-evidence indicates that soil exposures have little impact on human body burdens of dioxin-like compounds (DLCs), with dietary sources being the greatest contributor. In spite of this, USEPA recently proposed substantially lower preliminary remediation goals (PRGs) for soil based on their new oral reference dose (RfD) for dioxin. As such, it is important to understand how these lower soil PRGs compare to background concentrations in urban/suburban and rural soils. The objective of this evaluation was to conduct a comprehensive review of available data concerning background levels of DLCs in U.S. soils. There was substantial variability in how the soil dioxin data were presented (e.g., raw vs. summary data, congener vs. toxic equivalency [TEQ] concentration, number of DLC congeners reported, etc.). In cases where TEQ estimates were based on outdated TEFs and congener-specific data was provided, TEQ concentrations were recalculated using the current WHO2006 TEFs. The data available for rural soils were generally more robust than for urban/suburban soils. Not surprisingly, background levels of DLCs in urban/suburban soils were higher and more variable than in rural soils: 0.1-186 vs. 0.1-22.9 ng/kg TEQ, respectively. In several cases, incomplete soil DLC data were available (e.g., DL-PCBs not included) and, as such, calculated TEQ concentrations likely underestimate actual background levels. Though the current data are somewhat limited, these findings indicate that background DLC concentrations in urban/suburban soils may exceed the USEPA's updated PRGs based on the oral RfD, and are expected to substantially exceed future PRGs to be developed based on the forthcoming dioxin cancer slope factor. This demonstrates a need to characterize anthropogenic background DLCs in non-rural areas across the US to avoid establishing soil screening levels and PRGs that are lower than background concentrations.