Answers are grounded in this paper and cite their sources.
Understanding background concentrations of chemicals in the environment can be important for delineating the extent of impacts from known or suspected sources and for assessing cleanup responsibilities. Recent statewide surveys carried out in Vermont (VT), New Hampshire (NH), Maine (ME), and Massachusetts (MA) measured per- and polyfluoroalkyl substances (PFAS) in soils to better understand the presence of anthropogenic background concentrations. We combined data from these statewide surveys and used multivariate statistical analysis techniques to compare PFAS fingerprints in surface soils (top 0–6 inches) across the four states. Total PFAS concentrations in the combined dataset ranged from 153 to 34,034 ng/kg, with a mean of 3,945 ng/kg. The results show that PFAS fingerprints in the New England region were generally similar, suggesting influence from widespread, diffuse sources of PFAS. Principal component analysis (PCA) showed that perfluorooctanesulfonic acid (PFOS), perfluorooctanoic acid (PFOA), and perfluorobutanoic acid (PFBA) were the main contributors to the variability in the combined dataset. Hierarchical cluster analysis (HCA) confirmed the findings from PCA and identified three distinct clusters: one predominated by PFOS, another by PFOS and PFOA, and a third by PFOS, PFOA, and PFBA. The spatial variations observed in the clusters suggest that in some areas, background PFAS concentrations may reflect a combination of diffuse global and regional PFAS signatures and may additionally suggest differential PFAS transport in shallow soils due to age-related weathering. Incorporation of Total Oxidizable Precursor (TOP) assay data into PCA for the NH samples showed a notable increase in PFBA following oxidation, indicating that a portion of the background PFAS loading may be due to transformation of shorter-chain precursors. Overall, the results indicate that the source of background PFAS in New England soils is consistent with long-range atmospheric transport.
No takes yet. Share an insight, caveat, or question.
Mozumder et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: