Formatted Title

It's Raining PFAS: A Nationwide Study of PFAS in Rain

Background/Objectives

Poly- and perfluoroalkyl substances (PFAS) have elicited widespread interest due to their ubiquity and potential toxicity. The primary concern to date has been PFAS impacts to groundwater from localized point sources such as fire training areas. More recently, the deposition of airborne PFAS has drawn attention due to evidence that some PFAS species can undergo long-range atmospheric transport and result in impacts surface waters, soils, and groundwater (Schroeder et al. 2021). Several recent studies investigating air deposition of PFAS have been published, highlighting the importance of this developing topic:

• Shimizu et al. (2021) collected wet deposition (i.e., rain) and dry deposition samples over a one-year period in Wilmington, NC and found that there was a 20 times greater yearly mass flux of PFAS in rain compared to dry deposition.
• D’Ambro et al. (2021) collected rain samples at five locations over a year near a Chemours facility that emits PFAS. By combining the rain data with atmospheric modeling, the authors concluded that 95% of the mass emitted is deposited greater than 150 kilometers away from the facility.
• Schroeder et al. (2021) collected surface soil and groundwater seep samples from five regions in New England that were downwind from known PFAS sources. The authors found a statistically significant enrichment of PFOA in areas downwind of factories that emitted PFAS during significant rain events.
• Pike et al. (2021) collected rain samples at six locations in Ohio and Indiana and one location in Wyoming and found that there was a significant difference in the profile of PFAS that were measured in the mid-west locations compared to Wyoming.

While the previous studies are significant, the majority are limited because they measured a small number of PFAS analytes (e.g., Shimizu et al. [2021] measured only six analytes, Schroeder et al. [2021] measured 14 analytes, and Pike et al. [2021] measured 15) or because they were limited to specific geographic areas. Due to the limited analyte lists and limited geographic areas in previous studies, there is an incomplete understanding of spatial and temporal variation in PFAS concentrations in rain across the United States.

Approach/Activities

The purpose of this presentation is to share results of a study monitoring PFAS in rain at six locations across the United States: California, Texas, Washington, Michigan, Colorado, and New Hampshire. Because no standard protocol exists for collecting PFAS samples in rain, the project team devised a method using large surface area high-density polyethylene (HDPE) trays. The HDPE tray size allowed the project team to mobilize and collect sufficient sample volume from less than 0.1 inches of rain. Six to seven samples were collected at each location along with equipment blanks to evaluate decontamination procedures. Samples were submitted to Eurofins Sacramento for analysis using their laboratory-defined modified Method 537 expanded list of 72 analytes.

Results/Lessons Learned

PFAS were detected in 66% of samples, which is an increase compared to previous studies that had about 30% detection frequency, likely due to the use of an expanded analyte list. Analytes that are not included in Method 1633 dominated at many locations, with the percent of total non-1633 analytes ranging from 18% to 100%. In 11 samples, non-1633 analytes were the only PFAS detected. The most frequently detected analyte was perfluoropropionic acid (PFPrA), an ultrashort chain perfluorocarboxylic acid. PFPrA was also the most frequently detected compound in two previous studies evaluating PFAS in bottled water (Chow et al., 2021) and in water supplies used for drinking water (Pelch et al., 2023). A preliminary statistical evaluation on PFPrA concentrations suggest that there was no significant difference in PFPrA concentrations across locations. Two hypotheses for this outcome are proposed by the project team. The first hypothesis is that there are a wide range of sources of PFPrA across the country. This hypothesis is possible because, while the full range of PFPrA sources is unknown, it is a known breakdown product of chlorofluorocarbon (CFC) replacement compounds that were introduced in response to the Montreal Protocol. The second hypothesis is that the mobility of PFPrA results in its widespread distribution in the atmosphere.

“Total PFAS concentrations” calculated as the sum of 72 analytes varied in space and time, ranging from less than 1 nanogram per liter (ng/L) to more than 240 ng/L. In samples where PFAS were detected, 32% detected compounds with proposed federal Maximum Contaminant Levels (MCLs), including perfluorononanoic acid (PFNA), perfluorooctanoic acid (PFOA), perfluorooctane sulfonate (PFOS), and hexafluoropropylene oxide-dimer acid (HFPO-DA, also known by its trade name GenX). The majority of samples did not exhibit concentrations exceeding proposed MCLs, but one sample in Michigan contained 33 ng/L PFOS, exceeding the proposed federal MCL by more than eight times.

The results indicated that multiple analytes that are not included on typical PFAS analyte lists are widespread in rain throughout the United States, suggesting that current monitoring efforts may be inadequate to fully understand the scale of PFAS contributions from nonpoint sources. The results also found significant temporal and spatial variability in the type and concentrations of PFAS detected, suggesting that local and regional point sources contribute to PFAS in rain, and that site- or region-specific rain monitoring is necessary to evaluate contribution of PFAS from nonpoint sources. Finally, the results highlight the significance of nonpoint source contributions of PFAS to the water cycle.