Five general groups of threatened and endangered receptors were identified based on where animals live and what they feed on: terrestrial plants, terrestrial invertebrates, terrestrial wildlife, aquatic wildlife and aquatic biota (fish, aquatic invertebrates, aquatic plants, amphibians). Risk-based screening levels can be developed for each group following U.S. Environmental Protection Agency (EPA) methodologies.

• For aquatic biota, such as fish, aquatic invertebrates and aquatic plants, recommended water quality risk-based screening levels were assumed to represent risk-based screening levels protective of aquatic life. These are derived by applying the EPA Great Lakes Initiative methodology to toxicity data identified in the literature.
• For terrestrial plant and invertebrate receptors, EPA methodology uses literature-based allowable concentrations in soil to develop risk-based screening levels, referred to as soil screening levels in this report.
• To develop terrestrial and aquatic wildlife risk-based screening levels, the EPA combines toxicity benchmarks and estimates of potential exposure. Wildlife toxicity benchmarks, called toxicity reference values, are developed separately for birds and mammals and are derived based on toxicity information identified in literature. Potential exposures are modeled based on the dietary preferences of each receptor and information on the bioaccumulation of PFAS from soil, sediment, and surface water into the dietary items comprising the diet of each receptor. 

Following a comprehensive literature search for aquatic toxicity data, recommended water quality risk-based screening levels protective of aquatic plants, aquatic invertebrates, fish, and amphibians were calculated following the EPA’s Great Lakes Initiative approach and the compiled aquatic toxicity data. Screening levels were developed for 23 PFAS. The majority of available toxicity data pertained to perfluorooctanesulfonic acid (PFOS) and perfluorooctanoic acid (PFOA), lending higher confidence in these values. Other PFAS screening levels are based on few toxicity data and, therefore, are more likely to evolve as additional toxicity information become available.

Terrestrial and aquatic toxicity data from a comprehensive literature review were collected to support derivation of plant and invertebrate soil screening levels and bird and mammal riskbased screening levels. Reptile toxicity data were not identified and is an important focus for future PFAS toxicity work.

Terrestrial plant and terrestrial invertebrate soil screening levels were calculated as geometric means of the available toxicity data based on growth, reproductive, or survival endpoints.

Data on the direct toxicity of PFAS to threatened and endangered species and on the exposure of threatened and endangered species to PFAS are very limited, necessitating the use of surrogate species selected to be representative of threatened and endangered species when developing bird and mammal risk-based screening levels. Exposure parameters for each species, concentrations of individual PFAS in dietary items based on bioaccumulation factors, and calculated toxicity reference values comprise the basis of the model to determine final bird and mammal risk-based screening levels. These screening levels were calculated for each media relevant to a terrestrial receptor (soil and surface water) and each media relevant to an aquatic receptor (sediment and surface water). Risk-based screening levels for mammals were calculated for six PFAS: perfluorononanoic acid, PFOS, PFOA, perfluorohexanoic acid, perfluorobutane sulfonic acid (PFBS), and pentafluorobenzoic acid (PFBA).

Risk-based screening levels for birds were calculated for two PFAS: PFOS and PFBS. The PFOS no observed adverse effect level-based risk-based screening levels were the lowest of the six PFAS for soil, sediment, and surface water. Among all six PFAS, the insectivore or invertivore receptors were most often associated with the lowest calculated screening level. The lowest surface water no-observed adverse effect level risk-based screening levels protective of wildlife were lower than the chronic recommended water quality risk-based screening levels for all six PFAS except for PFBA. Thus, the final appropriate screening level for surface water, either wildlife or recommended water quality risk-based screening level, will depend on the presence of sensitive aquatic bird or mammal receptors. 

The screening levels calculated in this evaluation represent default screening values for Superfund risk assessments at DoD sites. These values will provide an initial assessment of the potential ecological risk associated with PFAS measured at various sites and assist with prioritizing areas of potential concern. In combination, the recommended water quality risk-based screening levels, plant and invertebrate soil screening levels, and wildlife riskbased screening level for surface water, sediment, and soil account for each environmental media that would be sampled and each feeding guild that would be considered in an ecological risk assessment. (Project Completion - 2020)