From Emissions to Ecosystems: PFAS Bioaccumulation and its Drivers in Aquatic Ecosystems

Abstract

Per- and polyfluoroalkyl substances (PFAS) are a large and chemically diverse class of highly fluorinated, anthropogenic organic chemicals. PFAS have been detected in environments globally, which has raised significant concern about crossing planetary boundaries for chemical pollution. Exposure to PFAS has been linked to many adverse human health impacts, including immune deficiencies, metabolic disorders, and various cancers. Many PFAS bioaccumulate in aquatic food webs, and fish and other shellfish are a major dietary source of PFAS in many populations. However, PFAS exhibit unique bioaccumulation behaviors that vary across classes and structures, and much remains unknown about the mechanisms driving these bioaccumulation patterns across compounds, organisms, and environments. While some US states have developed fish consumption advisories focused primarily on long-chained perfluoroalkyl acids (PFAA) with eight or more perfluorinated carbons (≥C8), voluntary and regulatory-driven phase-outs have led to production shifts towards short-chained PFAS and novel PFAS chemistries that are even less well understood. There is therefore a need for improved tools and monitoring data to understand the drivers of observed PFAS bioaccumulation patterns in aquatic ecosystems, which can inform our understanding of ecosystem responsiveness to current and potential future management actions. In this thesis, I use a combination of modeling and analytical approaches to characterize patterns of PFAS bioaccumulation in aquatic ecosystems and investigate its chemical, biological, and environmental drivers. In chapter 2 of my thesis, I adapt a well-established food web bioaccumulation model for neutral POPs to predict the bioaccumulation behavior of six PFAA and two perfluoroalkyl ether acids (HFPO-DA, 9Cl-PF3ONS) produced as PFAA replacements. The new model includes sorption to blood plasma proteins and phospholipids, empirically parameterized membrane transport, and renal elimination for PFAA. Improved performance relative to prior models without these updates is shown by comparing simulations to field and lab measurements. Results suggest bioaccumulation of ≥C8 PFAS is primarily driven by phospholipid partitioning, and that renal elimination is negligible for these compounds. However, specific protein binding mechanisms are important for reproducing the observed tissued concentrations of many shorter-chain PFAA, including protein transporter-mediated renal elimination. Results also highlight the importance of dietary exposure for aquatic food webs, particularly for turbid or benthic ecosystems. For ≥C8 PFAS (e.g. C8 perfluorosulfonic acid (PFSA), C10-C11 perfluorocarboxylic acid (PFCA), 9Cl-PF3ONS), which are often the most abundant PFAS in aquatic food webs, the new model reproduces observed bioaccumulation potential within a factor of two for \textgreater{}80% of fish species. These results indicate the readiness of this model to support development of fish consumption advisories for these compounds. In chapter 3, I examine PFAS bioaccumulation trends across broader range of legacy and novel PFAS compounds to evaluate their response to regulatory interventions. I apply the bioaccumulation model developed in chapter 2 along with statistical modeling approaches to investigate drivers of the observed patterns. Specifically, I examine temporal trends in extractable organofluorine (EOF), targeted PFAS, and tentatively identified PFAS detected by suspect screening in juvenile male North Atlantic pilot whales from the Subarctic between the late 1980s to 2023. The majority (\textasciitilde{}85%) of EOF could be explained by a relatively limited subset of targeted PFAS including the eight perfluorinated carbon chain (C8) sulfonamide (FASA), its terminal acid perfluorooctane sulfonate (PFOS), and C9 and greater PFCA. Concentrations of both EOF and targeted PFAS have significantly declined by over 60% since their peak concentrations in the early 2010s and early 2000s, respectively. Among individual PFAS, the timing of declines varies based on PFAS class, which drives the primary pathway of transport (atmospheric or oceanic), as well as chain-length dependent relationship, which affects fate and transport in open ocean ecosystems. For ionizable PFAS transported primarily by oceanic circulation, a decadal scale lag between declines in chemical production and PFAS concentrations was observed in pilot whale liver and muscle tissues. Among all targeted and suspect PFAS detected, only short chained sulfonamides (C4-C5 FASA), which are still widely used in modern commerce, showed significant increases over the study period. These findings reinforce the efficacy of shifts in chemical production and stringent global regulations for protecting ecosystem and wildlife health. Overall, this work enhances our understanding of the drivers of PFAS bioaccumulation in aquatic ecosystems, as well as their responsiveness to regulatory interventions. The tools and insights developed in this work can inform the development and prioritization of management strategies to mitigate human and ecosystem exposures to both legacy and future releases of PFAS.