Per and polyfluoroalkyl substances (PFAS) in marine and anadromous species sampled in the Southern Salish Sea

In 2013 the Washington Department of Fish and Wildlife (WDFW)’s Toxics
Biological Observation System (TBiOS) began a preliminary study to
determine the need for monitoring of per- and polyfluoroalkyl substances
(PFAS) in response to concerns about the potential exposure in Puget
Sound. Here we make publicly available a dataset for 13 PFAS that TBiOS
tested for in four indicator species: caged bay mussels, juvenile and
adult Chinook salmon, Pacific herring, and English sole. These data were
collected during the winter and spring of 2013, 2016, 2017, and 2018 and
included 126 samples spanning 64 unique sites. Samples consisted of
individuals (adult Chinook salmon) or composites of multiple individuals
(mussels, herring, English sole, and juvenile Chinook salmon), and
included various matrices: muscle, liver, and whole bodies. All samples
were tested for 13 distinct PFAS chemicals by SGS AXYS using Method
MLA-043 Revision 8. The data made available here were filtered to remove
samples that did not pass quality control specifications. This included
the removal of K, H, or MAX data which indicated qualitative concentration
estimates or that surrogate or authentic recoveries were not within the
method control limits set by the analytical laboratory. SGS AXYS provided
one method blank for every lab set, and each lab set consisted of a
maximum of 20 composite samples. No analytes were detected in the method
blanks in this dataset. These data have been referenced in the Washington
State PFAS Action Plan and were also contributed to the publication by
James et al., in 2023: “The screening and prioritization of contaminants
of emerging concern in the marine environment based on multiple biological
response measures.” Analytical methods and reporting limits have improved
markedly for PFAS since these preliminary studies were conducted and
expanded SGS AXYS/EPA methods now cover 40 analytes. Sample hold times,
the length of time between sample collection and final chemical analysis,
were also variable. For these reasons, we caution that this preliminary
dataset is best suited for prioritizing future work rather than for
definitive conclusions in risk assessment.