Bioaccumulation of poly- and perfluoroalkyl substances (PFASs) and mercury in European perch (Perca fluviatilis) : Spatial distribution and forest clear-cut (CC) effects in Swedish lakes

Abstract: Recently poly- and perfluoroalkyl substances (PFASs) have gained increasing attention due to their ubiquitous distribution in the environment and adverse effects to human and wildlife. PFASs have unique chemical and physical properties such as surfactant characteristics and are used for many industrial applications (e.g. firefighting, textile processing etc.). PFASs are not easily replaceable thus there have been efforts to use rather short chain PFASs than long chained ones (chain length of six or more perfluorinated carbons). Short chained PFASs are less likely to be toxic and bioaccumulative but still complex mixtures of PFASs can be found in animals and the environment. Little is known about the environmental dynamics of PFASs and how Hg and specific PFASs correlate in biota. This study aims to assess interactions and accumulation patterns of PFASs and Hg in freshwater fish and if forest clear-cut (CC) and other factors (e.g. age, δ13C etc.) can influence bioaccumulation. In this study samples of European perch (Perca fluviatilis) were taken after CC from five different lakes throughout Sweden (Björntjärn, Brobo-Kroktjärn, Gårdsjön, Långtjärn and Umeå 4) and from lake Kroktjärn European perch samples were collected before and after forest CC. In addition, the correlation between Hg and PFASs was investigated using Hg data from a previous master thesis study. The results showed that long chained PFCAs (PFUnDA, PFTriDA, PFDoDA, PFDA and PFTeDA) and PFOS were most present in all samples with detection frequencies between 68 and 99%. Also PFAS levels in fish were not elevated after CC with ΣPFASs of 1.23±0.622 ng/g ww (mean±standard deviation) before CC compared to 1.13±0.146 ng/g ww after CC and showed no significant difference (ANOVA, R2=0.0012, p>0.05). Furthermore, one conclusion in this study is that PFAS levels in fish were impacted by atmospheric deposition with a decreasing gradient from south (urban/industrial areas) to north (remote areas) (ANOVA, R2=0.72, p<0.0001). Fish from lakes located in more populated areas (lake Gårdsjön) showed elevated levels of PFASs with a maximum of ΣPFASs=4.2 ng/g ww. Moreover, the PFASs-latitude correlations showed significant correlations for all PFASs (except 6:2 FTSA) (p<0.0001), indicating that the location of the lake is important for the accumulation of PFASs in fish. All PFASs correlated significantly with each other (except 6:2 FTSA) (p<0.0001). Four significant correlations were found between PFASs (PFDA (p<0.001), PFUnDA, PFDoDA and ΣPFCAs (p<0.05)) and Hg, thus they might have similar accumulation and transport pattern.