Mercury in fish from the North East Atlantic: sources, bioaccumulation dynamics and co-occurrence with selenium

Abstract

Mercury (Hg) is a global neurotoxin distributed at trace levels in the earth’s crust. Although Hg input from anthropogenic sources has been reduced in North America and Europe, in some other parts of the world the emission is still high. Considering the longrange transport and long atmospheric half-life, Hg and particularly its most toxic form monomethylmercury (MeHg), remains an environmental concern at the global level causing threat to both wildlife and human health. In general, seafood is the main source of MeHg exposure to humans and Hg is the main reason for seafood consumption advisories. Therefore, measuring the Hg levels in seafood species and understanding the processes governing the variation of Hg levels are very important for seafood safety and security. Synthesis, bioaccumulation and biomagnification of MeHg are very critical processes controlling the MeHg levels in the environment and the biota. The main goals of this study were to investigate how Hg levels vary between different fish species as well as between different communities in offshore, fjord and coastal areas of the North East Atlantic Ocean (NEAO). The contribution from different Hg sources and parameters influencing these variations were also investigated. Large variation in Hg levels between fish species form NEAO was found (Paper I). The pelagic species including Atlantic mackerel (Scomber scombrus) and blue whiting (Micromesistius poutassou) with mean value of 0.04 mg kg-1 ww had the lowest Hg concentrations. Blue ling (Molva dypterygia) had the highest Hg levels with a mean of 0.72 mg kg-1 ww. Selenium (Se) varied in a smaller range compared to Hg, with mean concentrations from 0.27 mg kg−1 ww in Atlantic cod (Gadus morhua) to 0.56 mg kg−1 ww in redfish (Sebastes spp.). The Hg level in fish increased from north towards south in most species and this process was independent of Hg pollution in the environment (sediment). It was hypothesized that a gradual increase in water temperature and primary production duration from the north towards the south are the main parameters governing the intraspecific geographical variation. Fish species collected from fjords and coastal areas contained higher Hg levels compared to the same species sampled offshore. High levels of organic matter and atmospherically deposited Hg washed from the catchment inducing high MeHg production and high Hg bioavailability in fjords and coastal areas were suggested as the main drivers. The Hg variation between species was mostly driven by Hg trophodynamics and δ15N as a proxy for trophic position explained the Hg variation between fish species from different areas in NEAO. The results indicated that the fjord and coastal areas and the Barents Sea had lower Hg levels at the base of the food web while demonstrating higher trophic magnification rates compared to the other areas (PhD thesis). In order to investigate the effect of an industrial point source on environmental Hg levels in a fjord, levels of Hg and MeHg were measured in seawater, sediment and seafood species (fish and crustaceans) close to industrial point source of Hg pollution in Hardangerfjord ecosystem (Paper II). Elevated levels of Hg and MeHg were found in all compartments with increasing levels towards the point source in Sørfjord. However, in predatory species, tusk (Brosme brosme), Hg was accumulated at the same level in the sidearm of Eidfjord, where Hg contamination in sediment is low. Thus, organic matter and atmospheric Hg from the catchment area were suggested as other important drivers of Hg variation in biota in fjord ecosystems. In a continuation of this study, a similar investigation was conducted in Sognefjord with no major pollution source. There, Hg in tusk increased from offshore North Sea to the coast and further into outer and inner Sognefjord, while Hg levels measured in sediment samples were at the background level. Measurements of δ13C, as a proxy for energy/carbon source, showed that the contribution of allochthonous carbon to the food web increases towards the inner fjord and explained the majority of the Hg variation in tusk (Paper III). It is suggested that surplus Se may provide protection against Hg toxicity for consumers. In most fish species from NEAO, Hg and Se were correlated and particularly in species with high Hg levels, this correlation was stronger (Paper I). All species from NEAO on average had higher molar concentrations of Se than Hg, and Se health benefit values (HBVSe) were above 2. In predatory species including tusk and blue ling from the inner part of Hardangerfjord, mean Hg levels were above the European maximum level (EUML) and the HBVSe were negative, indicating higher molar concentration of Hg than Se with a relatively high risk for consumers. Although tusk from Sognefjord also had mean Hg levels exceeding the EUML, while HBVSe values were above 3 in the fillet. Overall in the NEAO, only blue ling had an average Hg level above EUML, and the Hg exposure assessment showed that consumers having two servings of blue ling, tusk and/or Atlantic halibut per week will exceed the tolerable weekly intake (TWI) of MeHg. Consumption of all species from NEAO except Haddock (Melanogrammus aeglefinus), common ling (Molva molva), tusk and blue ling, on average provide more benefit from essential fatty acids than risk from MeHg. In the Hardangerfjord and Sognefjord studies both total mercury (THg) and MeHg were measured in tusk fillet and liver. The MeHg to THg ratio (%MeHg) decreased when THg levels increased in tusk fillet and liver in both fjords, indicating MeHg demethylation as a response to MeHg accumulation (Paper II and III). Our results suggest that inorganic Hg (iHg) produced from MeHg demethylation can bind Se and be stored in fish liver. Discovering the details of demethylation process in marine fish may help better understand the Hg fate and cycling in the food web with implication for food safety and security.