Accepted Preprint
Environmental Toxicology A BROAD COCKTAIL OF ENVIRONMENTAL POLLUTANTS FOUND IN EGGS OF THREE SEABIRD SPECIES FROM REMOTE COLONIES IN NORWAY
SANDRA HUBER, NICHOLAS A. WARNER, TORGEIR NYGÅRD, MIKAEL REMBERGER, MIKAEL HARJU, HILDE T. UGGERUD, LENNART KAJ, and LINDA HANSSEN
Environ Toxicol Chem., Accepted Article • DOI: 10.1002/etc.2956
Accepted Article
"Accepted Articles" are peer-reviewed, accepted manuscripts that have not been edited, formatted, or in any way altered by the authors since acceptance. They are citable by the Digital Object Identifier (DOI). After the manuscript is edited and formatted, it will be removed from the “Accepted Articles” Web site and published as an Early View article. Note that editing may introduce changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. SETAC cannot be held responsible for errors or consequences arising from the use of information contained in these manuscripts.
This article is protected by copyright. All rights reserved
Accepted Preprint
Environmental Toxicology
Environmental Toxicology and Chemistry DOI 10.1002/etc.2956
A BROAD COCKTAIL OF ENVIRONMENTAL POLLUTANTS FOUND IN EGGS OF THREE SEABIRD SPECIES FROM REMOTE COLONIES IN NORWAY
Running title: A broad cocktail of environmental pollutants in seabirds
SANDRA HUBER, *† NICHOLAS A. WARNER, † TORGEIR NYGÅRD, ‡ MIKAEL REMBERGER, § MIKAEL HARJU, † HILDE T. UGGERUD, || LENNART KAJ, § and LINDA HANSSEN †
† Department of Environmental Chemistry, Fram Centre, Norwegian Institute for Air Research, Tromsø, Norway ‡ Norwegian Institute for Nature Research (NINA), Trondheim, Norway § Swedish Environmental Research Institute (IVL), Stockholm, Sweden
|| Department of Environmental Chemistry, Norwegian Institute for Air Research, Kjeller,
Norway
*Address correspondence to [email protected]
Additional Supporting Information may be found in the online version of this article.
This article is protected by copyright. All rights reserved Submitted 8 October 2014; Returned for Revision 17 February 2015; Accepted 22 February 2015
This article is protected by copyright. All rights reserved
Accepted Preprint
Abstract: Three seabird species, Common eider (Somateria mollisima), European shag
(Phalacrocorax aristotelis aristotelis), and European herring gull (Larus argentatus), were selected to survey for a broad range of legacy and emerging pollutants to assess chemical mixture exposure profiles of seabirds from the Norwegian marine environment. A total of 201 chemical substances were targeted for analysis ranging from metals, organotin compounds, polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and associated metabolites, chlorinated paraffins, chlorinated and non-chlorinated organic pesticides, per- and polyfluroalkyl
substances (PFAS), dechlorane plus, octachlorostyrene, brominated flame retardants, organophosphorous compounds (OPCs), brominated and alkyl phenols, cyclic siloxanes, and
phthalates. Of the chemicals targeted, 149 substances were found above the detection limits with metals dominating the contaminant profile and comprising 60% of the total contaminant load. PCBs, pesticides, OPCs and PFAS were the dominant contaminant classes of organic pollutants found within seabird species with the highest loads occurring in herring gulls followed by shag and common eider. New generation pollutants (e.g., PFAS, OPCs and some alkylphenols) were detected at similar or higher concentrations than the legacy persistent organic pollutants (POPs). Time trends of reported concentrations of legacy POPs appear to have decreased in the last decades from the Norwegian coastal environment. Concentrations of detected pollutants do not appear to have a negative effect on seabird population development within the sampling area. However, additional stress caused by pollutants may affect seabird health more at the individual level. This article is protected by copyright. All rights reserved
Keywords: Contaminants, Emerging pollutants, Persistent organic pollutants, Heavy metals, Ecotoxicology
This article is protected by copyright. All rights reserved
Accepted Preprint
INTRODUCTION The industrial revolution made a significant impact on the living standards for humans.
Particularly the use of man-made chemicals, which continue to be produced and used for various industrial and household applications, have provided safer working and living environments (e.g.
flame retardants), simplified daily living tasks (e.g. Teflon coated cookware) or increased protection against unwanted organisms (e.g. pesticides). However, large-scale use of such compounds has resulted in release and exposure to a broad mixture of chemicals which have, and still may significantly impact ecosystem health. Effects of "cocktail" exposure have already been documented for selected compound groups (e.g., PCBs) [1] and highlight the possible risk which ecosystems are facing from exposure to various mixtures of chemicals. In recent years, cocktail
mixtures of chemicals and environmental pollutants have received growing attention. Evidence has shown exposure to single compounds that do not induce an adverse effect can give negative synergistic and additive effects in organisms when exposure occurs as part of a mixture [2-6]. Environmental pollutants, such as polycyclic aromatic hydrocarbons (PAHs), polychlorinated
biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs) and organochlorine pesticides are classified as harmful compounds and the majority of them have been banned from the market in most countries due to their persistence, toxicity and potential for bioaccumulation in the entire food chain. These legacy persistent organic pollutants (POPs) have been under investigation for
decades, but toxicity mechanisms are still not fully understood, thus driving research to further investigate the knowledge gaps surrounding potential cocktail effects. Research on newer generation pollutants such as: per- and polyfluoroalkyl substances (PFASs), alternative brominated flame retardants, organophosphorous compounds (OPCs), alkyl- and bromophenols, siloxanes or phthalates has not come as far compared to the legacy POPs. However, their risk to
This article is protected by copyright. All rights reserved
Accepted Preprint
environmental health is important to investigate. Exposure to inorganic contaminants (i.e.,
metals) must also not be ignored. Natural exposure to metals occurs where they can act as essential components in the functioning of environmental and biological processes. However, many human activities including mining and refining have resulted in greater release of metals into the environment, potentially inducing acute or chronic toxicity. Exposure to metals at low concentrations is also of concern as bioaccumulation of several metal species has been well
documented [7,8]. How interactions between these different groups of chemicals will affect biological organisms is unclear. This highlights the importance of identifying the ingredients in these cocktail mixtures for future toxicity assessment. Accessibility to tools of non-target analysis for identification of new and unknown pollutants has become easier in recent years as new technology and public available databases have entered the market. However, the use of this approach has limitations due to its complexity of raw extracts and enormous time consumption in interpretation of large data sets together with relatively high thresholds which do not allow a detection of analytes present in low concentrations. Therefore, targeted analysis must still be
relied upon to help address questions surrounding chemical composition and potential effects posed by pollutant cocktail mixtures. In the present study, an assessment of the state of pollution at two Norwegian remote coastal
sites was carried out to evaluate the chemicals to which wildlife may be exposed. Remote islands were chosen in order to estimate background concentrations and to exclude direct anthropogenic influence on pollution from neighboring settlements. Three wide spread seabird species in
Norway, Common eider (Somateria mollisima), European shag (Phalacrocorax aristotelis aristotelis), and European herring gull (Larus argentatus), were selected, as they represent different migration patterns and overwintering strategies, thus representing different exposure
This article is protected by copyright. All rights reserved
Accepted Preprint
scenarios. A survey for a broad range of legacy (i.e., POPs and metals) and emerging pollutants was done. Temporal patterns on contaminant load and distribution within seabirds were also
investigated for selected contaminants (e.g. PCBs, organochlorine pesticides, PFAS) to assess pollution status along Norwegian coastlines. To our knowledge, this is the first time that such a comprehensive study on a broad spectrum of environmental pollutants on seabird populations has
been performed. MATERIAL AND METHODS Bird species and characteristics The Common eider (Somateria mollisima) is distributed all over the Norwegian coastline
during its breeding season and the birds from the sampled colonies in this study overwinter in the same region) [9]. The Common eider is a mid-trophic predator of the marine benthic food web with its diet consisting mostly of crustaceans and mollusks. The European shag (Phalacrocorax aristotelis aristotelis) preys entirely upon small marine
fish. It inhabits the Norwegian coastline throughout the entire year [9]. The European herring gull (Larus argentatus) is an opportunistic species filling a broad
ecological niche with varied feeding patterns and is a common bird species along the Norwegian coast. The herring gull is often used for monitoring studies due its widely spread presence as for example the American herring gull (Larus smithsonianus) at the Great Lakes in North America
[10].
Sampling areas and sampling Seabird eggs from the selected species were collected at two remote islands (Figure 1),
Sklinna (11º 00’E, 65º11’N) and Røst (12º00’E, 67º30’N) on the Norwegian coast during the
breeding season between May and June 2012. These sampling sites have been utilized over
This article is protected by copyright. All rights reserved
Accepted Preprint
several decades to monitor seabird population health in Norway. Monitoring contaminant levels in seabird eggs is an excellent method for evaluating contaminant exposure to seabirds [11-13].
This provides a means of assessing contaminant exposure, particularly in sensitive populations, without sacrificing birds. Within the present study, six and 12 eggs were collected for each species at Sklinna and Røst, respectively. Sampling date, species, coordinates on geographical location (data not shown) and clutch size were noted. Length, width, weight and condition of the eggs were determined prior to the homogenization of sample material (Supplemental data Table S1). Eggs were individually placed in polyethylene bags and stored at +4ºC in a refrigerator until transport to the Norwegian Institute for Nature Research (NINA) in Trondheim.
Homogenisation of sample material Homogenisation of the sample material was performed at an outdoor site beside the Fram
Centre (Tromsø, Norway) in order to avoid cross contamination of problematic indoor air
contaminants (i.e., siloxanes and phthalates). Personnel handling the samples refrained from using personal care products before and during the homogenisation process and sample preparation. Egg content was classified by embryonic development and condition (Supplemental data Table S1). Three eggs were pooled and homogenised in a glass beaker with a stainless steel Ultra-turax homogeniser. Pooled samples AS-1 and AS-2 were homogenised with a stainless steel mixer because of large and partially developed embryos. All equipment was cleaned thoroughly between each sample with ultrapure water, acetone and then hexane to avoid cross contamination between samples. Additional procedures were taken by using a pooled sample of three homogenised hen eggs to provide a field blank for the homogenisation process. Hen egg homogenate was treated the same way as the seabird eggs and was used to assess any
This article is protected by copyright. All rights reserved
Accepted Preprint
contamination introduced during the homogenisation process. All sample containers were frozen at -20ºC and shipped to the respective laboratories for analysis. Analysed parameters A total of 201 different chemical substances were analysed, which included 11 metals, four
organotin compounds and 186 organic compounds belonging to the groups of polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and associated metabolites, chlorinated paraffins, chlorinated and non-chlorinated organic pesticides, per- and polyfluoroalkyl substances (PFASs), dechlorane plus, octachlorostyrene (OCS),
organophosphorous chemicals (OPCs), brominated flame retardants (BFRs), bromo- and alkyl phenols, siloxanes, and phthalates (see Supplemental data Table S2 for details). Analysis of inorganic and organic compounds was performed in the laboratories of the
Norwegian Institute for Air Research (NILU) at Kjeller and Tromsø, Norway, and the Swedish
Environmental Research Institute (IVL) at Stockholm, Sweden. In addition, extractable organic material (EOM) and stable isotopes of N and C were assessed. Stable isotope measurements were performed at the Institute of Energy Technology at Kjeller, Norway, and EOM estimations at NILU Kjeller. Sample preparation and analyses For details on sample preparation and analyses see Supplemental data S1. The eggs were
investigated for moisture loss [14,15] since fresh and rotten eggs and eggs containing embryos were collected (Supplemental data Table S1). All eggs had a specific gravity > 95 and therefore an adjustment for loss of moisture and lipid was not necessary [15].
This article is protected by copyright. All rights reserved
Accepted Preprint
Statistical method Interspecific differences in egg concentrations were assessed using IBM SPSS © v.22
statistics software. Conservative non-parametric tests (Kruskall-Wallis and Mann-Whitney) were applied.
Temporal trends Temporal trends of environmental pollutants were estimated by collecting data from previous
published studies. Results have to be interpreted with care since analyses were conducted at different laboratories and numbers of eggs included in the different studies were varying thus detailed information on single concentrations was not always available to assess statistical significance of temporal trends. Principal component analysis Principal component analysis (PCA) was applied in order to investigate relations between all
variables in Simca (vs 11.5, Umetrics Inc). In this case, the loading bi-plots were used to display both the concentration levels of environmental pollutants and stable isotopes (applied as quantitative data/variables) expressed as correlation coefficients and scores related to species and locations (expressed as passive variables). The chemical concentration and isotope data were preprocessed by using univariate scaling. In addition, the chemical concentration data was logarithmically transformed, while birds’ species and location only were considered as passive
variables. Concentrations found as significant outliers and which were outside the Hotellings T2 (95% confidence interval) were excluded from the model to give a better normal distribution of the data. RESULTS AND DISCUSSION Concentrations and distribution patterns between the species
This article is protected by copyright. All rights reserved
Accepted Preprint
In total, 201 different compounds were analysed of which 149 substances were found at
concentrations above the limits of detection (LODs). Results on concentrations are reported in wet weight. Details on average concentrations and concentration ranges are listed in Table 1 and
single concentrations of the individual compounds and pooled samples are listed in the Supplemental data in Table S3 - S13. Metals and organotin Selected metals (Cr, Co, Ni, Cu, Zn, As, Ag, Cd, Sn/organotin, Pb, total-Hg) were found in all
samples analysed with the exception of Sn/organotin (Supplemental data Table S3). Metal distribution patterns were dominated by Zn, Cu, As, and total Hg (Figure 2). Significant differences in mean concentrations between bird species were shown for Zn, Cu and As
(Supplemental data Table S3 and S14). The remaining metals were in most cases close to detection limits and are not further discussed. Dominance of Zn and Cu in the present study was
expected due to their greater abundance within the natural environmental and that they are vital in physiological processes within organisms. However, it cannot be demonstrated if the presented concentrations are related to biomagnification through the food chain or if they are within normal
physiological ranges. In general, the highest metal concentrations were detected in common eider eggs (except for
total Hg; Table 1 and Supplementary data Table S3). Higher concentrations of metals are generally associated with benthic dwelling organisms (i.e., bivalves and crustaceans) due to their high assimilation of metals from environmental and dietary uptake [16]. Thus, higher concentrations found in common eider eggs may be a result of high concentrations of metals in
their prey or a reflection of this species having higher assimilation efficiency and/or lower elimination rates compared to other bird species investigated [16]. Unlike the other elements,
This article is protected by copyright. All rights reserved
Accepted Preprint
higher concentrations of Hg were found in eggs from shag compared to common eiders, but the difference was statistically not significant (Supplementary data Table S14). This is not surprising as Hg can biomagnify through aquatic foodwebs [7,8] as it is extremely bioavailable in its
methylated form. Thus, pelagic seabirds (i.e., shags) are likely to receive greater inputs of Hg through their diet compared to benthic feeding seabirds (i.e., common eiders) due to bioaccumulation through longer food chain length [17,18]. Based on stable nitrogen isotope signatures, herring gull and shag appear to feed at similar
trophic levels (Supplemental data Table S13). However, total Hg concentrations were significant
lower in eggs from herring gulls compared to shag (Supplemental data Table S14 and Figure S1 A). This may be a reflection of differences between species regarding their assimilation/elimination rates as has previously been observed in seabirds from both the
Canadian Arctic and Barents Sea region [19]. The feeding and migratory behaviors between these bird species must also be considered. Shags inhabit and feed along the Norwegian coastline all year round, whereas herring gulls have a much greater migratory range that could reflect Hg concentrations accumulated from a less polluted location. However, concentrations reported for Hg within this study are associated with background levels of Hg and are below lowest observed adverse effect levels (< 0.6 µg/ww) [20]. The concentrations of As found in eggs from common eider are comparable to those found in
common eider eggs from the Aleutian Islands in a previous study [21]. Bioaccumulation of As has been reported in seabirds from Arctic regions, but was found to be region specific. The results of the present study are in agreement to those reported by Borgå et al. (2006) [19] who showed no bioaccumulation potential for As in seabirds from the Barents Sea region.
This article is protected by copyright. All rights reserved
Accepted Preprint
Additional data on metal concentrations at various trophic levels throughout the entire food web is needed to further assess bioaccumulation potential. The results of the present study suggest that diet and assimilation/elimination efficiencies are driving factors in controlling metal concentrations within these bird species. Organic compounds In total 186 organic compounds were analysed and 137 were found above the LODs
(Supplementary data Tables S4-S8).
Legacy POPs The exposure of legacy POPs to seabird eggs takes place through maternal transfer and not
through exposure of the eggs itself. Analysing seabird eggs will therefore give an actual status on exposure of the maternal birds to environmental contaminants. Polychlorinated biphenyls Most of the PCB congeners analysed were above the LODs for all samples. Common eiders
had the lowest average sum concentrations of 10.9 and 76.9 ng/g at Sklinna and Røst respectively. These concentrations were lower than those found in shag (169 and 161 ng/g) and in herring gull (706 and 1012 ng/g; Table 1) with the exception of one pooled common eider sample from Røst (AR-6, Supplemental data Table S4). Differences in PCB sum concentrations were statistically significant between the species (Supplemental data Table S14 and Figure S1 B). In common eiders, penta- and hexa-congeners dominated the congener profile (CB-82 to CB-
169; Supplemental data Table S4). Similar profiles were observed in shag and herring gull eggs where penta-, hexa-, and hepta congeners (CB-82 to CB-192; Supplemental data Table S4) dominated the CB distribution pattern with more than half of the sum PCB load consisting of
hexa-CB congeners (CB-128 to CB-169; Supplemental data Table S4).
This article is protected by copyright. All rights reserved
Accepted Preprint
Previous studies on PCBs in common eider and shag from Sklinna in 2003-2004 showed
higher concentrations of individual PCBs [22]. Similar findings were observed for shag where concentrations were in general higher in 2003-2004, while only CB-187 was slightly higher in
the present study. Time trends indicate stable concentrations for PCBs in eggs from common
eiders since the 1990s (Supplemental data Figure S2 A). For herring gull and shag, PCB concentrations decreased and are approximately 15 – 25% below concentrations reported in the 1970s. Previous studies from shag and herring gull collected at Røst in 1983, 1993 and 2003 indicate a decrease of the sum-CB concentrations in eggs over time [10,23]. Average sum-PCB concentrations in shag samples were 420 and 280 ng/g in 1983 and 1993 respectively [11, 23], and are higher compared to concentrations from 2012 (161 ng/g) reported in the present study. For herring gulls, the decline in concentrations is far greater compared to shags with levels dropping from 7070 ng/g in 1983 [11], to 1530 ng/g in 1993 [23] and 1012 ng/g in 2012 (present
study). The rapid decline between 1983 and 1993 most likely reflects a decrease in exposure due to restrictions placed on production and use of PCBs during this time period. The decline in concentrations between 1993 and 2012 is not as great compared to earlier years, especially for herring gulls, and is likely a reflection of exposure to secondary sources of PCBs and their slow environmental degradation in addition to simple first order kinetics as concentrations get lower and the elimination starts to slow. This findings indicate that bird populations are not endangered
by PCB exposure as effects on reproductive performance have been reported to occur at much higher concentrations (34.1 μg/g ww) [24]. However, the effects observed by Fernie et al. [24] are not directly comparable to the bird species investigated in the present study, since a terrestrial raptor species (American kestrel) was exposed to PCBs in laboratory experiments. If the same
This article is protected by copyright. All rights reserved
Accepted Preprint
effects can be shown in wildlife together with natural exposure pathways and synergistic, antagonistic or additive effects of cocktail mixtures of pollutants is questionable. Metabolites of polychlorinated biphenyls Metabolites of PCBs have over the last decade drawn attention due to their negative
contribution in early development [25]. As with the parent PCB congeners, seabird eggs are also exposed to the associated metabolites of PCBs (MeSO2 and OH- CBs) through maternal transfer [26].
For the MeSO2-CB congener profile of common eider, shag and herring gull the penta and
hexa chlorinated MeSO2-CBs (MeSO2-CBs 101, 110, 132, and 149; Supplemental data Table S5) were the most abundant congeners. This is in accordance with the results for the parent
compounds where the penta- and hexa-CB congeners (CB-82 to CB-169) dominated in all seabird species. MeSO2-CB load was highest in herring gull, followed by shag and common
eider (Table 1 and Supporting data Figure S1 C). The ratio of
19MeSO2-CBs
to parent PCB are an indicator of formation and retention
capacity of MeSO2-CBs among different species. Ten of the parent PCBs were analysed in the present study, where MeSO2-CBs 52, 101 and 149 (3 and 4-MeSO2-CB) accounts for 59% of the 19MeSO2-CB
fraction. Comparison of the MeSO2-CB-concentrations to their respective parent
PCB congener concentration (PCBs 52, 101 and 149) provides a rough approximation of the metabolic transformation. The ratio of
6MeSO2-CBs/ 3ParentPCBs
was 0.20 in common eider
and 0.12 in herring gull eggs, similar to other studies [25-27]. Despite large variations in MeSO2-CB concentrations between species, the ratio of
19MeSO2-CBs
to parent PCB was
consistent, indicating similar formation and retention capacity. However, the ratio in shags was 0.78 which suggest that shag may have a higher expression of CYP-like metabolizing enzymes
This article is protected by copyright. All rights reserved
Accepted Preprint
which leads to MeSO2-CB formation or a higher retention capacity for metabolites in shag. Additional analyses of eggs and plasma samples of all relevant parent PCBs are needed to verify these results. Detailed information regarding the expression of CYP-like metabolizing enzymes between these species would help to provide insight on metabolic differences between these
species.
The feeding patterns/behavior between bird species will also affect accumulation of MeSO2-
CB-metabolites. This is evident where concentrations of MeSO2-CB are higher and PCB
concentrations are significantly lower (i.e. in common eider and shag compared to herring gull). It might be attributed to a combination of feeding pattern differences and a lower total PCB load for shag, as well as a higher metabolisation rate of PCBs. In a study by
.
(2010) [28], a clear difference in metabolic capacity between bird species was observed. This is also supported by previous investigations by Warner et al. (2005) [29] which observed large variation in metabolic selectivity for enantiomers of chiral PCBs between different Arctic seabird species.
The concentrations of the OH-CBs were lower compared to the MeSO2 –CBs in all three bird
species. On average, common eiders had the lowest OH-CB concentrations (Table 1). The most prominent OH-CB was 4-OH-CB 187, comprising on average 64%, 31% and 49% of
10OH-
CBs in the eggs from common eider, shag and herring gull, respectively (Supplemental data Table S5). Organochlorine pesticides A total of 16 organochlorine pesticides (OC-pesticides) were analysed in this study (Supplemental data Table S6). Common eiders showed the lowest Σ OC-pesticide concentrations
with average sum concentrations of 8.4 and 54.8 ng/g at Sklinna and Røst respectively, compared
This article is protected by copyright. All rights reserved
Accepted Preprint
to shag (86.0 and 80.5 ng/g) and herring gull with 380 and 535 ng/g respectively (Table 1, Supplemental data Figure S1 D). The pattern of OC-pesticide concentrations from highest to lowest was found to be similar in all three species: p,p’-DDE > HCB > trans-NC > cis-NC > βHCH > other pesticides (Supplemental data Table S6). Significant statistical differences in concentrations between all species were found for cis-chlordane, cis-nonachlor, β-HCH, HCB, p,p’-DDD and o,p’-DDT (Supplemental data Table S14). Previous studies have reported similar or higher concentration of p,p’-DDT in eggs from
common eiders at Sklinna collected in 2004 and 2005 [22]. o,p’-DDT was not comparable due to
the high LODs in the previous study, whereas for eiders and shags, p,p’-DDE concentrations were slightly higher than in the present study. Time trends indicate stable concentrations for DDTs in eggs from common eiders since the 1990s (Supplemental data Figure S2 B). For herring gull and shag, DDT concentrations decreased and are approximately 15 – 25% below
concentrations reported in the 1970s. Similar as found for PCBs was a decline for p,p’-DDE in shag and herring gull concentrations observed between 1993 and 2012. Concentrations within shag (230 ng/g) and herring gulls (1050 ng/g) in 1983 decreased approximately 50% by 1993 [11,23] followed by a slower decline to 2012 (shag: 50 ng/g, herring gull: 413 ng/g; present study). This indicates that bird populations in this region are not endangered by DDT exposure as effects of eggshell thinning on predator birds have been reported to occur at much higher concentrations (> 1000 ng/g) [30-33]. Similar results were found for p,p’-DDE in ducks, where eggshell thinning is induced at high concentrations (µg/g levels) [33]. Shag levels were
decreasing in minor degrees between 2000 and 2009. β-HCH was the dominating HCH isomer in the present study and was approximately four
times higher in shag and herring gull eggs compared to common eider (Supplemental data Table
This article is protected by copyright. All rights reserved
Accepted Preprint
S6). HCH appear to have stabilized over the last 10 years (Supplemental data Figure S2 E), while chlordanes appear to have decreased within the same time period (Supplemental data Figure S2 D).
In the present study, concentrations of HCB were highest in herring gull followed by shag
and common eider (Supplemental data Table S6). Trends for HCB are difficult to interpret based on temporal data assessments (Supplemental data Figure S2 C), but concentrations of HCB in herring gull have decreased by approximately 80% since the 1970s. Shag eggs from 2004 and 2002 had lower concentrations [22,34] than measured here. This may be a reflection of increased
environmental exposure as air concentrations of HCB have increased over the past decade within the Canadian and Norwegian Arctic [35]. Hypothesized sources of this increase have been due to HCB contamination in current use chlorinated pesticides as well as re-volatilization from secondary sources [35]. However, such sources would also be expected to impact herring gulls and eider ducks in a similar manner, which is not observed. This indicates further monitoring is needed to confirm increasing trends in HCB concentrations within shags and the environment. Unlike HCB, PeCB followed a similar pattern as observed for PCBs with generally lower concentrations in common eider and higher concentrations detected in herring gull (Table 1). Polybrominated diphenyl ethers In the present study, all of the analysed 17 PBDEs were found in the egg samples
(Supplemental data Table S7). Σ PBDE concentrations were lower compared to organochlorine contaminants investigated within this study but followed similar patterns between the species (Table 1 and Figure 3). Lowest concentrations were found in common eiders with average summed concentrations of 1.09 and 0.84 ng/g at Sklinna and Røst respectively, followed by shag
(2.89 and 3.29 ng/g) and herring gull (11.2 and 13.9 ng/g) (Table 1) with concentrations being
This article is protected by copyright. All rights reserved
Accepted Preprint
statistically different between species (Supplemental data Table S14). The concentration levels for BDE-99 and BDE-100 were lower in this study compared to previous results reported in
common eider and shag eggs in 2004 [22]. BDE-153 was slightly higher in shag eggs in 2012 compared to 2004, but not in eider eggs, where the opposite was the case. This might be a
reflection of analytical or spatial variation, but cannot be proven at this time. Murvoll et al. (2006) [34] reported similar PBDE concentrations in shag at Sklinna in 2002 compared to what
was found in a 2004 study [22], while a former study on contaminants in herring gull eggs at Røst in 2003 reported Σ PBDE concentrations as high as 60 ng/g [26]. The latter is approximately 3.5 times greater than the highest Σ PBDE concentration reported in the present
study. BDE-100 was detected in five eggs from Sklinna in 2004 where concentrations were relatively high with up to 1.8 ng/g for BDE-100 [36]. In general, PBDE concentrations measured in eggs collected in 2002 and 2004 were higher than eggs from 2012, which indicates a decline in PBDE concentrations/exposure over time and especially over the last 10 years (Supplemental
data Figure S2 F). Previous studies found a consistent association with reproductive changes in American kestrels and concentrations of PBDE-congeners [37]. PBDE concentrations of the investigated species of the present study were much lower compared to Fernie et al. [37]. However, a direct comparison of effects is not possible (see section on PCBs for details). Polycyclic aromatic hydrocarbons Five of 16 analysed PAHs (naphthalene, anthracene, fluoranthene, pyrene and chrysene) were
found above the LODs. Detected sum-concentrations were relatively low (≤ 1.4 ng/g) with the exception of a herring gull sample from Røst, where naphthalene and pyrene accounted for most
of the 29.6 ng/g (Table 1 and Supplemental data Table S8). PAHs are less prone to bioaccumulation or biomagnification than the organochlorines, partly because of metabolic
This article is protected by copyright. All rights reserved
Accepted Preprint
degradation of PAHs in top predators and their prey [38]. However, a broad range of PAHs in seabirds from different islands surrounding Great Britain have been previously reported [39]. Due to the remoteness of the sampling locations within the present study, concentrations found likely reflect background levels. The islands investigated by Shore et al. (1999) [39] are probably influenced by anthropogenic sources due to their closer proximity to settlements on the mainland of Great Britain. However, in any future work it is recommended to include alkylated PAHs since they have to be found to be of greater importance in egg samples than their parent compounds [40]. New generation and emerging environmental contaminants Per- and polyfluoroalkylated substances. A total of 24 PFAS were analysed within the
present study, with 19 compounds detected above LOD. Unlike findings observed for the legacy
contaminants, individual single PFAS compounds showed similar concentrations between species and sampling locations (Supplemental data Table S9). However, for PFAS sum
concentrations of common eider eggs at Sklinna had lowest average Σ PFAS concentration (11.2 ng/g, Table 1 and Supplemental data Figure S1 G), followed by shag (25.5 ng/g) and herring gull
(58.9 ng/g). In the distribution patterns of PFAS, PFOS (3.92 – 42.3 ng/g) was predominant in all samples followed by PFNA (0.31– 5.64 ng/g), PFUnDA (0.974 – 7.0 ng/g), PFTrDA (0.97 – 8.55 ng/g), PFHxS (0.231 – 1.58 ng/g) and PFDA (0.107 to 1.13 ng/g; Supplemental data Table S9). Differences between species were not statistically significant for most of the PFAS compounds (Supplemental data Table S14). FOSA, a pre-cursor compound for PFOS and PFOA, was detected in most samples with concentrations ranging from below the detection limit to 0.663
ng/g and it was significantly higher in eggs from shag compared to herring gull (Supplemental
data Tables S9 and S14). Results cannot be explained by biomagnification dynamics and/or
This article is protected by copyright. All rights reserved
Accepted Preprint
trophic position of bird species only as PFAS exhibit different properties (hydrophobicity and lipophobicity) and different accumulation pathways compared to the legacy POPs. There is evidence that a majority of the PFAS exposure and uptake occurs through the diet. However, species-specific metabolism of PFAS-precursor compounds will also affect the overall accumulated body burden by presently unknown rates [41]. Previous studies conducted in this region also found PFOS to dominate the PFAS profile within common eider eggs in 2003 and 2004 [22]. Concentrations were generally higher in the earlier studies from Sklinna. Former studies of herring gull eggs at Røst showed average PFOS concentrations of 42.2 ng/g in 2003 [42]. The PFOS concentration has decreased since 2002, which reflects phase out initiatives within the US- and European markets due to environmental health concerns. The present study reports lower concentrations for FOSA, PFDS, PFOA and PFNA, similar concentrations for PFDA, PFDoDA and PFTeDA, and higher concentrations for PFUnDA, PFTrDA and PFPeDA
compared to Verreault et al. (2007) [42]. This likely reflects the change in PFAS production and applications from C8 based PFAS compounds to short and longer chain PFAS in the recent years.
Due to high LODs in the study from 2003 [42], PFHxA and PFHpA could not be evaluated together with the present data set. In general, PFAS in herring gull appears to have stabilized
within the last 10 years (Supplemental data Figure S2 G) while for common eider and shag only small changes were observed since 2000. For future research it is important to include temporal trend studies in general in order to get reliable time trends with state-of-the art methodology in
sample preparation as well as instrumental analysis. Organophosphorous compounds The most frequently detected OPCs in the egg samples were TBP (1-9 ng/g), TCEP (6-39
ng/g), TCPP (4-28 ng/g), DBPhP (0.2 -1.5 ng/g), EHDPP ( others; for European shag: α-HBCDD > β-TBECH > γ/δ-TBECH = DBDPE > others, and herring gull: α-HBCDD = γ/δ-TBECH > others. (Supplemental data Table S7). Statistically significant differences between species were
This article is protected by copyright. All rights reserved
Accepted Preprint
found for α-TBECH and for β-TBECH only between common eider and herring gull as well as shag and herring gull (Supplemental data Table S14). Of the alternative BFRs investigated, HBCDD has received the most attention regarding its persistence and bioaccumulative capacity.
In this study the concentrations in herring gull eggs ranged from 5.99 to 7.65 ng/g. Analysis of herring gull eggs from Lofoten, Norway in 2005 [26], showed relatively higher amounts of HBCDD compared to the present study with an average concentration of 11.3 ng/g and a
maximum of 26.5 ng/g for Σ HBCDD isomers. Within the same study average concentrations of
10 ng/g and maximum concentrations of 17.8 ng/g were found in eggs from the Vardø area in Northern Norway. Higher concentrations of HBCDD have been reported near point sources within Norway with concentration ranges between 4 and 216 ng/g in common eider eggs [50].
The remote nature of the sampling sites of Sklinna and Røst, combined with local feeding, likely accounts for the lower concentrations observed in the present study. Previous studies found a consistent association with reproductive changes in American kestrels and concentrations of αHBCDD [51]. HBCDD concentrations were in similar concentrations in herring gulls of the present study. However, a direct comparison of effects is not possible due to the reasons mentioned in the PCB section. The low concentrations found in the present study are also supported by results from recent
monitoring surveys in other remote regions where BTBPE, BEHTBP, EHTeBB and DBDPE were detected in eggs of polar Guillemot from Kongsfjorden at Spitzbergen and Bear Island [52]. DBDPE has also been reported in common eider eggs from the Norwegian marine environment and Svalbard with concentrations ranging from 0.3 – 5.9 ng/g [53] compared to 0.41 – 0.79 ng/g
in the current study. Tribromanisole
This article is protected by copyright. All rights reserved
Accepted Preprint
In the present study tribromanisole (TBA) was found in all bird egg samples with
concentrations up to 1.91 ng/g in (Supplemental data Table S7). While low, it still dominated the BFR profile in common eiders at Sklinna. TBA is the O-methylated product of 2,4,6-TBP, which is naturally produced by sponges and has been frequently detected in the marine environment
[54,55]. It has also been proven that aerobic bacteria in the environment may produce TBA by Omethylation of 2,4,6-TBP [56,57]. This reaction increases the bioaccumulation potential of the compound. The de-O-methylation of TBA to the phenol 2,4,6-TBP may occur by bacteria in anaerobic environments such as sediments and sludge but also occurs in biota, e.g. fish liver [5860]. 2,4,6-TBP may therefore have different routes and sources in the environment. In previous studies, TBA was detected in four out of five eggs from Sklinna in 2004 where concentrations were higher than the present study with maximum concentrations reaching 11.4 ng/g [36]. Bromo- and alkylphenols Bromophenols were found in highest concentrations in common eider at Sklinna (11.5 ng/g),
while the average sum concentration was 10 times lower at Røst (1.5 ng/g). Average sum
concentrations of bromophenols in shag and herring gull were less than 0.34 ng/g (Table 1). The most prominent compound was 2,4,6-TBP, which was also responsible for the elevated concentrations in common eiders from Sklinna. 2,4-DBP- and pentabromophenol were determined in 28% and 6 % of the samples and in concentrations below 2.9 ng/g and 0.38 ng/g,
respectively (Supplemental data Table S11). 2,4-DBP is an industrial BFR [61] but also naturally formed in the marine environment [62]. It is produced at high volumes and has been estimated to have a high potential for bio-accumulation and long-range transport. It has been detected in a wide range of environmental samples (i.e., indoor/outdoor air, water, sediment, biota and humans) in Australia, Asia, Pacific Ocean, Laurentian Great Lakes (U.S.), and Europe and in the
This article is protected by copyright. All rights reserved
Accepted Preprint
North and Baltic Seas [63-66]. TBP is also naturally formed in marine organisms such as sponges and algae [55,62]. The compound 2,4,6-TBP was the only ―new‖ BFR found in previous studies and was detected in common eider (0.09 ng/g ww) from the Arctic [52] as well as sea bird eggs from Skúvoy, Koltur (Faroe Islands) and Stora Karlsö (Sweden) [66]. Reported
concentrations of 2,4,6-TBP and PBP were in the same range as in the present investigation (0.14 – 21 ng/g and < LOD – 0.38 ng/g respectively). Alkylphenols (APs) and bisphenol A (BPA) showed extremely high average sum
concentrations in herring gull (254 ng/g) at Sklinna with BPA being the dominant compound (Table 1, Supplemental data Table S11). The other APs and BPA showed low detection frequency with shag eggs having the lowest concentrations of APs. In a previous study investigating biomagnification potential of 4-NP, livers of piscivorous birds were also included
and concentrations up to 359 ng/g ww were found [67]. Results showed that biomagnification of 4-NP was not apparent for trophic level consumers such as seabirds. However, results of the present study indicate maternal transfer of alkylphenols and bisphenol A to seabird eggs may occur.
Chlorinated Paraffins Detectable concentrations of chlorinated paraffins (CPs) were found in all bird eggs analysed
although medium chained CPs (MCCPs) were found at a much higher frequency (80%) compared to short chained CPs (SCCPs; 40%) with concentrations ranging below the limit of detection to 17.5 ng/g and 4.8 ng/g, respectively (Supplemental data Table S12). This may be
attributed to restrictions placed on SCCP production/usage in Europe in the 1990s based on concerns over toxicity and environmental persistence [68]. Such restrictions led to increased production/use of MCCPs [69], resulting in increased emissions of MCCPs into the environment.
This article is protected by copyright. All rights reserved
Accepted Preprint
Higher detection frequency of MCCPs may also be explained by slower metabolic/elimination rates as longer carbon chain length and higher degrees of chlorination have been shown to slow metabolism/degradation [69,70]. On average, sum concentrations of CPs were found to be higher in eggs from common eiders compared to eggs from other bird species investigated (Table 1), which may be related to differences in feeding behavior as common eiders are primarily benthic feeders. CPs will accumulate within the sediment compartment as they partition strongly to organic matter [71,72]. This will cause increased exposure/accumulation to benthic dwelling organisms and result in increased exposure to benthic feeding predators. On the other hand
higher levels in common eiders may reflect the lower metabolic capacity of their prey (i.e. molluscs) to transform CPs compared to the main prey items of herring gulls and shags (i.e. fish). SCCPs concentrations detected in eggs from the three species were comparable (Supplemental data Figure S1 J). However, patterns in MCCP concentration indicate that higher exposure may occur at Røst (Supplemental data Table S12). Limited environmental monitoring data exists regarding CP concentrations for birds. SCCPs and MCCPs have both been detected in little auks and kittiwakes collected from Bjørnøya in the European Arctic with concentrations ranging between 5 – 88 ng/g and 5 - 55 ng/g, respectively [73]. These concentrations are higher
compared to results from the present study. However, high body burdens in seabirds from Bjørnøya have been previously documented for various POPs [74] and may in the case of
Bjørnoya not accurately reflect CP exposure from other remote locations. Phthalates Phtalates were detected above the LODs in all species and concentrations ranged from < 3 to
42 ng/g which were driven by DEHP only. DEHP was detected in all herring gull samples,
whereas the other species also showed one (shag) or several (common eider) non-detects
This article is protected by copyright. All rights reserved
Accepted Preprint
(Supplemental data Table S12). There is an indication for DEHP to be present in higher concentrations in eggs from herring gull at Røst compared to Sklinna. In previous surveys DEHP
was investigated in black guillemot eggs collected from Skúvoy, Koltur (Faroe Islands) and
guillemot eggs from Stora Karlsö (Sweden). However, no samples were found above the LOD (4 ng/g) [75]. Cyclic Siloxanes Cyclic volatile methyl siloxanes (cVMS) were found below LODs in the majority of bird
eggs. Highest concentrations of cVMS were detected in eggs from common eider from Sklinna (3.4 ng/g) where both D5 and D6 were present (Table 1 and Supplemental data Table S12). This may be attributed to dietary differences between the bird species as common eiders are primarily benthic feeders whereas the other bird species feed at higher levels within the food chain. The sediment compartment is considered the dominant storage compartment for cVMS in the aquatic environment due to favourable partitioning to suspended particulate matter [76] and deposition to the sediment. Summed cVMS concentrations detected in bird eggs was dominated by D5 (
Environmental Toxicology A BROAD COCKTAIL OF ENVIRONMENTAL POLLUTANTS FOUND IN EGGS OF THREE SEABIRD SPECIES FROM REMOTE COLONIES IN NORWAY
SANDRA HUBER, NICHOLAS A. WARNER, TORGEIR NYGÅRD, MIKAEL REMBERGER, MIKAEL HARJU, HILDE T. UGGERUD, LENNART KAJ, and LINDA HANSSEN
Environ Toxicol Chem., Accepted Article • DOI: 10.1002/etc.2956
Accepted Article
"Accepted Articles" are peer-reviewed, accepted manuscripts that have not been edited, formatted, or in any way altered by the authors since acceptance. They are citable by the Digital Object Identifier (DOI). After the manuscript is edited and formatted, it will be removed from the “Accepted Articles” Web site and published as an Early View article. Note that editing may introduce changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. SETAC cannot be held responsible for errors or consequences arising from the use of information contained in these manuscripts.
This article is protected by copyright. All rights reserved
Accepted Preprint
Environmental Toxicology
Environmental Toxicology and Chemistry DOI 10.1002/etc.2956
A BROAD COCKTAIL OF ENVIRONMENTAL POLLUTANTS FOUND IN EGGS OF THREE SEABIRD SPECIES FROM REMOTE COLONIES IN NORWAY
Running title: A broad cocktail of environmental pollutants in seabirds
SANDRA HUBER, *† NICHOLAS A. WARNER, † TORGEIR NYGÅRD, ‡ MIKAEL REMBERGER, § MIKAEL HARJU, † HILDE T. UGGERUD, || LENNART KAJ, § and LINDA HANSSEN †
† Department of Environmental Chemistry, Fram Centre, Norwegian Institute for Air Research, Tromsø, Norway ‡ Norwegian Institute for Nature Research (NINA), Trondheim, Norway § Swedish Environmental Research Institute (IVL), Stockholm, Sweden
|| Department of Environmental Chemistry, Norwegian Institute for Air Research, Kjeller,
Norway
*Address correspondence to [email protected]
Additional Supporting Information may be found in the online version of this article.
This article is protected by copyright. All rights reserved Submitted 8 October 2014; Returned for Revision 17 February 2015; Accepted 22 February 2015
This article is protected by copyright. All rights reserved
Accepted Preprint
Abstract: Three seabird species, Common eider (Somateria mollisima), European shag
(Phalacrocorax aristotelis aristotelis), and European herring gull (Larus argentatus), were selected to survey for a broad range of legacy and emerging pollutants to assess chemical mixture exposure profiles of seabirds from the Norwegian marine environment. A total of 201 chemical substances were targeted for analysis ranging from metals, organotin compounds, polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and associated metabolites, chlorinated paraffins, chlorinated and non-chlorinated organic pesticides, per- and polyfluroalkyl
substances (PFAS), dechlorane plus, octachlorostyrene, brominated flame retardants, organophosphorous compounds (OPCs), brominated and alkyl phenols, cyclic siloxanes, and
phthalates. Of the chemicals targeted, 149 substances were found above the detection limits with metals dominating the contaminant profile and comprising 60% of the total contaminant load. PCBs, pesticides, OPCs and PFAS were the dominant contaminant classes of organic pollutants found within seabird species with the highest loads occurring in herring gulls followed by shag and common eider. New generation pollutants (e.g., PFAS, OPCs and some alkylphenols) were detected at similar or higher concentrations than the legacy persistent organic pollutants (POPs). Time trends of reported concentrations of legacy POPs appear to have decreased in the last decades from the Norwegian coastal environment. Concentrations of detected pollutants do not appear to have a negative effect on seabird population development within the sampling area. However, additional stress caused by pollutants may affect seabird health more at the individual level. This article is protected by copyright. All rights reserved
Keywords: Contaminants, Emerging pollutants, Persistent organic pollutants, Heavy metals, Ecotoxicology
This article is protected by copyright. All rights reserved
Accepted Preprint
INTRODUCTION The industrial revolution made a significant impact on the living standards for humans.
Particularly the use of man-made chemicals, which continue to be produced and used for various industrial and household applications, have provided safer working and living environments (e.g.
flame retardants), simplified daily living tasks (e.g. Teflon coated cookware) or increased protection against unwanted organisms (e.g. pesticides). However, large-scale use of such compounds has resulted in release and exposure to a broad mixture of chemicals which have, and still may significantly impact ecosystem health. Effects of "cocktail" exposure have already been documented for selected compound groups (e.g., PCBs) [1] and highlight the possible risk which ecosystems are facing from exposure to various mixtures of chemicals. In recent years, cocktail
mixtures of chemicals and environmental pollutants have received growing attention. Evidence has shown exposure to single compounds that do not induce an adverse effect can give negative synergistic and additive effects in organisms when exposure occurs as part of a mixture [2-6]. Environmental pollutants, such as polycyclic aromatic hydrocarbons (PAHs), polychlorinated
biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs) and organochlorine pesticides are classified as harmful compounds and the majority of them have been banned from the market in most countries due to their persistence, toxicity and potential for bioaccumulation in the entire food chain. These legacy persistent organic pollutants (POPs) have been under investigation for
decades, but toxicity mechanisms are still not fully understood, thus driving research to further investigate the knowledge gaps surrounding potential cocktail effects. Research on newer generation pollutants such as: per- and polyfluoroalkyl substances (PFASs), alternative brominated flame retardants, organophosphorous compounds (OPCs), alkyl- and bromophenols, siloxanes or phthalates has not come as far compared to the legacy POPs. However, their risk to
This article is protected by copyright. All rights reserved
Accepted Preprint
environmental health is important to investigate. Exposure to inorganic contaminants (i.e.,
metals) must also not be ignored. Natural exposure to metals occurs where they can act as essential components in the functioning of environmental and biological processes. However, many human activities including mining and refining have resulted in greater release of metals into the environment, potentially inducing acute or chronic toxicity. Exposure to metals at low concentrations is also of concern as bioaccumulation of several metal species has been well
documented [7,8]. How interactions between these different groups of chemicals will affect biological organisms is unclear. This highlights the importance of identifying the ingredients in these cocktail mixtures for future toxicity assessment. Accessibility to tools of non-target analysis for identification of new and unknown pollutants has become easier in recent years as new technology and public available databases have entered the market. However, the use of this approach has limitations due to its complexity of raw extracts and enormous time consumption in interpretation of large data sets together with relatively high thresholds which do not allow a detection of analytes present in low concentrations. Therefore, targeted analysis must still be
relied upon to help address questions surrounding chemical composition and potential effects posed by pollutant cocktail mixtures. In the present study, an assessment of the state of pollution at two Norwegian remote coastal
sites was carried out to evaluate the chemicals to which wildlife may be exposed. Remote islands were chosen in order to estimate background concentrations and to exclude direct anthropogenic influence on pollution from neighboring settlements. Three wide spread seabird species in
Norway, Common eider (Somateria mollisima), European shag (Phalacrocorax aristotelis aristotelis), and European herring gull (Larus argentatus), were selected, as they represent different migration patterns and overwintering strategies, thus representing different exposure
This article is protected by copyright. All rights reserved
Accepted Preprint
scenarios. A survey for a broad range of legacy (i.e., POPs and metals) and emerging pollutants was done. Temporal patterns on contaminant load and distribution within seabirds were also
investigated for selected contaminants (e.g. PCBs, organochlorine pesticides, PFAS) to assess pollution status along Norwegian coastlines. To our knowledge, this is the first time that such a comprehensive study on a broad spectrum of environmental pollutants on seabird populations has
been performed. MATERIAL AND METHODS Bird species and characteristics The Common eider (Somateria mollisima) is distributed all over the Norwegian coastline
during its breeding season and the birds from the sampled colonies in this study overwinter in the same region) [9]. The Common eider is a mid-trophic predator of the marine benthic food web with its diet consisting mostly of crustaceans and mollusks. The European shag (Phalacrocorax aristotelis aristotelis) preys entirely upon small marine
fish. It inhabits the Norwegian coastline throughout the entire year [9]. The European herring gull (Larus argentatus) is an opportunistic species filling a broad
ecological niche with varied feeding patterns and is a common bird species along the Norwegian coast. The herring gull is often used for monitoring studies due its widely spread presence as for example the American herring gull (Larus smithsonianus) at the Great Lakes in North America
[10].
Sampling areas and sampling Seabird eggs from the selected species were collected at two remote islands (Figure 1),
Sklinna (11º 00’E, 65º11’N) and Røst (12º00’E, 67º30’N) on the Norwegian coast during the
breeding season between May and June 2012. These sampling sites have been utilized over
This article is protected by copyright. All rights reserved
Accepted Preprint
several decades to monitor seabird population health in Norway. Monitoring contaminant levels in seabird eggs is an excellent method for evaluating contaminant exposure to seabirds [11-13].
This provides a means of assessing contaminant exposure, particularly in sensitive populations, without sacrificing birds. Within the present study, six and 12 eggs were collected for each species at Sklinna and Røst, respectively. Sampling date, species, coordinates on geographical location (data not shown) and clutch size were noted. Length, width, weight and condition of the eggs were determined prior to the homogenization of sample material (Supplemental data Table S1). Eggs were individually placed in polyethylene bags and stored at +4ºC in a refrigerator until transport to the Norwegian Institute for Nature Research (NINA) in Trondheim.
Homogenisation of sample material Homogenisation of the sample material was performed at an outdoor site beside the Fram
Centre (Tromsø, Norway) in order to avoid cross contamination of problematic indoor air
contaminants (i.e., siloxanes and phthalates). Personnel handling the samples refrained from using personal care products before and during the homogenisation process and sample preparation. Egg content was classified by embryonic development and condition (Supplemental data Table S1). Three eggs were pooled and homogenised in a glass beaker with a stainless steel Ultra-turax homogeniser. Pooled samples AS-1 and AS-2 were homogenised with a stainless steel mixer because of large and partially developed embryos. All equipment was cleaned thoroughly between each sample with ultrapure water, acetone and then hexane to avoid cross contamination between samples. Additional procedures were taken by using a pooled sample of three homogenised hen eggs to provide a field blank for the homogenisation process. Hen egg homogenate was treated the same way as the seabird eggs and was used to assess any
This article is protected by copyright. All rights reserved
Accepted Preprint
contamination introduced during the homogenisation process. All sample containers were frozen at -20ºC and shipped to the respective laboratories for analysis. Analysed parameters A total of 201 different chemical substances were analysed, which included 11 metals, four
organotin compounds and 186 organic compounds belonging to the groups of polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and associated metabolites, chlorinated paraffins, chlorinated and non-chlorinated organic pesticides, per- and polyfluoroalkyl substances (PFASs), dechlorane plus, octachlorostyrene (OCS),
organophosphorous chemicals (OPCs), brominated flame retardants (BFRs), bromo- and alkyl phenols, siloxanes, and phthalates (see Supplemental data Table S2 for details). Analysis of inorganic and organic compounds was performed in the laboratories of the
Norwegian Institute for Air Research (NILU) at Kjeller and Tromsø, Norway, and the Swedish
Environmental Research Institute (IVL) at Stockholm, Sweden. In addition, extractable organic material (EOM) and stable isotopes of N and C were assessed. Stable isotope measurements were performed at the Institute of Energy Technology at Kjeller, Norway, and EOM estimations at NILU Kjeller. Sample preparation and analyses For details on sample preparation and analyses see Supplemental data S1. The eggs were
investigated for moisture loss [14,15] since fresh and rotten eggs and eggs containing embryos were collected (Supplemental data Table S1). All eggs had a specific gravity > 95 and therefore an adjustment for loss of moisture and lipid was not necessary [15].
This article is protected by copyright. All rights reserved
Accepted Preprint
Statistical method Interspecific differences in egg concentrations were assessed using IBM SPSS © v.22
statistics software. Conservative non-parametric tests (Kruskall-Wallis and Mann-Whitney) were applied.
Temporal trends Temporal trends of environmental pollutants were estimated by collecting data from previous
published studies. Results have to be interpreted with care since analyses were conducted at different laboratories and numbers of eggs included in the different studies were varying thus detailed information on single concentrations was not always available to assess statistical significance of temporal trends. Principal component analysis Principal component analysis (PCA) was applied in order to investigate relations between all
variables in Simca (vs 11.5, Umetrics Inc). In this case, the loading bi-plots were used to display both the concentration levels of environmental pollutants and stable isotopes (applied as quantitative data/variables) expressed as correlation coefficients and scores related to species and locations (expressed as passive variables). The chemical concentration and isotope data were preprocessed by using univariate scaling. In addition, the chemical concentration data was logarithmically transformed, while birds’ species and location only were considered as passive
variables. Concentrations found as significant outliers and which were outside the Hotellings T2 (95% confidence interval) were excluded from the model to give a better normal distribution of the data. RESULTS AND DISCUSSION Concentrations and distribution patterns between the species
This article is protected by copyright. All rights reserved
Accepted Preprint
In total, 201 different compounds were analysed of which 149 substances were found at
concentrations above the limits of detection (LODs). Results on concentrations are reported in wet weight. Details on average concentrations and concentration ranges are listed in Table 1 and
single concentrations of the individual compounds and pooled samples are listed in the Supplemental data in Table S3 - S13. Metals and organotin Selected metals (Cr, Co, Ni, Cu, Zn, As, Ag, Cd, Sn/organotin, Pb, total-Hg) were found in all
samples analysed with the exception of Sn/organotin (Supplemental data Table S3). Metal distribution patterns were dominated by Zn, Cu, As, and total Hg (Figure 2). Significant differences in mean concentrations between bird species were shown for Zn, Cu and As
(Supplemental data Table S3 and S14). The remaining metals were in most cases close to detection limits and are not further discussed. Dominance of Zn and Cu in the present study was
expected due to their greater abundance within the natural environmental and that they are vital in physiological processes within organisms. However, it cannot be demonstrated if the presented concentrations are related to biomagnification through the food chain or if they are within normal
physiological ranges. In general, the highest metal concentrations were detected in common eider eggs (except for
total Hg; Table 1 and Supplementary data Table S3). Higher concentrations of metals are generally associated with benthic dwelling organisms (i.e., bivalves and crustaceans) due to their high assimilation of metals from environmental and dietary uptake [16]. Thus, higher concentrations found in common eider eggs may be a result of high concentrations of metals in
their prey or a reflection of this species having higher assimilation efficiency and/or lower elimination rates compared to other bird species investigated [16]. Unlike the other elements,
This article is protected by copyright. All rights reserved
Accepted Preprint
higher concentrations of Hg were found in eggs from shag compared to common eiders, but the difference was statistically not significant (Supplementary data Table S14). This is not surprising as Hg can biomagnify through aquatic foodwebs [7,8] as it is extremely bioavailable in its
methylated form. Thus, pelagic seabirds (i.e., shags) are likely to receive greater inputs of Hg through their diet compared to benthic feeding seabirds (i.e., common eiders) due to bioaccumulation through longer food chain length [17,18]. Based on stable nitrogen isotope signatures, herring gull and shag appear to feed at similar
trophic levels (Supplemental data Table S13). However, total Hg concentrations were significant
lower in eggs from herring gulls compared to shag (Supplemental data Table S14 and Figure S1 A). This may be a reflection of differences between species regarding their assimilation/elimination rates as has previously been observed in seabirds from both the
Canadian Arctic and Barents Sea region [19]. The feeding and migratory behaviors between these bird species must also be considered. Shags inhabit and feed along the Norwegian coastline all year round, whereas herring gulls have a much greater migratory range that could reflect Hg concentrations accumulated from a less polluted location. However, concentrations reported for Hg within this study are associated with background levels of Hg and are below lowest observed adverse effect levels (< 0.6 µg/ww) [20]. The concentrations of As found in eggs from common eider are comparable to those found in
common eider eggs from the Aleutian Islands in a previous study [21]. Bioaccumulation of As has been reported in seabirds from Arctic regions, but was found to be region specific. The results of the present study are in agreement to those reported by Borgå et al. (2006) [19] who showed no bioaccumulation potential for As in seabirds from the Barents Sea region.
This article is protected by copyright. All rights reserved
Accepted Preprint
Additional data on metal concentrations at various trophic levels throughout the entire food web is needed to further assess bioaccumulation potential. The results of the present study suggest that diet and assimilation/elimination efficiencies are driving factors in controlling metal concentrations within these bird species. Organic compounds In total 186 organic compounds were analysed and 137 were found above the LODs
(Supplementary data Tables S4-S8).
Legacy POPs The exposure of legacy POPs to seabird eggs takes place through maternal transfer and not
through exposure of the eggs itself. Analysing seabird eggs will therefore give an actual status on exposure of the maternal birds to environmental contaminants. Polychlorinated biphenyls Most of the PCB congeners analysed were above the LODs for all samples. Common eiders
had the lowest average sum concentrations of 10.9 and 76.9 ng/g at Sklinna and Røst respectively. These concentrations were lower than those found in shag (169 and 161 ng/g) and in herring gull (706 and 1012 ng/g; Table 1) with the exception of one pooled common eider sample from Røst (AR-6, Supplemental data Table S4). Differences in PCB sum concentrations were statistically significant between the species (Supplemental data Table S14 and Figure S1 B). In common eiders, penta- and hexa-congeners dominated the congener profile (CB-82 to CB-
169; Supplemental data Table S4). Similar profiles were observed in shag and herring gull eggs where penta-, hexa-, and hepta congeners (CB-82 to CB-192; Supplemental data Table S4) dominated the CB distribution pattern with more than half of the sum PCB load consisting of
hexa-CB congeners (CB-128 to CB-169; Supplemental data Table S4).
This article is protected by copyright. All rights reserved
Accepted Preprint
Previous studies on PCBs in common eider and shag from Sklinna in 2003-2004 showed
higher concentrations of individual PCBs [22]. Similar findings were observed for shag where concentrations were in general higher in 2003-2004, while only CB-187 was slightly higher in
the present study. Time trends indicate stable concentrations for PCBs in eggs from common
eiders since the 1990s (Supplemental data Figure S2 A). For herring gull and shag, PCB concentrations decreased and are approximately 15 – 25% below concentrations reported in the 1970s. Previous studies from shag and herring gull collected at Røst in 1983, 1993 and 2003 indicate a decrease of the sum-CB concentrations in eggs over time [10,23]. Average sum-PCB concentrations in shag samples were 420 and 280 ng/g in 1983 and 1993 respectively [11, 23], and are higher compared to concentrations from 2012 (161 ng/g) reported in the present study. For herring gulls, the decline in concentrations is far greater compared to shags with levels dropping from 7070 ng/g in 1983 [11], to 1530 ng/g in 1993 [23] and 1012 ng/g in 2012 (present
study). The rapid decline between 1983 and 1993 most likely reflects a decrease in exposure due to restrictions placed on production and use of PCBs during this time period. The decline in concentrations between 1993 and 2012 is not as great compared to earlier years, especially for herring gulls, and is likely a reflection of exposure to secondary sources of PCBs and their slow environmental degradation in addition to simple first order kinetics as concentrations get lower and the elimination starts to slow. This findings indicate that bird populations are not endangered
by PCB exposure as effects on reproductive performance have been reported to occur at much higher concentrations (34.1 μg/g ww) [24]. However, the effects observed by Fernie et al. [24] are not directly comparable to the bird species investigated in the present study, since a terrestrial raptor species (American kestrel) was exposed to PCBs in laboratory experiments. If the same
This article is protected by copyright. All rights reserved
Accepted Preprint
effects can be shown in wildlife together with natural exposure pathways and synergistic, antagonistic or additive effects of cocktail mixtures of pollutants is questionable. Metabolites of polychlorinated biphenyls Metabolites of PCBs have over the last decade drawn attention due to their negative
contribution in early development [25]. As with the parent PCB congeners, seabird eggs are also exposed to the associated metabolites of PCBs (MeSO2 and OH- CBs) through maternal transfer [26].
For the MeSO2-CB congener profile of common eider, shag and herring gull the penta and
hexa chlorinated MeSO2-CBs (MeSO2-CBs 101, 110, 132, and 149; Supplemental data Table S5) were the most abundant congeners. This is in accordance with the results for the parent
compounds where the penta- and hexa-CB congeners (CB-82 to CB-169) dominated in all seabird species. MeSO2-CB load was highest in herring gull, followed by shag and common
eider (Table 1 and Supporting data Figure S1 C). The ratio of
19MeSO2-CBs
to parent PCB are an indicator of formation and retention
capacity of MeSO2-CBs among different species. Ten of the parent PCBs were analysed in the present study, where MeSO2-CBs 52, 101 and 149 (3 and 4-MeSO2-CB) accounts for 59% of the 19MeSO2-CB
fraction. Comparison of the MeSO2-CB-concentrations to their respective parent
PCB congener concentration (PCBs 52, 101 and 149) provides a rough approximation of the metabolic transformation. The ratio of
6MeSO2-CBs/ 3ParentPCBs
was 0.20 in common eider
and 0.12 in herring gull eggs, similar to other studies [25-27]. Despite large variations in MeSO2-CB concentrations between species, the ratio of
19MeSO2-CBs
to parent PCB was
consistent, indicating similar formation and retention capacity. However, the ratio in shags was 0.78 which suggest that shag may have a higher expression of CYP-like metabolizing enzymes
This article is protected by copyright. All rights reserved
Accepted Preprint
which leads to MeSO2-CB formation or a higher retention capacity for metabolites in shag. Additional analyses of eggs and plasma samples of all relevant parent PCBs are needed to verify these results. Detailed information regarding the expression of CYP-like metabolizing enzymes between these species would help to provide insight on metabolic differences between these
species.
The feeding patterns/behavior between bird species will also affect accumulation of MeSO2-
CB-metabolites. This is evident where concentrations of MeSO2-CB are higher and PCB
concentrations are significantly lower (i.e. in common eider and shag compared to herring gull). It might be attributed to a combination of feeding pattern differences and a lower total PCB load for shag, as well as a higher metabolisation rate of PCBs. In a study by
.
(2010) [28], a clear difference in metabolic capacity between bird species was observed. This is also supported by previous investigations by Warner et al. (2005) [29] which observed large variation in metabolic selectivity for enantiomers of chiral PCBs between different Arctic seabird species.
The concentrations of the OH-CBs were lower compared to the MeSO2 –CBs in all three bird
species. On average, common eiders had the lowest OH-CB concentrations (Table 1). The most prominent OH-CB was 4-OH-CB 187, comprising on average 64%, 31% and 49% of
10OH-
CBs in the eggs from common eider, shag and herring gull, respectively (Supplemental data Table S5). Organochlorine pesticides A total of 16 organochlorine pesticides (OC-pesticides) were analysed in this study (Supplemental data Table S6). Common eiders showed the lowest Σ OC-pesticide concentrations
with average sum concentrations of 8.4 and 54.8 ng/g at Sklinna and Røst respectively, compared
This article is protected by copyright. All rights reserved
Accepted Preprint
to shag (86.0 and 80.5 ng/g) and herring gull with 380 and 535 ng/g respectively (Table 1, Supplemental data Figure S1 D). The pattern of OC-pesticide concentrations from highest to lowest was found to be similar in all three species: p,p’-DDE > HCB > trans-NC > cis-NC > βHCH > other pesticides (Supplemental data Table S6). Significant statistical differences in concentrations between all species were found for cis-chlordane, cis-nonachlor, β-HCH, HCB, p,p’-DDD and o,p’-DDT (Supplemental data Table S14). Previous studies have reported similar or higher concentration of p,p’-DDT in eggs from
common eiders at Sklinna collected in 2004 and 2005 [22]. o,p’-DDT was not comparable due to
the high LODs in the previous study, whereas for eiders and shags, p,p’-DDE concentrations were slightly higher than in the present study. Time trends indicate stable concentrations for DDTs in eggs from common eiders since the 1990s (Supplemental data Figure S2 B). For herring gull and shag, DDT concentrations decreased and are approximately 15 – 25% below
concentrations reported in the 1970s. Similar as found for PCBs was a decline for p,p’-DDE in shag and herring gull concentrations observed between 1993 and 2012. Concentrations within shag (230 ng/g) and herring gulls (1050 ng/g) in 1983 decreased approximately 50% by 1993 [11,23] followed by a slower decline to 2012 (shag: 50 ng/g, herring gull: 413 ng/g; present study). This indicates that bird populations in this region are not endangered by DDT exposure as effects of eggshell thinning on predator birds have been reported to occur at much higher concentrations (> 1000 ng/g) [30-33]. Similar results were found for p,p’-DDE in ducks, where eggshell thinning is induced at high concentrations (µg/g levels) [33]. Shag levels were
decreasing in minor degrees between 2000 and 2009. β-HCH was the dominating HCH isomer in the present study and was approximately four
times higher in shag and herring gull eggs compared to common eider (Supplemental data Table
This article is protected by copyright. All rights reserved
Accepted Preprint
S6). HCH appear to have stabilized over the last 10 years (Supplemental data Figure S2 E), while chlordanes appear to have decreased within the same time period (Supplemental data Figure S2 D).
In the present study, concentrations of HCB were highest in herring gull followed by shag
and common eider (Supplemental data Table S6). Trends for HCB are difficult to interpret based on temporal data assessments (Supplemental data Figure S2 C), but concentrations of HCB in herring gull have decreased by approximately 80% since the 1970s. Shag eggs from 2004 and 2002 had lower concentrations [22,34] than measured here. This may be a reflection of increased
environmental exposure as air concentrations of HCB have increased over the past decade within the Canadian and Norwegian Arctic [35]. Hypothesized sources of this increase have been due to HCB contamination in current use chlorinated pesticides as well as re-volatilization from secondary sources [35]. However, such sources would also be expected to impact herring gulls and eider ducks in a similar manner, which is not observed. This indicates further monitoring is needed to confirm increasing trends in HCB concentrations within shags and the environment. Unlike HCB, PeCB followed a similar pattern as observed for PCBs with generally lower concentrations in common eider and higher concentrations detected in herring gull (Table 1). Polybrominated diphenyl ethers In the present study, all of the analysed 17 PBDEs were found in the egg samples
(Supplemental data Table S7). Σ PBDE concentrations were lower compared to organochlorine contaminants investigated within this study but followed similar patterns between the species (Table 1 and Figure 3). Lowest concentrations were found in common eiders with average summed concentrations of 1.09 and 0.84 ng/g at Sklinna and Røst respectively, followed by shag
(2.89 and 3.29 ng/g) and herring gull (11.2 and 13.9 ng/g) (Table 1) with concentrations being
This article is protected by copyright. All rights reserved
Accepted Preprint
statistically different between species (Supplemental data Table S14). The concentration levels for BDE-99 and BDE-100 were lower in this study compared to previous results reported in
common eider and shag eggs in 2004 [22]. BDE-153 was slightly higher in shag eggs in 2012 compared to 2004, but not in eider eggs, where the opposite was the case. This might be a
reflection of analytical or spatial variation, but cannot be proven at this time. Murvoll et al. (2006) [34] reported similar PBDE concentrations in shag at Sklinna in 2002 compared to what
was found in a 2004 study [22], while a former study on contaminants in herring gull eggs at Røst in 2003 reported Σ PBDE concentrations as high as 60 ng/g [26]. The latter is approximately 3.5 times greater than the highest Σ PBDE concentration reported in the present
study. BDE-100 was detected in five eggs from Sklinna in 2004 where concentrations were relatively high with up to 1.8 ng/g for BDE-100 [36]. In general, PBDE concentrations measured in eggs collected in 2002 and 2004 were higher than eggs from 2012, which indicates a decline in PBDE concentrations/exposure over time and especially over the last 10 years (Supplemental
data Figure S2 F). Previous studies found a consistent association with reproductive changes in American kestrels and concentrations of PBDE-congeners [37]. PBDE concentrations of the investigated species of the present study were much lower compared to Fernie et al. [37]. However, a direct comparison of effects is not possible (see section on PCBs for details). Polycyclic aromatic hydrocarbons Five of 16 analysed PAHs (naphthalene, anthracene, fluoranthene, pyrene and chrysene) were
found above the LODs. Detected sum-concentrations were relatively low (≤ 1.4 ng/g) with the exception of a herring gull sample from Røst, where naphthalene and pyrene accounted for most
of the 29.6 ng/g (Table 1 and Supplemental data Table S8). PAHs are less prone to bioaccumulation or biomagnification than the organochlorines, partly because of metabolic
This article is protected by copyright. All rights reserved
Accepted Preprint
degradation of PAHs in top predators and their prey [38]. However, a broad range of PAHs in seabirds from different islands surrounding Great Britain have been previously reported [39]. Due to the remoteness of the sampling locations within the present study, concentrations found likely reflect background levels. The islands investigated by Shore et al. (1999) [39] are probably influenced by anthropogenic sources due to their closer proximity to settlements on the mainland of Great Britain. However, in any future work it is recommended to include alkylated PAHs since they have to be found to be of greater importance in egg samples than their parent compounds [40]. New generation and emerging environmental contaminants Per- and polyfluoroalkylated substances. A total of 24 PFAS were analysed within the
present study, with 19 compounds detected above LOD. Unlike findings observed for the legacy
contaminants, individual single PFAS compounds showed similar concentrations between species and sampling locations (Supplemental data Table S9). However, for PFAS sum
concentrations of common eider eggs at Sklinna had lowest average Σ PFAS concentration (11.2 ng/g, Table 1 and Supplemental data Figure S1 G), followed by shag (25.5 ng/g) and herring gull
(58.9 ng/g). In the distribution patterns of PFAS, PFOS (3.92 – 42.3 ng/g) was predominant in all samples followed by PFNA (0.31– 5.64 ng/g), PFUnDA (0.974 – 7.0 ng/g), PFTrDA (0.97 – 8.55 ng/g), PFHxS (0.231 – 1.58 ng/g) and PFDA (0.107 to 1.13 ng/g; Supplemental data Table S9). Differences between species were not statistically significant for most of the PFAS compounds (Supplemental data Table S14). FOSA, a pre-cursor compound for PFOS and PFOA, was detected in most samples with concentrations ranging from below the detection limit to 0.663
ng/g and it was significantly higher in eggs from shag compared to herring gull (Supplemental
data Tables S9 and S14). Results cannot be explained by biomagnification dynamics and/or
This article is protected by copyright. All rights reserved
Accepted Preprint
trophic position of bird species only as PFAS exhibit different properties (hydrophobicity and lipophobicity) and different accumulation pathways compared to the legacy POPs. There is evidence that a majority of the PFAS exposure and uptake occurs through the diet. However, species-specific metabolism of PFAS-precursor compounds will also affect the overall accumulated body burden by presently unknown rates [41]. Previous studies conducted in this region also found PFOS to dominate the PFAS profile within common eider eggs in 2003 and 2004 [22]. Concentrations were generally higher in the earlier studies from Sklinna. Former studies of herring gull eggs at Røst showed average PFOS concentrations of 42.2 ng/g in 2003 [42]. The PFOS concentration has decreased since 2002, which reflects phase out initiatives within the US- and European markets due to environmental health concerns. The present study reports lower concentrations for FOSA, PFDS, PFOA and PFNA, similar concentrations for PFDA, PFDoDA and PFTeDA, and higher concentrations for PFUnDA, PFTrDA and PFPeDA
compared to Verreault et al. (2007) [42]. This likely reflects the change in PFAS production and applications from C8 based PFAS compounds to short and longer chain PFAS in the recent years.
Due to high LODs in the study from 2003 [42], PFHxA and PFHpA could not be evaluated together with the present data set. In general, PFAS in herring gull appears to have stabilized
within the last 10 years (Supplemental data Figure S2 G) while for common eider and shag only small changes were observed since 2000. For future research it is important to include temporal trend studies in general in order to get reliable time trends with state-of-the art methodology in
sample preparation as well as instrumental analysis. Organophosphorous compounds The most frequently detected OPCs in the egg samples were TBP (1-9 ng/g), TCEP (6-39
ng/g), TCPP (4-28 ng/g), DBPhP (0.2 -1.5 ng/g), EHDPP ( others; for European shag: α-HBCDD > β-TBECH > γ/δ-TBECH = DBDPE > others, and herring gull: α-HBCDD = γ/δ-TBECH > others. (Supplemental data Table S7). Statistically significant differences between species were
This article is protected by copyright. All rights reserved
Accepted Preprint
found for α-TBECH and for β-TBECH only between common eider and herring gull as well as shag and herring gull (Supplemental data Table S14). Of the alternative BFRs investigated, HBCDD has received the most attention regarding its persistence and bioaccumulative capacity.
In this study the concentrations in herring gull eggs ranged from 5.99 to 7.65 ng/g. Analysis of herring gull eggs from Lofoten, Norway in 2005 [26], showed relatively higher amounts of HBCDD compared to the present study with an average concentration of 11.3 ng/g and a
maximum of 26.5 ng/g for Σ HBCDD isomers. Within the same study average concentrations of
10 ng/g and maximum concentrations of 17.8 ng/g were found in eggs from the Vardø area in Northern Norway. Higher concentrations of HBCDD have been reported near point sources within Norway with concentration ranges between 4 and 216 ng/g in common eider eggs [50].
The remote nature of the sampling sites of Sklinna and Røst, combined with local feeding, likely accounts for the lower concentrations observed in the present study. Previous studies found a consistent association with reproductive changes in American kestrels and concentrations of αHBCDD [51]. HBCDD concentrations were in similar concentrations in herring gulls of the present study. However, a direct comparison of effects is not possible due to the reasons mentioned in the PCB section. The low concentrations found in the present study are also supported by results from recent
monitoring surveys in other remote regions where BTBPE, BEHTBP, EHTeBB and DBDPE were detected in eggs of polar Guillemot from Kongsfjorden at Spitzbergen and Bear Island [52]. DBDPE has also been reported in common eider eggs from the Norwegian marine environment and Svalbard with concentrations ranging from 0.3 – 5.9 ng/g [53] compared to 0.41 – 0.79 ng/g
in the current study. Tribromanisole
This article is protected by copyright. All rights reserved
Accepted Preprint
In the present study tribromanisole (TBA) was found in all bird egg samples with
concentrations up to 1.91 ng/g in (Supplemental data Table S7). While low, it still dominated the BFR profile in common eiders at Sklinna. TBA is the O-methylated product of 2,4,6-TBP, which is naturally produced by sponges and has been frequently detected in the marine environment
[54,55]. It has also been proven that aerobic bacteria in the environment may produce TBA by Omethylation of 2,4,6-TBP [56,57]. This reaction increases the bioaccumulation potential of the compound. The de-O-methylation of TBA to the phenol 2,4,6-TBP may occur by bacteria in anaerobic environments such as sediments and sludge but also occurs in biota, e.g. fish liver [5860]. 2,4,6-TBP may therefore have different routes and sources in the environment. In previous studies, TBA was detected in four out of five eggs from Sklinna in 2004 where concentrations were higher than the present study with maximum concentrations reaching 11.4 ng/g [36]. Bromo- and alkylphenols Bromophenols were found in highest concentrations in common eider at Sklinna (11.5 ng/g),
while the average sum concentration was 10 times lower at Røst (1.5 ng/g). Average sum
concentrations of bromophenols in shag and herring gull were less than 0.34 ng/g (Table 1). The most prominent compound was 2,4,6-TBP, which was also responsible for the elevated concentrations in common eiders from Sklinna. 2,4-DBP- and pentabromophenol were determined in 28% and 6 % of the samples and in concentrations below 2.9 ng/g and 0.38 ng/g,
respectively (Supplemental data Table S11). 2,4-DBP is an industrial BFR [61] but also naturally formed in the marine environment [62]. It is produced at high volumes and has been estimated to have a high potential for bio-accumulation and long-range transport. It has been detected in a wide range of environmental samples (i.e., indoor/outdoor air, water, sediment, biota and humans) in Australia, Asia, Pacific Ocean, Laurentian Great Lakes (U.S.), and Europe and in the
This article is protected by copyright. All rights reserved
Accepted Preprint
North and Baltic Seas [63-66]. TBP is also naturally formed in marine organisms such as sponges and algae [55,62]. The compound 2,4,6-TBP was the only ―new‖ BFR found in previous studies and was detected in common eider (0.09 ng/g ww) from the Arctic [52] as well as sea bird eggs from Skúvoy, Koltur (Faroe Islands) and Stora Karlsö (Sweden) [66]. Reported
concentrations of 2,4,6-TBP and PBP were in the same range as in the present investigation (0.14 – 21 ng/g and < LOD – 0.38 ng/g respectively). Alkylphenols (APs) and bisphenol A (BPA) showed extremely high average sum
concentrations in herring gull (254 ng/g) at Sklinna with BPA being the dominant compound (Table 1, Supplemental data Table S11). The other APs and BPA showed low detection frequency with shag eggs having the lowest concentrations of APs. In a previous study investigating biomagnification potential of 4-NP, livers of piscivorous birds were also included
and concentrations up to 359 ng/g ww were found [67]. Results showed that biomagnification of 4-NP was not apparent for trophic level consumers such as seabirds. However, results of the present study indicate maternal transfer of alkylphenols and bisphenol A to seabird eggs may occur.
Chlorinated Paraffins Detectable concentrations of chlorinated paraffins (CPs) were found in all bird eggs analysed
although medium chained CPs (MCCPs) were found at a much higher frequency (80%) compared to short chained CPs (SCCPs; 40%) with concentrations ranging below the limit of detection to 17.5 ng/g and 4.8 ng/g, respectively (Supplemental data Table S12). This may be
attributed to restrictions placed on SCCP production/usage in Europe in the 1990s based on concerns over toxicity and environmental persistence [68]. Such restrictions led to increased production/use of MCCPs [69], resulting in increased emissions of MCCPs into the environment.
This article is protected by copyright. All rights reserved
Accepted Preprint
Higher detection frequency of MCCPs may also be explained by slower metabolic/elimination rates as longer carbon chain length and higher degrees of chlorination have been shown to slow metabolism/degradation [69,70]. On average, sum concentrations of CPs were found to be higher in eggs from common eiders compared to eggs from other bird species investigated (Table 1), which may be related to differences in feeding behavior as common eiders are primarily benthic feeders. CPs will accumulate within the sediment compartment as they partition strongly to organic matter [71,72]. This will cause increased exposure/accumulation to benthic dwelling organisms and result in increased exposure to benthic feeding predators. On the other hand
higher levels in common eiders may reflect the lower metabolic capacity of their prey (i.e. molluscs) to transform CPs compared to the main prey items of herring gulls and shags (i.e. fish). SCCPs concentrations detected in eggs from the three species were comparable (Supplemental data Figure S1 J). However, patterns in MCCP concentration indicate that higher exposure may occur at Røst (Supplemental data Table S12). Limited environmental monitoring data exists regarding CP concentrations for birds. SCCPs and MCCPs have both been detected in little auks and kittiwakes collected from Bjørnøya in the European Arctic with concentrations ranging between 5 – 88 ng/g and 5 - 55 ng/g, respectively [73]. These concentrations are higher
compared to results from the present study. However, high body burdens in seabirds from Bjørnøya have been previously documented for various POPs [74] and may in the case of
Bjørnoya not accurately reflect CP exposure from other remote locations. Phthalates Phtalates were detected above the LODs in all species and concentrations ranged from < 3 to
42 ng/g which were driven by DEHP only. DEHP was detected in all herring gull samples,
whereas the other species also showed one (shag) or several (common eider) non-detects
This article is protected by copyright. All rights reserved
Accepted Preprint
(Supplemental data Table S12). There is an indication for DEHP to be present in higher concentrations in eggs from herring gull at Røst compared to Sklinna. In previous surveys DEHP
was investigated in black guillemot eggs collected from Skúvoy, Koltur (Faroe Islands) and
guillemot eggs from Stora Karlsö (Sweden). However, no samples were found above the LOD (4 ng/g) [75]. Cyclic Siloxanes Cyclic volatile methyl siloxanes (cVMS) were found below LODs in the majority of bird
eggs. Highest concentrations of cVMS were detected in eggs from common eider from Sklinna (3.4 ng/g) where both D5 and D6 were present (Table 1 and Supplemental data Table S12). This may be attributed to dietary differences between the bird species as common eiders are primarily benthic feeders whereas the other bird species feed at higher levels within the food chain. The sediment compartment is considered the dominant storage compartment for cVMS in the aquatic environment due to favourable partitioning to suspended particulate matter [76] and deposition to the sediment. Summed cVMS concentrations detected in bird eggs was dominated by D5 (
