Assessment of Pesticide, Dioxin-like Polychlorinated Biphenyl, and Polycyclic Aromatic Hydrocarbon Existence around Galindez Island, and Comparison with the Organic Pollution Bibliography of the Antarctic Peninsula

Abstract

Sediment, notothenioid fish, and moss samples were collected from the vicinity of Galindez Island, Antarctic Peninsula during the austral autumn of 2016 and 2017. Pesticide, Polycyclic Aromatic Hydrocarbon (PAH), and dioxin-like Polychlorinated Biphenyl (dl-PCB) concentrations were measured using High-Resolution Gas and Liquid Chromatography. Pollutant concentrations were below detection limits in sediment and moss samples. However, pesticides, PAH, and dl-PCB congeners were detected in the muscle tissue of fishes. Pesticide concentrations varied between 0.46 and 12.2 ng/g-dw, and Mecarbam was the dominant compound. Kresoxim-methyl, Mecarbam, Procymidone, Pyridaben, and Quinoxyfen were reported in the muscle tissue of the fishes, for the first time from the Antarctic. PCB-118, PCB-105, and PCB-156 were dominant dl-PCBs. The ∑12-dl-PCB concentration was 160,929 pg/g-dw, and WHO-TEQ-total dl-PCB was 8.30 pg/g-dw in Trematomus bernachii, over the consumable limit in fishes according to the European Commission. The PCB-126 concentration was 36 pg/g-dw in the muscle tissue of fish, the first reported from the Antarctic. Phenanthrene was the dominant PAH congener. The ∑16-PAH concentration was 22.5 ng/g-dw. PAH sources were local and petrogenic in the fishes, likely after long-term bioaccumulation. The flow rate is rather low around Galindez Island; accordingly, contaminant removal takes time and may demonstrate long-lasting effects including bioaccumulation in the marine food web.

1. Introduction

Both the Arctic and Antarctic are considered relatively non-polluted environments. The Antarctic continent, in particular, is further away from anthropogenic influences. However, in the last few decades, organic pollutants have started to show up in these environments [1,2,3,4,5,6,7]. Therefore, the level of existing pollutants in the continent, whether new ones are transported, and, if they are, the transportation pathways, are issues that need to be investigated.

Organic contaminants resistant to environmental degradation to a varying degree, which possess toxic features and a risk of causing adverse effects to biota and humans, are termed persistent organic pollutants (POPs) (e.g., dl-PCB’s, DDT, Aldrin, Beta Hexachlorocyclohexane, Chlordane, Dieldrin, Endrin, Heptachlor, Hexachlorobenzene [HCB], Lindane, Mirex, Toxaphene, etc.) [8]. POPs, such as dioxin-like Polychlorinated Biphenyls (dl-PCBs), and some pesticides can be used to understand the long-range transport of pollutants in regions such as the Antarctic, where local sources of pollution are almost absent [9]. Their resistance to biological, chemical, and photolytic processes allows them to travel across intercontinental boundaries by water currents and air movements, which may deposit them far from where they originate [10,11].

Even though pesticides are considered useful for crop yield, some of them are highly toxic to living organisms, including humans. Indeed, 9 of the 12 initial POPs and 16 of all 30 listed POPs declared by the Stockholm Convention [10] are pesticides [12]. dl-PCBs, which are commercially used in capacitors, transformers, building sealants, etc., have generally high lipophilicity, a persistent nature, and toxicity [13,14]. dl-PCBs tend to spread into wide areas by binding to different vectors in air and fall on both land and sea surfaces as snowflakes or raindrops via precipitation in the Antarctic [15]. Although most of the listed POPs are no longer produced, their traces are still present worldwide due to global cross-border movements [16]. On the other hand, Polycyclic Aromatic Hydrocarbons (PAHs) are another one of the important toxic organic pollutant groups, due to the mutagenic and/or carcinogenic properties of some of their congeners on animals, as well as humans [17]. Different from the other chemicals investigated in the present study, PAHs are emitted locally into the atmosphere, particularly during fuel combustion for heating purposes in research bases established in the Antarctic [18].

After long-range transport or local accumulations, the pollutants can directly settle on the sea or be deposited through permanent conservation in the moss, soil, snow, ice, or firm-ice cover. Accumulated contaminants may be released from melting ice and snow that sink into oceanic waters and sediments [8] and then may join the marine food web [5,6].

The present study was conducted as a part of ecological research on the western Antarctic Peninsula [19,20] in the frame of the first Turkish polar program [21]. The study aims to better understand the contamination level and transportation pathway of some potentially dangerous organic pollutants for aquatic and terrestrial biota, including POPs, in living and non-living constituents in the Antarctic. Therefore, in the present study, we investigated 262 Pesticides’, 12 dl-PCBs’, and 16 PAHs’ congener concentrations in the surface sediment, fish (to evaluate transport, in particular by marine water), and moss (to evaluate atmospheric deposition) samples, collected from the vicinity of Galindez Island (Antarctic Peninsula), and assessed their possible sources.

2. Material and Methods

2.1. Study Locations and Characterization of Samples

The Antarctic Peninsula is the part of Antarctica nearest to South America. Because of its location and milder climate, today, there are several permanent and seasonal research stations established throughout. The Ukrainian Vernadsky Research Base (formerly the British Faraday Station) is one of the oldest research bases in the Antarctic Peninsula and is located at Marina Point on Galindez Island, where the present study was conducted around its vicinity.

In the frame of the present study, fish, sediment, and moss samples were collected around Galindez Island (Antarctic Peninsula) to analyze 262 Pesticides’, 12 dl-PCBs’, and 16 PAHs’ congener concentrations, including some POPs. The sediment and fish samples were collected during the first Turkish Joint Antarctic Expedition in April 2016, and the moss samples in April 2017 (Figure 1).

To trap fishes, a cone-type fyke net was anchored in two sites (Stations 2 and 3) at Galindez Island offshore. In total, nine adult fish, i.e., three Trematomus bernacchii Boulenger, 1902 (emerald rock-cod) and six Notothenia coriiceps Richardson, 1844 (black rock-cod), were trapped and used for analysis. Due to unfavorable weather conditions and difficulties with fishing, a limited number of fish could be caught. However, the number of samples taken was sufficient for analytical measurements and recovery procedures. Both fishes are benthic, typically live in shallow coastal waters, and feed on various organisms that live at the bottom. The total lipid contents of flesh were reported as 6.1% and 4.6% dw for N. coriiceps and T. bernacchii, respectively, and 23% dw in the liver for N. coriiceps [22]. Since the tissue structures are important in the bioaccumulation of lipophilic chemicals, both species that have notable fatty tissue were used in the analysis. These species are also known as having restricted home ranges that may help to understand nearby contamination [22,23]. Surface sediment samples were collected from five sites (i.e., Stations 2, 3, 4, 5, and 6) using a Van Veen-type grab sampler thrown by hand from a boat, and the moss samples (Polytrichum briedel and Andreaea regularis) were collected manually in Station 1 from a rocky area located a hundred meters north of the Vernadsky Research Base (Figure 1).

2.2. Analytical Preparation of the Samples and Extraction

The muscle tissues of the fishes were filleted using a porcelain knife, divided into small pieces, and the tissues belonging to each fish species were mixed. Sediment samples were desalted three times using distilled water, and coarse particles (>1 mm) were removed using a stainless-steel sieve. Rhizoid and gametophyte parts of the moss samples were separated from each other. After pretreatment, all samples were weighed and freeze-dried at −55 °C for 72 h using Christ Alpha 1–2 LD+ equipped with RZ 2.5 Rotary vane pump. Before homogenization, the samples were transported to a desiccator to remove moisture and weighed. Dry samples were homogenized using a porcelain mortar and pestle and stored at −20 °C until analysis.

Preparations and extractions of all samples were carried out according to the methods US-EPA 429 [24], ISO 11338 [25], and IP 346 [26] for the fish, US-EPA 429 and modified STN EN 16619 [27] for the moss, and US-EPA 429 [24], ISO 11338 [28], and US-EPA 3540 [29] for the sediment samples to determine the PAH concentrations, whereas the methods used to determine the dl-PCB concentrations were modified US-EPA 1668 [30] for the fish, modified US-EPA 1668 and modified CSN P CEN/TS 16190 [31] for the moss, and JIS K0311 [32] for the sediment samples, using an isotope dilution technique via High-Resolution Gas Chromatography–High-Resolution Mass Spectrometry (HRGC-HRMS).

A modified AOAC 2007.01 [33] method was used for pesticide residue analysis in sediment by acetonitrile extraction and partitioning with magnesium sulfate, by gas and liquid chromatography–tandem mass spectrometers (GC-MS/MS and LC-MS/MS). A certain amount of the sediment sample (15 g) was homogenized by adding distilled water. Fifteen milliliters of 1% acetic acid (HOAc) and buffered acetonitrile (MeCN) were added to the sample and transferred to a centrifuge tube. Six grams of anhydrous magnesium sulfate and 1.5 g of sodium acetate (MgSO₄/NaOAc—4/1, w/w) were added and vortexed. The mixture was centrifuged for 2 min at 4000 rpm. The upper layer of the MeCN extract was added to the 15 mL centrifuge tube containing 250 mg MgSO₄ and 50 mg PSA (5/1, w/w per 1 mL extract) and was centrifuged again for 2 min at 4000 rpm. The final extract was transferred to autosampler vials after filtration through a 0.45 µm filter paper for analysis by GC-MSMS and LC-MSMS. For pesticide residue analysis in biota, a modified method, namely PAM Vol. 1-82-1 [34], was used. The biota samples (10 g) were mixed with 50 mL of petroleum ether and shaken vigorously. One milliliter of the extract was passed through a petroleum ether-conditioned florosil cartridge, then 5 mL of 6%, 15%, and 50% petroleum ether/diethyl ether (1:1, w/w) mixtures were passed through the cartridge. Collected extracts were dried under a slow stream of high-purity nitrogen and then dissolved in 5 mL of acetone and injected into both GC-MSMS and LC-MSMS instruments.

To calculate recovery values, a 2000 mg/kg pesticide congener mix (Accustandard) stock solution was prepared (80 µg/g in 1% acetic acid) and added to the dried N. coriiceps tissue (10 g, separated from the extracted sample) and analyzed via GC-MSMS and LC-MSMS. According to compound specifications (e.g., solubility, volatility, polarity, etc.), LC-MS/MS or GC-MS/MS was used to quantify a total of 262 pesticide compounds, including organochlorine, organobromine, and organophosphates. The concentrations of detectable pesticides in the muscle tissue of the T. bernacchii and N. coriiceps, thallus of P. briedel and A. regularis, and five surface sediment samples were measured. The concentrations obtained from the samples were subtracted from the concentrations obtained from the samples with the congener mix, and recovery values were calculated (Table S1-1). The recovery values for pesticide compounds ranged between 41.6% and 112% with an average of 77% in the muscle tissue of N. coriiceps; however, their concentrations were below the detection limits for all other moss and sediment samples (i.e., <1.0 ng/g dw for that analyzed in GC MS/MS and <2.0 ng/g dw for that analyzed in LC-MS/MS; Tables S1-1, S1-2). The compound-specific spectrometer settings (i.e., ion rates, collision energies, polarities, and chromatography operating conditions) were given in the Supplementary Material (Tables S1-3, S1-4, S1-5, S1-6). To perform the external calibration, pesticide mix standards were used and correlation coefficients for calibration curves were calculated higher than 0.99 for all the congeners. In both the sediment and biota, 186 pesticide congeners were analyzed by GC-MS/MS, and 76 by LC-MS/MS (Tables S1-5, S1-6).

The dried sample was spiked after homogenization with an extraction containing ¹³C₁₂-labeled PCB congeners and ²H_x-labeled PAH congeners. The spiked sample was mixed with anhydrous sodium sulfate/silica (1/1) and transferred into an extraction thimble. Extraction with DCM/hexane (1/1, v/v) in a Soxhlet extractor took 20 h. Then, the remaining was concentrated in the rotary vacuum evaporator. For the cleanup of PAHs, the raw extract was pre-cleaned on the column containing silica-gel impregnated with NaOH solution. The DCM/hexane (2/1, v/v) fraction from the column was concentrated using the modified Kuderna–Danish concentrator up to 0.5–1 mL. For the cleanup of dl-PCBs, the final extract was transferred to an Erlenmeyer flask, diluted in hexane, and shaken with silica-gel impregnated with H₂SO₄. Then, the whole mixture was transferred to the top of a multi-layer silica-gel column and eluted with hexane/DCM (95/5, v/v). The extract was concentrated using the modified Kuderna–Danish concentrator up to 0.5–1 mL. For the fractionation florisil column and the combination of hexane and dichloromethane, the elution of two different PCBs fractions was used. The first hexane fraction contained mono-ortho PCBs and non-dioxin-like PCBs. The dichloromethane fraction contained non-ortho PCBs. Both fractions were concentrated using a Kuderna–Danish concentrator to a volume of approximately 0.5 mL. The final extract containing PAHs is spiked with a specific ²H_x-labeled injection standard, and 2–4 µL are injected into HRGC-HRMS. The final extract containing mono-ortho PCBs and non-dioxin-like PCBs is spiked with an injection standard containing specific ¹³C₁₂-labeled PCBs, and 2–4 µL are injected into HRGC-HRMS. The final extract containing non-ortho PCBs is concentrated up to dryness under a nitrogen stream and then spiked with an injection standard containing specific ¹³C₁₂-labeled PCBs, and 2–4 µL are injected into HRGC-HRMS. Some details (e.g., the sample amount, dry matter, final extract-injection volume, detection limits, etc.) regarding PAH and dl-PCB analysis are provided in the Supplementary Material (Tables S2-1, S2-2, S2-3, S2-4, S2-5, S2-6).

The measurement uncertainty was expressed as a double relative standard deviation (RSD%) and corresponded to a 95% confidence interval. The estimation of measurement uncertainty for each PAH and dl-PCB was 30%, while for WHO-TEQs, it was 20%. The sums of measured values were determined as ∑16-PAH and ∑12-dl-PCB. For dl-PCB-WHO-TEQ calculation in biota, the concentrations were multiplied with the toxic equivalency factor (TEF), which have been developed to compare the toxicity and assess the combined effect of PCDD/Fs and dl-PCBs [35], revised by WHO in 2005 [36], while the concentrations were multiplied with TEFs according to Ahlborg et al. [37] for dl-PCB toxic equivalency (dl-PCB-TEQ) calculations in sediment samples. The limit of quantifications (LOQs) was defined on the base of the blank level and used for non-detectable PCBs to calculate WHO-TEQ and TEQ values (Tables S2-2a, S2-2b, S2-4a, S2-4b, S2-6a, S2-6b, S2-6c, S2-6d, S2-6e). Since the PAH and dl-PCB congener concentrations were measured below the detection limits in the thallus and rhizoid of P. briedel (i.e., ranged between <0.24 and <6.30 ng/g dw for PAHs, and <0.037 and <9.70 pg/g dw for dl-PCBs) and sediment (i.e., ranged between <0.00026 and <0.15 ng/g dw for PAHs, and <0.00071 and <2.6 ng/g dw for dl-PCBs) samples, measurement details are provided in the Supplementary Material (Tables S2-3, S2-4a, S2-4b, S2-5a, S2-5b, S2-5c).

2.3. PAH Source Estimation

Proportions between certain PAH congeners, referred to as the diagnostic PAH ratio, can be used to understand particular emission sources. Although these ratios are usually preferred for sediment, Guinan et al. [38] reported that the PAH profiles of sediment and mussels are often similar, and diagnostic PAH ratios in biota samples such as fish and pine needles as well as water, air, and dust samples can be used for the PAH source partitioning [39]. Therefore,

$$Flu/\left(Flu+Pyr\right) \mathrm{and} Ant/\left(Ant+Phe\right)$$

ratios were used to estimate PAH sources, considering certain PAH congener concentrations in the muscle tissue of fish. In the formulas, Flu, Pyr, Ant, and Phe represent fluoranthene, pyrene, anthracene, and phenanthrene, respectively.

2.4. Oceanographic Measurements

A data logger (Hydrolab DS-5) was used to measure the temperature, salinity, chlorophyll-a (Chl-a), dissolved oxygen (DO), and pH at Station 3 along a 20 m water column. The data logger was calibrated before deployment, waiting two minutes for it to warm up, and was then submerged at a constant speed (i.e., 0.2 m/s) in the water column. The device was adjusted for one measurement per second.

Pearson correlations (two-tailed) were used to determine linear correlations between two variables (e.g., congener concentrations in T. bernacchii and N. coriiceps) at 95% and 99% confidence levels where necessary.

3. Results

3.1. Oceanographic Conditions

The average values of water temperature (°C), salinity, DO (mg/L), Chl-a (µg/L), pH, and turbidity (NTU) through the water column were −0.68, 30.2, 11.4, 7.72, 0.47, and 0.005, respectively (Figure 2). The surface water temperature represented late autumn conditions with −0.8 °C, and slightly increased by depth. Salinity and DO concentrations have a similar increasing pattern by depth, while pH and turbidity were almost the same along the water column. In the vicinity of the stations, no ice cover was present during the study.

3.2. Pesticides in Biota and Sediment Samples

In total, 21 pesticide compounds were determined in the muscle tissue of fish. While two fungicides, two herbicides, and twelve insecticides were found in T. bernacchii, four fungicides, one herbicide, and eleven insecticides were found in N. coriiceps. Nineteen pesticide compounds were organochlorines and organophosphates, whereas two were organobromines. Sixteen pesticide compounds were detected in both fish species, of which ten of them were common (

Table 1. Pesticide concentrations (ng/g-dw) in the muscle tissue of T. bernacchii and N. coriiceps, gametophyte of P. briedel and A. regularis, and surface sediment samples collected from the vicinity of Galindez Island and Vernadsky Research Base.

Pesticide Congeners	Fish		Moss		Sediment a
	T. bernacchii	N. coriiceps	P. briedel	A. regularis	Sediment
Hexachlorobenzene (HCB) f,#	1.76	1.86	<1.0	<1.0	<1.0
Kresoxim-methyl f,*	-	5.80	<2.0	<2.0	<2.0
Procymidone f,#	6.03	6.09	<1.0	<1.0	<1.0
Quinoxyfen f,#	-	7.62	<1.0	<1.0	<1.0
Trifluralin h,#	0.99	1.32	<1.0	<1.0	<1.0
Lindane (δ-BHC) h,#	3.20	-	<1.0	<1.0	<1.0
Lindane (γ-HCH) i,#	1.46	-	<1.0	<1.0	<1.0
2,4′-DDD (o,p’-DDD) i,#	4.20	6.03	<1.0	<1.0	<1.0
2,4′-DDE (o,p’-DDE) i	3.90	6.07	<1.0	<1.0	<1.0
2,4′-DDT (o,p’-DDT) i,#	8.05	11.2	<1.0	<1.0	<1.0
4,4′-DDD (p,p’-DDD) i,#	7.08	-	<1.0	<1.0	<1.0
4,4′-DDE (p,p’-DDE) i,#	4.11	6.30	<1.0	<1.0	<1.0
4,4′-DDT (p,p’-DDT) i,#	8.50	4.70	<1.0	<1.0	<1.0
Bromophos methyl i,#	2.72	0.72	<1.0	<1.0	<1.0
Bromophos-ethyl i,#	1.97	1.96	<1.0	<1.0	<1.0
Chlordan gamma i,#	-	1.49	<1.0	<1.0	<1.0
Chlorpyrifos-ethyl i,#	1.14	-	<1.0	<1.0	<1.0
Heptachlor i,#	1.48	1.65	<1.0	<1.0	<1.0
Mecarbam i,*	-	12.2	<2.0	<2.0	<2.0
Methoxychlor i,#	1.35	-	<1.0	<1.0	<1.0
Pyridaben i,*	-	0.46	<2.0	<2.0	<2.0
TOTAL	57.9	75.5

^f: Fungicide, ^h: Herbicide, ⁱ: Insecticide, ^#: by GC-MS/MS, *: by LC-MS/MS, ^a: Since concentrations of pesticide compounds in the sediment samples were below the detection limit in all stations, only one column is used in the table.

). The detectable concentration ranges varied between 0.99 ng/g dw (Trifluralin) and 8.50 ng/g dw (p,p’-DDT) in the muscle tissue of T. bernacchii. In N. coriiceps, Mecarbam had the highest concentration (12.2 ng/g dw). The ratios of ∑(DDD + DDE)/∑DDT were calculated as 0.86 for both fish species. Total pesticide concentrations were 57.9 ng/g dw and 75.5 ng/g dw in T. bernacchii and N. coriiceps, respectively (Table 1). On the other hand, those concentrations were below the detection limits in the thallus of P. briedel and A. regularis, and sediment samples were collected from five stations in the vicinity of Galindez Island (Table S1-2).

3.3. dl-PCB and PAH Concentrations in Biota and Sediment Samples

Dioxin-like-PCBs were found in detectable levels merely in the muscle tissues of the fish species (

Table 2. dl-PCB concentrations (±uncertainty) in the muscle tissue of fishes, gametophyte, and rhizoid of moss and sediment samples collected from the vicinity of Galindez Island (pg/g-dw).

dl-PCB Congeners	Fish Species		Bryophyta Species		Sediment	WHO 2005 TEFs
	T. bernacchii	N. coriiceps	P. briedel (gametopyhte)	P. briedel (Rhizoid)	Sediment a
PCB 77	83.0 ± 30%	<37.0	<3.20	<3.40	<88.0	0.0001
PCB 81	<8.80	<2.80	<0.31	<0.56	<0.83	0.0003
PCB 126	36.0 ± 30%	<4.50	<0.51	<0.48	<0.71	0.1
PCB 169	<0.58	<3.10	<0.04	<0.30	<1.10	0.03
PCB 105	32,000 ± 30%	5500 ± 30%	<5.80	<9.60	<1300	0.00003
PCB 114	680 ± 30%	<29.0	<0.04	<0.30	<29.0	0.00003
PCB 118	110,000 ± 30%	15,000 ± 30%	<9.20	<9.70	<2600	0.00003
PCB 123	600 ± 30%	<18.0	<0.44	<0.71	<39.0	0.00003
PCB 156	12,000 ± 30%	4000 ± 30%	<2.70	<5.10	<170	0.00003
PCB 157	1900 ± 30%	1000 ± 30%	<0.28	<1.20	<63.0	0.00003
PCB 167	3400 ± 30%	1000 ± 30%	<0.67	<3.00	<97.0	0.00003
PCB 189	230 ± 30%	93.0 ± 30%	<1.90	<0.33	<1.80	0.00003
WHO-TEQ (dl-PCB) Lowerbound	8.30 160,929 pg/g	0.81 26,593 pg/g	0	0	0
WHO-TEQ (dl-PCB) Upperbound	8.30	1.40	0.053	0.058	0.0007

^a: Since concentrations of dl-PCB compounds in the sediment samples were below the detection limit in all stations, only one column is used in the table.

) whereas all of the dl-PCB congeners were below detection limits in the moss and surface sediment samples (Tables S2-4a, S2-4b, S2-6a, S2-6b, S2-6c, S2-6d, S2-6e). Elevated levels of dl-PCBs were found in both fishes’ muscle tissue. With a 30% measurement uncertainty, ∑12-dl-PCB was calculated as 160,929 pg/g dw and 26,593 pg/g dw in the muscle tissue of T. bernacchii and N. coriiceps, respectively. PCB 118, 105, and 156 were the dominant congeners with contributions of 56.4–68.4%, 19.9–20.7%, and 7.5–15.0%, respectively (Table 2). Lower-bound WHO-TEQ levels from quantified dl-PCB concentrations were calculated as 8.30 and 0.81 for T. bernacchii and N. coriiceps, respectively (Tables S2-2a, S2-2b). Ten of the twelve dl-PCB congeners were detected in the muscle tissue of the fishes. However, none of them were detected in thallus and rhizoide of P. briedel as well as surface sediment samples (Tables S2-4a, S2-4b, S2-6a, S2-6b, S2-6c, S2-6d, S2-6e).

Among the analyzed PAH congeners, Phenanthrene Fluoranthene and Pyrene were found in detectable levels in the muscle tissue of T. bernacchii and N. coriiceps, while the other congeners were below the detection limits (

Table 3. PAH concentrations (± uncertainty) in the muscle tissue of fishes, gametophyte, and rhizoid of moss and sediment samples were collected from the vicinity of Galindez Island (ng/g-dw).

PAH Congeners	Fish		Moss		Sediment a
	T. bernacchii	N. coriiceps	P. briedel (thallus)	P. briedel (Rhizoide)	Sediment
Naphthalene	<4.30	<4.60	<4.10	<5.30	<4.90
Acenaphthylene	<0.39	<0.62	<0.24	<0.33	<3.60 <0
Acenaphthene	<0.85	<0.62	<0.27	<0.43	<4.90
Fluorene	<3.10	<3.10	<0.27	<1.40	<61.0
Phenanthrene (Phe)	9.80 ± 30%	10.0± 30%	<2.20	<6.30	<300
Anthracene (Ant)	<0.46	<0.46	<0.36	<1.00	<3.00
Fluoranthene (Flu)	3.40 ± 30%	4.70 ± 30%	<0.88	<2.33	<20.0
Pyrene (Pyr)	7.70 ± 30%	7.80 ± 30%	<0.90	<4.30	<18.0
Ant/(Ant + Phe)	0.06	0.04
Flu/(Flu + Pyr)	0.31	0.38
Benz[a]anthracene	<0.36	<0.46	<0.24	<0.38	<6.10
Chrysene	<0.21	<0.46	<0.36	<0.38	<7.30
Benzo[b]fluoranthene	<0.21	<0.31	<0.24	<0.38	<8.50
Benzo[k]fluoranthene	<0.21	<0.31	<0.24	<0.38	<6.70
Benzo[a]pyrene	<0.14	<0.31	<0.24	<0.25	<6.70
Indeno[1,2,3-cd]pyrene	<0.14	<0.31	<0.24	<0.25	<6.10
Dibenzo[a,h]anthracene	<0.14	<0.31	<0.24	<0.25	<2.40
Benzo[ghi]perylene	<0.21	<0.31	<0.24	<0.25	<5.10
∑16 PAH-Lowerbound	20.9	22.5	0	0	0
∑16 PAH-Upperbound	31.6	34.7	11.0	24.0	464

^a: Since concentrations of PAH compounds in the sediment samples were below the detection limit in all stations, only one column is used in the table.

). All of the PAH congeners were found to be below the detection limits in the moss and surface sediment samples (Tables S2-3, S2-5). In the muscle tissue of T. bernacchii and N. coriiceps, ∑16-PAH concentrations were calculated as 20.9 ng/g dw and 22.5 ng/g dw, respectively. Phenanthrene was the dominant congener, while fluoranthene had the lowest value among detectable PAH congeners in both fish species (Table 3).

4. Discussion

The Antarctic, which is known as a pristine environment, receives pollutants from point sources and non-point sources in remote areas. Therefore, terrestrial mosses, surface sediment, and fish samples were analyzed in the present study to discriminate possible transport pathways of these chemicals, to understand whether they came via atmospheric deposition or water currents. Moreover, we measured some oceanographic parameters simultaneously with the samplings to understand the living conditions of the examined fish.

The measured oceanographic parameters represented convenient conditions for marine organisms around the island (Figure 2). The presence of Chl-a through the water column indicates primer productivity. Considerable living activity, including phytoplankton and zooplankton, was reported from the vicinity of Galindez Island on the same date as the sampling of the present study [20].

Results of the present study indicate that a significant amount of pesticides, dl-PCBs, and PAHs, including POPs, have contaminated the Antarctic marine environment, and new contaminants have been added to the list. In studies conducted around Elephant Island in 1987 and 1996, HCB concentrations in the liver of the same Notothenioid fish species were found to measure an average of 24 ng/g in Extracted Organic Material (EOM) and 19 ng/g in EOM, respectively [40], while in the same location, HCB concentrations were 17 ng/g in EOM and 30 ng/g in EOM during the same period in the muscle tissue of another Notothenioid fish species (Goerke et al., 2004). In the present study, measured HCB concentrations (i.e., 1.76 ng/g dw and 1.86 ng/g dw in the muscle tissue of T. bernacchii and N. coriiceps, respectively) were higher than the concentrations measured in the previous studies in the muscle tissue, whereas they were lower than the concentrations in the liver, considering a wet weight basis. The γ-HCH concentration was 3.44 ng/g lw in the muscle tissue of N. coriiceps between 2008 and 2011 in the Antarctic peninsula (Table 4) [41]. Its concentration was measured as 1.46 ng/g dw in 2016 in the present study for the same fish species and tissue; therefore, the γ-HCH concentration, measured in the present study, was higher than the previous study, on a wet weight basis. Chlordan-gamma, another highly poisonous banned insecticide, was 1.49 ng/g dw in T. bernacchii in the present study. These data suggest that HCB, γ-HCH, and Chlordan-gamma maintain their presence in the Antarctic Peninsula and Notothenioid fish, despite their banning over time since 1985.

Despite being banned since the 1970s, DDT and its derivatives joined circulation in the world and maintained its existence due to its long half-life and environmental persistence [42]. Results of previous studies from the Antarctic Peninsula suggest that while the final degradation product of p,p′-DDT (i.e., p,p′-DDE) concentrations are increasing, p,p′-DDT levels in fish tissue have been relatively decreasing since 1985 [41,43,44,45] (Table 4). However, in the present study, p,p′-DDT, and ∑DDT levels were measured as 4.70 ng/g dw and 34.3 ng/g dw, respectively, whereas their concentrations were found to be 2.14 ng/g lw and 9.95 ng/g lw, respectively, in the muscle tissue of N. coriiceps during 2008–2011 in the Antarctic Peninsula [41] (Table 4). In the same study, p,p′-DDT and ∑DDT concentrations were also measured in a trematomus species’ (i.e., Trematomus newnesi) muscle tissue as 4.70 ng/g lw, and 11.90 ng/g lw, respectively, while their concentrations were found to be 8.50 ng/g dw and 35.8 ng/g dw, respectively, in Trematomus bernachii muscle tissue in the present study (Table 4). Since the present p,p′-DDT and ∑DDT levels are higher than the previous studies (on a wet weight basis), it is thought that DDT is still transported outside of the continent. Eventually, transportation and deposition of contaminants by global distillation and cold condensation [5,46] and subsequent transport through the food chain still cause ∑DDT concentrations to increase in the muscle tissue of Antarctic fish. Besides DDT and its derivatives, the abovementioned compounds (i.e., HCB, γ-HCH, Chlordan isomers) were also reported from the Arctic fish [6,7], indicating the global impact of pesticide pollution.

Table 4. ∑16 PAH, ∑12 dl-PCB, and OCp concentrations detected in several moss, fish, and sediment samples around the Antarctic Peninsula (ng/g except for dl-∑PCBs, which is in pg/g).

Pollutant	Moss		Fish		Sediment
	Conc. (ng/g)	Date Site, Species, Reference	Conc. (ng/g)	Date Site, Species (tissue), Reference	Conc. (ng/g)	Date Site, Reference
∑16 PAH	4.400–34.00	2009 L.isl, Su, [45]	20.90–22.50 257.1 268.0 398.0–596.0	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2004–05 P.cov, Nc (liver), [47] 1991–93 D.bay, Nc (liver), [48] 1991–93 P.sta, Nc (liver), [48]	32.70–1900 nd-14.49 200.0–1.650 14.00–280.0 nd-202.0 nd-252.8	2005 P.cov, [47] 1989 P.sta, [49] 1989 P.sta, [50] 1988 S.isl, [51] 2004 A.bay, [18] 1997–00 A.bay, [52]
∑12 dl-PCB (pg/g-dw)	10.40–812.0 8.000–79.00	2008 F.pen, ns, [14] 2009 L.isl, Su, [45]	26,593–160,929 21.180 32.200–191.00	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2015 KG.isl, Th (muscle), [14] 2010 KG.isl, Tb (muscle), [53]	nad
Hexachlorobenzene	0.021–0.120 0.300 0.450 0.800 0.680 0.139–0.663 0.779 1.06 0.811	2009 L.isl, Su, [45] 1985 PL, Ba, [44] 1985 G.isl, Du, [44] 1985 R.pt, Ar, [44] 1985 La.isl, Du, [44] 2009–10 KG.isl, ns, [43] 2004–05 KG.isl, Bs, [54] 2004–05 KG.isl, Sp, [54] 2004–05 KG.isl, Su, [54]	1.760–1.860 17.00, 30.00 17.10, 15.50 26.00, 20.40 22.60, 18.60	2016 G.isl, Tb (muscle)-Nc (muscle), TS 1987, 1996 E.isl, Te (muscle), [55] 1987, 1996 E.isl, Ca (liver), [40] 1987, 1996 E.isl, Cg (liver), [40] 1987, 1996 E.isl, Gg (liver), [40]	16.70 14.90 3.400 0.300 0.400 0.058 0.950–4.000	2008 SOoL.isl, [56] 2008 OoP.sta, [56] 2008 OoLv.isl, [56] 2008 OoA.isl, [56] 2008 SOoAP, [56] 2009–10 KG.isl, [43] 2005 JR.isl, [28]
2,4’-DDD (o,p’-DDD)	nad		4.20–6.03 <LOD <LOD <LOD	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	nad
4,4’-DDD (p,p’-DDD)	nad		7.080 2.630 0.940 0.870	2016 G.isl, Tb (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	nad
4,4’-DDE (p,p’-DDE)	0.005–0.040 0.170 0.320 0.330 0.530 0.020–0.228	2009 L.isl, Su, [45] 1985 PL, Ba, [44] 1985 G.isl, Du, [44] 1985 R.pt, Ar, [44] 1985 La.isl, Du, [44] 2009–10 KG.isl, ns, [44]	4.110–6.300 4.200, 7.700 7.200, 14.50 4.000, 5.000 3.700, 7.500 4.590 6.860 7.310	2016 G.isl, Tb (muscle)-Nc (muscle), TS 1987, 1996 E.isl, Te (muscle), [55] 1987, 1996 E.isl, Ca (liver), [40] 1987, 1996 E.isl, Cg (liver), [40] 1987, 1996 E.isl, Gg (liver), [40] 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	1.400 2.800 1.600 38.10	2008 SOoL.isl, [56] 2008 OoP.sta, [56] 2008 OoL.isl, [56] 2009–10 KG.isl, [43]
4,4’-DDT (p,p’-DDT)	<LOQ-0.014 0.300 0.250 0.440 0.550 nd-0.051	2009 L.isl, Su, [45] 1985 PL, Ba, [44] 1985 G.isl, Du, [44] 1985 R.pt, Ar, [44] 1985 La.isl, Du, [44] 2009–10 KG.isl, ns, [43]	8.500–4.700 4.700 2.140 3.610	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	0.313	2009–10 KG.isl, [43]
∑DDTs	0.020–0.324 1.220 1.730 1.620	2009–10 KG.isl, ns, [43] 2004–05 KG.isl, Bs, [54] 2004–05 KG.isl, Sp, [54] 2004–05 KG.isl, Su, [54]	35.80–34.30 11.90 9.950 11.80	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	0.577 0.190–1.150	2009–10 KG.isl, [43] 2005 JR.isl, [28]
Chlordan gamma	nad		1.490 nad	2016 G.isl, Nc (muscle), TS nad	3.600 1.000	2008 OoP.sta, [56] 2008 OoL.isl, [56]
Lindane (γ-HCH)	0.400 1.060 1.700 1.160 nd-0.079	1985 PL, Ba, [44] 1985 G.isl, Du, [44] 1985 R.pt, Ar, [44] 1985 La.isl, Du, [44] 2009–10 KG.isl, ns, [43]	1.460 2.340 3.440 3.900	2016 G.isl, Tb (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	nd	2009–10 KG.isl, [43]

dw: Dry weight, ww: Wet weight, nd: Not detected, nad: No available data, LOD: Limit of Detection, TS: This study. Ar: Andreaea regularis, Ba: Bryum algens, Bs: Brachitecyum sp., Ca: Chaenocephalus aceratus, Cg: Champsocephalus gunnari, Du: Drepanocladus uncinatus, Gg: Gobionotothen gibberifrons, Nc: Notothenia coriiceps, Nr: Notothenia rossii, Su: Sanionia uncinata, Sp: Syntrichia princeps, Te: Trematomus eulepidotus, Tb: Trematomus bernacchii, Th: Trematomus hansoni, Tn: Trematomus newnesi, ns: Not specified A.bay: Admiralty bay, D.bay: Dalman bay, E.isl: Elephant island, F.pen: Fildes peninsula, G.isl: Galindez island, JR.isl: James Ross island, KG.isl: King George island, La.isl: Lagoon island, L.isl: Livingstone island, OoA.isl: Offshore of Adelaide island, OoLv.isl: Offshore of Lavosier island, OoP.sta: Offshore of Palmer station, P.cov: Potter cove, P.sta: Palmer station, PL: Port Lockroy, R.pt: Rothera point, S.isl: Signy island, SOoAP: Southern Offshore of Antarctic Peninsula, SOoL.isl: Southern offshore of Livingstone island,

Table 4. ∑16 PAH, ∑12 dl-PCB, and OCp concentrations detected in several moss, fish, and sediment samples around the Antarctic Peninsula (ng/g except for dl-∑PCBs, which is in pg/g).

Pollutant	Moss		Fish		Sediment
	Conc. (ng/g)	Date Site, Species, Reference	Conc. (ng/g)	Date Site, Species (tissue), Reference	Conc. (ng/g)	Date Site, Reference
∑16 PAH	4.400–34.00	2009 L.isl, Su, [45]	20.90–22.50 257.1 268.0 398.0–596.0	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2004–05 P.cov, Nc (liver), [47] 1991–93 D.bay, Nc (liver), [48] 1991–93 P.sta, Nc (liver), [48]	32.70–1900 nd-14.49 200.0–1.650 14.00–280.0 nd-202.0 nd-252.8	2005 P.cov, [47] 1989 P.sta, [49] 1989 P.sta, [50] 1988 S.isl, [51] 2004 A.bay, [18] 1997–00 A.bay, [52]
∑12 dl-PCB (pg/g-dw)	10.40–812.0 8.000–79.00	2008 F.pen, ns, [14] 2009 L.isl, Su, [45]	26,593–160,929 21.180 32.200–191.00	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2015 KG.isl, Th (muscle), [14] 2010 KG.isl, Tb (muscle), [53]	nad
Hexachlorobenzene	0.021–0.120 0.300 0.450 0.800 0.680 0.139–0.663 0.779 1.06 0.811	2009 L.isl, Su, [45] 1985 PL, Ba, [44] 1985 G.isl, Du, [44] 1985 R.pt, Ar, [44] 1985 La.isl, Du, [44] 2009–10 KG.isl, ns, [43] 2004–05 KG.isl, Bs, [54] 2004–05 KG.isl, Sp, [54] 2004–05 KG.isl, Su, [54]	1.760–1.860 17.00, 30.00 17.10, 15.50 26.00, 20.40 22.60, 18.60	2016 G.isl, Tb (muscle)-Nc (muscle), TS 1987, 1996 E.isl, Te (muscle), [55] 1987, 1996 E.isl, Ca (liver), [40] 1987, 1996 E.isl, Cg (liver), [40] 1987, 1996 E.isl, Gg (liver), [40]	16.70 14.90 3.400 0.300 0.400 0.058 0.950–4.000	2008 SOoL.isl, [56] 2008 OoP.sta, [56] 2008 OoLv.isl, [56] 2008 OoA.isl, [56] 2008 SOoAP, [56] 2009–10 KG.isl, [43] 2005 JR.isl, [28]
2,4’-DDD (o,p’-DDD)	nad		4.20–6.03 <LOD <LOD <LOD	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	nad
4,4’-DDD (p,p’-DDD)	nad		7.080 2.630 0.940 0.870	2016 G.isl, Tb (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	nad
4,4’-DDE (p,p’-DDE)	0.005–0.040 0.170 0.320 0.330 0.530 0.020–0.228	2009 L.isl, Su, [45] 1985 PL, Ba, [44] 1985 G.isl, Du, [44] 1985 R.pt, Ar, [44] 1985 La.isl, Du, [44] 2009–10 KG.isl, ns, [44]	4.110–6.300 4.200, 7.700 7.200, 14.50 4.000, 5.000 3.700, 7.500 4.590 6.860 7.310	2016 G.isl, Tb (muscle)-Nc (muscle), TS 1987, 1996 E.isl, Te (muscle), [55] 1987, 1996 E.isl, Ca (liver), [40] 1987, 1996 E.isl, Cg (liver), [40] 1987, 1996 E.isl, Gg (liver), [40] 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	1.400 2.800 1.600 38.10	2008 SOoL.isl, [56] 2008 OoP.sta, [56] 2008 OoL.isl, [56] 2009–10 KG.isl, [43]
4,4’-DDT (p,p’-DDT)	<LOQ-0.014 0.300 0.250 0.440 0.550 nd-0.051	2009 L.isl, Su, [45] 1985 PL, Ba, [44] 1985 G.isl, Du, [44] 1985 R.pt, Ar, [44] 1985 La.isl, Du, [44] 2009–10 KG.isl, ns, [43]	8.500–4.700 4.700 2.140 3.610	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	0.313	2009–10 KG.isl, [43]
∑DDTs	0.020–0.324 1.220 1.730 1.620	2009–10 KG.isl, ns, [43] 2004–05 KG.isl, Bs, [54] 2004–05 KG.isl, Sp, [54] 2004–05 KG.isl, Su, [54]	35.80–34.30 11.90 9.950 11.80	2016 G.isl, Tb (muscle)-Nc (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	0.577 0.190–1.150	2009–10 KG.isl, [43] 2005 JR.isl, [28]
Chlordan gamma	nad		1.490 nad	2016 G.isl, Nc (muscle), TS nad	3.600 1.000	2008 OoP.sta, [56] 2008 OoL.isl, [56]
Lindane (γ-HCH)	0.400 1.060 1.700 1.160 nd-0.079	1985 PL, Ba, [44] 1985 G.isl, Du, [44] 1985 R.pt, Ar, [44] 1985 La.isl, Du, [44] 2009–10 KG.isl, ns, [43]	1.460 2.340 3.440 3.900	2016 G.isl, Tb (muscle), TS 2008–11 P.cov, Tn (muscle), [41] 2008–11 P.cov, Nc (muscle), [41] 2008–11 P.cov, Nr (muscle), [41]	nd	2009–10 KG.isl, [43]

dw: Dry weight, ww: Wet weight, nd: Not detected, nad: No available data, LOD: Limit of Detection, TS: This study. Ar: Andreaea regularis, Ba: Bryum algens, Bs: Brachitecyum sp., Ca: Chaenocephalus aceratus, Cg: Champsocephalus gunnari, Du: Drepanocladus uncinatus, Gg: Gobionotothen gibberifrons, Nc: Notothenia coriiceps, Nr: Notothenia rossii, Su: Sanionia uncinata, Sp: Syntrichia princeps, Te: Trematomus eulepidotus, Tb: Trematomus bernacchii, Th: Trematomus hansoni, Tn: Trematomus newnesi, ns: Not specified A.bay: Admiralty bay, D.bay: Dalman bay, E.isl: Elephant island, F.pen: Fildes peninsula, G.isl: Galindez island, JR.isl: James Ross island, KG.isl: King George island, La.isl: Lagoon island, L.isl: Livingstone island, OoA.isl: Offshore of Adelaide island, OoLv.isl: Offshore of Lavosier island, OoP.sta: Offshore of Palmer station, P.cov: Potter cove, P.sta: Palmer station, PL: Port Lockroy, R.pt: Rothera point, S.isl: Signy island, SOoAP: Southern Offshore of Antarctic Peninsula, SOoL.isl: Southern offshore of Livingstone island,

Among the detected pesticides in the present study, δ-BHC, Trifularin, Bromophos-ethyl, Bromophos-methyl, Chlorpyrifos-ethyl, Heptachlor, and Metoxychlor have been reported previously from the Antarctic [57] but not the Antarctic Peninsula. Furthermore, no available data were found in open literature regarding residues of Kresoxim-methyl, Mecarbam, Procymidone, Pyridaben, and Quinoxyfen from the Antarctic Peninsula, and to our knowledge, these pesticides are reported from the Antarctic for the first time by the present study. These new pesticides are described as very toxic and hazardous to the aquatic environment, and among them, Kresoxim-methyl is suspected to cause cancer [58]. Since these compounds have low and/or moderate volatility, their transportation mechanism is presumably atmospheric after adsorption into the particles. However, concentrations of these pesticides were below the detection limits in the moss samples of the present study.

There were no available data in the open literature regarding dl-PCB contamination around Galindez Island. However, elevated dl-PCB congener concentrations (e.g., PCB 118, 105, 156, 167, 157, and 189) were found in the muscle tissue of T. bernacchii and N. coriiceps in the present study (Table 2). It is striking that the WHO-TEQ dl-PCB value (i.e., 8.30 pg/g) of the present study is over the consumable limit in fishes (i.e., 4.5 pg/g WHO-TEQ dl-PCB) according to the European Commission directive [59]. In another aspect, since the Scientific Committee on Food of the European Union established a tolerable weekly intake of 14 pg TEQ/kg body weight [35], the amount measured in the present study (i.e., 8.30 pg/g) is over the limit for an average (i.e., 70 kg) adult that consumes 120 g of fish weekly. Moreover, PCB 126, which has the highest toxicity among dl-PCBs (i.e., TEF = 0.1) [36], was found in the muscle tissue of T. bernacchii at a level of 36 pg/g dw. These concentrations are the highest dl-PCB concentrations in the muscle tissue of fish samples ever reported from the continent, and hundreds to thousands-fold higher than previous studies [14,53] carried out in the Antarctic Peninsula (Table 4). Significant positive correlations (in terms of detected common dl-PCB congener concentrations) between T. bernacchii and N. coriiceps muscle tissue (r = 0.99, p < 0.01, n = 6) indicate that both fish species were exposed to dl-PCB contamination via the same pathway. Since all dl-PCB congener concentrations were below detection limits both in the moss and the sediment samples, it is thought that the contaminants are not of airborne origin and are very likely from local contamination sources.

Diagnostic PAH ratios, i.e., Flu/(Flu + Pyr) and Ant/(Ant + Phe), obtained from fish tissues were used to estimate PAH sources. Both ratios in T. bernacchii (i.e., 0.31 and 0.06, respectively) and N. coriiceps (i.e., 0.38 and 0.04, respectively) indicate that PAHs around Galindez Island originated from petrogenic sources (Figure 3).

Since diesel PAHs mainly comprise low molecular PAHs (i.e., 2 and 3 ringed) and contain very low amounts of anthracenes due to refining processes [60], detected PAH congeners in the present study are very likely to be related to diesel contamination. In fact, the Ukrainian Vernadsky research base in Galindez Island is one of the oldest bases in the Antarctic and the oldest operational one in the Argentina Island archipelago in the Peninsula, operating year-round since 1934 [61]. Furthermore, the base host’s non-governmental cruise vessels and thousands of tourists annually since 1996 [62], as well as unwanted leakages from small, motorized transportation vehicles or marine vessels may contribute to PAH contamination. Although insignificant (p > 0.05), the remarkable correlation (r = 0.99) calculated for three detectable PAH congener concentrations between T. bernacchii and N. coriiceps suggests that the fish were exposed to the same contamination sources.

Figure 3. Source partitioning of PAHs around the vicinity of Galindez Island. Values < 0.5 of Flu/(Flu + Pyr) indicate that the major PAH input is from petrogenic sources [63], also values <0.1 of Ant/(Ant + Phe) are attributed to petrogenic sources [64].

Figure 3. Source partitioning of PAHs around the vicinity of Galindez Island. Values < 0.5 of Flu/(Flu + Pyr) indicate that the major PAH input is from petrogenic sources [63], also values <0.1 of Ant/(Ant + Phe) are attributed to petrogenic sources [64].

In previous studies, besides atmospheric long-range transportation, numerous PAH contamination issues have been reported to be linked to fuel storage tanks, ship and boating activities or accidents, construction, station runoff, decommissioned dump sites, past disposal practices, etc., from the scientific bases [48,49]. Previous studies around the peninsula [47,50] have shown that PAHs reached elevated levels in the liver, while its concentrations did not present a detectable level in muscle, which needs a relatively long time for accumulation. In the present study, ∑16-PAH concentrations were 20.9 ng/g dw and 22.5 ng/g dw in the muscle tissue of T. bernacchii and N. coriiceps, respectively. Therefore, the present results represent long-term PAH bioaccumulation because of previous contamination rather than a recent accident. It should be noted that, in the present study, phenanthrene was the dominant PAH congener (i.e., 9.80 ng/g dw, and 10.0 ng/g dw, in the muscle tissue of T. bernacchii and N. coriiceps, respectively), similar to mentioned studies. This result is probably related to the high adsorption capacity of phenanthrene on suspended particular matter and subsequent bioaccumulation by fish.

Although POPs released to the atmosphere are mainly transported through air movements [5], water currents may play important roles in the fate of all the POPs on a local scale. It is reported that the movements around Galindez Island are in the northern direction in the surface zone and in the southern direction in the lower layer, the average current speed in both regions is 4 cm/sec, and the tidal currents are generally less than 10 cm/s [47,65,66]. Therefore, compared with open waters, removal of the contaminants may take a long time, and the increased retention time of the pollutants around the island makes it easier for marine organisms, including fish, to intake contaminants.

5. Conclusions

Comparisons between the present results and long-term data regarding persistent organic pollutant concentrations revealed that the Antarctic continent has lost its pristine features over time. The Argentina Island Archipelago and Galindez Island, in particular, are under the influence of environmental contaminants, and alongside nonpoint sources, increased human activities cause pollutant accumulation, including POPs, in living organisms. Since the renewal capacity of the surrounding water is rather low around the island, the intake of pollutants may increase. After their occurrence, some pollutants may show long-lasting effects around Galindez Island, including bioaccumulation and biomagnification stages in the marine food web. In addition, the availability of contaminants in fish tissue suggests that T. bernacchii and N. coriiceps are ideal bioindicators of local contamination because of their feeding habits and restricted home ranges. Consequently, the implementation of monitoring studies using appropriate bioindicator organisms to determine the pollution trend would be beneficial for Antarctic regions such as Galindez Island that are under the influence of human activities and the effect of contaminants.