In the present study, muscle and bile samples from a total of 33 flatfish were examined for different EC. The study is an expansion of the investigation reported in Maser et al. (2023b) to include a larger geographical area and dump sites, and the results from the present study are consistent with those in Maser et al. (2023b). The mean concentrations measured in bile for each of the EC (TNT, 2-ADNT, 4-ADNT, 2,4-DNT, and 1,3-DNB) were between 0 and 1.25 ng/mL, with higher concentrations of 4- and 2-ADNT than of TNT itself. Koske et al. (2020) also detected EC in the one- to two-digit ng/mL range in the bile of flatfish from the Kolberger Heide, which is a munition dumping area in the western Baltic Sea in Germany. Comparable to our results, the TNT metabolites 4-ADNT (mean 17.06 ng/mL) and 2-ADNT (mean 1.60 ng/mL) were found in higher concentrations, while TNT concentrations were low (around 0.1 ng/mL) (Koske et al. 2020).

In our investigation, EC residues could also be measured in the muscle samples of the flatfish. While the TNT metabolites 2- and 4-ADNT were measured in concentrations of a maximum of 1 ng/g (dry weight), TNT itself was found in the flatfish muscle (mean per dry weight) of 4 ng/g in the Borkum region, 2 ng/g in the Baltrum region, and 3 ng/g in the Außenweser region. Hence, TNT accumulated at higher concentrations in muscle tissue than in the bile. This is interesting, because we have here a reversal of the ratios of unmetabolized TNT compared to its metabolites 2- and 4-ADNT in muscle versus bile. From this, it could be deduced that TNT can accumulate better in the muscle than in the bile, and that there is an alternative uptake of EC into the muscle apart from the uptake via diet. Lotufo (2011) exposed Sheepshead minnows (Cyprinodon variegatus) to radiolabeled isotopes of TNT and detected in total 46% of the TNT metabolites ADNTs resided in the liver and 64% of the parent compound in the viscera. Tissue-specific concentrations were determined with 6 µmol/kg for liver versus 280 µmol/kg in the viscera (Lotufo 2011). Since bile is produced in the liver, similar ratios of TNT to ADNTs should be expected in both compartments.

Further evidence that fish accumulated TNT via the dietary route were provided by Belden et al. (2005), who exposed channel catfish (Ictalurus punctatus) via food pellets containing different concentrations of TNT, and Lotufo and Blackburn (2010), using the amphipods Leptocheirus plumulosus as prey and the fish Cyprinodon variegatus as predator. Houston and Lotufo (2005) exposed the oligochaete worm Lumbriculus variegatus to 14C-labeled TNT for 5 h in water and, after frozen into meal-size packages, fed them to individual juvenile fathead minnows (Pimephales promelas).

As a matter of fact, xenobiotics that are ingested into organisms via diet are absorbed from the intestinal tract into the venous blood stream and distribute in various tissues. Depending on their lipophilicity, these compounds may accumulate in marine organisms along the marine food chain. According to the principles of toxicokinetics, these substances undergo a first pass metabolism in the liver or intestine which leads to the excretion of the metabolites via the bile or urine, or their redistribution within different organs and/or tissues via the blood stream. For example, the TNT metabolites ADNTs and DANT were found in various fish organs, especially DANT in the liver (Beck et al. 2022). Mariussen et al. (2018) showed that TNT is excreted by salmon through the gall bladder, and that TNT transformation products accumulate in bile. This is the reason why bile samples of the fish investigated show higher concentrations of the TNT metabolites 2-ADNT and 4-ADNT compared to TNT (Lotufo 2011) and corresponds to the findings of Ek et al. (2008) who detected mainly 2- and 4-ADNT in fish bile rather than TNT itself. Ownby et al. (2005) found that TNT metabolite accumulation in fish viscera during aqueous exposure was higher than TNT (Ownby et al. 2005). From our results, that fillet contains higher concentrations of non-metabolized TNT compared to both ADNT metabolites, lead us to conclude that TNT enters the fish through additional routes other than diet, possibly through gills. Organ-specific uptake and depuration in Atlantic salmon (Salmo salar) exposed to TNT was studied by Mariussen et al. (2018). They indeed found that TNT is taken up primarily by the gills and rapidly excreted from fish via the bile. Importantly, TNT and the metabolites 2-ADNT and 4-ADNT were found in the muscle tissue, whereas only 2-ADNT and 4-ADNT were found in the bile.

As fish gills represent the major interface between water and the body of a fish, and are strongly perfused with blood, a xenobiotic transfer may occur across the gill lamellae, such that the branchial route should be considered as a toxicant uptake. It has been demonstrated that there is a significant relationship between toxicant uptake and fish oxygen uptake regardless of chemical hydrophobicity and fish species. These results support the view that the main route of toxicant entry for fish is across the gills, where gas exchange occurs. Exchange across the gills is fast and toxicant intake via other sources, e.g., feeding, is much less important (Yang et al. 2000) than generally postulated for water breathing animals. This fits with previous own research that also found higher TNT concentrations in the fillets compared to bile samples in fish caught near a munitions-containing shipwreck (Maser et al. 2023b).

Aquarium studies showing TNT uptake in fish via contaminated water were performed by Lotufo et al. (2016), Yoo et al (2006), and Lotufo and Lydy (2005), while the present study was a field study taking sediment contamination into account. Flounders (Platichthys flesus) are a group of flatfish species that are found at the ocean floor and hide themselves into the sediment as protection against predators. Sediments of dumping sites that contain EC may therefore serve as source of EC contamination of flatfish. Studies investigating sediment as route of exposure (Lotufo et al. 2010) concluded that direct contact with the sediment bed or resuspended sediment is a not relevant route of EC exposure to near-bottom fish. Other studies provided no evidence that sediment contamination with EC is a causative factor for the induction of adverse biological effects in near-bottom fish (Bernet et al. 2011; Rosen and Lotufo 2010). While the body burden of the fish of Lower Saxony with TNT and its metabolites were obvious (Figs. 3 and 4), there were no signs of biological malfunctions visible. In the present investigation, the measured concentrations in sediment samples were near the LoQ (Table 4) at a maximum level of 1.5 ng/kg sediment (Table 5), thereby resembling EC concentrations near the John Mahn wreck site in Belgian waters (Maser et al. 2023b). Higher concentrations (in the ng per g range) have been measured in the Kolberger Heide in the Baltic Sea of Germany (Jansen et al. 2011) or in Eastern Scheldt in The Netherlands (den Otter et al. 2023).

Laboratory studies reveal that EC, especially TNT and its metabolites, have acute and chronic negative effects on various marine species. With regard to fish, a variety of acutely toxic concentrations have been described. Koske et al. (2019) determined an LC50 of 4.5 mg/L for TNT in zebrafish embryos, as well as 13.4 mg/L for 2-ADNT and 14.4 mg/L for 4-ADNT. For the eyespot lyrefish (Synchiropus ocellatus), Nipper et al. (2001) described an LOEC of 10.8 mg/L TNT, based on the survival of the fish larvae. Juhasz and Naidu (2007) published an EC50 of 8.2 mg/L for TNT for the red drummer (Sciaenops ocellatus) and an EC50 of 2.3 mg/L TNT for the gemfish (Cyprinodon variegatus) based on fish mortality. Talmage et al. (1999) reported LC50 concentrations in the range of 0.8 to 3.7 mg/L TNT for four fish species in their review. The authors independently agree in their publications that fish are among the most sensitive organisms to exposure to EC. Furthermore, Lotufo et al. (2010) showed in a laboratory study that more than 90% of juvenile sheepshead minnows (Cyprinodon variegatus) survive a 4-day exposure to sediment spiked with 7 mg/kg TNT, but when the TNT concentration in the sediment reached 340 mg/kg did all the fish die within 24 h.

Other studies reported lethal and sublethal effects at TNT concentrations below 1 mg/L. For example, Koske et al. (2019) showed that TNT and its metabolites 2- and 4-ADNT damage the DNA of zebrafish embryos even at the lowest tested concentrations (0.1 mg/L for TNT and 1 mg/L for 2- and 4-ADNT), while Liu et al. (1983) determined lethal concentrations of TNT to fish in the range of 0.8–5.0 mg/L water. Behavioral responses of the fathead minnow, such as lethargy and loss of motor control, have also been observed after exposure to TNT for 96 h at a concentration of 0.46 mg/L (Smock et al. 1976).

For an ecotoxicological risk assessment, the actual EC concentrations in the marine environment must be taken into account. In the present study, the measured concentrations in sediment samples from Lower Saxony were a maximum of 1.5 ng/kg sediment. These concentrations are several orders of magnitude lower than the acute effect concentrations shown above for fish in laboratory studies. Seen from this perspective, the measured EC concentrations in our study do not appear to pose an acute health problem for the fish living there.

However, a direct extrapolation from laboratory studies at relatively high EC concentrations should only be done with caution, as so far little is known about the long-term effects of low EC concentrations. In this context, the so-called cocktail effect must also been considered. This phenomenon describes the enhancement of toxic properties of various individual substances or groups of substances through additive effects within an organism, even if the measured concentration of an individual substance is below its previously known effect threshold. Negative effects on fish cannot therefore be completely ruled out. Mariussen et al. (2018) examined the effects of TNT on juvenile Atlantic salmon (Salmo salar). Fish were exposed to dissolved TNT with an initial concentration of 1 mg/L for 48 h. At the end of the exposure experiment, the mortality of the fish was increased compared to the control. All salmon, including those that survived the experiment, were found to have severe bleeding in the back-muscle tissue near the spine, as well as effects on blood parameters, such as glucose, urea, hematocrit, and hemoglobin. The authors concluded that if the exposure period had been extended, all fish would have died from the severe effects of TNT. Leffler et al. (2014) exposed alevins of Atlantic salmon to TNT wastewater for 40 days. In the high exposure group (2.1 mg/L TNT), they observed approximately 25% mortality after 14 days and 100% mortality after 40 days. In the group exposed to 0.41 mg/L, they observed approximately 30% mortality after 40 days.

In a field study in the Kolberger Heide munitions dumping area (Kiel Bight, western Baltic Sea, Germany), low TNT concentrations between 0.5 and 51.5 ng/L were found in the water with a median concentration of 3 ng/L (Esposito et al. 2023). Interestingly, poorer health status was demonstrated in flatfish from the same area than in reference areas, with the fish also having, for example, higher rates of liver nodules and tumors (Straumer and Lang 2019).

Figure 2 shows an overview of the sampling locations in Lower Saxony with the munition-contaminated areas, suspected areas, and munition dumping areas. Several hot spots regarding the release of EC are emerging here. These include the regions in the Jade area and the Jade Bay as well as the East Frisian Islands, especially Borkum and Baltrum. In the vicinity of the Borkum and Baltrum fishing areas, estimated 2000 metric tons of mines, grenades, bombs, torpedoes, rocket-propelled grenades, small ammunition, as well as plate and sea mines were dumped in these areas (Böttcher et al. 2011). In addition, because of so-called on-route dumping at that time munition-contaminated sites can also be expected outside the declared areas. The Jade Bay and Jade collection region also overlap with a munitions-contaminated or munitions dumping area (Böttcher et al. 2011). According to estimates, between 650,000 and 1.2 million metric tons of conventional munitions are located in these areas (Böttcher et al. 2011). Only in the Wurster Watt and Spiekeroog, sampling spots are no known munition dumpsites or suspected areas within a radius of three kilometers (Böttcher et al. 2011). These were also the only areas investigated in which no evidence of EC could be found in the sediment samples in this study.

With regard to food safety, it is of interest in how far the carcinogenic EC accumulate in the edible part of sea-food species. Bioaccumulation of chemicals from one compartment to another, or from one species to another, is defined as concentrations increasing by a factor of higher than 1000-fold, while values below 1000-fold are regarded as bioconcentration (Lotufo et al. 2013; Ownby et al. 2005). Whether or not a substance has the potential to bioaccumulate is dependent on its logKow (Arnot and Gobas 2006). The bioconcentration factor for TNT varies widely from 0.3 to 9.7 mL g−1 for various marine and freshwater invertebrates and fish species and was until today regarded as a compound that bioconcentrates rather than bioaccumulates (Lotufo et al. 2013). However, when considering the transfer of TNT from sediment samples in the range of 1 ng/kg of sediment, or even below, on the one side, and the occurrence of TNT in the fish fillets in the range between 2000 and 4000 ng/kg, then the demand of bioaccumulation has been fulfilled in our findings. Here, it is interesting to speculate that flatfish living in and feeding from the sediment are continuously exposed to EC. Moreover, flatfish hide in the sediment and stir up sediment in search of food, thereby mobilizing EC.

The question now arises as to whether the consumption of this TNT-contaminated fish poses a health risk for humans. Particularly noteworthy here is the chronic toxic risk of consuming low-contaminated fish and seafood in terms of the potential carcinogenicity of TNT (Bolt et al. 2006). Calculations show that, even with a lifelong daily consumption of an average consumption of approx. 39 g (FIZ 2017) of the fillet of the fish examined in this project, no negative health effects should be expected for the human consumer (Maser and Strehse 2021), as the TNT concentrations were in the single-digit nanogram range per gram of dry weight.

However, this could worsen in the coming years as the corrosion of submerged munitions continues, thereby increasing the release of EC into the marine environment. Studies in the Kolberger Heide munition dumping area in the Baltic Sea near Kiel (Germany) have shown an increased uptake of EC in blue mussels exposed to free-lying chunks of hexanite (German Schiesswolle), when compared to mussels mounted near corroding moored mines with more or less intact metal shells (Appel et al. 2018; Strehse et al. 2017). From the EC concentrations found in the highly contaminated blue mussels, it was concluded that they were no longer suitable for consumption due to the carcinogenic risk (Maser and Strehse 2021). Targeted blast-in-place detonations of munitions underwater also lead to a drastic increase in EC concentrations in the surrounding sediment and water (Maser et al. 2023a), which may also enhance the contamination of fish living in the nearby area.

In summary, the present investigations in the areas of the Lower Saxony North Sea showed that there is a correlation between munition deposits and the occurrence of EC in sediment and flatfish living there and that a transfer of EC from the munition items into the fish is obvious. So far, the EC concentrations found in the sediments and flatfish are low, but due to live-long exposure, there is a risk that the fish themselves will experience negative effects on their health. Whatsoever, the EC concentrations in the fillet as the edible portion of the fish are so low that there is no danger to humans if they consume these fish. However, the further corrosion of the munition bodies could lead to toxic levels of EC in fish in the future. This would particularly affect flatfish, which are relatively stationary and prefer to stay in the sediments of the seabed.