Mussels document loss of bioavailable polycyclic aromatic hydrocarbons and the return to baseline conditions for oiled shorelines in Prince William Sound, Alaska

Polycyclic aromatic hydrocarbons (PAH) were measured in mussels (Mytilus trossulus) collected between 1990 and 2002 from 11 sites on the shores of Prince William Sound (PWS), Alaska, that were heavily oiled by the 1989 Exxon Valdez oil spill (EVOS). This study, utilizing the methods of the NOAA Status and Trends Mussel Watch Program, found that concentrations of PAH released from spill remnants have decreased dramatically with time and by 2002 were at or near the range of total PAH (TPAH) of 3-355 ng/g dry weight obtained for mussels from unoiled reference sites in PWS. Time-series TPAH data indicate a mean TPAH half-life in mussel tissues of 2.4 years with a range from 1.4 to 5.3, yielding an annual mean loss of bioaccumulated TPAH of 25%. The petroleum-derived TPAH fraction in mussel tissues has decreased with time, reflecting the decreasing release of EVOS residues in shoreline sediments. These results show that PAH from EVOS residues that remain buried in shoreline sediments after the early 1990s are in a form and at locations that have a low accessibility to mussels living in the intertidal zone.     

Semipermeable membrane devices link site-specific contaminants to effects: Part 1 - Induction of CYP1A in rainbow trout from contaminants in Prince William Sound, Alaska

Extracts from semi-permeable membrane devices (SPMDs) deployed on beaches in Prince William Sound (PWS), Alaska, were used to evaluate if complex contaminant mixtures from different sources can be distinguished by the resulting cytochrome P450 1A (CYP1A) activity in exposed test animals. Deployment sites included canneries, salmon hatcheries, and beaches where lingering oil remains from discharges during the 1964 earthquake or the 1989 Exxon Valdez oil spill. Other sites were selected at random to evaluate region-wide contaminant inputs or were located in salmon streams to evaluate contaminants carried and released by migrating salmon carcasses following reproduction. Following standard deployments of approximately 28 d, an aliquot of the accumulated contaminants was intraperitoneally injected without cleanup into juvenile rainbow trout (Oncorhynchus mykiss). After 2 d and 7 d, the activity of CYP1A was measured by the ethoxyresorufin-o-deethylase (EROD) assay. Exposure to extracts from the oiled sites and one hatchery site with numerous creosote pilings elicited strong EROD responses, whereas fish exposed to salmon stream extracts elicited weak but significant responses during late summer compared to late spring. Responses from the other sites were not significant, indicating contaminants from these sources are unlikely to cause CYP1A induction in resident biota. Rather than simply assessing extant contaminants, this method evaluates the potency of the different sites for bringing about aryl hydrocarbon receptor responses in resident biota.     

Persistence of oiling in mussel beds after the Exxon Valdez oil spill

Persistence and weathering of Exxon Valdez oil in intertidal mussel (Mytilus trossulus) beds in Prince William Sound (PWS) and along the Gulf of Alaska was monitored from 1992 to 1995. Beds with significant contamination included most previously oiled areas in PWS, particularly within the Knight Island group and the Kenai Peninsula. In sediments, yearly mean concentrations of total petroleum hydrocarbons ranged from < 60 micrograms/g in reference beds to 62,258 micrograms/g wet wt., or approximately 0 to 523 micrograms/g dry wt. total polynuclear aromatic hydrocarbons (TPAHs). In mussels, mean TPAH concentrations ranged up to 8.1 micrograms/g dry wt. Hydrocarbon concentrations declined significantly with time in some, but not all mussels and sediments, and should reach background levels within three decades of the spill in most beds. In 1995, mean hydrocarbon concentration was greater than twice background concentration in sediments from 27 of 34 sites, and in mussels from 18 of 31 sites.     

Restoration of oiled mussel beds in Prince William Sound, Alaska

Natural loss of hydrocarbons was often low from mussel (Mytilus trossulus) beds (which were typically not cleaned after the Exxon Valdez oil spill), thus this habitat remained a long-term source of oil. Consequently, experimental restoration of nine contaminated beds was attempted in 1994; mussels were removed, contaminated surface sediment was replaced (33 metric tons), and original mussels were returned. Hydrocarbon concentrations and mussel populations were monitored for 5 years thereafter. Post-restoration mussel population fluctuations were indistinguishable from regional changes. Increased short-term oil loss was apparent, but long-term (5 year) improvement was equivocal and difficult to distinguish from natural losses. By 1999, oil concentrations in mussels were typically at baseline levels in restored and oiled reference beds; concentrations in replaced sediment were elevated in one third of restored beds, indicating recontamination from underlying or surrounding sediment. Our results suggest mussel relocation is feasible but suggest oil might more effectively be removed from sediment mechanically or chemically than manually.     

Retrospective analysis: bile hydrocarbons and histopathology of demersal rockfish in Prince William Sound,Alaska, after the Exxon Valdez oil spill

Demersal rockfish are the only fish species that have been found dead in significant numbers after major oil spills, but the link between oil exposure and effect has not been well established. After the 1989 Exxon Valdez oil spill in Prince William Sound, Alaska, several species of rockfish (Sebastes spp.) from oiled and reference sites were analyzed for hydrocarbon metabolites in bile (1989-1991) and for microscopic lesions (1990 and 1991). Biliary hydrocarbons consistent with exposure to Exxon Valdez oil were elevated in 1989, but not in 1990 or 1991. Significant microscopic findings included pigmented macrophage aggregates and hepatic megalocytosis, fibrosis, and lipid accumulation. Site differences in microscopic findings were significant with respect to previous oil exposure in 1991 (P=0.038), but not in 1990. However, differences in microscopic findings were highly significant with respect to age and species in both years (P<0.001). We conclude that demersal rockfish were exposed to Exxon Valdez oil in 1989, but differences in microscopic changes in 1990 and 1991 were related more to age and species differences than to previous oil exposure.     

Evaluation of bioavailable hydrocarbon sources and their induction potential in Prince William Sound, Alaska.

To realistically evaluate the consequences of exposure to a complex mixture, we modified a passive sampler technology, the semipermeable membrane device (SPMD), which absorbs the bioavailable hydrophobic organic compounds present in an environment. These samplers were deployed in Prince William Sound (PWS), Alaska, at locations selected as potential sites of hydrocarbon deposition, as well as in random sites for regional assessment.Some of these sites were affected by previous human activity, such as canneries and salmon hatcheries, while others were sites of oil discharge as a consequence of the 1964 earthquake or the oil spill of T/V Exxon Valdez in 1989. The SPMDs were deployed for 27-28 d, processed, and then split, with one aliquot dedicated to chemical analysis and the other injected into juvenile rainbow trout (Oncorhynchus mykiss), along with the proper controls including a solvent control, field blank, and positive control. Trout fry were sacrificed after 2 or 7d, and their livers assayed for CYP1A induction by the standard bioassay for hydrocarbon exposure, the ethoxyresorufin-o-deethylase (EROD) assay. The results of this study were consistent and reproducible and showed that oil, whether deposited in 1964 or 1989, is still bioavailable as it can elicit as sustained response. Also, the same oil deposited in different sites of the same region has degraded differently, which is demonstrated by this method. Other putative sources of hydrocarbons, such as oil seeps, were dismissed as regional sources of induction agents as the responses following injection of modified SPMD extract from those sites did not differ significantly from the solvent control. This is a flexible, sensitive method that assesses the response to site-specific bioavailable contaminants and does so within the normal physiological response range of the target.     

Impact of crude oil on sulphate reduction and methane production in sediments impacted by the Amoco Cadiz oil spill

The activities of methane-producing and sulphate-reducing bacteria in intertidal sediments along the Brittany coast of France were examined in order to determine the effect of the Amoco Cadiz oil spill on sediment microbial processes. Porewater chemistry, methane production, sulphate rate and [2?¹⁴C]-acetate metabolism did not vary significantly between beach, estuary, and marsh sites, oiled or unoiled, after the Amoco Cadiz spill. Oiled sediments contained highly weathered oil of Amoco Cadiz origin, but unoiled comparison sites also contained hydrocarbons from anthropogenic sources. The additions of weathered Amoco Cadiz mousse, fresh and slightly weathered light Arabian crude oil, benzene or toluene to sediments from the oiled and unoiled marsh site did not significantly affect rates of sulphate reduction or methane production. The oxidation of [2?^14C]-acetate to ¹⁴CO₂, however, was significantly decreased when mousse, crude oil, benzene or toluene was added to sediments from the unoiled site. Inhibition seemed to be proportional to the extent of weathering. Sediments recently exposed to Amoco Cadiz crude oil were less affected by a second oiling, suggesting that the indigenous populations may have been replaced by organisms more resistant to oil toxicity. These results suggested that the Amoco Cadiz oil spill did not have major long-term effects on sulphate reduction or methane production.     

Hepatic CYP1A levels and EROD activity in English sole: biomonitoring of marine contaminants in Vancouver Harbour

To assess chemical contaminant stress in the marine environment, ethoxyresorufin-O-deethylase (EROD) activity and cytochrome P450 1A (CYP1A) expression were measured in 88 English Sole (Pleuronectes vetulus) collected during May and June 1999 from four sites in Vancouver Harbour and at an expected reference site outside the harbour. Hepatic microsomes were prepared from the fish and analyzed for total CYP content, EROD activity, and CYP1A protein levels. Hepatic EROD activity and CYP1A protein levels were elevated in fish from two sites in the inner harbour. A comparison with sediment chemistry data showed that fish with increased EROD activity and CYP1A levels came from sites containing relatively high levels of polycyclic aromatic hydrocarbons and polychlorinated biphenyls. Unexpectedly high levels of EROD activity and CYP1A protein were also found in fish from a reference site near Gibsons, in Howe Sound. The elevated EROD activity and CYP1A expression in fish from this site cannot be explained by the chemical analysis data collected.     

Using cytochrome P450 to monitor the aquatic environment: Initial results from regional and national surveys

We are currently analyzing hepatic cytochrome P4501A and associated monooxygenase activities in fish sampled in several regional and national monitoring programs, including the National Benthic Surveillance Project of NOAA's Status and Trends Program, damage assessment studies of the Exxon Valdez oil spill, and intensive surveys of specific embayments, such as Puget Sound, Washington. Thus far, apparent contaminant-related increases in the activities of cytochrome P4501A-dependent monooxygenases have been readily measured in most test species. The results presented in this paper show that, for II species of fish, there is excellent concordance between hepatic activities of aryl hydrocarbon hydroxylase (AHH) and ethoxyresorufin-O-deethylase (EROD); moreover, levels of cytochrome P4501A determined by an enzyme-linked immunosorbent assay (ELISA) are also generally concordant with results from catalytic assays. The use of both a catalytic assay and immunoquantitation is recommended, because of the additional quality assurance provided by concurrent use of an immunoquantitation technique, which is desirable in large monitoring programs.     

Apparent lack of CYP1A response to high PCB body burdens in fish from a chronically contaminated PCB site

Chronic exposure to organic contaminants such as polychlorinated biphenyls (PCBs) can lead to the development of resistance to these chemicals, a condition associated with reduced response of CYP1A1, a pollutant-inducible biomarker. We measured CYP1A activity (ethoxyresorufin o-deethylase, EROD) and PCB concentrations in feral fish from the Town Branch/Mud River system (Logan County, KY), a stream historically contaminated with PCBs and partially remediated. As a first step in evaluating the possible development of resistant populations in this system, we measured CYP1A expression and PCB body burdens in resident fish from sites we previously characterized as containing biologically significant levels of CYP1A inducing compounds. Mean PCB concentrations in edible flesh ranged from 75.2 to 16.7 microg/g in fish collected from Town Branch remediated sites and were relatively low (1.23 microg/g) in Town Branch reference site fish. However, hepatic CYP1A activity was similar among individuals of most species collected from reference and contaminated/remediated sites. The absence of elevated CYP1A levels in resident fish species despite the presence of significant PCB body burdens may indicate these fish have developed reduced sensitivity to CYP1A induction, a condition associated with acquired resistance to toxicants.     

Semipermeable membrane devices link site-specific contaminants to effects: PART II - A comparison of lingering Exxon Valdez oil with other potential sources of CYP1A inducers in Prince William Sound, Alaska

We deployed semipermeable membrane devices (SPMDs) on beaches for 28 days at 53 sites in Prince William Sound (PWS), Alaska, to evaluate the induction potential from suspected sources of cytochrome P450 1A (CYP1A)-inducing contaminants. Sites were selected to assess known point sources, or were chosen randomly to evaluate the region-wide sources. After deployment, SPMD extracts were analyzed chemically for persistent organic pollutants (POPs) and polycyclic aromatic hydrocarbons (PAH). These results were compared with hepatic CYP1A enzyme activity of juvenile rainbow trout injected with the same extracts prior to clean-up for the chemical analyses. Increased CYP1A activity was strongly associated with PAH concentrations in extracts, especially chrysene homologues but was not associated with POPs. The only apparent sources of chrysene homologues were lingering oil from Exxon Valdez, asphalt and bunker fuels released from storage tanks during the 1964 Alaska earthquake, creosote leaching from numerous pilings at one site, and PAH-contaminated sediments at Cordova Harbor. Our results indicate that PWS is remarkably free of pollution from PAH when nearby sources are absent as well as from pesticides and PCBs generally.     

Semipermeable membrane devices link site-specific contaminants to effects: PART II – A comparison of lingering Exxon Valdez oil with other potential sources of CYP1A inducers in Prince William Sound,Alaska

We deployed semipermeable membrane devices (SPMDs) on beaches for 28 days at 53 sites in Prince William Sound (PWS), Alaska, to evaluate the induction potential from suspected sources of cytochrome P450 1A (CYP1A)-inducing contaminants. Sites were selected to assess known point sources, or were chosen randomly to evaluate the region-wide sources. After deployment, SPMD extracts were analyzed chemically for persistent organic pollutants (POPs) and polycyclic aromatic hydrocarbons (PAH). These results were compared with hepatic CYP1A enzyme activity of juvenile rainbow trout injected with the same extracts prior to clean-up for the chemical analyses. Increased CYP1A activity was strongly associated with PAH concentrations in extracts, especially chrysene homologues but was not associated with POPs. The only apparent sources of chrysene homologues were lingering oil from Exxon Valdez, asphalt and bunker fuels released from storage tanks during the 1964 Alaska earthquake, creosote leaching from numerous pilings at one site, and PAH-contaminated sediments at Cordova Harbor. Our results indicate that PWS is remarkably free of pollution from PAH when nearby sources are absent as well as from pesticides and PCBs generally.     

Impact of the Sea Empress oil spill on lysosomal stability in mussel blood cells.

Coastal zones are among the most productive and vulnerable areas on the planet. An example of impact on these fragile environments was shown in the case of the "Sea Empress" oil tanker, which ran aground in the Bristol Channel in 1996, spilling 72,000 tons of "Forties" crude oil. The objective was to investigate the sub-lethal cellular pathology and tissue hydrocarbon contamination in marine mussel populations, 4 months after the initial spill, using the neutral red retention (NRR) assay for lysosomal stability in blood cells. NRR was reduced in mussels, and indicative of cell injury, from the two sites closest to the spill in comparison with more distant and reference sites. Lysosomal stability was inversely correlated with polycyclic aromatic hydrocarbon concentrations in mussel tissues. Reduced lysosomal stability has previously been shown to contribute to impaired immunocompetence and to autophagic loss of body tissues. The use of this type of technique is discussed in the context of cost-effective, ecotoxicological tools for Integrated Coastal Zone Management.     

Hindcast of the Argo merchant spill using the uri oil spill fates model

A comprehensive oilspill fates model, developed by Cornillon and Spaulding at the University of Rhode Island, has been employed to hindcast the well-studied Argo Merchant spill. Using data collected during the spill to define the environmental conditions and the oil composition, properties and release rates, the oil spill fates model was used to predict the spatial and temporal distribution of spilled oil from December 16 to 31, 1976. Comparison of model predictions to observations for spilled oil mass balance, surface trajectories and oiled surface area was in general quite good. Improved predictions require a better definition of the wind and tidally-forced circulation in the vicinity of the wreck site, a more soundly based theory for oil entrainment or dispersion into the water column and an oil spreading theory which more accurately addresses the complex behaviour of a weathering multicomponent petroleum hydrocarbon.     

The expression of CYP1A,vitellogenin and zona radiata proteins in Atlantic salmon (Salmo salar) after oral dosing with two commercial PBDE flame retardant mixtures: absence of short-term responses

The short-term effects of the commercial PBDE flame retardant mixtures Penta-BDE and cta-BDE on the expression of cytochrome P450 1A (CYP1A), vitellogenin (Vtg) and zona radiata proteins (Zrp) were investigated in juvenile salmon (Salmo salar). For this purpose, groups of fish were dosed twice (oral intake at days I and 4) with 10 and 50 mg/kg body weight of both commercial mixtures. The fishes were sacrificed at day 7 (n = 5 for each group) and 14 (n = 6 for each group), and blood, liver, fillet, and brain were collected. Blanks and positive controls were also part of the experiment. The expressions of Vtg, Zrp, and CYPIA were measured with several techniques (EROD, ELISA, Western, Northern and Slot Blot). The values in the groups of fish treated with Penta-BDE or Octa-BDE did not significantly differ from the reference group for any of the parameters tested. In contrast, the positive control groups treated with estradiol-17beta for Vtg and Zrp expression, and beta-naphthoflavone for CYP1A expression did show a significant response, indicating the potential sensitivity of the fishes for the parameters measured. Since the results of the chemical analyses showed concentrations of a number of PBDE congeners in liver, fillet, and brain that were about three orders of magnitude above those of fish from the North Sea, it is concluded that the short-term toxicity of both commercial PBDE mixtures for these endpoints was low.     

Initial impact of the Treasure oil spill on seabirds off western South Africa

On 23 June 2000, the bulk ore carrier MV Treasure sank off western South Africa between Dassen and Robben islands, which individually currently support the largest and 3^rd largest colonies of African penguins Spheniscus demersus. Subsequently, more than 19 000 penguins were oiled, almost twice the previous highest number of seabirds oiled during a single event in southern Africa (10 000 penguins after the sinking of the Apollo Sea in June 1994). About 19 000 oiled penguins were collected for cleaning and care and about 150 oiled adults died in the wild. Some 19 500 unoiled penguins were caught at Dassen and Robben islands and relocated to Port Elizabeth, 800 km to the east, to remove them from waters affected by the oil. Of all penguins caught, which amounted to 20% of the total species population, less than 2 000 died within the first month, considerably less than in the Apollo Sea spill. This can be attributed to improved transport of penguins and the rapid arrival at rescue centres of experts able to administer emergency care. However, resources were severely extended and mortality would probably have been considerably higher had large numbers of birds not been removed from the area affected by the oil. Many relocated birds returned to their home islands within a month of being released, but considerable disruption of pair bonds is expected to result from mortality, different periods in captivity and disruption of moult cycles. This is likely to result in decreased breeding success. Recruitment to colonies will also be reduced by substantial loss of chicks and eggs. Although more than 3 000 orphaned chicks were collected for captive rearing, an estimated 4 000 died at the islands before they could be rescued. Up to 20% of bank cormorants Phalacrocorax neglectus at Robben Island, the 3^rd largest colony of the species in South Africa, died. There was low success in catching oiled cormorants and in saving those that were caught. Of 53 grown birds of four species of cormorant that were oiled and caught, only 17 survived. Captive rearing of bank cormorant chicks, which it was feared may have been orphaned, proved more successful. Spilt oil had minor impact on gulls, terns and shorebirds in the region.     

The expression of CYP1A, vitellogenin and zona radiata proteins in Atlantic salmon (Salmo salar) after oral dosing with two commercial PBDE flame retardant mixtures: absence of short-term responses

The short-term effects of the commercial PBDE flame retardant mixtures Penta-BDE and Octa-BDE on the expression of cytochrome P450 1A (CYP1A), vitellogenin (Vtg) and zona radiata proteins (Zrp) were investigated in juvenile salmon (Salmo salar). For this purpose, groups of fish were dosed twice (oral intake at days 1 and 4) with 10 and 50 mg/kg body weight of both commercial mixtures. The fishes were sacrificed at day 7 (n=5 for each group) and 14 (n=6 for each group), and blood, liver, fillet, and brain were collected. Blanks and positive controls were also part of the experiment. The expressions of Vtg, Zrp, and CYP1A were measured with several techniques (EROD, ELISA, Western, Northern and Slot Blot). The values in the groups of fish treated with Penta-BDE or Octa-BDE did not significantly differ from the reference group for any of the parameters tested. In contrast, the positive control groups treated with estradiol-17β for Vtg and Zrp expression, and β-naphthoflavone for CYP1A expression did show a significant response, indicating the potential sensitivity of the fishes for the parameters measured. Since the results of the chemical analyses showed concentrations of a number of PBDE congeners in liver, fillet, and brain that were about three orders of magnitude above those of fish from the North Sea, it is concluded that the short-term toxicity of both commercial PBDE mixtures for these endpoints was low.     

Differential gene expression in mummichogs (Fundulus heteroclitus) following treatment with pyrene: comparison to a creosote contaminated site

Mummichogs, Fundulus heteroclitus, an estuarine fish with a relatively small home range found along the eastern coast of the United States are well-suited to monitoring contaminant effects, including those of polycyclic aromatic hydrocarbons (PAHs). One of the common PAHs in estuaries is pyrene. We report here on efforts to develop multiple biomarkers of pyrene exposure in this species. Adult male mummichogs were exposed in the laboratory to the weak aryl hydrocarbon receptor (AhR) agonist pyrene at 0, 30, or 50 microg/L in 7-day static renewal exposures. The RNA was extracted from livers and alterations in mRNA expression were assessed by subtractive hybridization and differential display in order to produce multiple biomarkers of pyrene exposure. Genes demonstrating differential expression were confirmed by quantitative-PCR (Q-PCR) and include cytochrome P-450 1A (CYP1A), a putative hepatocyte growth factor activator, a X-ray inducible retrotransposon, and several expressed sequenced tags (ESTs). Some of these genes represent new biomarkers of pyrene exposure and potential biomarkers of PAH exposure. Therefore, similar changes were investigated at a Superfund site in Charleston, SC. Mummichogs from a creosote contaminated site and from a reference site (North Inlet National Estuarine Research Reserve near Georgetown, SC) were trapped, RNA extracted from the livers, and Q-PCR performed. Many of the genes differentially expressed following pyrene exposure were not altered at the creosote contaminated site in comparison to the reference site. However, CYP1A and an EST were induced. CYP1A induction at Diesel Creek indicates that this population of fish does not demonstrate refractory CYP1A phenotypes observed at several sites with high levels of AhR agonists. Ultimately, we anticipate that the use of multiple biomarkers of PAH exposure will provide useful information on the potential effects of toxicants.     

Assessment of pollution in the West Black Sea Coast of Turkey using biomarker responses in fish

Aim of this study was to determine the extent of pollution in the West Black Sea Coast of Turkey by measuring CYP1A associated EROD activity, phase II enzyme, glutathione S-transferase and antioxidant enzymes, catalase and glutathione reductase activities and immunochemical detection of CYP1A protein level in the liver of mullet. The fish samples were caught from six locations having a varying degree of pollution in the West Black Sea Region of Turkey in August 2005, 2006 and 2007. Mullets caught from Zonguldak Harbour, Ere?li Harbour and Gülüç Stream’s Mouth displayed 6–9-fold higher EROD, 2–4-fold higher glutathione S-transferase and 2–3-fold higher catalase activities than the reference site, Amasra. Total polyaromatic hydrocarbon levels in mullets caught from these locations were also significantly higher (2–4-fold) than Amasra. The results of this study indicate that Zonguldak Harbour, Ere?li Harbour and Gülüç Stream are highly polluted by polycyclic aromatic hydrocarbons and related contaminants.     

Assessing and monitoring local and long-range-transported hydrocarbons as potential stressors to fish stocks

The Arctic is poorly described and understood compared to the North Sea and other boreal areas. We need to learn how knowledge obtained at boreal conditions may be applied at Arctic conditions. Increasing activity of the hydrocarbon industry may cause exposure stress from discharges. Within a joint industry research programme (Biosea) we studied how responses in fish from oil exposure may differ at North Sea boreal and Barents Sea Arctic conditions. Hydrocarbon uptake, metabolites, and enzymatic and genotoxic biomarker type of effect responses were measured in cod (Gadus morhua L.). Hydrocarbon metabolites remain longer in fish bile than original hydrocarbons, which are eliminated fast from tissues. The metabolites may be measured to background concentrations. They describe exposure and they constitute a link to other effects. Body burden, biliary polyaromatic hydrocarbon (PAH) metabolite concentration, and quantities of Cytochrome P-450 1A (CYP1A) and DNA adducts increased with oil in water concentration. The extent of biomarker expression was lower for some biomarkers and elimination was slower at the lowest temperature. The results show that several factors have to be accounted for if warm-water biomarker data are to be applied in cold water.