Measurement of stress effects (scope for growth) and contaminant levels in mussels (Mytilus edulis) collected from the Irish Sea

Marine Environmental Research 53 (2002) 327–356 www.elsevier.com/locate/marenvrev

Measurement of stress effects (scope for growth) and contaminant levels in mussels (Mytilus edulis) collected from the Irish Sea J. Widdowsa,*, P. Donkina,1, F.J. Staffa, P. Matthiessenb,2, R.J. Lawb, Y.T. Allenb, J.E. Thainb, C.R. Allchinb, B.R. Jonesb a Plymouth Marine Laboratory, Prospect Place, The Hoe, Plymouth, PL1 3DH, UK Centre for Environment, Fisheries and Aquaculture Science, Burnham Laboratory, Remembrance Avenue, Burnham-on-Crouch, Essex, CM0 8HA, UK

b

Received 27 June 2001; received in revised form 7 August 2001; accepted 31 August 2001

Abstract The objective of this research was to quantify the impact of pollution along the coastlines of the Irish Sea. Pollution assessment was based on the combined measurement of scope for growth (SFG), and chemical contaminants in the tissues of mussels (Mytilus edulis) collected from 38 coastal sites around the Irish Sea during June–July in 1996 and 1997. On the UK mainland coast, the SFG showed a general trend with a significant decline in water quality in the Liverpool and Morecambe Bay region. High water quality was recorded along the west coast of Wales, as well as southwest England and northwest Scotland (clean reference sites outside the Irish Sea). Along the coast of Ireland there was a similar trend with reduced SFG within the Irish Sea region. SFG was generally low north of Duncannon and then improved north of Belfast. The poor water quality on both sides of the Irish Sea is consistent with the prevailing hydrodynamics and the spatial distribution of contaminants associated with urban/ industrial development. The decline in SFG of mussels on both sides of the Irish Sea was associated with a general increase in contaminant levels in the mussels. Certain contaminants, including PAHs, TBT,
DDT, Dieldrin, g-HCH, PCBs, and a few of the metals (Cd, Se, Ag, Pb), showed elevated concentrations. Many of these contaminants were particularly elevated in the coastal margins of Liverpool Bay, Morecambe Bay and Dublin Bay. A quantitative * Corresponding author. Tel.: +44-1752-633100; fax: +44-1752-633101. E-mail address: [email protected] (J. Widdows). 1 Present address: Department of Environmental Sciences, University of Plymouth, Drake Circus, Plymouth, PL4 8AA, UK. 2 Present address: CEH Windermere, The Ferry House, Far Sawrey, Ambleside, Cumbria, LA22 0LP, UK. 0141-1136/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0141-1136(01)00120-9

328

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

toxicological interpretation (QTI) of the combined tissue residue chemistry and SFG measurements indicated that at the majority of coastal sites, c. 50 to >80% of the observed decline in SFG was due to PAHs as a result of fossil fuel combustion and oil spills. TBT levels were highest at major ports and harbours, but these concentrations only made a minor contribution to the overall reduction in SFG. At no sites were individual metals accumulated to concentrations that could cause a significant effect on SFG. The study identified many sites where the observed reduction in SFG was far greater than predicted from the limited number of chemical contaminants analysed, thus indicating the presence of additional ‘unknown toxicants’. Sewage (containing domestic, agricultural and industrial components) appears to be an important contributor to reduced SFG and linear alkylbenzenes (LABs) and As may provide suitable ‘sewage markers’. There was a highly significant positive correlation between SFG and As (P< 0.001). This relationship may be due to reduced As uptake by algal food material and mussels at sites with elevated PO4 concentrations (e.g. at sites with sewage inputs). Phosphate is a known competitive inhibitor of As accumulation, at least in algae. The results highlight that further research is required on ‘sewage markers’ in mussels. The SFG approach therefore provides a rapid, cost-effective and quantitative measure of pollution impact, as well as a means of identifying the causes through a QTI of tissue contaminants levels. It also serves to identify the presence of unidentified toxicants and areas that require further study. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Irish Sea; Mussels; Scope for growth; Contaminants; Pollution monitoring; Effects-physiology; Hydrocarbons; Metals; Sewage

1. Introduction The Irish Sea is generally defined as the area north of a line from St. Anne’s Head (Wales) across to Hook Head (Ireland), and south of a line from Rathlin Island (Northern Ireland) to Mull of Kintyre (Scotland). It is a relatively small, enclosed sea area of c. 52,000 km2, with shallow coastal waters of < 50 m depth representing 60% of the area. The Irish Sea is subject to a wide range of human uses and activities including shipping, oil terminals, direct sewage input, industrial waste, cooling water for nuclear and non-nuclear power stations, gravel extraction and fishing (Shaw, 1990a, 1990b, 1990c, 1990d). The hydrodynamics within the Irish Sea, driven primarily by tides and wind, consists of a general northerly flow of water from the Celtic Sea, through the St. George’s Channel in the southern Irish Sea, and out through the North Channel into the Atlantic (Bowden, 1980). There is also a tidally propagated input along the coastline of Northern Ireland into the Irish Sea through the North Channel. The time-averaged circulation of the Irish Sea is relatively weak and with no particularly coherent directionality over large areas (models by Proctor, 1981). Stratification of the water column occurs due to summer heating in the area to the southwest of the Isle of Man, and to a lesser extent in Liverpool Bay. In the winter and spring there is also significant stratification in the Liverpool Bay area due to river inputs along the Lancashire and Cumbrian coasts (reviewed by Dickson & Boelens, 1988). Recent studies have highlighted a summer gyre in the western Irish Sea with some important management implications (Hill, Brown, & Fernand, 1997).

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

329

To date, Irish Sea monitoring programmes have focused primarily on the concentrations of a few selected metals and organic contaminants present in the water, and accumulated in sediments and biota (Achterberg & Van den Berg, 1996; Fitzsimmons, Rashid, Riley, & Wolf, 1995; Franklin, 1987, 1992; Gardner, 1978; Laslett, 1995; Law, 1981; Law et al., 1991; Leah, Evans, & Johnson, 1992; Leah, Johnson, Connor, & Levene,1997; MPMMG, 1998; Nixon, McLaughlin, Boelins, & O’Sullivan, 1991; Nixon, McLaughlin, Rowe, & Smyth, 1995; Nixon, Rowe, Smith, McLoughlin, & Silke, 1994; O’Sullivan, Nixon, McLaughlin, O’Sullivan, & O’Sullivan, 1991 Service, Mitchell, & Oliver, 1996; Thompson, Allen, Dodoo, Hunter, Hawkins, & Wolff, 1996). Biological studies have largely been limited in spatial scale and concerned with the marine ecology, particularly benthic macrofauna analysis (Shaw, 1990b), and acute bioassays (e.g. oyster embryo bioassay; MPMMG, 1998). There is little published data on the biological effects of chemical contaminants in the Irish Sea other than localised studies following pollution incidents, such as the ‘Sea Empress’ oil spill (Edwards & Sime, 1998; Law, Kelly, & Brown, 1998). Recent advances in environmental toxicology involving the close coupling of the sensitive stress response (scope for growth—SFG) and contaminant levels in the tissues of mussels (Mytilus edulis) have provided a powerful and cost-effective method of assessing environmental pollution (Widdows, 1998; Widdows & Donkin, 1992). This approach serves to complement the established chemical monitoring programmes by providing a means of assessing whether the recorded contaminant levels are causing deleterious effects, and whether all relevant toxicants are being measured. In a previous study (Widdows, Donkin, Brinsley et al., 1995), the approach was successfully extended and applied over a large spatial scale, consisting of > 1000 km of UK North Sea coastline. SFG measurements were able to detect and quantify changes in environmental quality, as well as identify some of the causes of pollution through quantitative toxicological interpretation (QTI) of contaminant concentrations in the tissues. The QTI is based on established Quantitative Structure-Activity Relationships (QSARs) and cause-effect relationships (i.e. between concentration of contaminants in mussel tissues and the SFG response; for details see Widdows & Donkin, 1992; Widdows, Donkin, Brinsley et al., 1995). Consequently, the additive toxicological contribution of different groups of contaminants towards the decline in SFG can be assessed. Research has confirmed that the toxicities of structurally related compounds, forming a single QSAR line, are simply additive, and that toxic effects of the majority of structurally unrelated compounds also tend to be additive (i.e. on the basis of proportional additivity; Widdows & Donkin, 1991, 1992). Following the North Sea study, a similar monitoring programme was designed to quantify the degree of pollution in the Irish Sea and to provide input into the Quality Status Report (Irish Sea Quality Status Report, 2000). The specific objectives of this study were: 1. to quantify the degree of pollution by measuring the physiological stress responses (SFG) and contaminant concentrations in the tissues of mussels collected from Irish Sea coastal sites; and

330

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

2. to provide a QTI of the tissue residue data based on established relationships between the tissue concentrations of toxicants and SFG.

2. Materials and methods 2.1. Irish Sea sampling sites Fig. 1 shows the location of the 38 Irish Sea sites that were sampled and measured for scope for growth (SFG) and chemical contaminants in the tissues of mussels (Mytilus edulis). The sampling and measurement of SFG was carried out in June and July, after the main spawning period (March–May) and during the summer period of maximum growth. Since it was not possible to visit all sites in one summer, sampling was divided into two phases:

Fig. 1. Location of mussel sampling sites for the Irish Sea survey in June–July 1996 and June–July 1997, as well as location of the Plymouth Marine Laboratory.

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

331

Phase I (June–July 1996) 24 sites on the eastern side of the Irish Sea (southwest England and northwest England, Wales and southwest Scotland. Phase II (June–July 1997) 14 Sites on the western side of the Irish Sea (South and east Ireland, Northern Ireland and the Isle of Man), as well as four repeated and two additional sites on the eastern side. Six potentially ‘clean reference’ sites were selected outside the Irish Sea, two sites along the north Cornwall coast at the southern entrance to the Irish Sea, two along the west coast of Scotland and two on the northern coast of Ireland. The aim and design of the study was not to examine known polluted sites or ‘hot spots’ within estuaries. Instead, sites were selected to represent areas of open coastline and the mouths of estuaries (i.e. areas that were not in the immediate vicinity of industrial or sewage discharges, based on information provided by CEFAS). The majority of sites were > 5 km from any known discharge and all were > 1 km from a known discharge. During phase II in 1997 some UK sites were repeated for comparison with 1996 and for purposes of examining small scale spatial differences (e.g. Port Quin; Dale and St. Ishmaels in Milford Haven; New Brighton; and northern/outer and southern/inner Loch Creran). In addition, two sites in southern Ireland (i.e. Wexford and Duncannon) were repeated to confirm the initial SFG measurements which were low. Mussels were collected from approximately low water neap-tide level and from five sub-samples at least 10 m apart, and then pooled in order to provide an averaged assessment of contamination and SFG at each site. Environmental parameters such as temperature, salinity, and substrate were recorded and the site photographed. Two-hundred mussels of a standard size (4 cm shell length) were packed in insulated containers and transported cool (c. 5
C) and air exposed, via overnight express delivery (Datapost) to the Plymouth Marine Laboratory. Mussels were collected from one site per day. 2.2. Scope for growth measurements SFG measurements were carried out using the ‘standard operating procedures’ developed for the North Sea mussels study (Widdows, Donkin, Brinsley et al., 1995). Twenty mussels were selected for physiological measurements and the remainder (n=180) were used for analysis of chemical contaminants. SFG measurements were performed after a period of 24 h recovery (i.e. recovery from air exposure/transportation) in clean offshore sea water (collected from Eddystone, SW England). Physiological responses (clearance rate, respiration rate and food absorption efficiency) of mussels were measured under standardised laboratory conditions (i.e. 15
C, 33ppt, and a standard algal ration of Isochrysis galbana). The standard operating procedures are designed to minimise stress/disturbance of the mussels during measurement, and to ensure sufficient replication (n=15 individuals per site) for the detection of statistically significant differences amongst sites (Widdows, Donkin, Brinsley et al., 1995). The QA/QC involved the collection of high

332

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

quality offshore sea water immediately prior to the Irish Sea mussel study and the testing of water quality in the 3 m3 aquarium in terms of chemical and biological parameters (i.e. Cu concentration in the sea water and clearance rate of mussels from a well established clean site). In addition, an inter-laboratory comparison of SFG, using the ‘standard operating procedures’, provided a very consistent response amongst 10 laboratories in Europe (16% C.V.; Widdows et al., unpublished data). 2.3. Analysis of chemical contaminants Mussels for chemical analysis were depurated (6 h) in clean offshore sea water to enable sediment to be evacuated/depurated from the mantle cavity and intestine, and then frozen and stored at 20
C prior to analysis by PML and CEFAS Burnham. Frozen mussel samples were transported to CEFAS Burnham at the beginning of August in 1996 and 1997. Aromatic hydrocarbons, polar organics and linear alkylbenzenes were extracted from the tissues by cyclic steam distillation into iso-hexane as in our previous North Sea study (Widdows, Donkin, Brinsley et al., 1995). 2.3.1. Aromatic hydrocarbons The concentration of 2- and 3-ringed aromatic hydrocarbons in the distillate was determined by separating the aromatic hydrocarbons by ring number on an aminocyano high performance liquid chromatography (HPLC) column and detecting them by UV absorbance. The 2- and 3-ringed aromatic hydrocarbon groups were quantified by reference to 2,3-dimethylnaphthalene and 1-methyl phenanthrene respectively (Donkin & Evans, 1984; Widdows, Donkin, Brinsley et al., 1995). 2.3.2. Linear alkylbenzenes. Linear alkylbenzenes (LABs) are established markers of sewage pollution in coastal environments and are readily accumulated by mussels (Eganhouse, Blumfield, & Kaplan, 1983; Murray, Richardson, & Gibb, 1991; Zeng, Khan, & Tran, 1997). The steam distillates in iso-hexane were reduced in volume and analysed by gas chromatography–mass spectrometry (GC–MS). The LABs were detected and quantitative comparisons made by selected ion monitoring using the LABcharacteristic ions m/z 91 and 105 and the molecular ions (m/z 218, 232, 246, 260, and 274) of the C-10, 11, 12, 13 and 14 alkyl-substituted benzenes. Confirmation of identity was also made using MS/MS by selecting the molecular ion as the primary ion and monitoring for the formation of 91 and 105 daughter ions, and by full-scan MS. The methods were calibrated using C-1 substituted phenyl n-alkane (C-10 to C14) standards. LAB data are only available for phase II in 1997. 2.3.3. Metals, organotins and organochlorines Mussel tissue homogenates were analysed by CEFAS for organotin compounds (TBT and DBT), metals and related elements (Cr, Ni, Cu, Zn, As, Ag, Se, Cd, Hg and Pb) and organochlorine compounds (a- and g-HCH, HCB, dieldrin, p,p0 -DDE, p,p0 -TDE, and p,p0 -DDT; 25 individual polychlorinated biphenyl (PCB) congeners).

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

333

The analyses were conducted using established methodology and under appropriate quality control protocols involving the analysis of blanks and reference materials with each batch of samples (Allchin, Kelly, & Portmann, 1989; Jones & Laslett, 1994; Waldock, Waite, Miller, Smith, & Law, 1989). The final determinations were made by cold vapour/atomic fluorescence spectroscopy (Hg), inductively coupled plasma/mass spectrometry (other elements), capillary gas chromatography-flame photometric detection (organotin compounds) and capillary gas chromatographyelectron capture detection (organochlorine compounds). 2.4. Statistical analyses Statistical analysis of the data was by means of ANOVA and stepwise multiple regression techniques using Minitab Ver. 12.

3. Results 3.1. Irish Sea mussels—scope for growth The SFG results (mean  95% C.I.) are presented in Fig. 2A (Phase I, 1996) and Fig. 2B (Phase II, 1997). SFG can be initially classified into three general categories: High growth potential/Low stress (> 15 J g1 h1), Moderate growth potential/ Moderate stress (5–15 J g1 h1) and Low growth potential/High stress (< 5 J g1 h1). SFG values for the eastern side of the Irish Sea (Fig. 2A) show significant spatial differences that are generally consistent with the hydrodynamics and the major inputs of contaminants. There was a statistically significant (P < 0.001) and consistent reduction in SFG of mussels (i.e. < 10 J g1 h1) between Llandudno and Barrow, as the residual currents in the Irish Sea flow north along the northwest coast of England where there are the major urban and industrial developments. The general trend shows lowest SFG values in the central region and highest SFG values in the south and north. The SFG tends to increase (P < 0.001) north of Barrow before the water exits through the North Channel and mixes with water from the North Atlantic. The area of reduced SFG between Bangor and St. Bees (Liverpool Bay and Morecambe Bay) is also consistent with the water mass characterised by slightly reduced surface salinities (i.e. < 33 psu; Bowden, 1980), indicating reduced mixing and flows compared with the Central Channel around the Isle of Man. There were also significant reductions in SFG outside of the central region. Mussels from St. Brides and St. Ishmaels near the Milford Haven Oil terminal, had significantly reduced SFG values (St. Ishmaels < 10 J g1 h1) six months after the ‘Sea Empress’ oil spill in Milford Haven in January 1996. Aberporth mussels had SFG < 15 J g1 h1 and Llanddwyn Bay mussels had SFG values < 10 J g1 h1. Bootle, St. Bees and Silloth mussels have reduced SFG values of < 15 J g1 h1 and probably reflect the industrial discharges (Langston, Pope, & Burt, 1997) and sewage inputs along the Cumbrian coast, before they exit the Irish Sea through the North Channel. Sites in the North Channel (Kirkcolm) and outside the Irish Sea (Loch

334

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

Fig. 2. (A) Scope for growth of mussels collected in 1996 (mean95% C.I.). (B) Scope for growth of mussels collected in 1997 (mean95% C.I.).

Striven, Mull of Kintyre) all show mussel SFG values > 15 J g1 h1. The exception was Loch Creran (northern shore) which was originally selected as a ‘clean site’, but the low SFG values at this site appear to be consistent with the ‘strong odour’ recorded when dissecting these mussels and the appearance of heavy algal/microbial growth on the rocks and mussels at this particular site. Therefore these results appeared to be anomalous, reflecting the specific and localised site conditions rather than the general water quality of the area. Consequently, two Loch Creran mussel populations (southern and northern shores) were measured for SFG in 1997, and the

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

335

Fig. 2. (Continued).

former showed high SFG while lower SFG was confirmed in the latter. An additional sampling of Port Quin mussels at the end of the survey (Port Quin repeat; Fig. 2A) as well as at the beginning, demonstrated that there were no significant changes in the water quality of the PML seawater system during the 3 weeks of SFG measurement.

336

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

The SFG results for the western side of the Irish Sea (Phase II) are presented in Fig. 2B. Mussels sampled from the coastline of Ireland and Northern Ireland generally showed low SFG values (with the exception of Duncannon) along the southern and eastern coasts up to and including Belfast Lough ( < 10 J g1 h1), but increased along the north coast outside the Irish Sea (max. of 22 J g1 h1). Due to the unexpectedly low SFG values at some sites along the southern coastline (i.e. Celtic Sea region), various checks and repeat measurements were carried out in order to establish that these values were consistent and reflected the performance of mussels and water quality at the sampling sites in July 1997. Duncannon and Wexford sites were re-visited (2 weeks later) and the repeat SFG measurements confirmed the initial results (Fig. 2B). There were also additional checks on the water quality and QA/QC at PML (e.g. repeating Port Quin) and these clearly demonstrated and confirmed the high water quality at PML. Furthermore, mussels from the Mornington site were measured after one, two and three days in clean seawater at PML and they showed a significant but only partial recovery after 2 days (Fig. 2B), suggesting depuration of some major toxicants. Baltimore was originally selected as a clean reference site in the southwest of Ireland, but the mussels from this site had a low SFG (< 5 J g1 h1) reflecting a high stress condition. The subsequent analysis of chemical contaminants demonstrated that this site had the highest TBT levels and high PAH concentrations (Table 1B). These contaminants presumably originate from the local fishing industry (a ‘nonpoint’ source) and the elevated concentrations probably reflect the restricted water exchange in this inlet. Duncannon (initial and repeat sample) was found to be moderately stressed (10–12 J g1 h1), whereas mussels from Wexford through to Dublin and Mornington were highly stressed with SFG < 5 J g1 h1. With the exception of Dundrum Bay and Belfast Lough, the SFG begins to improve north of Carlingford Lough and is highest at Red Bay and Lough Swilly. Sites repeated in 1997 were found to be consistent with SFG measurements made in 1996 (Table 2). For example, Port Quin and St. Ishmaels in Milford Haven were similar to 1996 SFG values, but New Brighton at the mouth of the Mersey was significantly lower than in 1996 (ANOVA, P < 0.001). The Loch Creran (northern shore) population had a slightly higher SFG than in 1996 (but not significantly), and the Loch Creran (southern shore) population had a SFG considerably higher (P < 0.001) than those at the Loch Creran (northern) site (Fig. 2B). 3.2. Irish Sea mussels—chemical contaminants The concentration of chemical contaminants in the tissues of mussels sampled from sites around the Irish Sea are presented in Table 1A (Phase I, 1996) and B (Phase II, 1997). 3.2.1. Polyaromatic hydrocarbons The concentrations of 2- and 3-ring PAHs were highest in mussels from St. Ishmaels, Milford Haven, 6 months after the Sea Empress oil spill (Fig. 3). These concentrations were > 5-fold higher than the maximum values recorded at the oil

Table 1 Concentration (mg g1 dry wt.) of chemical contaminants in body tissues of Irish Sea mussels (mean values of duplicate samples) Mussel sites N to S

Metals Cr Ni

Detection limit

0.7 1.8 3.0 2.0 1.6 1.7 1.0 1.2 0.9 1.4 1.2 0.8 0.6 1.2 1.6 0.4 1.3 1.4 2.2 2.6 2.8 1.2 1.7 1.6

Zn

As

Se

9.3 13.6 5.7 5.8 5.5 3.8 3.8 3.8 10.2 8.0 7.8 8.9 5.0 7.5 8.0 4.3 7.4 7.8 6.8 8.5 9.7 7.6 7.4 10.3

54 108 78 76 65 96 75 67 131 142 101 147 50 70 69 52 108 114 124 113 79 71 202 116

9.3 17.9 11.4 15.2 7.5 13.3 7.5 8.4 12.5 15.9 10.1 8.9 7.9 10.6 14.3 6.5 14.8 15.6 11.3 31.1 16.9 16.2 20.2 17.8

3.9 4.8 15.6 16.3 5.0 20.8 3.9 3.2 6.2 35.9 14.0 3.3 10.4 13.6 4.0 2.5 5.7 6.0 5.3 23.4 5.4 17.7 44.8 5.5

Ag

Cd

Hg

0.83 1.14 1.34 1.15 1.24 2.59 1.23 0.85 1.10 0.57 0.37 0.37 nd nd 0.21 nd 1.27 1.19 1.47 1.59 1.08 1.02 1.05 1.25

0.6 ndc 0.11 0.8 nd 1.1 0.11 1.0 0.12 1.8 0.21 2.0 0.18 2.3 0.16 2.9 0.28 5.5 0.32 4.9 0.21 2.8 0.16 3.6 0.10 2.0 0.13 3.4 0.33 2.5 0.10 1.1 0.13 26.2 0.13 24.1 0.10 14.7 0.16 2.7 0.20 2.6 0.08 2.5 0.10 1.7 0.10 2.9

0.01 0.1

Pb

Organics

TBT

DBT

Dieldrin p, p0 DDE p, p0 -TDE HCB g-HCH
25CBsa PAHsa LABsb ratio

nd 0.04 0.05 0.15 nd 0.03 nd 0.05 0.11 0.04 0.08 0.09 nd 0.06 nd nd nd nd nd 0.16 0.04 0.38 0.05 0.06

nd 0.04 nd 0.06 nd nd nd 0.03 0.06 0.03 0.04 0.05 0.03 nd nd nd nd nd nd 0.06 0.04 0.11 0.04 0.04

0.005 nd 0.010 nd nd nd nd nd 0.011 nd 0.011 0.016 nd nd nd 0.005 0.003 0.004 0.003 0.004 0.002 0.003 0.003 nd

nd nd 0.005 nd nd nd nd nd 0.006 0.009 0.017 0.016 nd 0.001 nd 0.002 0.002 0.002 0.002 nd 0.002 0.002 nd nd

nd nd 0.005 nd 0.005 0.011 0.009 0.009 0.023 0.027 0.062 0.058 0.008 nd nd 0.004 0.001 0.001 nd nd nd nd nd nd

nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd

nd nd nd nd nd nd nd 0.004 nd nd nd nd 0.008 nd nd 0.004 0.002 0.003 0.005 0.004 0.002 0.002 nd 0.003

nd nd 0.083 0.029 nd nd nd nd 0.023 0.062 0.123 0.184 0.004 0.005 nd 0.013 0.009 0.020 0.007 0.010 0.015 0.009 nd nd

0.03

0.03

0.001

0.001

0.001

0.001 0.001

0.002

0.25 0.39 2.71 0.96 0.85 0.91 0.48 1.66 2.67 3.29 3.33 4.47 1.22 1.04 0.32 0.28 2.53 0.46 0.13 3.03 2.97 22.48 0.32 0.38

337

(Table continued on next page)

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

A. Phase I, July 1996 Loch Creran (N) 1.0 Mull of Kintyre 1.3 Loch Striven 3.0 Kirkcolm 1.8 Silloth 1.6 St. Bees 3.5 Bootle 1.3 Barrow 1.3 Morecambe 1.4 Blackpool 1.0 Southport 1.4 New Brighton 0.6 Llandudno 0.6 Bangor Flats 1.3 Hen Borth 1.5 Llanddwyn Bay 0.7 Abersoch 1.5 Borth 1.4 Aberporth 1.8 Fishguard 1.9 St. Brides 4.2 St. Ishmaels 1.2 Port Quin 1.7 Zennor 1.9

Cu

Organometals

338

Table 1 (continued) Mussel sites N to S

Metals Cr Ni

2.2 2.3 1.1 2.0 0.6 1.2 2.0 1.2 1.6 1.9 0.7 0.2 1.7 1.7 0.3

Isle of Man Loch Creran (N)g Loch Creran (S) St Ishmaelsg

0.6 0.9 nd nd

Detection limit a b c d e f g

0.9 1.2 1.2 1.1

0.02

Zn

As

8.1 6.2 6.7 10.5 8.4 5.6 8.0 5.8 7.8 8.8 8.7 7.5 8.7 8.6 7.1

95.0 67.1 106.8 128.3 65.2 60.8 99.0 49.8 93.2 83.2 49.4 48.7 123.4 104.5 70.9

12.2 6.7 11.8 14.0 8.9 8.9 13.0 6.2 10.9 10.8 7.0 7.1 10.9 9.2 10.9

8.6 131.3 13.7 5.0 30.4 41.3 3.2 20.3 41.1 6.3 32.1 36.5

Se

Organics

TBT

DBT

Dieldrin p, p0 DDE p, p0 -TDE HCB g-HCH
25CBsa PAHsa LABsb ratio

0.07 0.03 0.07 0.08 0.09 0.04 0.18 0.08 0.09 0.05 0.23 0.20 0.16 0.08 0.52

0.06 0.03 0.06 0.08 0.06 0.04 0.11 0.04 0.06 0.06 0.08 0.09 0.09 0.08 0.14

nd nd nd nd 0.005 0.005 nd nd 0.005 0.01 nd nd nd nd nd

nd nd nd nd nd nd nd nd nd 0.005 nd nd nd nd nd

nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd

nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd

nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd

nd nd nd 0.018 nd nd nd nd nd 0.01 nd nd nd nd nd

0.62 0.59 0.36 0.93 0.63 0.55 0.95 2.01 0.88 11.31 2.57 2.14 2.74 1.78 2.74

8 12 16 –d 29 20 14 4 9 20 10 n/af 6 n/a 6

0.06 nd 0.03 0.44

0.05 0.04 nd 0.23

nd nd nd nd

nd nd nd nd

nd nd nd nd

nd nd nd nd

nd nd nd nd

nd nd nd 0.013

1.02 0.28 0.26 7.87

7 105 10 15

0.05 0.03

0.03

0.005

0.005

0.005

0.005 0.005

Ag

Cd

Hg

6.1 3.5 5.1 4.3 4.4 5.1 6.8 4.6 5.2 4.1 3.5 2.3 6.7 5.2 5.2

0.02 0.02 0.02 0.24 0.05 0.05 0.04 0.06 0.10 3.16 0.04 0.01 0.03 0.02 0.05

0.77 0.54 0.62 0.92 0.47 0.50 0.78 0.28 0.66 0.47 0.21 0.16 1.06 0.87 0.40

0.14 nd 0.10 0.17 0.05 nd 0.13 nd 0.10 nd nd nd 0.11 nd nd

5.7 3.4 3.4 3.5

0.06 nd nd 0.06

1.10 0.10 nd 0.19

0.10 21.7 0.38 0.10 0.37 0.10 0.30 nd

0.01 0.01 0.1

Pb

0.8 0.3 1.4 4.0 0.8 1.1 3.4 0.7 2.9 4.0 0.7 0.7 3.1 2.4 0.9

0.005

PAHs, 2+3-ring polyaromatic hydrocarbons. Linear alklybenzenes (LABs) as sewage tracers. Data expressed as a ratio of the mean peak height of all LAB peaks to the mean peak height of background (blank) peaks. nd, Below limit of detection. Major unidentified peak (not LABs) interfered with analysis. Repeat sampling. n/a, Not analysed. Repeat sampling of 1996 sites.

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

B. Phase II, July 1997 Lough Swilly 2.2 Lough Foyle 1.5 Red Bay 1.5 Belfast Lough 1.5 Strangford Lough 1.5 Dundrum 1.1 Carlingford Strand 2.0 Mornington 1.2 Rogerstown 1.3 Dublin 1.9 Wexford 0.7 1.0 Wexforde Duncannon 2.7 Duncannone 2.3 Baltimore 1.1

Cu

Organometals

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

339

Table 2 Comparison between scope for growth (J g1 h1) measurements on the same mussel populations in 1996 and 1997 (mean 95% C.I., n=16) Site

1996

1997

ANOVA

Port Quin St. Ishmaels New Brighton Loch Creran (northern shore)

16.332.31 9.501.12 8.711.96 7.031.70

18.802.34 7.112.85 3.170.97 8.361.89

n.s. n.s. P <0.001 n.s.

loading jetty of Sullom Voe oil terminal in the Shetlands (Widdows, Donkin, Evans, Page, & Salkeld, 1995), but half the maximum values recorded in the North Sea study at Blyth and Whitby (Widdows, Donkin, Brinsley et al., 1995). Levels of 2and 3-ring PAHs were also elevated at sites along the Lancashire coast from New Brighton (c. 10-fold higher than background) but gradually declining to Barrow. Similar elevated levels of PAHs were recorded at sites near to known sources, e.g. St. Brides (north of Milford Haven oil terminal), Fishguard (near harbour/port), Abersoch (near jetty), and Loch Striven (outer Firth of Clyde c. 1km from a NATO Naval oil loading jetty). In 1997 the concentrations of 2- and 3-ring PAHs in mussels from St. Ishmaels, Milford Haven, were still relatively high (4 to 7.8 mg g1 dry wt.), but had declined to values c. 33% of those recorded in July 1996, presumably reflecting the recovery after the Sea Empress oil spill. PAH levels in the mussels at Dale near the mouth of Milford Haven were 50% lower than St Ishmaels which is nearer the oil terminal. The 2- and 3-ring PAH concentrations near the Milford Haven oil terminal in 1997 had therefore declined to concentrations similar to the maximum values previously recorded at the Sullom Voe oil terminal (Widdows, Donkin, Evans et al., 1995). On the western side of the Irish Sea, the highest 2+3 ring PAH levels were found in Dublin Bay (11 mg g1 dry wt.) and these were 2 to 3-fold higher than those found at the mouth of the Mersey in 1996. At the majority of sites along the south and east coasts, between Baltimore and Mornington, PAH levels were significantly elevated (2–3 mg g1 dry wt.). Mussels from all sites north of Mornington and along the northern Ireland coast had low PAH levels (i.e. < 1mg g1 dry wt.). 3.2.2. Organotins The concentrations of organotin compounds were generally low at sites around the Irish Sea. On the UK mainland coast, the
BT ranged from ‘not detected’ to 0.49 mg g1 dry wt., with the highest value occurring in mussels from St. Ishmaels in Milford Haven in the vicinity of the oil terminal and loading jetties. Elevated TBT and DBT concentrations were also found in mussels near other ports/harbours (e.g. Fishguard, Mersey, Heysham, Stranraer/Kirkcolm). Along the coast of Ireland the concentrations of organotin compounds were generally low, but
BT ranged from 0.06 to 0.66 mg g1 dry wt. Highest TBT values (0.52 mg TBT g1 dry wt.) occurred in mussels from Baltimore in the SW of Ireland

340

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

Fig. 3. Concentrations of 2- and 3-ring polyaromatic hydrocarbons (PAH: mg g1 dry wt.) in the body tissues of Irish Sea mussels (Phase I, June–July 1996).

which also had high PAH levels, both probably originating from the local fishing industry. Elevated TBT and DBT concentrations were also found in mussels near the harbours of Wexford (0.23 mg TBT g1 dry wt.), Duncannon (0.16 mg TBT g1 dry wt.) and Carlingford (0.18 mg TBT g1 dry wt.). TBT levels were generally low north of Carlingford.

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

341

3.2.3. Organochlorines The concentrations of organochlorine pesticides were generally low, at or near the detection limits for the majority of sites on the UK mainland coast. DDT and degradation products were elevated at New Brighton and Southport (> 10-fold above the detection limit/background) and gradually declined to near background levels between New Brighton and Barrow (Fig. 4). This is consistent with the south to north water circulation in the Irish Sea. Dieldrin showed a similar elevation at New Brighton ( > 10-fold higher than background) and also at sites along the

Fig. 4. Concentrations of DDT and degradation products (mg g1 dry wt.) in the body tissues of Irish Sea mussels (Phase I, June–July 1996).

342

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

Lancashire coast. Dieldrin was c. 5-fold higher than background at Loch Striven and Loch Creran, and at sites in North Wales. g-HCH concentrations were the highest at Llandudno (10-fold higher than detection limit/background), but were also elevated at Barrow (5-fold) and some sites in Wales (Llanddwyn Bay, Aberporth). The concentrations of organochlorine pesticides were generally low, at or near the detection limits for the majority of sites along the western side of the Irish Sea. DDT and degradation products were only detectable at Dublin Bay and Strangford Lough (0.01 mg g1 dry wt.), but these concentrations were nearly an order of magnitude lower than Liverpool Bay. Dieldrin was above the detection limit at Dublin Bay, Rogerstown, Dundrum and Strangford Lough (0.005–0.01 mg g1 dry wt.). For PCBs the highest concentrations were seen close to the River Mersey (New Brighton; c. 100-fold higher than detection limits/background levels), with a gradual decline northwards along the Lancashire coast until PCBs were below the detection limit at Barrow. A secondary peak of concentration occurred in the outer Firth of Clyde (Loch Striven; c. 40-fold higher than background levels). On the Irish coast the highest PCB concentrations were recorded in Dublin Bay (0.01 mg g1 dry wt.) and Belfast Lough (0.018 mg g1 dry wt.). All other sites were below the detection limit. 3.2.4. LABs Traces of LABs were found in all the mussel populations studied (Phase II, 1997; Table 1B). This is not surprising because of the ubiquitous and large scale use of linear alkylbenzene-sulphonate surfactants. The presence of LABs was identified with a high level of certainty in most samples by full-scan GC–MS. However, precise quantification of LABs is difficult since they occur as complex mixtures which vary in composition from site to site and structurally identical standards are not readily available (Zeng & Yu, 1996). Since the role of LABs in our study was as a marker of sewage contamination (and LABs may not be toxic to mussels; Donkin, Widdows, Evans, & Brinsley, 1991), a simplified semi-quantitative site comparison was made based on the mean peak height of all LAB peaks in the m/z 91 ion chromatogram, expressed relative to background peak signal. Further work is required to achieve a more precise quantification. LABs with 10–14 alkyl carbon atoms were detected in the mussels, though the composition of the LAB residue varied spatially, probably reflecting differences in the type of surfactants used and differences in the degree of LAB degradation in relation to distance from the source. Typical sample to blank LAB ratios for relatively clean sites were below 10, whereas proximity to urban areas (e.g. Dublin) significantly elevated levels in mussel tissues. GC–MS chromatograms of extracts of mussels from Dublin displayed a complex distribution of LABs. In contrast, a simple unambiguous LAB peak group pattern occurred in mussels from Strangford Lough and Dundrum Bay, with molecular ions characteristic of C-10, 11, 12, 13 and 14 substitution detected, similar to the distribution in sewage sludge (Sweetman, 1994). The chromatograms from these two sites looked very similar suggesting either a common source of contamination or a similar source type. The distinctiveness of the chromatographic pattern suggests that the LABs had been little degraded, suggesting that the mussels were close to the source.

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

343

The highest level of LAB contamination observed in this study was 105 times background (Table 1A) at the Loch Creran (northern shore) site. Here the LABS were probably derived from a very small localised sewage input (e.g. isolated restaurant/bar; J. Gage, personal communication). The chromatograms closely resembled those from Strangford Lough and Dundrum on the Irish coast, again suggesting that the mussels were in close proximity to a source of wastewater containing linear alkylbenzenesulphonate surfactants. LABs were also detected at the Loch Creran (southern) site, but at much lower levels (10 times background). 3.2.5. Metals For metals and related elements the ranges of concentration in the tissues of mussels were quite restricted. Along the eastern side of the Irish Sea the recorded range was c. 3 to 5-fold other than for Cr (7-fold), Se (18-fold), Pb (40-fold) and Cd (200-fold): Cr 0.6–4.2; As 6.5–31; and Pb 0.6–26.2;

Ni 0.6–3.0; Cu 3.8–13.6 Se 2.5–44.8; Cd < 0.01–2.59; all as mg g1 dry tissue weight.

Zn 50–202; Hg < 0.1–0.33;

The highest concentrations of lead (15–26 mg g1) were found in Cardigan Bay, between Aberporth and Abersoch, and this reflects anthropogenic input originating from old lead mines dating back to the Roman period. Cr and Cd had the lowest values in the central region along the Lancashire coast and the highest values for Cr were at St. Brides, St. Bees and Loch Striven, and the highest Cd was also Loch Striven in the outer Firth of Clyde. Hg concentrations were elevated along the Lancashire and Cumbrian coast from Blackpool to St. Bees, and the highest Hg concentration was at Hen Borth on Anglesey. For mussels sampled from the western side of the Irish Sea the metals and related elements had a similarly restricted range in concentration (c. 2 to 3-fold) other than for Cd (7-fold), Ni (10-fold), Pb (70-fold) and Ag ( > 100-fold): Cr 0.7–2.7; As 6.7–14; Pb 0.3–22;

Ni 0.2–2.3; Se 2.3–6.8; Ag 0.01–3.16

Cu 5.6–10.5; Zn 49- 128; Cd < 0.16–1.1; Hg < 0.1–0.17; all as mg g1 dry tissue weight.

The highest concentrations of lead were found in the Isle of Man and probably reflects inputs from old lead mines. Cd was also highest in the Isle of Man, Ni was highest in the Lough Foyle and Lough Swilly area, all probably reflecting the natural geological inputs. Ag was particularly high in Dublin Bay, possibly due to inputs from the photographic industry.

4. Discussion Small scale pollution studies in estuaries, fjords and bays, generally demonstrate highly significant correlations between individual contaminants and SFG (typical correlation coefficients are r=0.9), due to the processes of dilution, dispersion and

344

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

degradation along relatively simple gradients (e.g. Langesundfjord in Norway, Widdows, & Johnson, 1988; Hamilton Harbour in Bermuda, Widdows, Burns, Menon, Page, & Soria, 1990; Sullom Voe in the Shetlands, Widdows, Donkin, Brindsley et al., 1995b; Venice Lagoon, Widdows, Nasci, & Fossato, 1997). However, over a larger spatial scale (such as the Irish Sea and North Sea coastlines) the composition of contaminant mixtures will be complex and vary from site to site. Consequently, contaminants are unlikely to co-vary or correlate with a single ‘marker’ of overall contamination. Hence the need to extend the analysis of data beyond statistical correlations and provide a quantitative toxicological interpretation of the chemical and biological data using established concentration-response realationships and QSARs. While it is possible to establish a database of concentration-response relationships for a limited number of individual contaminants, such as metals and organometals, clearly it is unrealistic to examine the sublethal effects of > 100,000 individual organic contaminants. A QSAR approach provides a means of overcoming this problem by establishing QSAR lines for representative groups of structurally related compounds. This facilitates the prediction of toxicological properties of organic compounds from their structure and physico-chemical properties (for details see Donkin & Widdows, 1990). While the database describing tissue contaminant concentration-sublethal response (SFG) relationships for mussels is substantial as a result of many years of research (i.e. > 30 toxicants and groups of compounds), and probably greater than for other aquatic organisms used in environmental monitoring, it still only represents a small proportion of the total contaminants entering the marine environment. Consequently, there are known toxicants whose toxic effects on the SFG of mussels still need to be established, as well as many more unknown toxicants which are entering the environment and remain to be identified, quantified and their toxicity determined. In contrast to SFG, the majority of other existing toxicological databases are for lethal effects in relation to toxicant concentrations in water (only occasionally tissue concentrations of contaminants) and are therefore not appropriate for ‘in situ’ and sublethal environmental monitoring studies. 4.1. Toxicological interpretation of Irish Sea mussels data 4.1.1. Hydrocarbons A large and significant contribution towards the observed decline in SFG in mussels along the Irish Sea coastline is caused by toxic hydrocarbons, acting through the mechanism of ‘non-specific narcosis’ resulting in the inhibition of mussel feeding rate. The effect of toxic hydrocarbons on SFG is calculated from the previously established linear relationship describing the reduction in SFG with the log increase in total toxic hydrocarbons concentration in mussel tissues (for details see Widdows, Donkin, Brinsley et al,. 1995).
SFG J g1 h1 ¼ 26:8  ð9:4  log toxic hydrocarbon concentrationÞ Their contribution to the observed decline in SFG can then be plotted (Fig. 5A and B). It is apparent that at the majority of the Irish Sea coastal sites, toxic

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

345

hydrocarbons are accumulated to concentrations that are capable of inducing a significant inhibition of SFG. For example, every order of magnitude increase in toxic hydrocarbons will reduce SFG by c. 10 J g1 h1. Consequently, the growth of mussels as well as other bivalves could be markedly increased by reducing hydrocarbon contamination in the marine environment. Hydrocarbon contamination of the marine environment is derived from several sources, including: 1. road runoff and atmospheric inputs from road transport (majority of sites); 2. industry and urban development (Liverpool Bay, Lancashire, Cumbria, Dublin Bay, Belfast); 3. harbours, ports and shipping activity (Fishguard, Abersoch, Mersey, Stranraer/Kirkcolm, Baltimore, Duncannon, Wexford, Dublin, Belfast); and 4. oil loading, oil terminals and refineries (Milford Haven, Mersey, old NATO oil loading jetty in Loch Striven, Dublin, Belfast). The only clean sites largely free of hydrocarbon contamination are in rural areas such as the SW of England (Zennor and Port Quin) and the SW of Scotland (Mull of Kintyre and Loch Creran). 4.1.2. Organotins Tributyltin (TBT) and its breakdown product (dibutyltin, DBT) were accumulated in the mussel tissues at most sites, with the exception of those on the west coast of Wales and those north of the Firth of Clyde in Scotland. The concentrations were particularly elevated at harbours (Baltimore, Duncannon, Rogerstown), ports (Fishguard, Heysham, Stranraer) and oil terminals (Milford Haven), enough to have a small effect on SFG. TBT’s primary mechanism of toxicity, at concentrations from 0.2 to 10 mg g1 dry wt. in the mussel, is the uncoupling of oxidative phosphorylation (Widdows & Page, 1993). None of the mussels from the Irish Sea sites had TBT concentrations capable of inducing inhibitory effects on feeding rate (i.e. > 4 mg g1 dry wt.). The relatively small effect of TBT on SFG was calculated as described by Widdows, Donkin, Brinsley et al. (1995) and its contribution to the observed reduction in SFG is illustrated in Fig. 5A and B. 4.1.3. Organochlorines Selected persistent organochlorines, which form part of many routine chemical monitoring programmes, were not detected at many of the Irish Sea sites. Dieldrin, DDTs and PCBs were particularly elevated at the mouth of the Mersey (New Brighton) and at the sites immediately north along the Lancashire coast (Southport, Blackpool and Morecambe/Heysham). This is consistent with the hydrodynamics and the northerly flow. PCBs were only detected at Dublin and Belfast on the western coastline of the Irish Sea, but these and the other organochlorines were not particularly elevated at any of the sites on the coast of Ireland. To date, there is no evidence that the recorded levels of organochlorines in mussels are likely to have an adverse effect on the SFG of mussels (Donkin, Widdows, Evans, Staff, & Yan, 1997).

346

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

Fig. 5. (A) Quantitative toxicological interpretation and partitioning of the reduced scope for growth of mussels from the eastern Irish Sea coastal sites (June–July 1996), together with additional observations/ information. The dashed line at c. 27 J g1 h1 denotes the approximate maximum growth potential of unstressed mussels measured under ‘standardised operating procedures’. (B) Quantitative toxicological interpretation and partitioning of the reduced scope for growth of mussels from the western and eastern Irish Sea coastal sites (June–July 1997), together with additional observations/information. The dashed line at c. 27 J g1 h1 denotes the approximate maximum growth potential of unstressed mussels measured under ‘standardised operating procedures’.

4.1.4. Metals At no sites were individual metals accumulated to concentrations that could cause a significant reduction in SFG of mussels. Tissue concentrations were all considerably below the recorded ‘no observed effect thresholds’ (Widdows, Donkin, Brinsley et al., 1995). However, there was an unexpected significant positive relationship between SFG and As (P < 0.001; see later).

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

347

Fig. 5. (Continued).

4.1.5. Unexplained component Following the initial process of quantitative toxicological interpretation of the combined SFG and chemical contaminant data, it is then possible to identify those sites and regions with a large ‘unexplained component’ to the low SFG values (i.e. where the ‘observed’ reduction in SFG is considerably greater than ‘predicted’ from the toxic contaminants analysed in the tissues). This ‘unexplained component’ is represented in Fig. 5A and B by the difference between the maximum growth potential of unstressed mussels (c. 27 J g1 h1) and the sum of measured SFG and the toxic effects induced by PAHs and TBT. A large unexplained component (i.e. more than twice the mean 95% C.I. for SFG) is indicative of the presence of significant amounts of additional toxicants in the marine environment which are at present unidentified, including sewage, industrial and agrochemical contaminants.

348

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

Stepwise multiple regression analysis of SFG and the contaminant concentrations did not show any significant relationships, with the exception of a positive correlation between SFG and As (P < 0.001), which was also expressed as a significant negative correlation (P < 0.001) between the ‘unexplained SFG’ and As. The range of As concentrations in mussels from the Irish Sea suggests that the strong positive correlation between SFG and As is unlikely to reflect its role as an essential element in animals, with deficiency leading to reduced growth (Anke, 1986). This unexpected correlation is probably due to a reduction in As accumulation in phytoplankton, macroalgae and mussels (either directly or indirectly via the diet) at sites with elevated PO4 concentrations (Froelich, Kaul, Byrd, Andreae, & Roe, 1985; Langston et al., 1997). Phosphate is a known competitive inhibitor of As uptake, at least in algae, because of its chemical similarity. Elevated PO4 concentrations are likely to occur at sites with significant sewage inputs, and consequently mussels will have lower As bioaccumulation (e.g. Loch Creran (northern), Silloth, Bootle, Barrow, Southport to Bangor, Llanddwyn Bay, Aberporth and many sites on western shores of the Irish Sea). The data suggest that this As vs. PO4 interaction tends to override any elevated inputs of As, whether geological (Cornwall, SW England) or anthropogenic (Mersey estuary; Langston, personal communication). The LABs also appear to be a useful indicator of contamination by wastewater as they were detected at differing concentrations in mussels at all sites investigated (Phase II, Table 1B). However, it is unlikely that the bioaccumulated levels of these compounds can be linked in a simple, direct way to the impact of sewage on mussels. 1-phenyldecane (i.e. C-10 substituted LAB) was not toxic to mussel feeding activity in laboratory exposure experiments (Donkin et al., 1991) and structure-activity considerations suggest that the C-11 to C-14 substituted benzenes are likely to be even less toxic. Sewage is a complex mixture of toxic and non-toxic molecules with different environmental half lives. It seems likely that LABs will have longer halflives in the marine environment than some of the smaller molecular weight and more volatile toxic sewage constituents (e.g. aminopropane; Fitzsimons et al., 1995). Therefore the LAB ‘‘signal’’ may persist after the major toxic components of the sewage have dissipated and degraded. However, the data obtained in this study suggest that the presence of an undegraded LAB pattern, as in Loch Creran (northern), Strangford Lough and Dundrum Bay, could be a good correlate of total sewage impact (i.e. sites with low SFG, Fig. 5B). In spite of this, there was no significant correlation between As and LABs, probably due to the limited data for LABs and the relatively narrow range for As concentrations in mussels along the Irish coast. 4.2. Assessment of water quality in the Irish Sea—stress and contaminant levels in mussels The combined measurement of SFG and contaminant levels in mussels collected from 38 sites has provided a broad spatial assessment of water quality in the outer estuaries and coastal zone of the Irish Sea. On the UK mainland coast, the general trend in scope for growth showed a significant decline in water quality in the

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

349

Mersey/Liverpool Bay region and north along the Lancashire coast. High water quality was recorded along the west coast of Wales, as well as the ‘clean reference’ sites in southwest England and Scotland. On the Irish coast there was a similar trend with low scope for growth within the Irish Sea, declining north of Duncannon and improving again north of Belfast Lough at sites along the northern Ireland coastline. These similar spatial trends in water quality on both sides of the Irish Sea are consistent with the prevailing hydrodynamics, with residual currents from south to north, and the spatial distribution of urban/industrial development (e.g. Liverpool Bay and Dublin Bay). Such findings are in contrast to the trend recorded in the North Sea mussel survey carried out in 1990/1991 (Widdows, Donkin, Brinsley et al., 1995), where water quality showed a general decline from north to south, which reflected the hydrodynamics of the North Sea and the general increase in urban/ industrial/agricultural activity from north to south. The decline in scope for growth of mussels on both sides of the Irish Sea was associated with a general increase in contaminant levels in the mussel tissues. Certain contaminants showed elevated concentrations (i.e. > 10-fold higher than background or detection limits). These included PAHs, TBT,
DDT, Dieldrin, g-HCH, PCBs, and a few of the metals (Cd, Se, Ag, Hg, Pb). Many of these contaminants are particularly elevated in the coastal margins of Liverpool Bay, Morecambe Bay and Dublin Bay. Petroleum hydrocarbons form an important component of reduced water quality, and the levels accumulated in mussels were sufficiently high to explain a significant part of the observed decline in their scope for growth (Fig. 5A and B). These high levels of hydrocarbons are derived from fossil fuel combustion in urban environments (e.g. coal burning, road transport), shipping, as well as releases from oil refineries and terminals. TBT is still an important contaminant of areas associated with significant shipping activity, such as the ports and harbours of Milford Haven, Fishguard, Heysham, Stranraer, Baltimore, Duncannon, Wexford and Carlingford. While these concentrations have little impact on the SFG of mussels they are at levels that impair the reproductive activity of the highly sensitive dogwhelk (Nucella lapillus; Gibbs, Bryan, Pascoe, & Burt, 1987). While metals were not accumulated to toxic levels in the tissues of mussels they do provide useful tracers of inputs. Lead provides evidence of contamination from a variety of sources including road transport, despite the fact that most vehicles in Europe now use lead-free petrol. For example, Pb levels are elevated at sites near urban developments, including Bangor, Cheshire and Lancashire coasts on the eastern side and at Duncannon, Dublin, Newry (Carlingford) and Belfast on the western side. In addition, there are elevated Pb concentrations, along the west and north coasts of Wales and on the Isle of Man, due to inputs from old mines dating back to the Roman period. Mercury was elevated in the tissues of mussels from the Liverpool and Morecambe Bay region. These findings are in agreement with previous studies by Gardner (1978), Rae and Aston (1981), Law et al. (1991), Leah, Evans, and Johnson (1991), and Leah et al. (1992) which demonstrated Hg contamination of sediments, fish and marine mammals in this area, primarily due to historical inputs from

350

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

chlor-alkali factories. The elevated concentrations of
DDT in mussels living in the Liverpool Bay region is consistent with earlier chemical monitoring studies using fish and shellfish (Franklin, 1987; Leah et al., 1997). An important feature of the Irish Sea and North Sea mussel studies is the extent to which the reduction in scope for growth at many sites cannot be readily explained by the rather limited number of contaminants (metals, TBT, PAHs and selected organochlorines) that are routinely analysed in chemical monitoring programmes. There are many sites where the observed reduction in scope for growth is far greater than predicted from the tissue concentrations of these contaminants, thus indicating the presence of additional ‘unknown or unquantified toxicants’. These sites are often near estuaries receiving significant amounts of sewage inputs, on both sides of the Irish Sea, but particularly along the Irish coast (Fig. 6A and B). The low As and

Fig. 6. (A) Major UK sewage outfalls to tidal waters in the Irish Sea (discharges in terms of dry weather flows in m3 day1). 1997 data from CEFAS. (B) Major sewage inputs into the marine environment along the coastline of Ireland (discharges in terms of dry weather flows in m3 day1). Data from the Northern Ireland Water Service and the Irish Department of the Marine.

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

351

Fig. 6. (Continued).

elevated LAB concentrations in mussels at certain sites provides evidence in support of the hypothesis that a significant proportion of the recorded decline in SFG (unexplained component) is due to sewage inputs. However, there is an urgent requirement for better sewage markers in mussel tissues that will provide a quantitative measure of exposure to sewage. Sewage is a complex mixture of chemical contaminants and there is evidence that mussels are sensitive to sewage. Previous studies have shown the short term toxic effects of sewage on SFG of mussels (Butler, Roddie, & Mainstone, 1990), the recovery in SFG of mussels following the introduction of sewage treatment works in Narragansett Bay (Nelson, 1990), and the improved fecundity/reproductive output of the clam, Macoma balthica, following the reduction of sewage inputs into San Francisco Bay (Hornberger et al., 2000). Recent hydrodynamic studies (Hill et al., 1997) have demonstrated that stratification occurs during the summer months in the relatively shallow coastal margins of the Irish Sea (c. 60% is < 50 m depth) with the formation of a summer gyre west of the Isle of Man. Consequently, sewage and other contaminants will be dispersed within a relatively small volume of water with low currents and a relatively long retention time. Hill et al. (1997) not only suggest that this gyre could act to retain contaminants but that this area could also be at disproportionate risk from

352

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

environmental damage. In addition, recent studies (Sherwin, Matthews, & Kennedy, 1997) have used synthetic aperture radar (SAR) on the ERS-1 satellite to demonstrate the development and persistence of effluent slicks in the Menai Straits. Even very small discharges (c. 100 m3 day1) are detected and extend for several km along the coast. Consequently, the larger sewage effluent inputs could have an impact along major lengths of the Irish Sea coastline, because they are not rapidly diluted and dispersed into the receiving waters. Major sewage inputs to the eastern side of the Irish Sea during the period 1996/ 1997 were from the North Wales, the Lancashire coast, particularly into Liverpool Bay, and from Workington (Fig. 6A). However, there is also the problem of smaller sewage inputs (both human and agricultural) which may have a more localised impact. For example, sewage was probably responsible for the reduced SFG at Aberporth, a rural site on the west coast of Wales, particularly as none of the measured contaminants were elevated at this site. At the time of sampling mussels from rocks at Aberporth it was noted that sediments in the adjacent gully had a ‘sewage odour’, and mussels have subsequently been shown to have relatively low As concentrations. Along the western side, sewage inputs are also significant from Cork, Waterford, Wexford, Dublin, Drogheda, Dundalk and Belfast (Fig. 6B). While human sewage input estimates are semi-quantitative, there is also increasing concern about unquantified sewage inputs from agriculture in both the UK and Ireland. Routine monitoring of toxic algal blooms and algal toxins in shellfish by CEFAS (S. Milligan, Lowestoft) and Aberdeen Marine Laboratory (G. Howard) in the UK and the Irish Marine Institute in Ireland (McMahon, 1998) has confirmed that there was no evidence of major toxic algal blooms in 1996/1997. Furthermore, during 1996 the degree of toxicity was unusually low throughout Europe. Consequently, toxic algal blooms can be eliminated as a contributor to the observed reduction in SFG of Irish Sea mussels. Another important feature of this combined biological/chemical monitoring programme is the identification of areas with ‘unexpected’ low water quality. For example, Llanddwyn Bay on Anglesey was expected to be a relatively ‘clean site’ with high mussel SFG. However, mussels were found to have a significantly reduced scope for growth (9.05 J g1 h1) and subsequent chemical analysis revealed elevated levels of dieldrin and g-HCH. The low water quality in this area may in part be related to the loss of MV Kimya with its cargo of sunflower oil in 1991, and the persistence of vegetable oils (and perhaps their associated contaminants) in the marine environment (Mudge, 1997; Mudge, Slagado, & East, 1993). Furthermore, the very low As concentrations in mussels from Llanddwyn Bay is also indicative of significant nutrient/sewage inputs. Caernarfon Bay to the south also receives chemical effluents (CEFAS data). Finally, mussel population studies at selected sites along the UK coastline of the Irish Sea have confirmed that the reduced performance at the individual level (i.e. lower SFG) is also reflected in important changes at the population level (Thain and Allen, unpublished data). In summary, sites in the Irish Sea, particularly in the Liverpool Bay and Morecambe Bay region, had very low growth rates (< 1 mm between June and November 1996), poor recruitment and a lower maximum shell

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

353

size, compared to the southern sites (> 3 mm shell growth) and northern-most sites (> 10 mm shell growth between June and November 1996). The SFG results indicate that mussels are living closer to their limit of growth and survival (i.e. their assimilative capacity) and this will affect not only their productivity but also reflect the general productivity and health of the Irish Sea system. At present a major phase of sewage treatment improvements aims to reduce the sewage inputs to the Irish Sea, particularly along the North Wales and Lancashire coasts. This is to meet the EU directives on bathing water quality and urban waste water treatment. Consequently, future studies should aim to test the hypothesis that the water quality and SFG of mussels in the Irish Sea will improve as a result of significant reductions in sewage inputs to the marine environment.

Acknowledgements This work was supported by the UK Department of the Environment, Transport and the Regions (now Department for Environment, Food and Rural Affairs; DETR contract No. EPG 1/9/89). We thank Mr. N. Bowley for his assistance in collecting the mussels, Dr. D. Morgans and Mr. B. Dawkins for assistance with the measurement of SFG, and Ms. S. Evans for assistance in the PAH analyses and Mr. A. Sheikh for assistance with LAB analyses.

References Achterberg, E. P., & Van den Berg, C. M. G. (1996). Automated monitoring of Ni, Cu and Zn in the Irish Sea. Marine Pollution Bulletin, 32, 471–479. Allchin, C. R., Kelly, C. A., & Portmann, J. E. (1989). Methods of analysis for chlorinated hydrocarbons in marine and other samples. Aquatic environment protection: analytical methods. MAFF Directorate of Fisheries Research, Lowestoft, 6. Anke, M. (1986). Arsenic. In W. Mertz (Ed.), Trace elements in human and animal nutrition (pp. 347–373). Orlando, Florida: Academic Press. Bowden, K. F. (1980). Physical and dynamical oceanography of the Irish Sea. In F. T. Banner, M. B. Collins, & K. S. Massie (Eds.), The North West European Shelf Seas, the bed and the sea in motion, Physical and chemical oceanography and physical resources, II (pp. 391–413). Elsevier. Butler, R., Roddie, B. D., & Mainstone, C. P. (1990). The effects of sewage sludge on two life-history stages of Mytilus edulis. Chemistry and Ecology, 4, 211–219. Dickson, R. R., & Boelens, R.G. (1988). The status of current knowledge on anthropogenic influences in the Irish Sea. ICES Cooperative Research Report No. 155. Donkin, P., & Evans, S. V. (1984). Application of steam distillation in the determination of petroleum hydrocarbons in water and mussels (Mytilus edulis) from dosing experiments with crude oil. Analytica Chimica Acta, 156, 207–219. Donkin, P., & Widdows, J. (1990). Quantitative structure-activity relationships in aquatic invertebrate toxicology. Reviews in Aquatic Sciences, 2, 375–398. Donkin, P., Widdows, J., Evans, S. V., & Brinsley, M. D. (1991). QSARs for the sublethal responses of marine mussels (Mytilus edulis). Science of the Total Environment, 109/110, 461–476. Donkin, P., Widdows, J., Evans, S. V., Staff, F. J., & Yan, T. (1997). Effect of neurotoxic pesticides on the feeding rate of marine mussels (Mytilus edulis). Pesticide Science, 49, 196–209.

354

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

Eganhouse, R. P., Blumfield, D. L., & Kaplan, I. R. (1983). Long-chain alkylbenzenes as molecular tracers of domestic wastes in the marine environment. Environmental Science and Technology, 17, 523–530. Edwards, R., & Sime, H. (1998). The Sea Empress oil spill. Lavenham, Suffolk: Terance Dalton. Fitzsimons, M. F., Rashid, M. K. A., Riley, J. P., & Wolf, G. A. (1995). Aminopropanone as a marker for raw sewage in natural waters. Marine Pollution Bulletin, 30, 306–312. Franklin, A. (1987). The concentration of metals, organochlorine pesticides and PCB residues in marine fish and shellfish: Results of MAFF fish and shellfish monitoring programmes, 1977–84. Aquatic Environment Monitoring Report No. 16. Lowestoft: MAFF. Franklin, A. (1992). Monitoring and surveillance of non-radioactive contaminants in the aquatic environment and activities regulating the disposal of waste at sea, 1990. Aquatic Environment Monitoring Report, No. 30. Lowestoft: MAFF. Froelich, P. N., Kaul, L. W., Byrd, J. T., Andreae, M. O., & Roe, K. K. (1985). Arsenic, barium, germanium, tin, dimethlysulfide and nutrient biogeochemistry in Charlotte Harbor, Florida, a phosphorusenriched estuary. Estuarine, Coastal and Shelf Science, 20, 239–264. Gardner, D. (1978). Mercury in fish and waters of the Irish Sea and other United Kingdom fishing grounds. Nature, 272, 49–51. Gibbs, P. E., Bryan, G. W., Pascoe, P. L., & Burt, G. R. (1987). The use of the dog-whelk, Nucellus lapillus as an indicator of tributyltin (TBT) contamination. Journal of the Marine Biological Association, UK, 67, 507–523. Hill, A. E., Brown, J., & Fernand, L. (1997). The summer gyre in the western Irish Sea: Shelf sea paradigms and management implications. Estuarine and Coastal Shelf Science, 44A, 83–95. Hornberger, M., Luoma, S.N, Cain, D., Parchaso, F., Brown, C., Bouse, R., Wellise, C., & Thompson, J. (2000). Linkage of bioaccumulation and biological effects to changes in pollutant loads in San Francisco Bay. Environmental Science and Technology, 34, 2401–2409. Irish Sea Quality Status Report (2000). Quality status report of the marine and coastal areas of the Irish Sea and Bristol Channel. (Law, R.J. & Jones, J.—Compilers). Contract Report to the Department of the Environment, Transport and the Regions (DETR). Jones, B. R., & Laslett, R. E. (1994). Methods for analysis of trace metals in marine and other samples. Aquatic environment protection: analytical methods. MAFF Direct. Fish. Res., Lowestoft 11 Langston, W. J., Pope, N. D., & Burt, G. R. (1997). Metals in biota of the Cumbrian coast. In P. D. Jones, & R. G. Chambers (Eds.), Coastal zone topics: process, ecology and management 2 (pp. 71–89). The Solway and Cumbrian coasts: Joint Nature Conservation Council. Laslett, R. E. (1995). Concentrations of dissolved and suspended particulate Cd, Cu, Mn, Ni, Pb and Zn in surface waters around the coasts of England and Wales and adjacent seas. Estuarine and Coastal Shelf Science, 40, 67–85. Law, R. J. (1981). Hydrocarbon concentrations in water and sediments from UK marine waters, determined by fluorescence spectroscopy. Marine Pollution Bulletin, 12, 153–157. Law, R. J., Fileman, C. F., Hopkins, A. D., Baker, J. R., Harwood, J., Jackson, D. B., Kennedy, S., Martin, A. R., & Morris, R. J. (1991). Concentrations of trace metals in the livers of marine mammals (seals, porpoises and dolphins) from waters around the British Isles. Marine Pollution Bulletin, 22, 183– 191. Law, R. J., Kelly, C. A., & Brown, J. (1998). The effects of a chemically-dispersed oil spill on fish and shellfish: Experience from the Sea Empress spill in Wales in 1996. In B. K. Trudel (Ed.), Dispersant use in Alaska: a technical update (pp. 189–202). Cordova, Alaska: Prince William Sound Oil Spill Recovery Instutute. Leah, R. T., Evans, S. J., & Johnson, M. S. (1991). Mercury in muscle tissue of lesser-spotted dogfish (Scyliorhinus caniculus L.) from the north-east Irish Sea. Science of the Total Environment, 108, 215–224. Leah, R. T., Evans, S. J., & Johnson, M. S. (1992). Mercury in flounder (Platichthys flesus L.) from estuaries and coastal waters of the north-east Irish Sea. Environmental Pollution, 75, 317–322. Leah, R. T., Johnson, M. S., Connor, L., & Levene, C. F. (1997). DDT group compounds in fish and shellfish from the Mersey Estuary and Liverpool Bay. Environmental Toxicology and Water Quality, 12, 223–230.

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

355

McMahon, T. (1998). Monitoring of algal toxins in shellfish. Aquaculture Newsletter, 25, 5 Dublin: Irish Sea Fisheries Board.. MPMMG (1998). Marine pollution monitoring management group, national monitoring programme, survey of the quality of UK coastal waters. Aberdeen: MPMMG. Mudge, S. M. (1997). Can vegetable oils outlast mineral oils in the marine environment? Marine Pollution Bulletin, 34, 213. Mudge, S. M., Slagado, M. A., & East, J. (1993). Preliminary investigations into sunflower oil contamination following the wreck of the M.V. Kimya. Marine Pollution Bulletin, 26, 40–44. Murray, A. P., Richardson, B. J., & Gibbs, C. F. (1991). Bioconcentration factors for petroleum hydrocarbons, PAHs, LABs and biogenic hydrocarbons in the blue mussel. Marine Pollution Bulletin, 22, 595–603. Nelson, W. G. (1990). Use of the blue mussel, Mytilus edulis, in water quality toxicity testing and in situ marine biological monitoring. In W. G. Landis, & W. H. van der Schalie (Eds.), Aquatic toxicology and risk assessment (Vol. 13), ASTM STP 1096 (pp. 167–175). Philadelphia: American Society for Testing and Materials. Nixon, E., McLaughlin, D., Boelens, R. G., & O’Sullivan, G. (1991). Contaminants in marine biota 1990 monitoring programme. Fisheries Leaflet 151. Dublin: Dept. of the Marine. Nixon, E., McLaughlin, D., Rowe, A., & Smyth, M. (1995). Monitoring of shellfish growing area–1994. Fisheries Leaflet 166. Dublin: Dept. of the Marine. Nixon, E., Rowe, A., Smith, M., McLoughlin, D., & Silke, J. (1994). Monitoring of shellfish growing area—1993. Fisheries Leaflet 160. Dublin: Dept. of the Marine. O’Sullivan, M. P., Nixon, E. R., McLaughlin, D., O’Sullivan, M. I., & O’Sullivan, D. (1991). Chemical contaminants in Irish estuarine and coastal waters, 1978–1988. Fisheries Bulletin 10. Dublin: Dept. of the Marine. Proctor, R. (1981). Tides and residual circulation in the Irish Sea: a numerical modelling approach. PhD thesis, University of Liverpool. Rae, J. E., & Aston, S. R. (1981). Mercury in coastal and estuarine sediments of the northeastern Irish Sea. Marine Pollution Bulletin, 12, 367–371. Service, M., Mitchell, S. H., & Oliver, W. T. (1996). Heavy metals in the superficial sediment of the N-W Irish Sea. Marine Pollution Bulletin, 32, 828–830. Shaw, D. F. (1990a). Irish Sea study group report. Part I. Nature conservation. Liverpool University Press. Liverpool. Shaw, D. F. (1990b). Irish Sea study group report. Part II. Waste inputs and pollution. Liverpool University Press. Liverpool. Shaw, D. F. (1990c). Irish Sea study group report. Part III. Exploitable living resources. Liverpool University Press. Liverpool. Shaw, D. F. (1990d). Irish Sea Study Group Report. Part IV. Planning, development and management. Liverpool University Press. Liverpool. Sherwin, T. J., Matthews, J. P., & Kennedy, F. (1997). Effluent slicks in the Menai Strait: a comparison of ERS-1 SAR signatures and model predictions. Marine Pollution Bulletin, 34, 264–268. Sweetman, A. J. (1994). Development and application of a multi-residue analytical method for the determination of n-alkanes, linear alkylbenzenes, polynuclear aromatic hydrocarbons and 4-nonylphenol in digested sewage sludges. Water Research, 28, 343–353. Thompson, A., Allen, J. R., Dodoo, D., Hunter, J., Hawkins, S. J., & Wolff, G. A. (1996). Distributions of chlorinated biphenyls in mussels and sediment from Great Britain and the Irish Sea coast. Marine Pollution Bulletin, 32, 232–237. Waldock, M. J., Waite, M. E., Miller, D., Smith, D. J., & Law, R. J. (1989). The determination of total tin and organotin compounds in environmental samples. Aquatic environment protection: analytical methods: 4. Lowestoft: Ministry of Agriculture, Fisheries and Food, Directorate of Fisheries Research. Widdows, J. (1998). Chapter 9: marine and estuarine invertebrate toxicity tests. In P. Calow (Ed.), Handbook of ecotoxicology (pp. 145–166). Oxford: Blackwell Science. Widdows, J., Burns, K. A., Menon, N. R., Page, D. S. & Soria, S. (1990). Measurement of physiological energetics (scope for growth) and chemical contaminants in mussel (Arca zebra) transplanted along a contamination gradient in Bermuda. Journal of Experimental Marine Biology and Ecology, 138, 99-117.

356

J. Widdows et al. / Marine Environmental Research 53 (2002) 327–356

Widdows, J., & Donkin, P. (1991). Role of physiological energetics in ecotoxicology. Comparative Biochemistry and Physiology, 100C, 69–75. Widdows, J., & Donkin, P. (1992). Chapter 8: mussels and environmental contaminants: bioaccumulation and physiological aspects. In E. Gosling (Ed.), The mussel Mytilus (pp. 383–424). Amsterdam: Elsevier Press. Widdows, J., Donkin, P., Brinsley, M. D., Evans, S. V., Salkeld, P. N., Franklin, A., Law, R. J., & Waldock, M. J. (1995). Scope for growth and contaminant levels in North Sea mussels Mytilus edulis. Marine Ecology Progress Series, 127, 131–148. Widdows, J., Donkin, P., Evans, S. V., Page, D. S., & Salkeld, P. N. (1995). Sublethal biological effects and chemical monitoring of Sullom Voe (Shetland) using mussels (Mytilus edulis). Proceedings of the Royal Society of Edinburgh, 103B, 99–112. Widdows, J., & Johnson, D. (1988). Physiological energetics of Mytilus edulis: scope for growth. Marine Ecology Progress Series, 46, 113–121. Widdows, J., Nasci, C., & Fossato, V. U. (1997). Effects of pollution on the scope for growth of mussels (Mytilus galloprovincialis) from the Venice lagoon. Marine Environmental Research, 43, 69–79. Widdows, J., & Page, D. S. (1993). Effects of tributyltin and dibutyltin on the physiological energetics of the mussel, Mytilus edulis. Marine Environmental Research, 35, 233–249. Zeng, E. Y., Khan, A. R., & Tran, K. (1997). Organic pollutants in the coastal marine environment off San Diego, California. 3. Using linear alkylbenzenes to trace sewage-derived organic materials. Environmental Toxicology & Chemistry, 16, 196–201. Zeng, E. Y., & Yu, C. C. (1996). Measurement of linear alkylbenzenes by GC/MS with interference from tetrapropylene-based alkylbenzenes: calculation of quantitation errors using a two-compartment model. Environmental Science and Technology, 30, 322–328.