- Email: [email protected]
Mercury in the stomach contents of dab (Limanda limanda) from the North East Irish Sea and Mersey Estuary
Environmental Pollution 72 (1991) 117 126
Mercury in the Stomach Contents of Dab (Limanda limanda) from the North East Irish Sea and Mersey Estuary R. T. Leah, Z. Y. Ma, S. J. Evans & M. S. Johnson Industrial Ecology Research Centre, Department of Environmental and Evolutionary Biology, University of Liverpool, PO Box 147, Liverpool, L69 3BX, UK (Received 20 August 1990: revised version received 28 November 1990; accepted 3 December 1990)
A BS TRA C T A study (?[mercuo' concentrations in the stomach contents o[i/~'shfrom the north-east h'ish Sea and Mersey Estuary has been shown to provide a means .[br surveillance ~['geographical and time-based changes in environmental exposure ~?['[4sh biota to mercury in marine and estuarine ecosystems. This paper describes data for the [tatfish dab ( Limanda limanda ), caught &tring the period 1986 88. The low degree of variability in the data enables confirmation of clear trends" in mercury concentration in stomach contents oeer time. As the inputs o f mercury to the sewage sludge dumping ground in Liverpool Bay have decreased, there has been a corresponding decrease in mercury in.fish Jbod items. The mean mercur)' t,alue in stomach contents around the dump site has declined to lOO ltg kg 1 (wet weight) which now predominates over the whole o[' Liverpool Bay. In 1986, mercury concentrations in stomach contents o f f i s h ranged to over 7501tgkg -l although the majority o f values were below 200 Itg kg-1. Most (?[ the sites within the Mersey Estuary produced mean concentrations which were similar to those in the open sea, except for Garston which is the site closest to an inland, and principal alternative, source (?['mercury.
INTRODUCTION The principal aim of this study was to investigate the mercury levels of the stomach contents of dab (Limanda limanda) both as a broad index of 117 Environ, Pollut. 0269-7491/91/$03.50 ~ 199l Elsevier Science Publishers Ltd, England. Printed in Great Britain
118
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson
geographical variation in environmental conditions and a more specific guide to dietary exposure of fish to trace pollutants. In particular, the intention was to investigate the use of fish stomach contents to clarify the spatial distribution of environmental mercury in Liverpool Bay including the Mersey Estuary, and their use to assess the general status of mercury in the marine environment over a timespan of several years. Historically, mercury has entered the Mersey Estuary and adjacent Irish Sea area from a number of separate sources including effluents from chemical industries and from the disposal of sewage sludge in Liverpool Bay. This has led to a progressive accumulation of mercury within sediments of the Mersey Estuary and parts of the Irish Sea (Airey & Jones, 1982; Campbell et al., 1986). Since the human health and environmental hazards associated with mercury exposure have been realised, there has been a concerted effort, worldwide, to set environmental quality objectives and to reduce substantially the inputs to natural ecosystems (CEC, 1976; Preston & Portmann, 1981). Subsequent changes in environmental conditions have been monitored in various ways, but surveillance of mercury in fish tissues has been of particular interest because of the relationship with human exposure via the food chain (Schreiber, 1983). Concentrations of mercury in fish from some locations have reduced in line with input controls, but, on a wider scale, there are areas where mercury levels remain relatively high (ICES, 1988). Data on contemporary concentrations of mercury in muscle tissue of fish from the study area show evidence of a slow decline in recent years (Franklin, 1987) and, in allied research, Leah et al. (in press), have shown consistent differences in the rates at which fish accumulate mercury according to the geographical location and species concerned. There is evidence of a decline in the mercury content of fish muscle tissue with distance from the River Mersey, a pattern that suggests fish populations remain fairly coherent as they move around the locality. This helps to preserve a 'record' of their previous mercury exposure. However, there remain considerable problems in defining the contributions of different sources of mercury to the food chain and to understanding the processes that control the equilibrium concentration of mercury in fish muscle. Fish are often used as monitoring organisms for trace elements such as mercury (e.g. Phillips et al., 1980) due to their propensity for bioaccumulation. As such they can be used to indicate spatial differences in the environmental burden or bioavailability of metallic pollutants. Although fish populations are sufficiently coherent to exhibit consistent differences in mercury accumulation over smalldistances, there is a lower limit to the scale of spatial discrimination that is possible because of the movement of fish stocks. Such movements also mean that trends through time are difficult to discern against this biological and environmental 'noise'. As much detailed
Mercury in the stomach contents of dab
119
information as possible is required for understanding complex environments and this paper reports upon recent research addressed to this problem and focussed on the Mersey Estuary and Liverpool Bay where there are multiple sources of mercury pollution and recent major reductions in inputs.
MATERIALS AND METHODS A programme of sampling was undertaken to test the hypothesis that the stomach contents of fish are representative of the food items from a relatively small area of sea in reasonably close proximity to the location of capture. As such, the mercury content of the consumed food ought to represent the contamination level of the local environment, allowing discrimination of small scale spatial differences. If this is the case, then monitoring of stomach contents should provide an effective means for following changes over time of mercury in the food chain, particularly around geographically precise locations such as sludge disposal grounds. Moreover, such data might clarify the relative contributions of different sources of mercury to the dietary exposure of fish populations. Stomach contents were analysed from samples of dab collected annually in the autumn between 1986 and 1988, a period when there was a continued reduction in the mercury being disposed of to the sludge dumping ground in Liverpool Bay (Dickson, 1987). Dab was chosen as the target species because of its widespread availability throughout the study area. Fish were sampled from various sites around the north-east Irish Sea (Fig. 1) using small inshore trawlers. Random samples of approximately thirty individuals above the commercial size limit (20 cm) were retained for analysis. The fish were blast frozen soon after capture, and were defrosted and dissected subsequently using stainless steel instruments. Muscle, liver tissue and the alimentary tract were excised for analysis, and the major food items identified. The material contained in the stomach (defined by the position of the pyloric sphincter) was analysed for total mercury by digestion in concentrated nitric acid in a beaker, a method that represents a slight modification of the tube digestion and cold vapour assay technique described by Evans et al. (19861. The changes were made due to frothing arising from the calcium carbonate in shell fragments. After 8 h at room temperature, the digestion was finished by evaporation from 30ml to approximately 10 ml on a hotplate at 130°C, a method which was verified by recovery of 94-106% of the mercury from reference tissues MAA-2 and MAM-2 supplied by the International Atomic Energy Agency, in Vienna. The limit of detection of this method at the tissue level is 10/tg k g - ~ under the conditions adopted. Early examination of a few samples showed that it was difficult to
120
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson
l~e of Man
Walney •
Fleetwood •
3 4 Dumping Grounds
Fig. 1. North East Irish Sea & MerseyEstuary. determine mercury reliably where the sample weight was low. Thus, in order to reduce analytical variability, only the contents of stomachs containing more than one gram of material are reported, these samples presumably representing fish which were actively feeding in the period immediately before capture. Data for mean mercury concentration in the stomach contents of dab are presented as histograms accompanied by standard deviation bars on which the replication is indicated. The data are presented by site so that changes through time at individual locations can be identified.
121
Mercury in the stomach contents of dab
RESULTS
Spatial and temporal differences in Liverpool Bay/Irish Sea There was a noticeable decline in the mean mercury concentration of the stomach contents of dab caught around the sludge dumping ground during the study period, as well as a significant reduction in the range of values encountered, reflected in the standard deviation of the means (Fig. 2). This pattern was recorded at sampling stations DG3 and DG4, close to the sludge dumping ground, where there was a notable reduction in the number of high values recorded over the sampling period. Site DG 4 showed the greater decline in mean mercury concentration, mainly because of the gradual disappearance of a small number of very elevated values (> 700 #g kg t l from the sample set. Near Fleetwood there was no significant difference %o> 0.05) in the mean value or range throughout the study period (Fig. 3), with the spread of values encountered remaining similar and fairly narrow. Similarly, the results for Walney show low and consistent values (Fig. 3).
Gradients in the River Mersey Data are available only for the estuary sites from the 1988 catch. Mean values did not vary much within the estuary but the range of values increased inland towards Garston (Fig. 4), which had a significantly higher mean concentration than most of the estuarine sites (p < 0.05). Overall, the range of values in samples from the river did not differ greatly from that in the area of sea immediately adjacent, but there were different frequency distributions across the sites. The most obvious difference was in 19 400
23
=~ 17
•
DG3
ii
DG4
200 •
-r-
19
100
O
~ 1986
1987
1988
Year
Fig. 2, Mercury concentrationin Dab stomach contents, dumping grounds 3 and 4.
122
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson 500
400
300
8 8
• Fwood [] Walney 200
27
o~
o~
16
100
0 1986
1987
1988
Year
Fig. 3.
Mercury concentration in Dab stomach contents, Fleetwood & Walney.
the sample from the Great Burbo Bank at the mouth of the estuary where the range was very narrow compared to elsewhere.
DISCUSSION Several methodological problems arise in the analysis of stomach contents, the most important being limitations imposed by the amount of food material present. The mercury in stomach contents of specimens containing only a small amount of food was apparently high, but results were discarded G.G.B,- Great Burbo Bank R.C,L.- Rock Channel Lagoon NB.+E.- New Brighton & EgremonL P.B.- Piuckington Bank
500
400
,~
soo
200
100
GB.B.
R.C.L.
N.B.+E.
P,B.
Gar ~ton
Sites
Fig. 4.
Mercury concentration in Dab stomach contents, Mersey Estuary, 1988.
Mercury in the stomach contents o[dah
123
as unreliable because the operating conditions were close to the analytical detection limit. A notable finding was the relative constancy of mercury concentrations and low statistical errors for data sets compiled from specimens with substantial food residues in the stomach. This suggests that the method has high repeatability and that it provides a means of integrating environmental heterogeneity. This low variation is especially important as it occurs over a broad area of sea and despite the wide background range of food items and extraneous material in the stomachs. This similarity in mercury concentration, despite varying species composition, is notable since Pellegrini and Barghigiani (1989) attributed the difference in mercury uptake by two species of flatfish to differences in their respective diets. In the present study of just one species, the overall range of values was very low, with wide variation occurring only at a small number of sites.
Spatial distribution of environmental mercury in the Irish Sea & Mersey Estuary The range of mean values for mercury in stomach contents of dab from different locations around the north-east Irish Sea was surprisingly small, with the modal class similar at all sites. Based on the 1988 dataset, a mean value of approximately 1 0 0 ~ g k g - I now prevails over the study area. Although there were values as high as 7 5 0 # g k g -1 in 1986, the m a x i m u m concentration evident in 1988 was around 350 #g k g - t. It is notable that the higher mean values and wider ranges are associated with sites near to the sewage sludge dump site. Overall, the range of mercury values for estuarine sites was similar to locations in Liverpool Bay, although there were very characteristic differences in the modal classes at the various sites. Significant features include the very narrow spread of values at Great Burbo Bank, on the seaward side of the estuary, and the greater range and elevated mean value for the inner estuary at Garston, closer to an industrial source. In 1988, most of the sites within the Mersey Estuary returned mean concentrations which were similar to those in the open sea, at approximately 100~gkg - t wet weight. However, the Garston site showed a mean value of 200 ~g k g - t
Changes in environmental mercury with time A significant outcome of the low variation within the data is that it is possible to identify convincing trends in mercury concentration over relatively short periods of time. The results show that, as the inputs of mercury to the sludge dumping ground decreased between 1986 and 1988, there was a coincident decrease in the mercury content of the prey of dab
124
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson
feeding in the area. Mercury in food items at the d u m p site decreased to approximately 100 #g kg-1, a value that now predominates over the whole of Liverpool Bay and further north at Fleetwood and Walney. The range of mercury concentrations in stomach contents at the d u m p sites decreased through the study period. In 1986, mercury concentrations in fish food ranged to over 750 #g kg-1 wet weight, although the majority of values were below 200 #g kg-1. Sites around the sludge dumping area provided the majority of high values, particularly site DG4 which is subject to residual flows from the dumping site. By 1988, high individual mercury values were more or less absent, resulting in very similar mean concentrations for all the truly marine sites.
Biotransfer The task of relating concentrations of mercury in fish to inputs into Liverpool Bay has been described as difficult because of the number of point sources (Murray & Norton, 1982). However, Murray and N o r t o n recognised that inputs from sewage sludges are more likely to provide mercury in a bioavailable form than are inputs from other sources. It is widely accepted that methylated mercury is the principal form of the metal involved in bioaccumulation in the marine food chain. The association of high individual mercury values in the stomach contents offish feeding around the sludge d u m p site suggests that mercury in that area is at a higher environmental concentration or, as it is already within the food chain, that is more bioavailable than elsewhere. With the recent reductions in input (Dickson, 1987), and the disappearance of high individual mercury concentrations in fish stomach contents, the rate of mercury transfer into the food chain appears to have decreased to a new equilibrium rate which is now similar over the whole area. This implies that by 1988 the rate of transfer was largely independent of geographical location throughout the study area, except perhaps for the upper Mersey Estuary. It is concluded, therefore, that within the Liverpool Bay/Mersey Estuary complex, the dumping of sewage sludge caused more mercury to be bioavailable in the area near to the d u m p site and was probably exerting a greater influence upon the concentration of this pollutant in fish food organisms than inputs of inorganic mercury from direct industrial sources entering via the estuary. However, this study does not cover the areas where dredge and industrial wastes have been deposited in the past. Since concentrations of mercury in fish stomach contents from most of the study area are now very similar, bioaccumulation rates are now largely independent of inputs in sewage sludge and as has been suggested elsewhere (Dickson, 1987) are unlikely to decline much further in the near term.
Mercury in the stomach contents q/"dab
125
It also appears that, in contrast with conventional views a b o u t the ability of mercury to magnify along food chains (Fimreite et al., 1971; Lyle, 1984), there is little evidence of biomagnification from the present study. Simple b i o a c c u m u l a t i o n is sufficient to explain the present data and the concentrations found in muscle tissue offish from the same area (Leah et al., in press).
ACKNOWLEDGEMENT The authors are grateful to Mrs Sally Collings for her assistance and advice in the preparation of this manuscript.
REFERENCES Airey, D. & Jones, P. D. (1982). Mercury in the River Mersey, its estuary and tributaries during 1973 and 1974. Water Res., 16, 565 77. Campbell, J. A., Chan, E. Y. L., Riley, J. P., Head, P. C. & Jones, P. D. (1986). The distribution of mercury in the Mersey Estuary. Mar. Pollut, Bull., 17, 36 40. Commission of the European Communities (CEC)(1976). Council Directive No. (4/5/76) on Pollution Caused by Dangerous Substances Discharged into the Aquatic Environment of the Community. Offl J. Eur. Commun., LI29, 23 29. Dickson, R. R. (ed.) (1987). Irish Sea Status Report ~t" the Marine Pollution Monitoring Management Group. Aquatic Environment Monitoring Report No. 17. Ministry of Agriculture, Fisheries & Food, Lowestoft. 83 pp. Evans, S. J., Johnson, M. S. & Leah, R. T. (1986). Determination of mercury in lish tissue: A rapid, automated technique for routine analysis. Varian Instruments at Work, Vol. AA-60. Varian-Techtron Pry, London. Fimreite, N., Holsworth, W. N., Keith, J. A., Pearce, P. A. & Gruchy, I. M. (1971). Mercury in fish and fish-eating birds near sites of industrial contamination in Canada. Can. Field Naturalist, 85, 211-20. Franklin, A. (1987). The concentration of metals, organochlorine pesticide and PCB residues in marine .fish and shellfish; results j~'om M A F F fish and shefilish monitoringprogrammes, 1977-1984. Aquatic Environment Monitoring Report No, 16. Ministry of Agriculture, Fisheries & Food, Lowestoft. 38 pp. International Council for the Exploration of the Sea (1988). Results ~?l"the 1985 baseline study of con taminan ts infish and shellfish, l C E S C o -opera tire Rese arc h Report No. 151. Leah, R. T., Evans, S. J., Johnson, M. S. & Collings, S. (in press). Spatial patterns in accumulation of mercury by fish from the North East Irish Sea. Mar. Pollut, Bull. Lyle, J. M. (1984). Mercury concentrations in four carcharhinid and three hammerhead sharks from coastal waters of the Northern Territory. Aust. J. Mar. Freshw. Res., 35, 441-51. Murray, A. J. & Norton, M. G. (1982). Thefield assessment of e/fects qldumping
126
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson
wastes at sea. Fisheries Technical Report No. 69. Ministry of Agriculture, Fisheries & Food, Lowestoft. 42 pp. Pellegrini, D. & Barghigiani, C. (1989). Feeding behaviour and mercury content in two flat fish in the Northern Tyrrhenian sea. Mar. Pollut. Bull., 20, 443-7. Phillips, G. R., Lenhart, T. E. & Gregory, R. W. (1980). Relation between trophic position and mercury accumulation among fishes from the Tongue River Reservoir, Montana. Environ. Res., 22, 73-80. Preston, A. & Portmann, J. E. (1981). Critical path analysis applied to the control of mercury inputs to United Kingdom coastal waters. Environ. Pollut. (Series B), 2, 451-64. Schreiber, W. (1983). Mercury content of fishery products: Data from the last decade. Sci. Total Environ., 31,283-300.
Mercury in the Stomach Contents of Dab (Limanda limanda) from the North East Irish Sea and Mersey Estuary R. T. Leah, Z. Y. Ma, S. J. Evans & M. S. Johnson Industrial Ecology Research Centre, Department of Environmental and Evolutionary Biology, University of Liverpool, PO Box 147, Liverpool, L69 3BX, UK (Received 20 August 1990: revised version received 28 November 1990; accepted 3 December 1990)
A BS TRA C T A study (?[mercuo' concentrations in the stomach contents o[i/~'shfrom the north-east h'ish Sea and Mersey Estuary has been shown to provide a means .[br surveillance ~['geographical and time-based changes in environmental exposure ~?['[4sh biota to mercury in marine and estuarine ecosystems. This paper describes data for the [tatfish dab ( Limanda limanda ), caught &tring the period 1986 88. The low degree of variability in the data enables confirmation of clear trends" in mercury concentration in stomach contents oeer time. As the inputs o f mercury to the sewage sludge dumping ground in Liverpool Bay have decreased, there has been a corresponding decrease in mercury in.fish Jbod items. The mean mercur)' t,alue in stomach contents around the dump site has declined to lOO ltg kg 1 (wet weight) which now predominates over the whole o[' Liverpool Bay. In 1986, mercury concentrations in stomach contents o f f i s h ranged to over 7501tgkg -l although the majority o f values were below 200 Itg kg-1. Most (?[ the sites within the Mersey Estuary produced mean concentrations which were similar to those in the open sea, except for Garston which is the site closest to an inland, and principal alternative, source (?['mercury.
INTRODUCTION The principal aim of this study was to investigate the mercury levels of the stomach contents of dab (Limanda limanda) both as a broad index of 117 Environ, Pollut. 0269-7491/91/$03.50 ~ 199l Elsevier Science Publishers Ltd, England. Printed in Great Britain
118
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson
geographical variation in environmental conditions and a more specific guide to dietary exposure of fish to trace pollutants. In particular, the intention was to investigate the use of fish stomach contents to clarify the spatial distribution of environmental mercury in Liverpool Bay including the Mersey Estuary, and their use to assess the general status of mercury in the marine environment over a timespan of several years. Historically, mercury has entered the Mersey Estuary and adjacent Irish Sea area from a number of separate sources including effluents from chemical industries and from the disposal of sewage sludge in Liverpool Bay. This has led to a progressive accumulation of mercury within sediments of the Mersey Estuary and parts of the Irish Sea (Airey & Jones, 1982; Campbell et al., 1986). Since the human health and environmental hazards associated with mercury exposure have been realised, there has been a concerted effort, worldwide, to set environmental quality objectives and to reduce substantially the inputs to natural ecosystems (CEC, 1976; Preston & Portmann, 1981). Subsequent changes in environmental conditions have been monitored in various ways, but surveillance of mercury in fish tissues has been of particular interest because of the relationship with human exposure via the food chain (Schreiber, 1983). Concentrations of mercury in fish from some locations have reduced in line with input controls, but, on a wider scale, there are areas where mercury levels remain relatively high (ICES, 1988). Data on contemporary concentrations of mercury in muscle tissue of fish from the study area show evidence of a slow decline in recent years (Franklin, 1987) and, in allied research, Leah et al. (in press), have shown consistent differences in the rates at which fish accumulate mercury according to the geographical location and species concerned. There is evidence of a decline in the mercury content of fish muscle tissue with distance from the River Mersey, a pattern that suggests fish populations remain fairly coherent as they move around the locality. This helps to preserve a 'record' of their previous mercury exposure. However, there remain considerable problems in defining the contributions of different sources of mercury to the food chain and to understanding the processes that control the equilibrium concentration of mercury in fish muscle. Fish are often used as monitoring organisms for trace elements such as mercury (e.g. Phillips et al., 1980) due to their propensity for bioaccumulation. As such they can be used to indicate spatial differences in the environmental burden or bioavailability of metallic pollutants. Although fish populations are sufficiently coherent to exhibit consistent differences in mercury accumulation over smalldistances, there is a lower limit to the scale of spatial discrimination that is possible because of the movement of fish stocks. Such movements also mean that trends through time are difficult to discern against this biological and environmental 'noise'. As much detailed
Mercury in the stomach contents of dab
119
information as possible is required for understanding complex environments and this paper reports upon recent research addressed to this problem and focussed on the Mersey Estuary and Liverpool Bay where there are multiple sources of mercury pollution and recent major reductions in inputs.
MATERIALS AND METHODS A programme of sampling was undertaken to test the hypothesis that the stomach contents of fish are representative of the food items from a relatively small area of sea in reasonably close proximity to the location of capture. As such, the mercury content of the consumed food ought to represent the contamination level of the local environment, allowing discrimination of small scale spatial differences. If this is the case, then monitoring of stomach contents should provide an effective means for following changes over time of mercury in the food chain, particularly around geographically precise locations such as sludge disposal grounds. Moreover, such data might clarify the relative contributions of different sources of mercury to the dietary exposure of fish populations. Stomach contents were analysed from samples of dab collected annually in the autumn between 1986 and 1988, a period when there was a continued reduction in the mercury being disposed of to the sludge dumping ground in Liverpool Bay (Dickson, 1987). Dab was chosen as the target species because of its widespread availability throughout the study area. Fish were sampled from various sites around the north-east Irish Sea (Fig. 1) using small inshore trawlers. Random samples of approximately thirty individuals above the commercial size limit (20 cm) were retained for analysis. The fish were blast frozen soon after capture, and were defrosted and dissected subsequently using stainless steel instruments. Muscle, liver tissue and the alimentary tract were excised for analysis, and the major food items identified. The material contained in the stomach (defined by the position of the pyloric sphincter) was analysed for total mercury by digestion in concentrated nitric acid in a beaker, a method that represents a slight modification of the tube digestion and cold vapour assay technique described by Evans et al. (19861. The changes were made due to frothing arising from the calcium carbonate in shell fragments. After 8 h at room temperature, the digestion was finished by evaporation from 30ml to approximately 10 ml on a hotplate at 130°C, a method which was verified by recovery of 94-106% of the mercury from reference tissues MAA-2 and MAM-2 supplied by the International Atomic Energy Agency, in Vienna. The limit of detection of this method at the tissue level is 10/tg k g - ~ under the conditions adopted. Early examination of a few samples showed that it was difficult to
120
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson
l~e of Man
Walney •
Fleetwood •
3 4 Dumping Grounds
Fig. 1. North East Irish Sea & MerseyEstuary. determine mercury reliably where the sample weight was low. Thus, in order to reduce analytical variability, only the contents of stomachs containing more than one gram of material are reported, these samples presumably representing fish which were actively feeding in the period immediately before capture. Data for mean mercury concentration in the stomach contents of dab are presented as histograms accompanied by standard deviation bars on which the replication is indicated. The data are presented by site so that changes through time at individual locations can be identified.
121
Mercury in the stomach contents of dab
RESULTS
Spatial and temporal differences in Liverpool Bay/Irish Sea There was a noticeable decline in the mean mercury concentration of the stomach contents of dab caught around the sludge dumping ground during the study period, as well as a significant reduction in the range of values encountered, reflected in the standard deviation of the means (Fig. 2). This pattern was recorded at sampling stations DG3 and DG4, close to the sludge dumping ground, where there was a notable reduction in the number of high values recorded over the sampling period. Site DG 4 showed the greater decline in mean mercury concentration, mainly because of the gradual disappearance of a small number of very elevated values (> 700 #g kg t l from the sample set. Near Fleetwood there was no significant difference %o> 0.05) in the mean value or range throughout the study period (Fig. 3), with the spread of values encountered remaining similar and fairly narrow. Similarly, the results for Walney show low and consistent values (Fig. 3).
Gradients in the River Mersey Data are available only for the estuary sites from the 1988 catch. Mean values did not vary much within the estuary but the range of values increased inland towards Garston (Fig. 4), which had a significantly higher mean concentration than most of the estuarine sites (p < 0.05). Overall, the range of values in samples from the river did not differ greatly from that in the area of sea immediately adjacent, but there were different frequency distributions across the sites. The most obvious difference was in 19 400
23
=~ 17
•
DG3
ii
DG4
200 •
-r-
19
100
O
~ 1986
1987
1988
Year
Fig. 2, Mercury concentrationin Dab stomach contents, dumping grounds 3 and 4.
122
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson 500
400
300
8 8
• Fwood [] Walney 200
27
o~
o~
16
100
0 1986
1987
1988
Year
Fig. 3.
Mercury concentration in Dab stomach contents, Fleetwood & Walney.
the sample from the Great Burbo Bank at the mouth of the estuary where the range was very narrow compared to elsewhere.
DISCUSSION Several methodological problems arise in the analysis of stomach contents, the most important being limitations imposed by the amount of food material present. The mercury in stomach contents of specimens containing only a small amount of food was apparently high, but results were discarded G.G.B,- Great Burbo Bank R.C,L.- Rock Channel Lagoon NB.+E.- New Brighton & EgremonL P.B.- Piuckington Bank
500
400
,~
soo
200
100
GB.B.
R.C.L.
N.B.+E.
P,B.
Gar ~ton
Sites
Fig. 4.
Mercury concentration in Dab stomach contents, Mersey Estuary, 1988.
Mercury in the stomach contents o[dah
123
as unreliable because the operating conditions were close to the analytical detection limit. A notable finding was the relative constancy of mercury concentrations and low statistical errors for data sets compiled from specimens with substantial food residues in the stomach. This suggests that the method has high repeatability and that it provides a means of integrating environmental heterogeneity. This low variation is especially important as it occurs over a broad area of sea and despite the wide background range of food items and extraneous material in the stomachs. This similarity in mercury concentration, despite varying species composition, is notable since Pellegrini and Barghigiani (1989) attributed the difference in mercury uptake by two species of flatfish to differences in their respective diets. In the present study of just one species, the overall range of values was very low, with wide variation occurring only at a small number of sites.
Spatial distribution of environmental mercury in the Irish Sea & Mersey Estuary The range of mean values for mercury in stomach contents of dab from different locations around the north-east Irish Sea was surprisingly small, with the modal class similar at all sites. Based on the 1988 dataset, a mean value of approximately 1 0 0 ~ g k g - I now prevails over the study area. Although there were values as high as 7 5 0 # g k g -1 in 1986, the m a x i m u m concentration evident in 1988 was around 350 #g k g - t. It is notable that the higher mean values and wider ranges are associated with sites near to the sewage sludge dump site. Overall, the range of mercury values for estuarine sites was similar to locations in Liverpool Bay, although there were very characteristic differences in the modal classes at the various sites. Significant features include the very narrow spread of values at Great Burbo Bank, on the seaward side of the estuary, and the greater range and elevated mean value for the inner estuary at Garston, closer to an industrial source. In 1988, most of the sites within the Mersey Estuary returned mean concentrations which were similar to those in the open sea, at approximately 100~gkg - t wet weight. However, the Garston site showed a mean value of 200 ~g k g - t
Changes in environmental mercury with time A significant outcome of the low variation within the data is that it is possible to identify convincing trends in mercury concentration over relatively short periods of time. The results show that, as the inputs of mercury to the sludge dumping ground decreased between 1986 and 1988, there was a coincident decrease in the mercury content of the prey of dab
124
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson
feeding in the area. Mercury in food items at the d u m p site decreased to approximately 100 #g kg-1, a value that now predominates over the whole of Liverpool Bay and further north at Fleetwood and Walney. The range of mercury concentrations in stomach contents at the d u m p sites decreased through the study period. In 1986, mercury concentrations in fish food ranged to over 750 #g kg-1 wet weight, although the majority of values were below 200 #g kg-1. Sites around the sludge dumping area provided the majority of high values, particularly site DG4 which is subject to residual flows from the dumping site. By 1988, high individual mercury values were more or less absent, resulting in very similar mean concentrations for all the truly marine sites.
Biotransfer The task of relating concentrations of mercury in fish to inputs into Liverpool Bay has been described as difficult because of the number of point sources (Murray & Norton, 1982). However, Murray and N o r t o n recognised that inputs from sewage sludges are more likely to provide mercury in a bioavailable form than are inputs from other sources. It is widely accepted that methylated mercury is the principal form of the metal involved in bioaccumulation in the marine food chain. The association of high individual mercury values in the stomach contents offish feeding around the sludge d u m p site suggests that mercury in that area is at a higher environmental concentration or, as it is already within the food chain, that is more bioavailable than elsewhere. With the recent reductions in input (Dickson, 1987), and the disappearance of high individual mercury concentrations in fish stomach contents, the rate of mercury transfer into the food chain appears to have decreased to a new equilibrium rate which is now similar over the whole area. This implies that by 1988 the rate of transfer was largely independent of geographical location throughout the study area, except perhaps for the upper Mersey Estuary. It is concluded, therefore, that within the Liverpool Bay/Mersey Estuary complex, the dumping of sewage sludge caused more mercury to be bioavailable in the area near to the d u m p site and was probably exerting a greater influence upon the concentration of this pollutant in fish food organisms than inputs of inorganic mercury from direct industrial sources entering via the estuary. However, this study does not cover the areas where dredge and industrial wastes have been deposited in the past. Since concentrations of mercury in fish stomach contents from most of the study area are now very similar, bioaccumulation rates are now largely independent of inputs in sewage sludge and as has been suggested elsewhere (Dickson, 1987) are unlikely to decline much further in the near term.
Mercury in the stomach contents q/"dab
125
It also appears that, in contrast with conventional views a b o u t the ability of mercury to magnify along food chains (Fimreite et al., 1971; Lyle, 1984), there is little evidence of biomagnification from the present study. Simple b i o a c c u m u l a t i o n is sufficient to explain the present data and the concentrations found in muscle tissue offish from the same area (Leah et al., in press).
ACKNOWLEDGEMENT The authors are grateful to Mrs Sally Collings for her assistance and advice in the preparation of this manuscript.
REFERENCES Airey, D. & Jones, P. D. (1982). Mercury in the River Mersey, its estuary and tributaries during 1973 and 1974. Water Res., 16, 565 77. Campbell, J. A., Chan, E. Y. L., Riley, J. P., Head, P. C. & Jones, P. D. (1986). The distribution of mercury in the Mersey Estuary. Mar. Pollut, Bull., 17, 36 40. Commission of the European Communities (CEC)(1976). Council Directive No. (4/5/76) on Pollution Caused by Dangerous Substances Discharged into the Aquatic Environment of the Community. Offl J. Eur. Commun., LI29, 23 29. Dickson, R. R. (ed.) (1987). Irish Sea Status Report ~t" the Marine Pollution Monitoring Management Group. Aquatic Environment Monitoring Report No. 17. Ministry of Agriculture, Fisheries & Food, Lowestoft. 83 pp. Evans, S. J., Johnson, M. S. & Leah, R. T. (1986). Determination of mercury in lish tissue: A rapid, automated technique for routine analysis. Varian Instruments at Work, Vol. AA-60. Varian-Techtron Pry, London. Fimreite, N., Holsworth, W. N., Keith, J. A., Pearce, P. A. & Gruchy, I. M. (1971). Mercury in fish and fish-eating birds near sites of industrial contamination in Canada. Can. Field Naturalist, 85, 211-20. Franklin, A. (1987). The concentration of metals, organochlorine pesticide and PCB residues in marine .fish and shellfish; results j~'om M A F F fish and shefilish monitoringprogrammes, 1977-1984. Aquatic Environment Monitoring Report No, 16. Ministry of Agriculture, Fisheries & Food, Lowestoft. 38 pp. International Council for the Exploration of the Sea (1988). Results ~?l"the 1985 baseline study of con taminan ts infish and shellfish, l C E S C o -opera tire Rese arc h Report No. 151. Leah, R. T., Evans, S. J., Johnson, M. S. & Collings, S. (in press). Spatial patterns in accumulation of mercury by fish from the North East Irish Sea. Mar. Pollut, Bull. Lyle, J. M. (1984). Mercury concentrations in four carcharhinid and three hammerhead sharks from coastal waters of the Northern Territory. Aust. J. Mar. Freshw. Res., 35, 441-51. Murray, A. J. & Norton, M. G. (1982). Thefield assessment of e/fects qldumping
126
R. T. Leah, Z. Y. Ma, S. J. Evans, M. S. Johnson
wastes at sea. Fisheries Technical Report No. 69. Ministry of Agriculture, Fisheries & Food, Lowestoft. 42 pp. Pellegrini, D. & Barghigiani, C. (1989). Feeding behaviour and mercury content in two flat fish in the Northern Tyrrhenian sea. Mar. Pollut. Bull., 20, 443-7. Phillips, G. R., Lenhart, T. E. & Gregory, R. W. (1980). Relation between trophic position and mercury accumulation among fishes from the Tongue River Reservoir, Montana. Environ. Res., 22, 73-80. Preston, A. & Portmann, J. E. (1981). Critical path analysis applied to the control of mercury inputs to United Kingdom coastal waters. Environ. Pollut. (Series B), 2, 451-64. Schreiber, W. (1983). Mercury content of fishery products: Data from the last decade. Sci. Total Environ., 31,283-300.