Heavy metal contamination in freshwater fish from the border region between Norway and Russia

Heavy metal contamination in freshwater fish from the border region between Norway and Russia

the Science of the -IbtalEnvIrolltnent F TheScienceof the TotalEnvironmentZOl(1997)211-224 : Heavy metal contamination in freshwater fish from the ...

3MB Sizes 0 Downloads 6 Views

the Science of the -IbtalEnvIrolltnent F

TheScienceof the TotalEnvironmentZOl(1997)211-224

:

Heavy metal contamination in freshwater fish from the border region between Norway and Russia Per-Arne Amundsenat* , Frode J. Staldvik”, Anatolij A. Lukinb, Nikolai A. Kashulinb, Olga A. Popova”, Yuri S. Reshetnikov” hlnstitute

“Norwegian College of Fishery Science, Universi~ of Troms& N-9037 [email protected], Norway of North Ecological Industrial Problems, Kola Science Center, 14 Fersman Str., Apatity, Murmansk ‘A.N. Severtzov Institute of Ecology and Animal Morphology, Russian Academy of Sciences, Moscow,

Region, Russia

Russia

Received12February1997;accepted 8 April 1997

Abstract The contentsof Cd, Cu, Cr, Hg, Ni and Zn in muscle,liver and gills were studiedin whitefish, perch, pike, brown trout, burbot and vendace from three iake localities in a watercoursein the border region between Norway and Russia,in the vicinity of mining activity and severalmetallurgic smelters.The contents of Cd and Ni in fish tissue increasedwith increasingproximity to the smelters,whereasthe other elementsshowedsimilarconcentrationsat the three localities. The recorded heavy metal concentrationsappearedto be within the rangesreported for fish from other metal-contaminatedlakes,and higher than comparableobservationsfrom unpolluted systems.The heavy metal concentrationswere usually lowest in muscleand highest in the liver or the gills. Significant differences in metal concentration levels were found between different fish species,but Hg was the only metal where these species differenceswere possiblyrelated to biomagnification.For the other elements,the concentrationsgenerally appeared to be inversely related to the trophic level of the fish species. 0 1997Elsevier ScienceB.V. Keywords:

Heavy metals;Freshwaterfish; Pasvik River system;Norway; Russia

___

1. Introduction

Emission of heavy metals from mining activity, smelters and industry is the source of serious

*Corresponding author.

environmental pollution (Kelly, 1988). In aquatic ecosystems, heavy metals have received considerable attention due to their toxicity and accumulation in biota (Mance, 1987; Mason, 1991). Some of these elements are toxic to living organisms even at quite low concentrations, whereas others are biologically essential and natural constituents

0048-9697/97/$17.00 0 1997ElsevierScienceB.V. All rightsreserved. PI1 SOO48-9697(97>00125-3

212

P.-A. Amundsen

et al. / The Science of the Total Environment

of the aquatic ecosystems and only become toxic at very high concentrations. In fish, the toxic effects of heavy metals may influence both physiological functions, individual growth rates, reproduction and mortality (Mance, 1987; Sorensen, 1991; Woodward et al., 1994, 1995; Farag et al., 1994, 1995). Heavy metals may enter fish bodies in three possible ways: through the body surface, the gills or the digestive tract (Dallinger et al., 1987; Pourang, 1995). The gills are regarded as the important site for direct uptake from the water (Hughes and Flos, 1978; Thomas et al., 1983; Dallinger et al., 19871, whereas the body surface is generally assumed to play a minor role in heavy metal uptake of fish (Varanasi and Markey, 1978; Dallinger et al., 1987; Pourang, 1995). Food may also be an important source for heavy metal accumulation (e.g. Aoyama et al., 1978; Patric and Loutit, 1978; Dallinger and Kautzky, 1985; Dallinger et al., 1987), potentially leading to biomagnification, the increase of pollutants up the food chain (Mason, 1991). Major metallurgic industries are located in the Murmansk Region in northwest Russia, with large emissions of dust, heavy metals and sulphur dioxide. In the border region between Russia and Norway, the main pollution sources are the smelters of the Pechenganickel Company located in the towns of Nikel and Zapolyarny. The Pasvik River system, which is the main border river between Russia and Norway, is located near the smelters. The lower part of the Pasvik River drains the Nikel smelters directly through Kuetsjarvi, a lake which is situated in a sidebranch to the main watercourse (Fig. 1). Pollutants may enter the water system both via deposition of atmospheric emissions and direct run-off from slag piles. High contaminations of heavy metals have been recorded in water and sediments in the vicinity of the smelters (Traaen et al., 1991; Duvalter, 1992, 1994; Moiseenko et al., 1995), possessing a potential threat to fish and other biota in the water system. Increased levels of heavy metals have also been observed in terrestrial plants, birds and mammals in areas adjacent to the smelters (Aamlid, 1992; KUs et al., 1995). In the present work, the heavy metal concentrations in freshwater fish from the Pasvik River

201 (1997) 211-224

system were investigated. Three lake localities with presumably different exposures to the pollutant sources were studied; Kuetsjarvi close to the smelters, Bjornevatn 13 km downstream and Vaggatem 45 km upstream in the main watercourse (Fig. 1). The contents of cadmium, copper, chromium, mercury, nickel and zinc were analyzed in muscle, liver and gills of the fish. 2. Study area and fish communities The sub-arctic Pasvik River system runs along the national border between Norway and Russia, originating from Lake Inari in Finland (Fig. 1). The Norwegian-Russian part of the river system is 120 km long, has a total area of 142 km*, a catchment area of 20.890 km2, and a mean waterflow of about 175 m3/s. There are altogether seven water impoundments in the Pasvik River system. Previous rapids and waterfalls have disappeared, and the river system is dominated by consecutive lakes and reservoirs. The water level fluctuations are small, and usually less than 80 cm. The ice-free season in the lakes and reservoirs last from May/June to October/November. Vaggatem (69” 14’ N, 29” 12’ E; 52 m above sea level) is a lake situated in the upper part of the main water course, Bjornevatn (69” 31’ N, 30” 7’ E; 22 m a.s.1.) in the lower part, and Kuetsjarvi (69” 27.’ N, 30“ 11’ E; 23 m a.s.1.) in a tributary to the main watercourse. The lakes are oligotrophic with humic impacts (see Nest et al. (1991) and Langeland (1993)). The geology in the region is dominated by bedrock, mainly containing gneiss. A birch- and pinewood landscape with stretches of boggy land surrounds the watercourse. The annual mean air temperature is low (-3‘0, and the minimum and maximum monthly mean temperatures are - 13.5”C and + 14.o”C, respectively. The precipitation in the area is low, with an annual mean of 358 mm. The smelters in Nikel and Zapolyarny were constructed for the processing of local ores, and have been in operation since 1932 and 1955, respectively. Since 1971, the smelters have processed copper and nickel ores from Norilsk, Central Siberia, which have a particularly high sulphur content (Duvalter, 1994). In 1989, the

P.-A. Amundsen et al. / The Science of the Total Environment 201 (1997) 211-224

213

Fig. 1. Map of the Pasvik River system.

annual emission of Ni and Cu to the atmosphere from these smelters was estimated to be 504 and 300 metric tons, respectively (Sivertsen et al., 1992). Altogether 15 different fish species have been recorded in the Pasvik River system, the most commonly occurring species in the lakes are whitefish (Coregonus lavaretus s.l.), perch (Perca JzuviutiZis), pike (Esox lucks), burbot (Lotu lota)

truttu). Vendace brown trout (Sulmo (Coregonusalbula) h as recently invaded the Pasvik

and

River system, after being introduced to Lake Inari in the 1960s. The whitefish consists of two different morphs, differentiated by the morphology and number of gill rakers, and referred to as densely and sparsely rakered whitefish. The densely rakered whitefish has numerous long and densely spaced gill rakers (mean number 33.01, whereas

214

P.-A. Amundren

et al. /The

Science

of the

the sparsely rakered form has fewer, shorter and more widely spaced rakers (mean number 23.1). According to Reshetnikov (1980), the two forms may be referred to as C. lavaretus lavaretus (densely rakered whitefish) and C. Zavaretzu pidschian (sparsely rakered), whereas Svfrdson (1957) (1979) described these forms as two different species; Coregonus lavaretus and C. nasus, respectively. 3. Materials

and methods

Fish were sampled both in 1991 and 1992 from Kuetsjarvi, Bjarnevatn and Vaggatem. In 1991, sampling was accomplished from July 14 to July 22, and from September 2 to September 15; and in 1992, from June 22 to July 4, and from August 31 to September 16. Fish were sampled both in the littoral, profundal and pelagic zones using gillnets. The gillnets were 40 m long containing eight sections of 5 m with different mesh sizes. The mesh sizes used were 10, 12.5, 15, 18.5, 22, 26, 35 and 45 mm (knot to knot). In the littoral and profundal zone, 1.5-m deep bottom nets were used, whereas in the pelagic zone, 4-m deep floating nets were employed. In addition, special nets for pike were used, with mesh sizes 35, 45, 52. 63 and 74 mm. The fish samples included densely and sparsely rakered whitefish, perch, pike, brown trout, burbot and vendace. The fish were weighed and fork length was measured. Stomachs were removed, conserved in 96% ethanol, and the contents later analyzed. Samples of fish tissue for heavy metal analysis were dissected from muscle, liver and gills using stainless steel implements, and frozen in sample tubes. The heavy metal analyses were conducted at the laboratory of The Norwegian Institute for Nature Research, and included cadmium (Cd), chromium (Cr), copper (CL& mercury (Hg), nickel (Ni) and zinc (Zn). The samples were freeze-dried for 24 h to a final pressure of 0.05 mbar at -53°C. From each sample (standard 0.3-0.4 g dry weight), the organic matter was extruded using concentrated nitric acid (HNO,), and the content of heavy metals was determined by atomic absorption spectroscopy (AAS) (Perkin Elmer

Total Environment

201 (1997)

211-224

Model 1100 B). Cd, Cu, Ni and Zn were analyzed by flame AAS, Cr by using graphite-furnace HGA 700 with automatic sampler AS 70, and Hg by using a hydride system (FIAS 200) with automatic sampler AS 90. NaBH, was used as reducing agent. The concentrations of heavy metals in fish are expressed as pug/g dry weight of tissue. The detection limits were 0.01 pg/g for Cd, 0.15 pg/g for Cr, 0.5 pg/g for Cu, 0.015 pg/g for Hg, 0.1 pg/g for Ni, and 0.5 pg/g for Zn. The accuracy of the analytical procedures was checked against the National Bureau of Standards (NBS) for dogfish muscle DOLT-l. The accuracy of the analytical procedures was satisfying for most metals, but for Cr, only 50-60% of the certified values could be detected, and for Hg 85% (S. Lierhagen, The Norwegian Institute for Nature Research, personal communication). The mean lengths of fish sampled for heavy metal analysis are presented in Table 1. Brown trout from Kuetsjarvi and burbot from Vaggatem have been omitted in the presentation of the results due to low observation numbers, whereas vendace was not caught at all in Bjornevatn and Kuetsjarvi. In Kuetsjarvi, no gill tissue samples from perch were analyzed, whereas in fish from Bjornevatn, the concentrations of Cr were not analyzed in any tissue and Cd only in liver. Cu concentrations were not analyzed in gills from any species. Distributions of heavy metal concentrations are frequently skewed and non-normal (Giesy and Wiener, 1977). Non-parametric statistics (MannWhitney and Kruskal-Wallis tests) have therefore been used in testing differences between samples. Simple linear regression analysis has been used to test for correlation between heavy metal concentrations and fish length. 4. Results 4.1. Tissue concentrations of hea y metals With a few exceptions, there were rarely any significant correlations between the tissue concentrations of heavy metals and the size of the individual fish. For Zn, however, a significant

P.-A. Amundsen

et al. / The Science of the Total Environment

215

201 (1997) 211-224

Table 1 Mean length with range (in cm) and number (n) of fish analyzed for heavy metal contents in Vaggatem, Bjarnevatn and Kuetsjarvi Vaggatem

Densely rakered whitefish Sparsely rakered whitefish Perch Pike Brown trout Burbot Vendace

Bjamevatn

Kuetsjarvi

Length

Range

n

Length

Range

n

Length

Range

n

17.5 25.2 22.1 38.2 30.0

lo-29 12-36 12-30 20-50 27-34

19.2 26.1 24.1 44.4 41.9 29.3

lo-29 lo-40 9-36 28-55 30-55 18-47

9-26 10-44 15-18 35-52 20-22 53-57

9-16

12 18 19 16 11 7 0

17.0 22.7 17.1 45.0 21.0 54.6

13.8

12 16 18 17 5 1 5

16 26 7 9 2 4 0

decrease in concentrations in muscle with increasing fish length was found for both densely rakered whitefish, sparsely rakered whitefish and perch. A similar significant decrease in liver concentrations of Zn was also found in densely and sparsely rakered whitefish. For the other elements there were rarely any significant differences with increasing fish lengths, and in the further presentation of the results, the mean values of heavy metal concentrations for each fish species have been used, irrespective of the size of the fish. The mean concentrations of heavy metals in muscle, liver and gills from the different fish species at the three different localities are presented in Fig. 2. Zinc made up the highest concentrations both in muscle, liver and gills of the fish. Considering all fish species and localities, the mean concentrations of heavy metals in muscle varied from 0.01-0.81 pg/g Cd, 0.17-0.45 pg/g Cr, 1.6-12.3 pg/g Cu, 0.16-0.89 pg/g Hg, 0.48-3.1 pg/g Ni, and 17-63 pg/g Zn. In the liver, the concentrations varied from 0.40-4.3 pg/g Cd, 0.19-0.46 pg/g Cr, 11-180 pg/g Cu, 0.14-1.1 pg/g Hg, 0.58-9.0 pg/g Ni, and 98-614 pg/g Zn, and in the gills from 0.02-0.28 pg/g Cd, 0.64-2.0 pg/g Cr, 0.03-0.10 pg/g Hg, 0.4-9.13 E*.g/g Ni, and 75-675 pg/g Zn (Fig. 2). The maximum heavy metal concentrations observed in the muscle of individual fish were (name of species in brackets): 3.0 pg/g Cd (densely rakered whitefish), 2.0 pg/g Cr (pike), 40.5 pg/g Cu (densely rakered whitefish), 2.0 pg/g Hg (perch), 8.0 pg/g Ni (densely rakered whitefish), and 430 pg/g Zn (pike). In the liver, the maximum recorded levels of individual fish

were 10 pg/g Cd (densely rakered whitefish), 1.0 pg/g Cr (densely rakered whitefish), 354 pg/g Cu (brown trout), 2.0 pg/g Hg (brown trout), 18 pg/g Ni (sparsely rakered whitefish), and 891 pg/g Zn (sparsely rakered whitefish), whereas for the gills, the highest observations were 0.69 pg/g Cd (densely rakered whitefish), 3.0 pg/g Cr (pike), 0.3 pg/g Hg (pike), 17.3 pg/g Ni (densely rakered whitefish), and 1406 pg/g Zn (densely rakered whitefish). For most elements, the lowest concentrations of heavy metals were found in the muscle (Table 2 and Fig. 2). For Cd, the highest concentrations were always found in the liver, with levels many times higher than in the muscle, although there were large variations both between the different fish species and the localities. The contents of Cr in the liver were generally twice as high as in the muscle, whereas the highest levels of Cr were found in the gills. The Cu concentrations were always higher in the liver than in the muscle, with particularly high ratios in sparsely rakered whitefish and brown trout. The recorded Hg concentrations were always lowest in the gills, whereas the liver contents of Hg generally were equal or slightly higher than in the muscle. For Ni, the levels in liver and gills generally were of similar magnitude, and a few times higher than in the muscle. The relative proportion of Ni in the gills appeared to increase with decreasing distance from the smelters. Zn had lowest concentrations in the muscle, and, with the exception of perch, the highest concentrations were found in the gills (Table 2).

216

P.-A. Amundsen et al. / The Science of the Total Environment 201 (1997) 211-224

Vaggatcm

3

Bjernevatn

Kuetsjavri

l!iiiJm’ll1l. 15” 1.0 0.5 0.0

1.0

.W z

0.8 0.6 0.4 0.2 0.0 10 8 6 4 2 *oi 600 400 200 0

1234567

1234567

1234567

Fish species Fig. 2. Mean concentrations (fig/g dry weight) of Cd, Cr, Cu, Hg, Ni and Zn in muscle (black bars), liver (white bars) and gills (hatched bars) in fish from three different lake localities in the Pasvik River system. The fish species are indicated by numbers: (1) densely rakered whitefish, (2) sparsely rakered whitefish, (3) perch, (4) pike, (5) brown trout, (6) burbot, and (7) vendace.

4.2. Heay metal contamination in j&h from different localities

For a few of the examined elements, the tissue concentrations of heavy metals were significantly different between the three investigated lake localities. In particular, the concentrations of Ni in liver and gills decreased with increasing distances from the smelters (Fig. 3). Similarly, the observed concentrations of Cd in muscle and gills were highest in Kuetsjarvi in the vicinity of the smelters, and lowest in Vaggatem at the upper part of the water course (Fig. 3). In the gills, the differences between the localities in concentrations of both

Ni and Cd were most profound for densely and sparsely rakered whitefish, whereas perch exhibited the strongest difference in muscle contents of Cd. For elements other than Ni and Cd, there were generally no significant differences in concentration levels in fish tissue between the three investigated lake localities. 4.3. Trophical statusand heay metal contamination of different fish species

The most important habitat and prey choices of lacustrine fish species in the Pasvik River system are presented in Table 3. Adults of both pike,

P.-A. Amundsen

et al. / The Science of the Total Enviromnt

Table 2 Ratios between muscle, liver and gills (M&G) localities Fish species and locality

217

201 (1997) 211-224

in concentrations of heavy metals in different fish species at the three different

Cd

Cr

cu

Hg

Ni

Zn

Densely rakered whitefish Vaggatem Bjamevatn Kuetsjarvi

1:3:3 1.5:9:1

1:2:8 1:2:3

1:2.5:1:2.5:1:6.5:-

3:7:1 l:l:2:2:1

1:4:1 3:4:2 1:6:9

1:7:15 1:6:14 1:5:10

Sparsely rakered whitefish Vaggatem Bjamevatn Kuetsjarvi

1:200:4 1:150:13

1:2:1:1.5:-

1:18:1:13:1:12:-

1:5:3 1:2:1:6:4

1:6:1 1:2:3 1:8:9

1:7:18 1:9:36 1:8:26

Perch Vaggatem Bjemevatn Kuetsjarvi

1:200:6 1:5:-

l:l:-

1:12:1:1.7:1:5:-

1:1:0.1 l:l:l:l:-

1:1:2 I:22 1:5:-

1:6:4 1:5:4 1:8:-

Pike Vaggatem Bjtimevatn Kuetsjarvi

1:100:2 1:4:0.5

l:l:1216

1:5:1:3:-

7:4:1 1:1.5:1 1.4:3:1

2:3:1 l:l:1:1.8:2.3

1:8:20 1:13:24 1:3:6

Brown trout Vaggatem Bjtimevatn

-

-

l:llo:1:15:-

1:1.6:1:2:-

1:1.4:1:2:-

1:7:1:7:-

Burbot Vaggatem Bjemevatn

-

-

-

-

-

1:3:5 -

Vendace Vaggatem

-

-

1:18:-

1:2:-

-

1:17:-

-

- indicates insufficient or missing observations.

perch and brown trout were predominantly piscivorous, and particularly pike and brown trout fed heavily upon whitefish. Burbot, which most commonly was caught in the profundal zone, feed both on zoobenthos and fish. The coregonids (densely and sparsely rakered whitefish and vendate) were all restricted to invertebrate feeding. The densely rakered whitefish occurred frequently in all major habitats of the lakes in the Pasvik river system, feeding both on zooplankton and benthic prey. The sparsely rakered whitefish were found to be restricted to the benthic habitats feeding on zoobenthos, whereas vendace almost exclusively fed upon zooplankton in the pelagic zone (Table 3). Significant differences in heavy metal contents

between different fish species were found for Hg and Zn in muscle, Hg, Zn, Cd, Ni, Cu and Cr in liver, and Hg, Zn and Cd in gills (Table 4; Kruskal-Wallis test, P < 0.05). For Hg, the muscle concentrations were in general highest for the piscivorous species, and lowest for the invertebrate feeders. Brown trout had the highest Hg concentrations in the liver, whereas pike had the highest Hg levels in the gills. Vendace and densely rakered whitefish had the highest concentrations of Zn in muscle and liver, and sparsely rakered whitefish in the gills, whereas perch generally had the lowest Zn contents. The coregonid species also had the highest levels of Cd both in liver and gills, whereas the piscivorous species had the lowest. Further, for both Ni and Cr, the

218

P.-A. Amunaken et al. / The Science of the Total Environment 201 (1997) 211-224

Cd

.I 0.8s ,*R 0.6 -

:: ”g 0.4 7 2% g 0.2 -

e---l - a--n -. Kuotrj.

Knotrj.

Vnggotcm

Dr. w.firh S.r. w.fish Porch Pike

Vaggotsm

I

I

Ni

e--l - a--n- --

Fig. 3. Contaminations of Ni and Cd ( pg/g dry weight) in tissue of fish from Kuetsjati downstream) and Vaggatem (45 km upstream).

Dr. w.fish S.r. w.firh Perch Pike

(close to the smelters), Bj0mevatn (13 km

Table 3 Predominant prey and habitat choices of lacustrine fish in the Pasvik River system Fish species

Dominant prey

Dominant habitat

Densely rakered whitefish Sparsely rakered whitefish Perch Pike Brown trout Burbot Vendace

Zooplankton, benthic chydorids, dipteran pupae and larvae Dipteran pupae and larvae, molluscs, benthic chydorids Fish (9-sp. sticklebacks and whitefish), insect larvae Fish (whitefish, perch) Fish (whitefish) Insect larvae, fish Zooplankton

Pelagic, profundal and littoral Littoral, profundal Littoral Littoral Littoral, pelagic Profundal, littoral Pelagic

liver concentrations were highest in the coregonids and lowest in the piscivorous species, whereas for Cu, the liver contents were highest for brown trout, intermediate for the coregonid species and lowest for pike and perch (Table 4). 5. Discussion The six heavy metals studied in the present investigation are all regarded as potential hazards that can endanger both animal and human health

(Duffus, 1980; Mance, 1987). Knowledge of their concentrations in fish is therefore important both with respect to nature management and human consumption of fish. The literature on metal uptake and assimilation in fish may, however, be both confusing and partly also conflicting (MCFarlane and Franzin, 1980). Chemical factors like acidity, buffer capacity and the presence of Ca and organic compounds, may, together with biological factors like habitat preferences, feeding behaviour and growth rates, influence the

P.-A. Amunaken

et al. / The Science of the Total Environment

201 (1997) 211-224

219

Table 4 Ranking order of different fish species in the Pasvik River system according to their heavy metal contents. (Only species series with significant differences are included; Kruskal-Wallis test, P < 0.05) Hg

Muscle Liver Gills

Burbot > Perch > B. trout = Pike > Vendace = D.r. whitefish = S.r. whitefish B. trout > Vendace > Perch > D.r. whitefish = S.r. whitefish > Pike Pike > Dr. whitefish > S.r. whitefish > Perch

Zn Muscle Liver Gills

D.r. whitefish > Vendace = Pike > Burbot > B. trout > Sr. whitefish = Perch Vendace > Pike = D.r. whitefish > Sr. whitefish = B. trout > Perch > Burbot S.r. whitefish > Pike = D.r. whitefish > Burbot > Perch

Cd Liver Gills

Vendace > Dr. whitefish > S.r. whitefish > Perch > Pike > Burbot > B. trout S.r. whitefish = D.r. whitefish > Perch > Pike

Cr Liver

D.r. whitefish > S.r. whitefish > Perch > Pike

cu Liver

B. trout > Sr. whitefish > Vendace > Dr. whitefish > Pike > Perch

Ni Liver

Sr. whitefish > D.r. whitefish > Burbot > Perch > Pike > B.trout

bioavailability and accumulation of heavy metals in fish (see e.g. McFarlane and Franzin, 1980; Campbell and Stokes, 1985; Bradley and Morris, 1986; Dallinger et al., 1987; Sprenger et al., 1988; Iivonen et al., 1992). Further, the predominant pathways for heavy metal uptake appear to be highly variable over the range of metals, contaminant levels and fish species studied (e.g. Williams and Giesy, 1978; Dallinger et al., 1987). There also seems to be large interspecific variation in the assimilation of heavy metals from water and food, as well as in the allocation of metals into different organs and tissue. Additionally, the sensitivity to heavy metals and the ability for homeostatic control, detoxification and rejection appear to be highly variable for different elements and fish species (e.g. Wiener and Giesy, 1979; Dallinger et al., 1987). Intraspecific differences related to e.g. age, size and physiological status of the fish may also be of importance. Moreover, numerous potential interactions between different elements may influence both the assimilation and toxicity of the metals (Sorensen, 1991). Thus, the heavy metal concentrations in fish tissue may appear to be the result of a

complex interaction of many factors (Wiener and Giesy, 1979; McFarlane and Franzin, 1980; Dallinger et al., 19871, and Dallinger et al. also emphasised the importance of taking the specific ecological situation of a given environment into consideration when investigating heavy metal pollution in fish. The studied fish species in the Pasvik River system are important constituents of the limnetic ecosystems in the lakes, and are also being exploited for human consumption. Coregonids, perch, pike, brown trout and burbot have circumpolar distributions, and although data on heavy metal concentrations in Arctic and temperate fish populations are limited, some comparable studies exist. Allen-Gil and Martynov (1995) studied heavy metal contents in muscle from fish in the Pechora River, northern Russia. The muscle contents of cadmium and copper found in the present study were many times higher, and zinc concentrations slightly higher, than those found in coregonids, perch and pike in the Pechora River. Further, the concentrations of cadmium, copper and zinc in muscle and liver of Pasvik fish were generally higher than those found at two localities

220

P.-A. Amundren

et al. / The Science of the Total Environment

in the Great Lakes (Brown and Chow, 1977), whereas mercury levels were of similar magnitude (conversions between dry and wet weight of fish have been done by assuming 78% water contents). Also, the concentrations of cadmium, copper and zinc in liver of pike, perch, whitefish and brown trout from the Pasvik River system were in general higher than average concentrations found in the same species from 14 unpolluted Finnish lakes; although in these lakes, the mercury levels in muscle were higher than in Pasvik (Iivonen et al., 1992). Bradley and Morris (1986) studied heavy metals in fish (including whitefish, perch and pike) in a series of lakes near Sudbury, Ontario, where large quantities of copper, nickel and other metals have been released into the environment from several smelters. The levels of copper, nickel and zinc in muscle and liver of fish from Pasvik were similar or slightly higher than those found in these metal-contaminated lakes. Pike from the Pasvik River systems did, on the other hand, have lower levels of cadmium and copper in the liver compared to the maximum contaminations recorded in pike from lakes near a base metal smelter at Flin Flon, Manitoba, whereas the mercury levels were of similar magnitude (McFarlane and Franzin, 19801. Concentrations of cadmium and nickel recorded in fish tissue in the Pasvik River system, exhibited significant increases with increasing proximity to the smelters. For both metals, the lowest concentrations were found in Vaggatem and the highest in Kuetsjarvi. The high levels of cadmium and nickel in fish from Kuetsjarvi reflect the elevated environmental levels of heavy metals that have been found in the vicinity of the Pechenganikel smelters (Duvalter, 1992, 1994; Moiseenko et al., 1995). Both deposition of atmospheric emissions from the smelters and direct run-off from slag piles and mines may contribute to the metal contaminations in the Pasvik watercourse. Pollutants may also be transported downstream by the water Aow, although rapid sedimentation probably restrains this mode of heavy metal dispersal. The cadmium contamination is likely to be most associated with surface run-offs from the slag piles, as elevated Cd concentrations in sediments have only

201 (1997) 211-224

been detected close to the Pechenganikel smelters (Duvalter, 1994). For nickel, on the other hand, the atmospheric emissions from the smelters are large (Sivertsen et al., 1992), and apparently the main pollution source for this element. Nickel concentrations in sediments have been found to be very high in Kuetsjarvi, intermediate in Bjornevatn and small in Vaggatem (Rognerud et al., 19931, and the declines in Ni levels in fish and sediments with increasing distance from the smelters, are probably related to rapid atmospheric fallouts and deposition. The increasing proportion of nickel in the gills with decreasing distance from the smelters, may indicate that the uptake over the gills becomes increasingly important as the Ni contents in the water increases (see also Sreedevi et al., 1992). For the other studied metals, the lack of significant differences in fish tissue concentrations with increasing distance from the pollution source may suggest that the exposures of these elements are similar throughout the three localities. However, for most of these metals, a decline in water and sediment concentrations with increasing distance from the smelters has been demonstrated (Duvalter, 1992, 1994; Moiseenko et al., 1995). The present findings do, therefore, more likely reflect a poor correlation in the concentrations of these elements between fish and environment. McFarlane and Franzin (1980) emphasised that high Ca concentrations in lake waters may decrease metal accumulation in fish. Due to large dust emissions from the smelters, the Ca concentration in Kuetsjarvi has been found to be very high (Traaen et al., 1991; Nest et al., 19911, and this may have moderated the metal accumulation of fish in the lake. As a consequence, the heavy metal levels in fish may not be a good indicator of the ambient pollution levels (Wiener and Giesy, 1979; Bradley and Morris, 1986). However, this also implies that for an evaluation of the biological impacts of heavy metal pollution, the actual metal concentrations in the fish, and not solely the metal levels in water and sediments, should be taken into consideration. The Zn concentrations in fish tissue decreased significantly with increasing length of the fish,

P.-A. Amundsen et al. / The Science of the Total Environment 201 (1997) 21 l-224

which has also been demonstrated in several other studies (e.g. Cross et al., 1973; Giesy and Wiener, 1977; Murphy et al., 1978; Wiener and Giesy, 1979). Patric and Loutit (1978) indicated that this may partly be due to different adsorption rates across the gut or more efficient excretion in older fish. Zinc is an essential trace metal that is presumably homeostatically controlled (Frieden, 1972; Cross et al., 1973; Giesy and Wiener, 1977), and negative correlations between metal concentrations and body size may also be a result of homeostatic regulation (Wiener and Giesy, 1979). These findings may, on the other hand, also be related to the absorption rates of zinc across the gills. Zinc was the only element that was found in highest concentrations in the gills relative to the muscle and liver contents. High concentrations of a metal in the gills are indicative of a predominant uptake from water through the gills (Dallinger et al., 1987). The area of the gills relative to body size decreases with increasing fish size (Pauly, 1981>, and this may result in a larger uptake of certain elements like zinc in the smaller fish. For all the elements studied, significant differences in concentration levels were found between different fish species, demonstrating that the attained concentration levels partly are species related. Generally, the two whitefish forms and vendace seemed to have the highest concentrations of zinc, cadmium, chromium and nickel, whereas brown trout, burbot, perch and pike had the highest concentrations of mercury. The salmonoid species had higher liver contents of copper than pike and perch, which was also found by Iivonen et al. (1992), and pike and perch seemed in general to have lower accumulations of most metals in the liver relative to the other species. To a large extent the species differences in heavy metal concentrations appeared to be related to the trophic status of the fish species. Mercury was, however, the only metal where these species differences may be due to biomagnification. Particularly with respect to the muscle concentrations, the mercury levels were highest in the piscivorous species and lowest in the invertebrate feeders. Biomagnification of heavy metals in fish

221

is regarded as a controversial subject (Dallinger et al., 1987), although biomagnification of mercury along aquatic food chains has frequently been reported (e.g. Jemlijv and Lann, 1971; Sarkka et al., 1978; Mason, 1991). Biomagnification of a pollutant may lead to toxic levels in species high up in the trophic chain (Moriarity, 19841, and in freshwater systems, fish are usually among the top consumers. In fish from the Pasvik River system, the recorded levels of mercury were, however, not high compared to other, comparable fish studies (e.g. Haakanson et al., 1988; Iivonen et al., 1992), and were well below the accepted limits for human consumption (1 pg/g wet weight). For the other elements, and especially zinc and cadmium, the concentrations appeared to be inversely related to the trophic status of the fish. Fish feeding on invertebrates have been found to have higher concentrations of cadmium and zinc than piscivorous species (Mathis and Cummings, 1973; Windom et al., 1973; Murphy et al., 1978). The results of the present study largely supports these observations, as whitefish and vendate usually had the highest levels of these two elements, and the piscivorous species the lowest. In conclusion, the heavy metal concentrations recorded in lacustrine fish from the Pasvik River system appear to be within the ranges reported for fish from metal-contaminated lakes, and higher than comparable observations from unpolluted systems. The concentrations were lowest in muscle, and the observed metal levels did not exceed established quality standards for fish flesh. The metal levels were usually several times higher in the liver, which is known to be an important detoxification centre in fish (McFarlane and Franzin, 1980). Nickel and cadmium exhibited significant increases in tissue concentrations with decreasing distances from the smelters, and had particularly high levels in Kuetsjarvi. Nickel may be harmful to the survival and productivity of freshwater fauna (Moore and Ramamoorthy, 19841, but studies of effects of Ni accumulation in fish are scarce (Sreedevi et al., 1992). Cadmium is a persistent material that may be toxic to aquatic organisms at relatively low concentrations, but

222

P.-A. Amudsen

et al. /The

Science of the Total Enuironment

Taylor (1983) considered the bioconcentration of Cd in aquatic vertebrates to be so low that it was likely of little significance. Dallinger et al. (19871, on the other hand, pointed out that most heavy metals are effective at very low concentrations, so even low assimilation rates may be sufficient to attain biologically significant or even harmful concentrations in tissue. There are some indications that the fish populations in Kuetsjarvi may suffer pollution stress from heavy metal accumulations, since pathological anomalies such as nephrocalcitosis and liver cirrhosis have been shown frequently to occur (Langeland, 1993; Moiseenko et al., 1995; Lukin and Kashulin, unpublished). There is also some evidence for recruitment failure of the perch and pike populations in the lake (unpublished data). Unfortunately, however, there is little empirical evidence about biologically harmful concentrations of heavy metals in fish, and it is, therefore, difficult to evaluate the possible ecological effects of the metal levels recorded in the present study. Adams et al. (1989) advocated the use of bioindicators for assessing the effects of pollutant stress in fish. For future research it may be fruitful to integrate studies of heavy metal concentrations in fish tissue with investigations of bioindicators like fish pathology and histology, as well as extensive ecological surveillance both at the population and community level. Acknowledgements Thanks are due to Syverin Lierhagen, The Norwegian Institute for Nature Research, for help with the heavy metal analyses, to Laina Dalsbpr, Jan Evjen and Elleke Wartena, The Norwegian College of Fishery Science, for assistance in field and laboratory, and to Armand Kuris, University of California, Santa Barbara, for correcting the English. The study was part of a research programme carried out under the Norwegian-Russian Environmental Protection Commission, and had financial support from The Norwegian Ministry of Environment Protection and The Directorate for Nature Management.

201 (1997) 211-224

References Aamlid D. The concentration of nickel and copper in some important plants in South-Varanger, Norway. In: Kismul V, Jerre, Ubersli E, editors. Effects of air pollution on terrestrial ecosystem in the border area between Russia and Norway. Oslo, Norway: State Pollution Control Authority, Document 92:04, 1992:202-209. Adams SM, Shepard KL, Greeley Jr MS et al. The use of bioindicators for assessing the effects of pollutant stress on fish. Mar Environ Res 1989;28:459-464. Allen-Gil SM, Martynov VG. Heavy metal burdens in nine species of freshwater and anadromous fish from the Pechora River, northern Russia. Sci Total Environ 1995;160/161:653-659. Aoyama I, Inoue Yos, Inoue Yor. Experimental study on the concentration process of trace element through a chain from the viewpoint of nutrition ecology. Water Res 1978;12:831-836. Bradley RW, Morris JR. Heavy metals in fish from a series of metal-contaminated lakes near Sudbury, Ontario. Water Air Soil Pollut 1986;27:341-354. Brown JR, Chow LY. Heavy metal concentrations in Ontario fish. Bull Environ Contam Toxic01 1977;17:190-195. Campbell PGC, Stokes PM. Acidification and toxicity of metals to aquatic biota. Reviews. Can J Fish Aquat Sci 1985;42:2034-2049. Cross FA, Hardy LH, Jones NY, Barber RT. Relation between total body weight and concentrations of manganese, iron, copper, zinc and mercury in white muscle of bluefish (Pomatomus saltatrix) and a bathyl-demersal fish Antitrwra rostrata. J Fish Res Bd Can 1973;30:1287-1291. Dallinger R, Kautzky H. The importance of contaminated food for the uptake of heavy metals by rainbow trout (a field study). Oecologia (Berlin) 1985;67:&2-89. Dallinger R, Prosi F, Senger H, Back H. Contaminated food and uptake of heavy metals by fish (a review and proposal for further research). Oecologia (Berlin) 1987;73:91-98. Duffus JH. Environmental toxicology. London: Edward Arnold Publishers, 1980. Duvalter V. Concentration of heavy metals in superficial lake sediments of Pechenga district, Murmansk region, Russia. Vatten 1992;48:141-145. Duvalter V. Heavy metals in lake sediments of the Kola Peninsula, Russia. Sci Total Environ 1994;158:51-61. Farag AM, Boese CJ, Woodward DF, Bergman HL. Physiological changes and metal tissue accumulation in rainbow trout exposed to food-borne and water-borne metals. Environ Toxic01 Chem 1994;13:2021-2029. Farag AM, Stansbury MA, Hogstrand C, MacConnell E, Bergman HL. The physiological impairment of free-ranging brown trout exposed to metals in the Clark Fork River, Montana. Can J Fish Aquat Sci 1995;52:2038-2050. Frieden E. The chemical elements of life. Sci Am 1972;227:52-60.

P.-A. Amundsen

et al. /The

Science of the Total Enuironment

Giesy JR, Wiener JC. Frequency distribution of trace metal concentrations in five freshwater fishes. Tram Am Fish Sot 1977;106:393-403. Haakanson L, Nilsson A, Andersson T. Mercury in fish in Swedish lakes. Environ Pollut 1988,49:145-162. Hughes GM, Flos R. Zinc content of the gills of rainbow trout 6. gairdneti) after treatment with zinc solutions under mormoxic and hypoxic conditions. J Fish Biol 1978;13:717-728. Iivonen P, Piepponen S, Verta M. Factors affecting trace-metal bioaccumulation in Finnish headwater lakes. Environ Pollut 1992;78:87-95. Jemlijv A, Lann H. Mercury accumulation in food chains. Oikos 1971;22:403. Kelly M. Mining and the freshwater environment. London: Elsevier, 1988. K%lb JA, Ringsby TH, Lierhagen S. Metals and selenium in wild animals from Norwegian areas close to Russian nickel smelters. Environ Monitor Assess 1995;36:251-270. Langeland A, editor. Pollution impact on freshwater communities in the border region between Russia and Norway. Baseline study 1990-1992. NINA Scientific Report 44/93. Trondheim, Norway: Norwegian Institute for Nature Research, 1993:53. Mance G. Pollution threat of heavy metals in aquatic environments. London: Elsevier, 1987. Mason CF. Biology of freshwater pollution. 2nd ed. Essex: Longman Scientific and Technical, 1991. Mathis BJ, Cummings TF. Selected elements in sediments, water and biota in the Illinois River. J Water Pollut Contam Fed 1973;45:1573-1583. McFarlane GA, Franxin WG. An examination of Cd, Cu and Hg concentrations in levers of northern pike, Esox Lucius and white sucker, Catostomus commersoni, from five lakes near a base metal smelter at Flin Flon, Manitoba. Can J Fish Aquat Sci 1980;37:1573-1578. Moiseenko TI, Kudtyavtseva LP, Rodyushkin IV, Duvalter VA, Lukin AA, Kashulin NA. Airborne contamination by heavy metals and aluminium in the freshwater ecosystems of the Kola Subarctic region (Russia). Sci Total Environ 1995;160/161:715-727. Moore JW, Ramamoorthy S. Nickel. In: Desano RS, editor. Heavy Metals in Industrial Waters. Applied Monitoring and Impact Assessment. New York: Springer Verlag, 1984:161-173. Moriarity F. Persistent contaminants, compartmental models and concentrations along food-chains. Ecol Bull 1984;36:35-45. Murphy BR, Atchison GJ, McIntosh AW. Cadmium and zinc content of fish from an industrially contaminated lake. J Fish Biol 1978;13:327-335. Nest T, Yakovlev V, Berger HM, Kahsulin N, Langeland A, Lukin A. Impacts of pollution on freshwater communities in the border region between Russia and Norway. NINA Scientific Report 26. Trondheim, Norway: Norwegian Institute for Nature Research, 1991:41.

201 (1997)

211-224

223

Patric FM, Loutit M. Passage of metals to freshwater fish from their food. Water Res 1978;12:395-398. Pauly D. The relationship between gill surface area and growth performance in fish: a generalization of von Bertalanffy’s theory of growth. Meeresforschung 1981;28:251-282. Pourang N. Heavy metal bioaccumulation in different tissues of two fish species with regards to their feeding habits and trophic levels. Environ Monitor Assess 1995;35:207-219. Reshetnikov YS. Ecology and Systematics of Coregonid Fish. Moscow: Nauka Press, 1980:300, (In Russian). Rognerud S, Norton S, Duvalter V. Heavy metal pollution in lake sediments in the border areas between Russia and Norway. Oslo, Norway: State Pollution Control Authority, Report 522/93, 1993:18. Sivertsen B, Makarova T, Hagen LO, Baklanov AA. Air polution in the border areas of Norway and Russia. NILU, OR, 8/92, Lillestrom, Norway: Norwegian Institute for Air Research, 1992: 14. Sorensen EM. Metal poisoning in fish. Boca Raton, FL: CRC Press, 1991. Sprenger MD, McIntosch AW, Hoenig S. Concentrations of trace elements in yellow perch (Perca flauescens) from six acidic lakes. Water Air Soil Pollut 1988;37:375-388. Sreedevi P, Suresh A, Sivaramakrishna B, Prabhavathi B, Radhakrishnaiah K. Bioaccumulation of nickel in the organs of the freshwater fish, Cyprinus carpio and the freshwater mussel, Lamellidens marginalis, under lethal and sublethal nickel stress. Chemosphere 19922429-36. Svkdson G. The Coregonid problem VI. The Palearctic species. Rep Inst Freshwater Res Drottningholm 1957;38:267-356. Svardson G. Speciation of Scandinavian Coregonus. Rep Inst Freshwater Res Drottningholm 1979;57:95. SIrMtii J, Hattula M-L, Paasivirta J, Janatuinen J. Mercury and chlorinated hydrocarbons in the food chain of Lake Plijlnne, Finland. Ho1 Ecol 1978;1:326-332. Taylor D. The significance of the accumulation of cadmium by aquatic organisms. Ecotoxicol Environ Safety 1983;7:33-42. Thomas DG, Cryer A, Solbe J, De LG, Kay J. A comparison and protein binding of environmental cadmium in the gills, kidney and liver of rainbow trout (Salmo gairdneti Richardson). Comp Biochem Physiol 1983;76:241-246. Traaen TS, Moiseenko T, Duvalter V, Rognerud S, Henriksen A, Kudravseva L. Acidification of surface waters, and nickel and copper in water and lake sediments in the RussianNorwegian border areas. Progress report for 1989-1990. The Norwegian-Russian Environmental Protection Commission, Oslo (Norway) and Apatity (Russia), 1991:20. Varanasi U, Markey D. Uptake and release of lead and cadmium in skin and mucus of coho salmon (Oncorynchus kisutch). Comp Biochem Physiol 1978;60:187-191. Wiener JG, Giesy Jr JP. Concentrations of Cd, Cu, Mn, Pb and Zn in fishes in a highly organic softwater pond. J Fish Res Bd Can 1979;36:270-279. Williams DR, Giesy Jr JP. Relative importance of food and water sources to cadmium uptake by Gambusia afinis (Poeciliidae). Environ Res 1978;16:326-332.

224

P.-A. Amundren et al. / The Science of the Total Environment 201 (1997) 211-224

Windom H, Stickney R, Smith R, White D, Taylor F. Arsenic, cadmium, copper, mercury and zinc in some species of North Atlantic &fish. J Fish Res Bd Can 1973;30:275-279. Woodward DF, Brumbaug WG, DeLonay AJ, Smith C. Effects on rainbow trout of a metals-contaminated diet of benthic invertebrates from the Clark Fork river, Montana. Tram Am Fish Sot 1994;123:51-62.

Woodward DF, Farag AM, Brumbaug WG, Smith CE, Bergmann HL. Metals-contaminated invertebrates in the Clark Fork River, Montana: effects on age-0 brown trout and rainbow trout. Can J Fish Aquat Sci 1995;52:1994-2004.