Fish community responses to pulp and paper mill effluents at the southern Lake Saimaa, Finland

Fish community responses to pulp and paper mill effluents at the southern Lake Saimaa, Finland

Environmental Pollution 116 (2002) 309–317 www.elsevier.com/locate/envpol Fish community responses to pulp and paper mill effluents at the southern Lak...

291KB Sizes 0 Downloads 2 Views

Environmental Pollution 116 (2002) 309–317 www.elsevier.com/locate/envpol

Fish community responses to pulp and paper mill effluents at the southern Lake Saimaa, Finland A.E. Karels a,*, A. Niemi b a Department of Biological and Environmental Science, University of Jyva¨skyla¨, PO Box 35, FIN-40351 Jyva¨skyla¨, Finland Employment and Economic Development Centre for South Eastern Finland, Fisheries services, PO Box 1041, FIN-45101 Kouvola, Finland

b

Received 14 January 2000; accepted 8 February 2001

‘‘Capsule’’: Fish responses to pulp and paper mill effluents appear related to habitat changes caused by nutrient enrichment. Abstract The fish community in sublittoral and profundal waters, at stony shores, and densities of vendace larvae were studied in the southern Lake Saimaa, Finland. The objective was to investigate the possible recovery of fish populations after modernizations at the pulp and paper mills discharging into the lake. Sublittoral and profundal waters were studied by gill net trial fishings, stony shores by electrofishing, and vendace larvae by beach seine. The research area was divided in a polluted (0.5–4.0% effluent), an intermediate (0.1–0.5%) and a ‘clean’ reference area. The fish community in sublittoral and profundal waters in the lake was dominated (>60%) by perch and roach. Relative abundance of fish was highest in the polluted area, and lowest in the reference area. The number of species caught was similar among areas. The abundance of bleak and ruffe was highest in the polluted area, while the abundance of vendace and whitefish was highest in the intermediate and reference area. The fish fauna of stony shores in the lake was dominated by bullhead, stone loach and minnow, densities were lowest at polluted shores and highest in the intermediate area. Minnow, apparently a more sensitive species to pulp mill effluents, were not caught at the most polluted shores. The catch of vendace larvae was similar among areas, larvae were also caught in the vicinity of the mills. Compared with before the modernization at the mills, the relative abundance of perch in the polluted area was increased, ruffe decreased, while populations of whitefish and vendace showed signs of recovery. These changes may be explained by the reduced nutrient load and toxicicity of pulp mill effluents. At present, the fish community in the polluted area can be considered typical for a moderate eutrophication. # 2001 Elsevier Science Ltd. All rights reserved. Keywords: Fish; Communities; Populations; Pulp mill effluents; Lake Saimaa

1. Introduction The recent developments in process technology and wastewater treatment in the Finnish pulp and paper industry have significantly decreased the environmental impacts on receiving water ecosystems. Also at the southern Lake Saimaa, affected by the pulp and paper industry since 1897, the amount of nutrients and chemical compounds discharged by the pulp and paper mills has decreased significantly during the 1990s (Laine and Minkkinen, 1998). The loading of wastewater into the lake reached its maximum in the 1960s. The first

* Corresponding author.

improvements in the effluent quality were achieved in the 1970 and 1980s, with the development in the process technology (black liquor recovery and burning) and the start of mechanical purification and the first generation of biological treatment of effluents. Despite the increasing production, a clear decrease of the amount of suspended solids, biological oxygen demand (BOD) and chemical oxygen demand (COD) in the effluent took place at the mills (Fig. 1). However, phosphorus and nitrogen concentrations increased until the early 1990s, when modern activated sludge treatment plants were built for all three mills discharging effluent into the lake (Laine and Minkkinen, 1998). The most recent technical alterations at the mills, particularly the introduction of elemental chlorine free (ECF) bleaching and activated sludge treatment of

0269-7491/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0269-7491(01)00120-8

310

A.E. Karels, A. Niemi / Environmental Pollution 116 (2002) 309–317

Fig. 1. Total loading of suspended solids (SS), chemical oxygen demand (COD), biological oxygen demand (BOD), nitrogen (N), phosphorus (P), adsorbable organic halogens (AOX) and sodium (Na) of the pulp and paper mill factories in Lappeenranta, Joutseno and Imatra from 1973 to 1997 (modified from Laine and Minkkinen, 1998).

effluents, had a relatively fast impact on the lake water quality at the southern Lake Saimaa. This was in accordance with results of biochemical and physiological responses in feral and caged fish exposed to the mill effluents (Oikari and Holmbom, 1996; Soimasuo et al., 1998; Karels et al., 1998, 2000a, 2000b, 2001). For instance, the liver mixed function oxygenase (MFO) activity and chlorophenolics concentrations in the bile of exposed fish approached the reference levels after the process alterations at the mills, indicating a decreased exposure to pulp mill effluent constituents. As responses at the sub-organism and organism levels is understood to precede changes at the population and community level (Bayne et al., 1985), we hypothesized that the improved water quality, as reported by physicochemical and biomarker responses, should lead to recovery of fish populations in the lake. Although local fisheries research data on fish populations in Finnish lakes polluted by pulp mill effluents exist, publications in scientific journals are very scarce. Hakkari (1992) reported that fish communities in Finnish lake waters polluted by pulp mill effluents were dominated by perch Perca fluviatilis (L.), roach Rutilus rutilus (L.), and ruffe Gymnocephalus cernuus (L.), while so called sensitive fish species (salmoniformes) have disappeared. Earlier, direct ecotoxicological effects of pulp and paper mill effluents on fish populations are often associated with the grade of eutrofication due to nutrient enrichment by pulp and paper mill effluents (Hansson, 1987; Neuman and Kara˚s, 1988; Sandstro¨m et al., 1991; Hakkari, 1992, Landner et al., 1994). Also, repelling compounds in the effluents are expected to have an effect on fish populations; for instance, whitefish Coregonus lavaretus s.l. (L.), vendace Coregonus albula (L.) and lake trout Salmo trutta m. lacustris (L.) avoid

polluted waters (Myllyvirta and Vuorinen, 1989; Hakkari, 1992). At present, with the relatively low load of toxicants and nutrients due to the new process techniques, there is less knowledge about effects on exposed fish communities. The primary aim of this research was to study the possible recovery of fish populations in the recovering lake area affected by pulp mill effluents. Therefore, we investigated species composition and abundance of fish at sublittoral and profundal waters by gill net trial fishing, and species composition and densities of fish at stony shores by electrofishing. Additionally, the reproductive success of vendace was investigated by measuring the abundance of vendace larvae by a beach seine survey. Fish were sampled in areas with different grades of pulp mill effluent contamination and in a suggested ‘‘clean’’ reference area. The results of this research are compared with earlier fish community studies in the southern Lake Saimaa, before any of the major process renewals at the pulp and paper mills involved (Heinonen and Falck, 1971; Sauvonsaari, 1974).

2. Materials and methods 2.1. Study area and pulp and paper mills The southern Lake Saimaa is a large oligotrophic and oligo mesohumic lake in southeast Finland. It has a surface area of 620 km2, a water volume of 5.2 km3 and a mean depth of 8.4 m. The main charge of water flows into the lake at the strait of Rastinvirta, being practically the same as the outflow through the River Vuoksi into the Lake Ladoga, averaging 596 m3/s (years 1961– 1990). The directions of the water flow in the lake are

311

A.E. Karels, A. Niemi / Environmental Pollution 116 (2002) 309–317

shown in Fig. 2. The retention time of water in the southern Lake Saimaa is about 60 days. The islands in the western and eastern part are more shattered than at other parts of the lake, which are more open. Its shores are usually barren, stony or sandy. Vegetative areas are small and situated in sheltered bays. At the present time, four pulp, paper and cardboard mill units, located in the cities of Lappeenranta, Joutseno and Imatra, discharge their effluents into the lake (Fig. 2). Together they discharge about 300,000 m3/day biologically and 50,000 m3/day chemically treated effluent in the lake area (1997). During 1995 and 1998 all mills used ECF

bleaching processes and wastewater was treated by activated sludge plants. For the purpose of this study, the whole lake area was divided in three different areas (A, B and C) based on the dilution of the pulp mill effluents. For the electrofishing at stony shores and the vendace larvae survey, study area A was further subdivided into an area close to the mills (1–5 km) and a more distant area (5–15 km) from the mills (Fig. 2). Lake water quality characteristics at these areas during 1995–1997 are given in Table 1, the estimated effluent dilutions in the study areas were, respectively, area A, 0.5–4% effluent (polluted); area B,

Fig. 2. The study area at the southern Lake Saimaa with pulp and paper mills, areas A, B and C, sampling sites of gill net, electrofishing and beach seine surveys and water quality monitoring sites. Arrows indicate the direction of the water flow in the area.

Table 1 Lake water quality characteristics in the study areas at the southern Lake Saimaa during 1995–1997a Area

Na (mg/l) Estimated effluent % pH Oxygen (mg/l) Conductivity (mS/m) Secchi-dept (m) Colour mg (Pt/l) COD, Mn (mg/l) Chlorophyll (mg l 1) Total P (mg/l) Total N (mg/l) a

A

B

C

Polluted

Intermediate

Reference

5 – 10 0.1 – 0.5 7.00.1 12.40.2 5.80.2 3.60.1 34.63.9 6.80.1 3.50.2 7.10.7 415.96.8

3–5 <0.1 6.90.1 12.00.4 5.80.5 4.70.4 31.96.0 6.70.1 1.90.1 7.51.2 423.410.0

1–5 km

5–15 km

15 – 20 3.0 – 4.0 7.00.1 11.20.3 11.61.2 2.20.1 48.84.4 9.70.3 7.30.4 18.01.2 523.520.1

10 – 15 0.5 – 3.0 7.10.1 11.90.2 9.10.8 2.50.2 46.08.0 8.60.2 8.00.6 15.31.9 450.319.5

The water samples were collected during the spring overturn in May–June, when the water column was overturned. Data are means (  S.E.M.) of water column values from 1 m to near bottom (data from the Saimaa Water Protection Association Inc., Lappeenranta).

312

A.E. Karels, A. Niemi / Environmental Pollution 116 (2002) 309–317

0.1–0.5% (intermediate area); area C, << 0.1% (‘clean’ reference). The dilution of effluents in the lake water was estimated by means of concentrations of the inert effluent tracer sodium measured in spring and fall from 1990 to 1998 (Karels et al., 2000b).

shore was fished twice at a depth of 0–0.5 m. Surface area per sampling site varied from 90 to 240 m2. Captured fish were sorted by species, and densities of fish (numbers of fish/100 m2) were determined for each sampling site.

2.2. Fish surveys

2.2.3. Vendace larvae beach seine survey The objective of the vendace beach seine survey was to investigate the reproductive success of vendace in the study areas by investigating the abundance of newly hatched vendace larvae. Vendace is assumed to be a more sensitive species to effects of pulp mill effluents, because it is an autumn spawning planktivore species. Their eggs have an incubation period of 6 months on bottom sediment during the period of ice cover. Therefore, this phase of the reproduction of vendace is probably more sensitive to effects of sediment and water pollution. The hatching of vendace usually begins a short time before ice-off (April–May) and larvae appear soon thereafter in littoral and pelagic areas. Beach seine surveys on vendace larvae, conducted about 2–6 weeks after the ice-off, have been commonly used in Finnish inland waters for estimating the reproductive success of vendace (Viljanen, 1988; Karjalainen et al., 2000). In the present study, vendace larvae were sampled between 22 May and 11 June in 1995 (44 hauls), 1996 (43), 1997 (60) and 1998 (26). Hauls were conducted at 14 different sampling sites on stony shores at a depth of 0–2 m. (Fig. 2). Fish were captured with a shore seine (length 15.5 m, height 1.5 m). The mesh size of the seine cod end was ca. 0.5 mm and the wings, 3 mm. The upper rope of the seine net was floating on the water surface. The net was swept from offshore to onshore hereby sampling an area of approximately 200 m2. Samples of fish larvae were preserved in 70% alcohol. Differences between vendace and whitefish larvae were determined on the basis of the number of myomeres and the size of the larvae (Karjalainen et al., 1992). The relative abundance of vendace larvae in numbers per unit of effort (NPUE, number of larvae/ haul) was determined.

In the period from 1995 to 1998, we conducted a gill net survey in sublittoral and profundal waters, an electrofishing at stony shores and a beach seine survey on vendace larvae in the southern Lake Saimaa. 2.2.1. Gill net survey in sublittoral and profundal waters The objective of the gill net survey was to investigate the fish community in sublittoral and profundal waters with different grades of effluent dilution. In order to compare present results with the situation before major technical alterations at the pulp and paper mills, the sampling sites and gill net series were kept the same as in an earlier study in 1972–1973 (Sauvonsaari, 1974). Gill net samplings were carried out from June to September in 1995 and 1996 at 13 sampling sites (Fig. 2). At each site, gill net series were kept both at the bottom (benthic) and 1.5 m below the surface of the water column (pelagic). Our aim was to sample each site three times (three benthic and three pelagic) per summer. The gill net series consisted of 10 nets with mesh sizes of 12, 14, 17, 21, 25, 30, 35, 40, 50 and 75 mm. Each net had a length of 30 m and height of 1.8 m. Nets were set in the evening and emptied in the morning, so the fishing effort was 1 night (about 12 h). Altogether a total of 136 gill net series samplings (=1360 net efforts) were conducted. Captured fish were sorted by species, and number and total weight per species were determined. The relative abundance was calculated as mean number of fish per unit of effort (NPUE, numbers/net series/night) and biomass per unit of effort (BPUE, kg/net series/ night). 2.2.2. Electrofishing survey at stony shores The objective of the electrofishing survey was to study the fish fauna at stony shores in areas with different grades of pulp mill effluent dilution. The fish fauna at stony shores in the lake consists of small, relatively stationary fish species. The fish fauna at stony shores is totally different from the fish community in deeper waters, and can give important information on local effects of pollutants. The selected stony shores were all gently sloping with medium-sized stones of 10–20-cm diameter. A backpack electrofishing was conducted between 12 August and 25 September 1996 at 13 different sampling sites (Fig. 2). Fish were captured with a Ge Omega a/s type FA-2 electrofishing gear, using a voltage of 600 V and current of 1.2 A. The samplings were made during the day and no stop nets were used. Each

2.3. Statistics The majority of the catch data of the species caught in the gill net, the electrofishing and the beach seine survey did not meet the requirements for equality of variance and normality, even after log-transformation. Therefore, differences among areas and years were tested with the non-parametric Kruskal–Wallis test. The significance of all tests was set at P < 0.05. In the tables, mean and the standard deviation of the mean (S.D.) are given. The data were analyzed using SPSS1 computer software (Statistical Product Service Solutions, Chicago, IL, USA).

313

A.E. Karels, A. Niemi / Environmental Pollution 116 (2002) 309–317

3. Results 3.1. Gill net survey in sublittoral and profundal waters A total of 14 fish species, 14,810 individuals and 574.1 kg were caught at 13 sampling sites during the surveys in 1995 and 1996 (Table 2). Perch, roach, whitefish and smelt Osmerus eperlanus (L.) were encountered in both benthic and pelagic catches. Ruffe, pike-perch Stizostedion lucioperca (L.) and burbot Lota lota (L.) were predominantly encountered in the benthic catches, and vendace and bleak Alburnus alburnus (L.) in the pelagic catches. The relative abundance of fish in the study areas is given in Fig. 3 (NPUE, numbers/net series/night) and Table 3 (BPUE, kg/net series/night), respectively. The total NPUE and BPUE of fish at the polluted area A was significantly higher than the clean area C, area B was not significantly different from areas A and C. Within the study areas, the total NPUE and BPUE and the relative abundance of the most dominating species perch, roach, ruffe, bleak, pike-perch, vendace and whitefish were not significantly different between years. Perch and roach dominated the catches in the three study areas (>60% of the total catch). The relative abundance of perch, roach, bleak and ruffe was highest in the polluted area A and lowest in the clean area C. The relative abundance of pike-perch was significantly higher in area B, compared to the other areas. The relative abundance of vendace, whitefish and burbot was highest at areas B and C. Bream Abramis brama (L.), white bream Blicca bjoerkna (L.), pike Esox lucius

(L.), and ziege Pelecus cultratus (L.) were occasionally caught in area A; one landlocked salmon Salmo salar (L.) was caught in area C. 3.2. Electrofishing at stony shores A total of five species and 790 individual fish (1342 g) were caught at 13 sampling sites (Table 2). Mean

Fig.3. The relative abundance in numbers of fish per unit of effort (NPUE, numbers/net series/night) of benthic and pelagic gill net catches in the study areas in the southern Lake Saimaa in 1995 and 1996.

Table 2 Total catch in biomass (kg and g) and numbers of fish (n) of the gill net survey in profundal and sublittoral waters and electrofishing at stony shores in the southern Lake Saimaa during 1995 and 1996 Species

Total catch Gill net survey Mass (kg)

Roach (Rutilus rutilus L.) Perch (Perca fluviatilis L.) Bleak (Alburnus alburnus L.) Pike-perch (Stizostedion lucioperca L.) Whitefish (Coregonus lavaretus s.l. L.) Burbot (Lota lota L.) Ruffe (Gymnocephalus cernuus L.) Vendace (Coregonus albula L.) Bream (Abramis brama L.) Smelt (Osmerus eperlanus L.) White bream (Blicca bjoerkna L.) Ziege (Pelecus cultratus L.) Landlocked salmon (Salmo salar L.) Pike (Esox lucius L.) Bullhead (Cottus gobio L.) Stone loach (Noemacheilus barb. L.) Minnow (Phoxinus phoxinus L.)

188.14 163.28 60.32 47.27 21.86 13.61 12.81 10.57 5.32 2.50 1.19 0.69 0.30 0.19

Sum

574.1

Electrofishing Mass (g)

n 3472 6698 2128 266 185 60 1459 343 6 161 25 4 1 2

14,810

n

1

1

224 15

12 2

455 470 177

530 119 126

1342

791

314

A.E. Karels, A. Niemi / Environmental Pollution 116 (2002) 309–317

Table 3 Relative abundance of fish in biomass per unit of effort (BPUE, kg/net series/night) in benthic and pelagic gill net catches at the study areas in the southern Lake Saimaa during 1995 and 1996a Species/area

Biomass per unit of effort (kg/net series/night ) Benthic

Perch Roach Pike-perch Bleak Whitefish Smelt Ruffe Vendace Burbot Bream Silver bream Ziege Pike Landlocked salmon Total catch No. of net series a

Pelagic

A

B

C

A

B

C

3.282.31 A 1.912.63 A 0.380.47 A 0.311.53 A 0.110.22 A 0.020.70 AB 0.360.35 A 0 A 0 A 0.200.69 A 0.040.08 A 0 0.0070.04 A 0

1.481.20 B 0.571.03 B 1.211.23 B 0.090.04 B 0.230.35 A 0.030.05 B 0.080.06 B 0 A 0.330.47 B 0 A 0 B 0 0 A 0

1.38 1.41 B 0.29 0.30 B 0.43 0.63 A 0.0010.006 B 0.42 0.56 A 0.0080.02 A 0.07 0.07 B 0.01 0.04 A 0.32 0.60 B 0 A 0 B 0 0 A 0

0.991.45 A 2.532.58 A 0.040.14 A 1.572.70 A 0.040.07 A 0.020.03 A 0 A 0.030.003 A 0 0 0.0050.02 A 0.020.06 A 0 0 A

0.420.43 A 2.894.51 A 0.030.13 A 0.370.16 B 0.050.06 A 0.020.04 A 0 A 0.340.49 B 0 0 0 A 0.0060.03 A 0 0 A

0.250.26 A 0.641.16 B 0.0060.02 A 0.070.20 C 0.230.40 B 0.0020.006 B 0.0020.006 A 0.130.16 B 0 0 0 A 0 A 0 0.020.08 A

3.24 1.89 B 20

5.244.38 A 26

6.624.94 A 30

4.332.39 AB 22

4.135.03 AB 22

1.341.36 B 16

Values are the meanS.D.; numbers within species and depth, sharing an alphabetical letter were not significantly different (P>0.05).

densities of typical stony shore fish species (numbers/ 100 m2) in the study areas are given in Table 4. Most abundant species were bullhead Cottus gobio (L.), stone loach Noemacheilus barb. (L.) and minnow Phoxinus phoxinus (L.; Table 4). Total densities and densities of bullhead were significantly higher at the intermediate area, and lower at shores close to the mills (1–5 km). At the intermediate area, bullhead strongly dominated and consisted of more than 85% of the catch. Densities of stone loach, burbot and ruffe were not significantly different among areas. Minnow was not caught at shores close to the mills (1–5 km). Densities of minnow among the other areas were not significantly different.

3.3. Vendace larvae beach seine survey The relative abundance of vendace larvae (NPUE, ind./haul) in the study areas from 1995 to 1998 is given in Table 5. Within the study areas, the relative abundance was different between years. The highest abundance of vendace larvae was measured in 1995 and 1996. The relative abundance of vendace larvae at the polluted area in 1995 and 1996 was similar to the reference area C. In 1997, no vendace larvae were caught close to the mills (1–5 km) and in the intermediate area. In 1998, the relative abundance was similar among areas, and vendace larvae were also caught at sites close to the mills.

Table 4 Mean densities (numbers of fish/100 m2 S.D.) of fish species caught by electrofishing at stony shores at the study areas in the southern Lake Saimaa in 1996a Species

Area A

B

C

Polluted

Intermediate

Reference

127.937.0 C 2.384.12 A 0.790.73 A 0A 15.2212.05 B 146.341.7 C 3 275

13.473.87 7.224.97 0.560.64 0A 8.759.01 30.013.8 3 720

1–5 km Bullhead Stone loach Burbot Ruffe Minnow Sum Sites (n) Area (m2) a

2.013.13 2.652.54 0.100.18 0.140.24 0A 4.912.97 3 650

5–15 km A A A A A

11.104.51 AB 10.4814.59 A 1.14 0.92 A 0.25 0.50 A 3.29 6.57 AB 26.2516.84 B 4 638

Differences between areas (P<0.05) within species, are denoted by different alphabetical letters.

B A A B B

315

A.E. Karels, A. Niemi / Environmental Pollution 116 (2002) 309–317

Table 5 Relative abundance of vendace larvae in numbers of fish per unit of effort (NPUE, numbers/beach seine haul) at the study areas in the southern Lake Saimaa in spring 1995, 1996, 1997 and 1998a Year

1995 1996 1997 1998

Area A

B

C

Polluted

Intermediate

Reference

3.04.1 (10) B1 31.134.9 (7) A2 0 (10) A3 0 (4) A3

35.8 67.4 (20) A1 10.8 9.9 (20) A1 9.4 16.5 (24) B2 2.0 1.8 (4) A2

1–5 km

5–15 km

ns ns 0 (12) A1 2.6 9.3 (19) A2

15.4  15.6 (14) A1 16.8 18.9 (12) A1 1.5 1.7 (14) B2 ns

a The number of hauls per area is given in brackets. Values are the mean S.D.; numbers within a year sharing an alphabetical letter were not significantly different. Differences between years, within an area, are denoted by different superscript numbers (ns=not sampled).

Fish, however, were very patchily distributed as shown by the high standard deviations.

4. Discussion 4.1. Sublittoral and profundal water fish community It is evident that fish community and populations still vary in relation to effluent loading by modern pulp and paper mills. In the present study, the relative abundance of fish was highest in the most loaded area. Perch and roach dominated and constituted more than 60% of the gill net catches in the study areas. The abundance of bleak and ruffe was highest in the most effluent exposed area while the abundance of vendace and whitefish was highest in the intermediate and reference area. Results agree with those of an acoustic fish survey and a fishery enquiry at the southern Lake Saimaa (Sundell, 1998). The observed fish community structure in the most effluent exposed area in the lake corresponds with a moderate eutrophication by pulp and paper mill effluents (Hartmann, 1977, Sandstro¨m et al., 1991). In general, a moderate eutrophication is considered to favor both cyprinids and perch while salmonids disappear. A stronger eutrophication affects percids negatively, especially perch, while ruffe is more tolerant to eutrophication than perch. When eutrophication proceeds, the total abundance of all fish is drastically reduced (Hartmann, 1977). Similar to the present study, eutrophication effects on the fish community were shown earlier in coastal areas in the Baltic Sea and in Finnish inland waters affected by pulp mill effluents (Hansson, 1987; Neuman and Kara˚s, 1988; Kara˚s et al., 1991; Sandstro¨m et al., 1991; Hakkari, 1992; Landner et al., 1994). In general, catches were mostly dominated by cyprinids (mainly roach), perch and ruffe, while other species occurred in lowered densities. Compared with earlier fish community studies in the southern Lake Saimaa (Heinonen and Falck, 1971; Sauvonsaari, 1974), the share of weight of perch and

whitefish in benthic gill net catches in area A increased after the modernizations at the mills in the early 1990s, while the share of ruffe and burbot decreased, and the share of roach and pike-perch remained unchanged (Table 6). This change is indicative of a reduced eutrophication after the process alterations at the mills. At the same time, measurements of biochemical and physiological markers in feral and caged fish showed that the exposure of fish to pulp mill effluent compounds was decreased to almost reference levels after the modernizations at the mills (Oikari and Holmbom, 1996; Soimasuo, 1998; Karels et al., 1998, 2000a, 2000b, 2001). This was indicative of a lower toxicity of the present pulp mill effluents compared with before the modernizations at the mills. However, altered steroid hormone levels and gonad size in feral fish in the vicinity of the mills indicated possible alterations on fish reproduction (Karels et al., 1998, 2001). Growth and condition of perch and roach close to the mills were similar to the references, implying no alterations in the nutritional status of perch and roach (Karels et al., 1998, 2001). The size and age distribution of the perch population in the recipient of mill A, however, was shifted towards smaller and younger fish (Karels et al., 2001), and Table 6 The share (% of wt.) of fish species in benthic gill net catches in the polluted area A in the southern Lake Saimaa in 1966–1967 (Heinonen and Falck, 1971), 1972–1973 (Sauvonsaari, 1974) and 1995–1996 (present study) Species/year

1966–1967

1972–1973

1995–1996

Perch Roach Ruffe Pike-perch Bream Burbot Whitefish Vendace Others

4.9 13.0 55.3 11.5 9.5 2.3 0 1.3 2.2

16.0 29.1 31.9 3.3 9.3 3.9 0 0 6.5

49.5 28.8 5.7 3.5 5.6 0 1.7 0 15.2

316

A.E. Karels, A. Niemi / Environmental Pollution 116 (2002) 309–317

only year-classes born after the major process alterations at mill A in 1992 were caught. This could be indicative of possible recruitment failures of perch downstream of the mill before 1992, or possible migration of older perch to other areas in the lake. The size and age distribution of the roach population, on the contrary, was similar to the references, and recruitment failure of perch and roach living close to mill A and B was not observed after the process changes at the mills roach (Karels et al., 2001). However, recruitment failure in perch and other fish species has been earlier observed downstream of pulp mills without modern effluent treatment methods. Effects were associated with a high BOD load, resulting in low oxygen concentrations in the water, and toxic effects of the effluent (Kara˚s et al., 1991; Hakkari, 1992). In conclusion, the effects at the fish community level may be to a major extent explained by habitat changes related to eutrophication. The eutrophication of the lake has decreased after the modernizations at the mill, and the fish community in the effluent-exposed areas can be considered as typical for a moderate eutrophication by pulp mill effluents. However, important factors as the observed reproductive effects at lower levels of biological organization and the biological and physical structure of the receiving waters make clear cause–effect relationships not easy to establish. 4.2. Stony shore fish community Stony shores are one of the main shore types at the southern Lake Saimaa. Most abundant species caught in the present study were bullhead, stone loach and minnow (Table 4). Densities of bullhead were highest at the intermediate area, and lowest at shores close to the mills (1–5 km), while minnow was not caught at shores close to the mills (1–5 km). The species composition in the present study is comparable with results of an earlier study in clean areas at Lake Saimaa (Bagge and Hakkari, 1985). However, the absence of minnow and the lower densities of bullhead at shores close to the mills could be indicative of possible habitat changes due to eutrophication and/or toxic effects (e.g. recruitment failure) caused by the mill effluents. A higher predation pressure because of high densities of perch could also be an important factor. Similar studies at stony shores at Lake Pa¨ija¨nne showed also an absence of minnow in areas polluted by pulp mill effluents (Bagge and Hakkari, 1992). Minnow thus could be considered as a more sensitive species to pulp mill effluents, apparently still when ECF bleaching and activated sludge treatment processes are applied at the mills. 4.3. Vendace larvae occurence After spawning, vendace eggs are incubated during the winter period for 6 months on bottom sediment

under ice cover. Many eggs usually remain unfertilized and a large proportion of eggs die during the incubation period (Viljanen, 1988; Helminen et al., 1997; Karjalainen et al., 2000). The abundance of larval vendace is in general positively related to the spawning stock (Viljanen, 1988; Karjalainen et al., 2000). However, a large part of the larval mortality (65% on average) occurs during the first 3 weeks after hatching. Estimates of larval abundance after this period are a reasonably reliable indicator of the subsequent recruitment (Karjalainen et al. 2000). Results of the present study showed that the abundance of vendace larvae were different between years. These inter-annual differences of vendace larvae densities are typical for vendace in Finnish lakes (Viljanen, 1988; Hakkari and Bagge, 1992; Karjalainen et al., 2000). In the present study, vendace larvae were caught in all the three study areas, and in 1998, also close to the mills (1–5 km). This indicates that vendace is spawning in the polluted areas and that eggs are able to develop and survive on the bottom of the polluted area during the winter period. We hypothesized that a recovery of the vendace population in the polluted area due to the improvement of the water quality could be possible. However, a recovery can be hindered by the low spawning stock of vendace, and a high larval and juvenile mortality due to high predation pressure by dense populations of perch and other species in the effluent exposed area. A study after modernizations at a pulp mill at Central Lake Pa¨ija¨nne, e.g. showed that the reproductive success of vendace was still low in the vicinity of the mill, despite improvement of the water quality (Hakkari and Bagge, 1992). In an earlier study at the southern Lake Saimaa, before major process changes at the mills, Viljanen (1985) reported that vendace also spawned in the intermediate area and reference area. However, the number of eggs deposited on the bottom and densities of larvae were lower in the intermediate area than in the reference area. Polluted sites close to the mills were not sampled in 1985. It can be concluded that in the present study at the southern Lake Saimaa, vendace larvae were regularly caught in the polluted area, which is an emerging sign of recovery as expressed as tolerance of larvae to the water and sediment quality in this area.

Acknowledgements The authors wish to thank Professor Aimo Oikari, Professor Juha Karjalainen and an anonymous reviewer for their constructive comments on the manuscript, and Jussi Ja¨mse`n, Tuomo Manner, Ismo Piiroinen, Kari Porevirta and Petri Silvennoinen for their assistance in the trial fishings. The fishing co-operative of Kattelussaari (Kattelussaaren Kalastuskunta), and the fishing

A.E. Karels, A. Niemi / Environmental Pollution 116 (2002) 309–317

club of Kaukopa¨a¨ (Kaukopa¨a¨n Kalamiehet ry) are greatly thanked for the use of their accommodation facilities during the field work period.

References Bagge, P., Hakkari, L., 1985. Fish fauna of stony shores of Lake Saimaa (southeastern Finland) before and during the floods (1980– 1982). Aqua Fennica 15, 237–244. Bagge, P., Hakkari, L., 1992. Effects of paper mill effluents on the fish fauna of stony shores of Lake Pa¨ija¨nne. Hydrobiologia 243/244, 413–420. Bayne, B.L., Brown, D.A., Burns, K., Dixon, D.R., Ivanovici, A., Livingstone, D.A., Lowe, D.M., Moore, M.N., Stebbing, A.R.D., Widdings, J., 1985. The Effects of Stress and Pollution on Marine Animals. Praeger, New York, USA. Hakkari, L., 1992. Effects of pulp and paper mill effluents on fish populations in Finland. Finnish Fisheries Research 13, 93–106. Hakkari, L., Bagge, P., 1992. Reproductive succes of Coregonus species in areas loaded by effluents from paper mills. Hydrobiologia 243/244, 405–412. Hansson, S., 1987. Effects of pulp and paper mill effluents on coastal fish communities in the Gulf of Bothnia, Baltic Sea. Ambio 16, 344–348. Hartmann, J., 1977. Fischereiliche Vera¨nderungen in kulturbedingt eutrophierenden Seen. Schweiz. Z. Hydrol. 39 (2), 243–251. Heinonen, P., Falck, P., 1971. The effect of waste water on fish. Aqua Fennica 11, 105–110. Helminen, H., Sarvala, J., Karjalainen, J., 1997. Patterns in vendace recruitment in Lake Pyha¨ja¨rvi south-west Finland. Journal of Fish Biology 51 (Suppl. A), 303–316. Kara˚s, P., Neuman, E., Sandstro¨m, O., 1991. Effects of pulp mill effluents on the population dynamics of perch Perca fluviatilis L. Canadian Journal of Fisheries and Aquatic Science 48, 28–34. Karels, A.E., Soimasuo, M., Lappivaara, J., Leppa¨nen, H., Aaltonen, T., Mellanen, P., Oikari, A.O.J., 1998. Effects of ECF-bleached kraft mill effluent on reproductive steroids and liver MFO activity in populations of perch and roach. Ecotoxicology 7, 123–132. Karels, A.E., Oikari, A.O.J., 2000a. Effects of pulp and paper mill effluents on the reproductive and physiological status of perch (Perca fluviatilis L.) and roach (Rutilus rutilus L.) during the spawning period. Annales Zoologici Fennici 37, 65–77. Karels, A., Soimasuo, M., Suutari, R., Oikari, A., 2000b. Monitoring the recovery of a polluted lake with fish biomarkers: responses of whitefish Coregonus lavaretus (L.) experimentally exposed to pulp and paper mill effluents. Boreal Environment Research 5 (1), 53–65. Karels, A.E., Markkula, E., Oikari, A.O.J., 2001. Reproductive, bio-

317

chemical, physiological and population responses in perch (Perca fluviatilis L.) and roach (Rutilus rutilus L.) downstream of two ECF pulp and paper mills. Environmental Toxicology and Chemistry 20(7). Karjalainen, J., Helminen, H., Hirvonen, A., Sarvala, J., Viljanen, M., 1992. Identification of vendace Coregonus albula (L.) in Saimaa lake system (Finland). Archiv fu¨r Hydrobiology 125, 167–173. Karjalainen, J., Auvinen, H., Helminen, H., Marjoma¨ki, T., Niva, T., Sarvala, J., M. Viljanen, 2000. Unpredictability of fish recruitment: interannual variation in young-of-the-year abundance. Journal of Fish Biology 56, 837–857. Laine, P., Minkkinen, P., 1998. Effects of Chemical Forest Industry Effluents to the Water Quality of Southern Lake Saimaa. Publications of Karelian Institute, University of Joensuu, Finland, 122. Landner, L., Grahn, O., Ha¨rdig, J., Lehtinen, K.-J., Monfelt, C., Tana, J., 1994. A field study on environmental impacts at a bleached kraft pulp mill site on the Baltic Sea coast. Ecotoxicology and Environmental Safety 27, 128–157. Myllyvirta, T.P., Vuorinen, P.J., 1989. Avoidance of bleached kraft mill effluent by pre-exposed Coregonus albula L. Water Research 23, 1219–1227. Neuman, E., Kara˚s, P., 1988. Effects of pulp mill effluent on a Baltic coastal fish community. Water Science and Technology 20, 95–106. Oikari, A., Holmbom, B., 1996. Ecotoxicological effects of process changes implemented in a pulp and paper mill: a Nordic case study. In: Servos, M.R., Munkittrick, K.R., Carey, J.H., Van Der Kraak, G.J. (Eds.), Environmental Fate and Effects of Pulp and Paper Mill Effluents. St. Lucie press,Delray Beach, FL, pp. 613–625. Sandstro¨m, O., Neuman, E., Kara˚s, P., 1991. Pulp mill effluent effects on species distribution and recruitment in Baltic coastal fish. Finnish Fisheries Research 12, 101–110. Sauvonsaari, J., 1974. Etela¨-Saimaan Kalatalousselvitys (Report Oy Vesi Hydro Ab 97) Helsinki, Finland (in Finnish). Soimasuo, M.R., Karels, A.E., Leppa¨nen, H., Santti, R., Oikari, A.O.J., 1998. Biomarker responses in whitefish Coregenus lavaretus (L.) experimentally exposed in a large lake receiving effluents from pulp and paper industry. Archives of Environmental Contamination and Toxicology 34, 69–80. Sundell, P., 1998. Etela¨-Saimaan Kalasto ja Kalastus Vuonna 1996 (Research Report 53). University of Jyva¨skyla¨, Environmental Research Centre, Jyva¨skyla¨, Finland (in Finnish). Viljanen, M., 1985. Distribution of Vendace Coregonus albula (L.) and Whitefish Coregonus lavaretus (L.) Eggs and Larvae in Lake Etela¨Saimaa (Report 71). Publications of Karelian Institute, University of Joensuu, Joensuu, Finland. Viljanen, M., 1988. Population Dynamics of Vendace Coregonus albula (L.) in Finland (No. 12). Publications in Sciences, University of Joensuu, Joensuu, Finland.