Year-class strength, physical fitness and recruitment cycles in vendace (Coregonus albula)

Fisheries Research 173 (2016) 61–69

Contents lists available at ScienceDirect

Fisheries Research journal homepage: www.elsevier.com/locate/fishres

Year-class strength, physical fitness and recruitment cycles in vendace (Coregonus albula) Thomas Axenrot a,∗ , Erik Degerman b a Institute of Freshwater Research, Department of Aquatic Resources, Swedish University of Agricultural Sciences, Stångholmsvägen 2, 178 93 Drottningholm, Sweden b Institute of Freshwater Research, Department of Aquatic Resources, Swedish University of Agricultural Sciences, Pappersbruksallén 22, 702 15 Örebro, Sweden

a r t i c l e

i n f o

Article history: Available online 14 April 2015 Keywords: Vendace Recruitment cycles Year-class strength Physical fitness

a b s t r a c t Vendace, an obligate zoo-planktivore through all life stages, is a key-stone species in many large lakes in northern Eurasia. Vendace often produces strong year-classes in cycles of various length. To better understand the factors that determine the emergence of strong year-classes we analyzed long timeseries hydroacoustic data on abundance of young-of-the-year and older vendace, and representative samples of vendace from trawl catches measured for length, weight and age. Data were collected in the three largest lakes in Sweden – Lakes Vänern, Vättern and Mälaren – for 1995–2012, 1992–2012 and 2008–2012, respectively. The lakes range from ultra-oligotrophic to mesotrophic. In L. Vänern there is an extensive commercial fishery on vendace, whereas the fishery, at present, is small in L. Mälaren and negligible in L. Vättern. Size of mature vendace, expressed as 90% of Lmax , differed between the lakes and was positively correlated with lake nutrient levels. Strong year-classes did not occur in synchrony between lakes, not even between the two main basins within oligotrophic L. Vänern, and were less frequent in the ultra-oligotrophic and commercially unfished L. Vättern. Strong year-classes negatively affected physical fitness (Fulton’s condition factor) of older fish. During years of low physical fitness in sexually mature fish no strong year-classes would appear. Recovery of physical fitness might take several years and depended on food availability, expressed as nutrient levels or population density. Not until fish had regained individual physical fitness and could allocate energy to gonads, other factors – as physical environmental conditions – might become important for the emergence of a strong year-class. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Vendace (Coregonus albula) inhabits lakes in northern Eurasia and is also found in weak brackish water in the Baltic Sea, e.g. in the Gulfs of Finland and Bothnia (Järvi, 1950; Enderlein, 1981a). Unlike most other fish, vendace is an obligate zoo-planktivore through all life stages, which means that all age-groups of vendace compete for the same food resource (Enderlein, 1981b; Viljanen, 1986; Northcote and Hammar, 2006). Vendace is a key-stone species in many large lakes. It has been commercially fished – principally for the roe from the late 1960s – in the three largest lakes in Sweden, but nowadays mainly in Lake Vänern (310 tonnes of vendace landed in 2012). Vendace is also an important prey fish

∗ Corresponding author. Tel.: +46 10 4784213; fax: +46 10 4784269. E-mail addresses: [email protected] (T. Axenrot), [email protected] (E. Degerman). http://dx.doi.org/10.1016/j.fishres.2015.03.017 0165-7836/© 2015 Elsevier B.V. All rights reserved.

for, e.g. pikeperch (L. Vänern and Mälaren), salmon (L. Vänern and Vättern) and Arctic char (L. Vättern; e.g. Nilsson, 1979), all important species in the inland commercial and recreational fisheries. Vendace recruitment has been described as occurring in cycles with strong year-classes followed by often several years of weak recruitment. This pattern has been described as caused by intraspecific, density-dependent competition (Aas, 1972; Hamrin and Persson, 1986; Marjomäki and Huolila, 2001), physical factors (Nyberg et al., 2001; Marjomäki et al., 2004; Sandström et al., 2014) or both (Helminen and Sarvala, 1994). Marjomäki et al. (2004) reported a spatial synchrony between Finnish lakes in inter-annual population variation. It has also been suggested that the pattern might be less obvious or changed as a result of predation pressure, inter-specific competition and fishing mortality (Karjalainen et al., 2000; Auvinen et al., 2004). Salojärvi (1987) proposed that vendace populations are regulated by fecundity, starvation of larvae and predation, while Haakana and Huuskonen (2009) found

62

T. Axenrot, E. Degerman / Fisheries Research 173 (2016) 61–69

Fig. 1. Lake Vänern (Sweden). Hydroacoustic transects in the two basins Dalbosjön and Värmlandssjön. Subareas are separated by a dotted line. Representative and depth stratified trawling was performed in all subareas.

that larval starvation and mortality was not much affected by food availability. This study used individual physical fitness (Fulton’s condition factor) and year-class strength in vendace together with abundance estimations from hydroacoustic surveys from three nutrient contrasting lakes to explore how these data fitted the pre-existing theories on recruitment cycles in vendace. In two of the lakes practically no fishing for vendace is carried out. Our hypothesis was that the recruitment cycles were primarily determined by intra-specific competition affecting the physical fitness of potential spawners, and secondarily mediated by physical environmental factors. 2. Material and methods 2.1. Lake Vänern

of total phosphorus in the offshore regions has stabilized since the mid-90s around historical reference levels (4.5–5.5 ␮g l−1 ) indicating oligotrophic conditions. Total nitrogen level is still heightened because of emissions from farming, mainly originating from rivers in the southern parts (Sonesten, 2013). L. Vänern is species-rich with 36 different species of fish (Degerman et al., 2001). The most important species in the commercial fishery are vendace and pikeperch (Sander lucioperca) although another 5–6 species are also targeted in the fishery. Salmon (Salmo salar) and trout (Salmo trutta) are the most important species in the recreational fishery. The most common pelagic species in numbers is smelt (Osmerus eperlanus). Results from L. Vänern are given for the two main basins (Dalbosjön and Värmlandssjön; Fig. 1) separately. 2.2. Lake Vättern

Lake Vänern is the largest lake in Sweden and in the European Union, and the third largest lake in Europe (Fig. 1, Table 1). The level Table 1 General information on lakes Vänern, Vättern and Mälaren, the largest lakes in Sweden. Lake

Latitude (outlet)

Longitude (outlet)

Area (km2 )

Depth (average, m)

Altitude (m, a.s.l.)

Vänern Vättern Mälaren

N 58.80 N 58.30 N 59.50

E 13.40 E 14.50 E 16.90

5648 1893 1096

27 40 13

44 88 1

Lake Vättern, 1893 km2 , is the second largest lake in Sweden (Fig. 2, Table 1). Recovering from eutrophication in the 1960s and 1970s, the total phosphorus level is now close to historical reference levels (3–5 ␮g l−1 ; Renberg et al., 2003) and the lake is at present regarded as ultra-oligotrophic. L. Vättern holds over thirty different species of fish (Degerman et al., 2001). The most important species for the commercial and recreational fishery are whitefish (Coregonus maraena), Arctic char (Salvelinus alpinus) and two introduced species – salmon and signal crayfish (Pasifastacus leniusculus). The fishery on vendace has been negligible since the 1980s because of low abundance. The most dominant pelagic species

T. Axenrot, E. Degerman / Fisheries Research 173 (2016) 61–69

63

in numbers are smelt and three-spined stickleback (Gasterosteus aculeatus). 2.3. Lake Mälaren Lake Mälaren, the third largest lake in Sweden (Table 1), is a heterogeneous lake with several discrete basins that have different physical properties (Fig. 3). All in all, the lake is described as eutrophic with a total phosphorus level over 10 ␮g l−1 in all parts of the lake. There is, however, a gradient of decreasing eutrophication as well as increasing maximum depths from west to east. Some of the western basins are eutrophic (>20 ␮g l−1 total phosphorous). L. Mälaren holds 36 species of fish (Degerman et al., 2001), and pelagic species are dominated in numbers by smelt. The most important species in the commercial fishery are pike-perch, eel (Anguilla anguilla) and pike (Esox lucius). Vendace used to be an important species in the fishery until the population crashed in the early 1990s (Nyberg et al., 2001). Examination of the age-structure in the population at the time of the crash revealed a lack of young year-classes (Nyberg et al., 2001). 2.4. Hydroacoustics and midwater trawling

Fig. 2. Lake Vättern (Sweden). Hydroacoustic transects and representative biological sampling subareas. Depth stratified trawling was performed in all subareas.

Long-term time-series data from hydroacoustic surveys in late August and in September, supported by midwater trawling, were used to detect trends in the recruitment of vendace in Lakes Vänern, Vättern and Mälaren. In L. Vänern survey data were available for 1995–2010 and in L. Vättern for 1992–2010. In L. Mälaren data for 2004 and 2008–2012 from three basins were used. For data processing and assigning trawl data to the hydroacoustics, the lakes were divided into subareas determined from natural boundaries and hydrographic conditions (Figs. 1–3). The hydroacoustic surveys were performed at night, in darkness, with an index of coverage (Aglen, 1983) ranging between 4 and 5 in all lakes (Figs. 1–3). The hydroacoustic data were collected using hull mounted Simrad transducers with echo sounders EY-M (1992–1995), EY 500 (1996–2005) and EK60 (2006–2012). From 1992 through 2005 a 70 kHz transducer was used (Simrad ES 70-11) and from 2006 a 120 kHz transducer (Simrad ES 120

Fig. 3. Lake Mälaren (Sweden). Depth stratified trawling was performed in the basins (subareas) where hydroacoustic data was collected.

64

T. Axenrot, E. Degerman / Fisheries Research 173 (2016) 61–69

200

350 300 Mean

Length (mm)

Length (mm)

150 Max

100

Min Lmax

50

250 200

Mean Min

150

Max

100 0 0

1

2

3

4

5 6 7 Year-class

8

9

10

11

Lmax

50

12

0 0

Fig. 4. Length at age for vendace in L. Vättern. Mean values are given with standard deviation. Age determined from otoliths on fish sampled 1992–2010. Lmax calculated according to von Bertalanffy.

a

1

2

3

6

7

8

9

Fig. 5. Length at age for vendace in L. Mälaren. Mean values are given with standard deviation. Age determined from otoliths on fish sampled 2008–2010. Lmax calculated according to von Bertalanffy.

1400

0+ >0+

1200 Density (numbers per hectare)

4 5 Year-class

Period mean >0+ Period mean 0+

1000 800 600 400 200 0 1995

b

1997

1999

2001

2003

2005

2007

2009

2011

1400 0+

Density, numbers per hectare

1200

>0+ Period mean >0+

1000

Period mean 0+ 800 600 400 200 0 1995

1997

1999

2001

2003

2005

2007

2009

2011

Fig. 6. Vendace abundance in Lake Vänern basins: (a) Dalbosjön and (b) Värmlandssjön 1995–2012, yearling-and-older (>0+) and young-of-the-year (0+).

T. Axenrot, E. Degerman / Fisheries Research 173 (2016) 61–69

7C). The echo sounders were calibrated according to recommendations by Foote et al. (1987) and the manufacturer. The technical settings for the hydroacoustics have varied over the years due to the technical development of hydroacoustic equipment resulting in substitution of old echo sounders and transducers. The threshold for target strength of single echo detections was set low, usually from −60 dB, to include small, young-of-the-year fish. Midwater trawling was used to assign species composition and size distribution to the hydroacoustic data. Trawling speed was 3 knots. The trawl hauls through the years have lasted for 20–30 min in lakes Vänern and Vättern. In L. Mälaren the trawl hauls have lasted for 10 min because of high densities of fish. The trawl mouth opened 5 m × 12 m (height × width). The codend was partitioned into two bags with mesh-size 5 and 7 mm allowing retention of juvenile fish. The trawl hauls were performed to be representative for lake subareas and depths, i.e. during the same night as the geographically corresponding hydroacoustic data were collected (Figs. 1–3). At the time of the surveys (August and September), all three lakes are thermally stratified which has importance for the spatial structure of the fish community. Consequently, the trawl hauls in each subarea were performed in three depth strata; the shallow part of the water column (5–10 m), around the thermocline (usually 10–20 m) and below the thermocline (>20 m). For subareas with bottom depths down to approximately 50 m, one trawl haul per depth stratum was performed (i.e. usually three trawl hauls per subarea). For subareas with bottom depths over approximately 50 m the deepest stratum (>20 m), representing the predominant water volume, was generally covered with 1–3 more trawl hauls depending on the size of the subarea. Trawl catches were identified to species and individually measured for total length. If the number of individuals of a certain species exceeded 500, a random subsample of 200 juveniles and 300 adults were measured per each trawl haul. The total catch of each species in each trawl haul was weighed. The trawl catch composition, i.e. species, size proportions and – for target species like vendace – the proportion of young-ofthe-year (0+) and older (>0+) were assigned to the hydroacoustic densities for corresponding depth strata for each subarea. In this process trawl data from the different strata in a subarea were partitioned into classes (species, individual fish lengths, trawl geo-position, fishing depth, and bottom depth). Fish lengths (total

65

length) were transformed to target strengths based on Love’s equation (1971). Species and size-groups within species from the trawl data were matched with densities for corresponding size-groups in the hydroacoustic data, extracted in intervals (elementary distance sampling unit determined to avoid autocorrelation) and layers (5 m). The final density for one species or size-group within a species for a discrete subarea was calculated from the mean of intervals and sum of layers. For early years in the time-series the procedure was principally the same regarding subareas, depth stratification, species and size-groups, but the hydroacoustic data was extracted in 2 m layers for full transects and subarea final densities were calculated from weighted (by distance) means of transects. Whole lake and lake-basin densities of vendace – adults as well as juveniles – were obtained by calculating acoustic coverage weighted mean densities for the different subareas. Classification of vendace young-of-the-year juveniles was based on annual size-distributions in the trawl catches (Salonen, 2004) supported by age determination from scales and otoliths. Thus, lake specific size thresholds were used to distinguish between 0+ and >0+ fish. The thresholds varied slightly between years due to annual differences in growth. A year-class was considered strong if the density of young-of-the-year vendace was over the mean value for the whole time period for each basin/lake, respectively. 2.5. Age determination and physical fitness Otoliths for age determination were collected from subsamples of the catch from each lake and subarea (70 individuals per subarea). These 70 individuals were chosen to represent the size distribution of the sample, from small to large individuals. More fish were selected from numerous size-groups. These fish were preserved frozen for later preparation in the laboratory where they also were measured for length and weight to determine length and weight at age and length–weight relationships. Age determined fish were available from Lakes Vättern (1992–2010) and Mälaren (2008–2010). Physical fitness was expressed as the Fulton condition factor (Nash et al., 2006). Lmax according to the von Bertalanffy growth function was estimated using a Ford-Walford plot (Sparre and Venema, 1998).

700 0+

Density (numbers per hectare)

600

>0+ Period mean >0+

500

Period mean 0+ 400 300 200 100

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1988

0

Fig. 7. Vendace abundance in Lake Vättern, yearling-and-older (>0+) and young-of-the-year (0+) 1988–2012. No data was collected in 1989 and 1991.

66

T. Axenrot, E. Degerman / Fisheries Research 173 (2016) 61–69

Table 2 Age determined vendace from trawl catches over the period 1992–2010 in L. Vättern. Strong year-classes that dominated the vendace population for long periods of time are marked in grey.

Yearclass 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 n

Sampling year 199 199 199 2 3 4 2 2 1 31 1 21 6 23

6

199 5

6 2 20 9 106

199 6

199 7

199 8

199 9

200 0

200 1

200 2

200 3

200 4

200 5

200 6

200 7

200 8

85

143

201 0

n

1 1 7

4 0 1 37 2 2 43 15 434 0 0 48 20 19 10 3 188 2 81 52 156 5 56 0 21 9 7

1 2 80

56

27

34

43

25

22

10

2

2

3

11 3 8

12 5 1 2

6 2

3

10 3 2 2

5 2 1

2 1

43

29

39

25

21

1

20

18 22 24

13 5 44 4

2 2

5 3 1 5

1

14 2 13 17 50 0 51

12 13 5 23 1 2

1

2 3 15 3

2

8

200 9

82

62

49

54

55

42

3. Results Data on individual length at a certain age from lakes Vättern and Mälaren showed that young-of-the-year vendace could be identified and separated from older vendace by length (Figs. 4 and 5). At the time of the year when the hydroacoustic surveys were performed, yearling-and-older vendace were almost exclusively found

82

48

59

94

87

148

56

18 8 3

49

in the colder water below the thermocline. In general, vendace had reached 90% of Lmax length at the age of three. Lmax was estimated to be 168 mm in L. Vättern and 249 mm in L. Mälaren (Figs. 4 and 5). In L. Vänern the hydroacoustic results showed strong yearclasses in Dalbosjön in years 2000, 2003, 2004, 2008 and 2011, and in Värmlandssjön in 1995, 1996, 2004, 2005 and 2008 (Fig. 6). In L. Vättern strong year-classes appeared in the hydroacoustic results

0+ 2400

Numbers per hectare

120

>0+

2100

Period mean >0+

1800

100

Period mean 0+ 80

1500 1200

60

900 40 600 20 300 0

0 2004

2005

2006

2007

9

2008

2009

2010

2011

2012

Fig. 8. Vendace abundance in Lake Mälaren 2004–2012, yearling-and-older (>0+) and young-of-the-year (0+). In 2005–2007 no data was collected.

T. Axenrot, E. Degerman / Fisheries Research 173 (2016) 61–69 Table 3 Age determined vendace from trawl catches over the period 2008–2010 in L. Mälaren. From these sampled years a strong year-class was observed in 2008 (marked in grey).

Sampling year Year-class

2008

2009

2010

Table 4 Occurrence of strong (s) and weak (w) year-classes of vendace in three Swedish lakes (two basins in L. Vänern presented). nd indicates years when no data is available. Year-class strength was determined from hydroacoustic and trawls data for L. Vänern, hydroacoustic and trawl data with age reading (otoliths) for L. Vättern, and age reading (otoliths) for L. Mälaren. Strong and weak were decided from the mean for available years per lake.

n

Year 1999

3

3

2000

3

3

2001

nd

5

1993 nd

nd

w

nd

7

1994 nd

nd

w

nd

5

1995 s

w

w

nd

1996 s

w

w

nd

1997 w

w

w

nd

1998 w

s

w

nd

1999 w

w

w

w

2000 w

s

s

w

2001 w

w

w

w

2002 w

w

w

s

2003 w

s

w

w

2004 s

s

s

w

2005 s

w

w

w

2006 w

w

w

s

2007 w

w

w

w

2008 s

s

w

s

2009 w

w

w

w

2010 w

w

w

w

2011 w

s

w

s

2012 w

w

w

w

1 2

5

2004

4

1

2005

2

4

2

8

2006

6

11

8

25

2007

5

2

5

12

2008

12

31

26

69

5

8

13

10

10

43

55

Värmlandssjön Dalbosjön s

2003

n

Vättern Mälaren

nd

3

3

2010

Vänern

1992 nd

3

2002

2009

1

67

65

in 1992, 2000 and 2004 (Fig. 7). These years with strong year-classes were confirmed by the age determination of fish samples over the same period of time (Table 2). In L. Mälaren the hydroacoustic data showed strong year-classes in 2004 and 2011 (Fig. 8). The available results from age determination of vendace from L. Mälaren for 2008–2011 confirmed a strong year-class for 2011 but also showed a strong year-class in 2008 (Table 3). In the hydroacoustic data the year-class of 2008 was slightly stronger than other years but not a strong year-class according to the definition used in this study. In L. Vänern basins, Dalbosjön and Värmlandssjön, the mean abundance (numbers per hectare) for young-of-the-year vendace 1995–2012 were 95 and 246, respectively. For L. Vättern the corresponding mean abundance was 61 (1988–2012) and for L. Mälaren 39 (results from the basin Prästfjärden 2004, 2008–2012). In conclusion, strong year-classes did not occur in synchrony between the lakes or even the basins in Lake Vänern (Table 4). After a strong recruitment event, adult fish showed low physical fitness (indicated by Fulton’s index) for a number of years while the young recruits tended to follow a regular length/weight relationship (example from L. Vättern; Fig. 9). The lower physical fitness was not specific for any particular part of the lake (L. Vättern), but could be observed in all areas (Fig. 10) indicating synchrony in strong year-classes within the lake. The long time-series in L. Vättern showed that the adult vendace stock regained physical fitness (Fulton’s index) after 3–4 years after a strong recruitment event (Fig. 11). However, regained physical fitness did not automatically result in a strong year-class the following year. The mean physical fitness index for years when the stock had regained weight was 0.65 in L. Vättern and 0.85 for the single available year in L. Mälaren (2008).

4. Discussion In southern Scandinavia, where lakes Vänern, Vättern and Mälaren are situated, vendace generally had reached 90% of Lmax at the age of three, indicating they had reached adult length and were sexually mature at this age. According to Aas (1972) and Sandlund et al. (1991), vendace in this geographic region reaches maturity at age 2+. Sandlund et al. (1991) also reported that growth ceased at maturation. This was in agreement with our results on length

68

T. Axenrot, E. Degerman / Fisheries Research 173 (2016) 61–69 45

50 Middle

40

North 35

South

1994

30

1995

25

30

1996 20

1997

15

1998

Weight (g)

Weight (g)

40

1993

20

1999 10

10

5 0 50

70

90

110

130

150

170

190

210

0 50

Length (mm)

Fulton´s condition factor

at a specific age. Further, our data from L. Vättern showed that young vendace (age 0+ to 2+) were not greatly affected by the emergence of a strong year-class, but followed the normal growth of vendace (weight/length; Fig. 9). Possible explanations to this could be that young vendace had competitive advantages by lower metabolic requirements, the small size of prey items (i.e. zooplankton; Enderlein, 1981a; Hamrin and Persson, 1986), and that they did not allocate energy on gonads or spawning (Koops et al., 2004). The nutrient status and predation pressure (including fishing) differed between the studied lakes, which might affect growth rate, adult length and age at maturity. The relatively small adult length in L. Vättern, as indicated by Lmax , could be explained by the ultraoligotrophic nutrient conditions in the lake, when compared with the larger adult length of vendace that were caught in mesotrophic parts of L. Mälaren. The larger adult size in L. Mälaren was comparable with vendace from mesotrophic polish lakes (Czerniejewski and Rybczyk, 2008; Czerniejewski and Wawrzyniak, 2008). Also physical fitness (Fulton’s condition factor) was considerably lower in L. Vättern than in L. Mälaren. Czerniejewski and Wawrzyniak (2008)

90

110

130 150 Length (mm)

170

190

210

Fig. 10. Low weight-at-length appeared in older vendace (>160 mm, >2+; Fig. 4) after recruitment events that had produced strong year-classes. These were found in all areas of the lake and consequently low weight-at-length could not be explained by regional differences in food availability. Data from trawl catches of vendace in L. Vättern 1993–2005.

found that fecundity was mainly related to length, and according to Gregersen et al. (2011) females could show great plasticity in their allocation of reproductive investment due to environmental conditions. Hence, factors affecting the food resource might influence the time needed for a female to reach maturity or regain physical fitness after spawning (Koops et al., 2004). Consequently, in lakes with low food availability one would expect longer cycles in the generation of strong year-classes than in lakes with high food availability. However, food availability can be affected by nutrient conditions as well as low abundance of vendace due to predation pressure (including fishing). Contrary to the findings of Marjomäki et al. (2004) in Finnish lakes, there was no obvious synchrony in the incidence of strong year-classes between our studied lakes or basins (Table 4). The lack of synchrony in strong year-classes indicated that factors within a lake, or even differences between lake basins as in the case with L. Vänern, were more important for the generation of strong yearclasses than large-scale physical environmental factors as climate. However, we argue that there was a combination of internal and

0.8

1000

0.7

800

0.6

600

0.5

400

0.4

200

0.3

0

Density 0+ per hectare

Fig. 9. Weight-at-length for vendace from trawl catches in L. Vättern 1993–1999. After the strong year-class in 1992 (Fig. 2) sexually mature vendace that presumably had participated in that recruitment event showed low weight relative to length for a number of years. The young recruits (age 1+ in trawl catches in 1993 and 2+ in 1994) followed the regular weight-at-length relationship.

70

0+ 1-2+ >2+

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

Fig. 11. Density of 0+ vendace (bar) and Fulton’s condition factor for two age groups of vendace – 1–2+ (circle) and >2+ (dash). For the latter two mean values are given with standard deviation. Strong recruitment was observed in 1992, 2000 and 2004, i.e. after regained physical fitness of potential spawners. The dotted line indicates the mean value after regained physical fitness. Data from trawl catches of vendace from L. Vättern 1992–2010 (in 2010 no vendace >2+ was caught).

T. Axenrot, E. Degerman / Fisheries Research 173 (2016) 61–69

external factors involved. When a strong year-class emerged it negatively affected physical fitness of older mature vendace as well as the potential of regaining physical fitness for the fish that took part in the previous spawning event. The negative effect might last for several years depending on the food availability in the lake. Through years of low physical fitness in older fish (i.e. sexually mature potential spawners) no strong year-classes would appear. After the potential spawners had regained physical fitness, or young fish had reached maturity, strong year-classes might appear but could still be lacking as an effect of physical environmental factors affecting recruitment success, such as climate, temperature, and Julian day of ice-break (Nyberg et al., 2001; Sandström et al., 2014). The data on physical fitness in mature fish (>2+) from L. Vättern indicated that in an ultra-oligotrophic lake with relatively low predation pressure the time required for vendace to regain physical fitness was 3–4 years. This time period practically coincides with the time needed to grow and become sexually mature for vendace in these parts of Scandinavia (Aas, 1972; Sandlund et al., 1991). In the case of L. Vättern such a scenario would significantly increase the number of potential spawners a given year and thus enhance recruitment success, provided that the physical environmental factors were beneficial. High fishing pressure would decrease food competition and shorten the time to regain physical fitness. In heavily exploited stocks of vendace strong yearclasses occur frequently, often every second year (Karjalainen et al., 2000). Consequently, in these cases recruitment success would mainly depend on environmental factors that could be expected to be similar in a geographically limited area. This might explain the spatial synchrony of strong year-classes between lakes found by Marjomäki et al. (2004). These explanations support the hypothesis that the observed recruitment cycles in vendace populations were primarily determined by lake internal factors affecting food availability. Secondarily, external conditions like physical environmental factors will affect the recruitment success a given year. Acknowledgements We are grateful to people who have initiated and participated on the surveys and the processing of material and data over the years, especially Olof Enderlein, Eva Bergstrand, Per Nyberg, Anders Asp, Malin Hällbom, Tanja Martins, Alfred Sandström and Henrik Ragnarsson Stabo. We are also grateful for funding through the years by the Swedish Board of Fisheries, the Swedish Agency for Marine and Water Management and the Lakes Vänern and Vättern Societies for Water Conservation. References Aas, P., 1972. Age determination and year-class fluctuations of cisco, Coregonus albula L., in the Mjösa hydroelectric reservoir, Norway. Rep. Inst. Freshw. Res. Drottningholm 52, 5–22. Aglen, A., 1983. Random errors of acoustic fish abundance estimates in relation to the survey grid density applied. FAO Fish. Rep., 293–298. Auvinen, H., Kolari, I., Pesonen, A., Jurvelius, J., 2004. Mortality of 0+ vendace (Coregonus albula) caused by predation and trawling. Ann. Zool. Fenn. 41 (1), 339–350. Czerniejewski, P., Rybczyk, A., 2008. Variations in age and length growth rates of vendace, Coregonus albula (L.) from selected lakes in western Pomerania. Arch. Pol. Fish. 16 (1), 63–74.

69

Czerniejewski, P., Wawrzyniak, W., 2008. Fecundity of vendace, Coregonus albula (L.), from several lakes in Western Pomerania. Arch. Pol. Fish. 16 (2), 135–146. Degerman, E., Hammar, J., Nyberg, P., Svärdson, G., 2001. Human impact on fish diversity in the four largest lakes of Sweden. Ambio 30 (8), 522–528. Enderlein, O., 1981a. Interspecific food competition between the three pelagic zooplanktonfeeders, cisco (Coregonus albula (L.)), smelt (Osmerus eperlanus (L.)) and herring (Clupea harengus L.) in the Norrbotten part of the Bothnian Bay. Rep. Inst. Freshw. Res. Drottningholm 59, 15–20. Enderlein, O., 1981b. When, where, what and how much does the adult cisco, Coregonus albula (L.) eat in the Bothnian Bay during the ice-free season. Rep. Inst. Freshw. Res. Drottningholm 59, 21–32. Foote, K.G., Knudsen, H.P., Vestnes, G., MacLennan, D.N., Simmonds, E.J., 1987. Calibration of acoustic instruments for fish density estimation: a practical guide. ICES Coop. Res. Rep. 144, 69. Gregersen, F., Völlestad, L.A., Östbye, K., Aass, P., Hegge, O., 2011. Temperature and food-level effects on reproductive investment and egg mass in vendace, Coregonus albula. Fish. Manage. Ecol. 18 (4), 263–269. Haakana, H., Huuskonen, H., 2009. Predation of smelt on vendace larvae: experimental and field studies. Ecol. Freshw. Fish 18, 226–233. Hamrin, S.F., Persson, L., 1986. Asymmetrical competition between age classes as a factor causing population oscillations in an obligate planktivorous fish species. Oikos 47, 223–232. Helminen, H., Sarvala, J., 1994. Population regulation of vendace (Coregonus albula) in Lake Pyhäjärvi, southwest Finland. J. Fish Biol. 45, 387–400. Järvi, T.H., 1950. Die Kleinmaränenbestände in ihren Beziehungen zu der Umwelt (Coregonus albula L.). Acta Zool. Fenn. 61, 1–116. Karjalainen, J., Auvinen, H., Helminen, H., Marjomäki, T.J., Niva, T., Sarvala, J., Viljanen, M., 2000. Unpredictability of fish recruitment: interannual variation in youngof-the-year abundance. J. Fish Biol. 56 (4), 837–857. Koops, M.A., Hutchings, J.A., McIntyre, T.M., 2004. Testing hypotheses about fecundity, body size and maternal condition in fishes. Fish Fish. 5, 120–130. Love, R.H., 1971. Dorsal aspect target strength of an individual fish. J. Acoust. Soc. Am. 49, 816–823. Marjomäki, T.J., Huolila, M., 2001. Long-term dynamics of pelagic fish and vendace (Coregonus albula (L.)) stocks in four zones of a lake differing in trawling intensity. Ecol. Freshw. Fish 10 (2), 65–74. Marjomäki, T.J., Auvinen, H., Helminen, H., Huusko, A., Sarvala, J., Valkeajärvi, P., Viljanen, M., Karjalainen, J., 2004. Spatial synchrony in the inter-annual population variation of vendace (Coregonus albula (L.)) in Finnish lakes. Ann. Zool. Fenn. 41, 225–240. Nash, R.D.M., Valencia, A.H., Geffen, A.J., 2006. The origin of Fulton’s condition factor – setting the record straight. Fisheries 31 (5), 236–238. Nilsson, N.A., 1979. Food and habitat of the fish community of the offshore region of Lake Vänern, Sweden. Rep. Inst. Freshw. Res. Drottningholm 58, 126–139. Northcote, T.G., Hammar, J., 2006. Feeding ecology of Coregonus albula and Osmerus eperlanus in the limnetic waters of Lake Mälaren, Sweden. Boreal Environ. Res. 11, 229–246. Nyberg, P., Bergstrand, E., Degerman, E., Enderlein, O., 2001. Recruitment of pelagic fish in an unstable climate; studies in Sweden’s four largest lakes. Ambio 30 (8), 559–564. Renberg, I., Bindler, R., Bradshaw, E., Emteryd, O., Englund, J., Leavitt, P., 2003. Paleolimnologiska undersökningar i Vättern och Vänern. Rapport nr 75, Vätternvårdsförbundet. Länsstyrelsen i Jönköping, Jönköping, pp. 1–28. Salojärvi, K., 1987. Why do vendace (Coregonus albula L.) populations fluctuate? Aqua Fenn. 17 (1), 17–26. Salonen, E., 2004. Estimation of vendace year-class strength with different methods in the subarctic lake Inari. Ann. Zool. Fenn. 41, 249–254. Sandlund, O.T., Jonsson, B., Naesje, T.F., Aass, P., 1991. Year-class fluctuations in vendace, Coregonus albula (Linnaeus): who’s got the upper hand in intraspecific competition? J. Fish Biol. 38, 873–885. Sandström, A., Ragnarsson-Stabo, H., Axenrot, T., Bergstrand, E., 2014. Has climate variability driven the trends and dynamics in recruitment of pelagic fish species in Swedish Lakes Vänern and Vättern in recent decades? Aquat. Ecosyst. Health Manage. 17 (4), 349–356. Sonesten, L., 2013. Vattenkvaliteten i Storvänern. In: Peilot, S. (Ed.), Vänern - årsskrift 2013. Rapport nr 77, Vänerns vattenvårdsförbund. E-print AB, Stockholm, pp. 28–31. Sparre, P., Venema, S.C., 1998. Introduction to tropical fish stock assessment. Part I: Manual. FAO Fisheries Technical Paper. 306/1. Rev. 2. Viljanen, M., 1986. Biology, propagation, exploitation and management of Vendace (Coregonus albula L.) in Finland. Arch. Hydrobiol. Beih. Ergebn. Limnol. 22, 73–97.