High stock density impairs growth, female condition and fecundity, but not quality of early reproductive stages in vendace (Coregonus albula)

Fisheries Research 186 (2017) 159–167

Contents lists available at ScienceDirect

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

High stock density impairs growth, female condition and fecundity, but not quality of early reproductive stages in vendace (Coregonus albula) Thomas Wanke a,∗ , Uwe Brämick a , Thomas Mehner b a b

Institute of Inland Fisheries, Im Königswald 2, 14471 Potsdam, Germany Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Müggelseedamm 310, 12587 Berlin, Germany

a r t i c l e

i n f o

Article history: Received 22 March 2016 Received in revised form 24 August 2016 Accepted 27 August 2016 Handled by George A. Rose Keywords: Maternal effects Spawning stock biomass Egg size Egg quantity Hatching success Larval size

a b s t r a c t In fisheries science, stock density is one of the fundamental factors that affect growth and condition. However, reproductive traits like fecundity and also the size and quality of offspring (via maternal effects) can be modified by stock density. In the short-living freshwater fish species vendace (Coregonus albula), the impact of stock density via maternal condition on quantity and quality of eggs and larvae was often proposed as a driver of population fluctuations, but such mechanisms have only rarely been evaluated. We systematically analysed growth, female condition and reproductive traits (fecundity, egg size, hatching rate and size of hatchlings) of vendace in relation to stock density across three lakes over two years. Across populations of contrasting density, length at first maturity, female condition and relative fecundity strongly reflected the difference in stock density, while size of eggs and larvae and hatching success at artificial breeding were not correlated with stock density. Our results indicate that density-dependent female spawner traits do not alter reproductive success, but vendace maintain egg size and hatching rate at the expense of egg quantity even at low per-capita resource availability. Accordingly, the reproductive potential of vendace stocks is primarily driven by spawning stock biomass and relative fecundity. We conclude that besides spawning stock biomass and relative fecundity, the additional inclusion of spawner condition into stock assessment has little potential to improve vendace fisheries management. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Stock density (respectively biomass) is the most surveyed and most managed parameter in marine and inland fisheries (Hilborn and Walters, 1992) and is known to directly or indirectly affect fundamental characteristics of a fish stock. Stock density is the key determinant for catch rate and fisheries yield but might also directly influence the reproductive potential of a stock (spawning stock recruitment theory; Ricker, 1954; Beverton and Holt, 1957). Moreover, stock density directly modulates intraspecific competition for resources, thus indirectly influencing individual somatic growth, condition and mortality, and also reproductive parameters like propagule and offspring quality (Green, 2008; Jakobsen, 2009). While the overall, species-unspecific influence of stock den-

∗ Corresponding author. E-mail addresses: [email protected] (T. Wanke), [email protected] (U. Brämick), [email protected] (T. Mehner). http://dx.doi.org/10.1016/j.fishres.2016.08.028 0165-7836/© 2016 Elsevier B.V. All rights reserved.

sity on somatic growth, mortality and condition is well understood and forms the basis of fisheries theory (surplus production; Russell, 1931; Hjort et al., 1933; Graham, 1935), the influence of stock density on egg quality via maternal effects (here defined as the non-genetic contribution of the female to offspring condition, see Reznick, 1991) is substantially more complex and less generalizable among species (Green, 2008). Maternal condition, and size and age of females, were shown to affect fecundity and offspring quality in many fish species (Solemdal, 1997; Marshall et al., 1998; Marteinsdottir and Begg, 2002) and incorporation of such female traits into recruitment models significantly improved their performance (Marteinsdottir and Thorarinsson, 1998; Marshall et al., 2003). Meta-analysis of time-series data of 25 marine species revealed that the importance of age- and size-related maternal effects increases with the reproductive life span of the populations (Venturelli et al., 2009), suggesting a low effect in short-living species due to their low variety in size and age of spawners. Furthermore, in species with short lifespan, offspring quality is de facto unrelated to the weakly

160

T. Wanke et al. / Fisheries Research 186 (2017) 159–167

varying size and age of females, but female condition may play a stronger role in affecting offspring quality. Female condition is primarily modified by stock density and the overall environmental conditions, and hence the reproductive potential of fish stocks with short lifespans is probably strongly density-dependent. Such a high impact of stock density on maternal condition and offspring quality was shown for short-living sardine (Sardinops sagax) (Kawasaki and Omori, 1995; Schwartzlose et al., 1999; Zwolinski and Demer, 2014). Vendace (Coregonus albula L.) is a short-lived freshwater fish species inhabiting cold, well-oxygenated lakes in North and Central Europe. In most vendace populations, the quantitative relationship between the size of the spawning stock and recruitment is weak (Viljanen, 1988; Marjomäki, 2004; Wanke et al., 2016) and poor reproductive success was often observed despite high parental stock density (Valtonen and Marjomäki, 1988). These strong interannual recruitment fluctuations have often been explained by compensatory effects of stock density via maternal condition on quality and quantity of propagules and offspring (Hamrin and Persson, 1986; Helminen and Sarvala, 1994; Auvinen, 1995). Furthermore, high inter-annual and inter-population variations in growth and condition of spawners are associated with variation of stock density (Christianus, 1995; Marjomäki and Kirjasniemi, 1995; Karjalainen et al., 2016). With the exception of smelt (Osmerus eperlanus L.) in some cases, vendace rarely co-occur with other abundant cold-water species in the hypolimnetic areas of lakes in the northern lowlands (Diekmann et al., 2005; Mehner et al., 2005), and the pelagic dominance of vendace hence makes it likely that density-dependent intraspecific competition is the primary driver of growth and condition of vendace spawners. This pronounced variability of condition makes it likely that besides fecundity variation, there is a strong maternal effect on propagule and offspring quality, in particular because vendace reach maturity almost invariably at age 1+ (Hamrin and Persson, 1986; Huusko and Hyvärinen, 2005) in contrast to fish species with longer life span, which delay maturity in response to low per-capita food resources (Trippel, 1995; de Roos et al., 2006). However, only one study conducted in the northernmost part of the distribution range of vendace (Karjalainen et al., 2016) systematically analysed the relationship between stock density, maternal condition and the quantity and quality of their eggs. A better understanding of the female-offspring relationship in vendace could furthermore help to explain the strong population fluctuations and to improve the management of vendace stocks. In this study we systematically analysed the effect of stock density on somatic and reproductive traits of female vendace across three lakes with a broad range of vendace stock densities over two successive years. We focused on females as they provide the majority of non-genetic material for embryonic development, even though paternal effects have also been demonstrated (Kamler, 2005; Green, 2008). We quantified size-at-age, length at maturity and water content of muscle tissue as a proxy for the energy status and condition of female vendace in relation to stock density (Lambert and Dutil, 1997; Green, 2008). Furthermore, we measured fecundity and the dry weight of ready-to-spawn eggs in all populations in both years. Finally, we ran incubation experiments to quantify the hatching success as a proxy for egg quality and measured the total length of hatchlings. We hypothesized that the strongly varying stock density of vendace across the three lakes has a corresponding impact on somatic and reproductive parameters. In accordance with other studies (Christianus, 1995; Marjomäki and Kirjasniemi, 1995), we predicted a strong negative correlation between stock density and growth and condition of female vendace. This inter-population difference in somatic traits was expected to be transmitted to quality and quantity of ovaries, propagules and offspring. Accordingly we expected that relative

fecundity, egg size, hatching rate and larval size-at-hatch is higher in low-density than in high-density stocks. 2. Methods 2.1. Study sites Lakes Breiter Luzin, Stechlin and Werbellin are located in northeast Germany in the ecoregion ‘Central Plains’ (Illies, 1978). This region forms the southern boundary of the European distribution area of vendace and hence vendace occur in this area only in deep, stratified lakes with a cold hypolimnion. All three lakes are characterized by an extended pelagic habitat, but differ in their productivity (Table 1). Commercial fisheries, as indicated by the annual fisheries yield in relation to standing stock biomass, are strongest in Lake Werbellin, low in Lake Stechlin, and absent in Lake Breiter Luzin (Table 2). Lake Stechlin is inhabited by a total of 13 fish species. The hypolimnetic fish community is dominated by vendace, but the numerically less abundant endemic Fontane cisco (Coregonus fontanae Schulz & Freyhof) co-occur with vendace (Mehner and Schulz, 2002; Helland et al., 2009). Lake Breiter Luzin is also inhabited by a species pair of coregonids, with vendace dominating the hypopelagic area, and the endemic cisco Coregonus lucinensis Thienemann being less abundant (Waterstraat, 1990). Lake Werbellin is inhabited by 13 fish species (Eckmann, 1995). Vendace and smelt are the dominating species found in the ˚ et al., 2012). hypopelagic zone (Juza 2.2. Hydroacoustics To estimate vendace areal density and biomass, hydroacoustic night-time surveys (beginning at least 1.5 h after sunset) using a SIMRAD EY-60 split beam echo sounder were conducted in all lakes during autumn 2012 and 2013. Surveys were designed as a set of systematic parallel transects with the total transect length adjusted to meet an acoustic degree of coverage of six (Breiter Luzin 11.3 km, 15 transects; Stechlin 13.9 km, 19 transects; Werbellin 18.5 km, 22 transects). Operating frequency was set to 120 kHz (7◦ × 7◦ circular transducer, pulse length 0.256 ms, 3.34 pings s−1 ). Before each field season, the echo sounder was calibrated with a standard copper sphere. Raw data were stored on a computer and analysed after conversion (−100 dB base Sv threshold) using the post-processing software Sonar5-Pro version 6.0.2 (Balk and Lindem, 2011). Areal density was calculated based on the single echo detection algorithm (max. angel std. dev. = 0.80, max. gain comp. = 3 dB (one way), echo length (relative to pulse length) = 0.6–1.6). Mean target strength (TS) of single echoes was converted into total length (TL) of fish following the vendace-specific formula TS (dB) = 25.5× log10 TL (cm) – 70.9 (Mehner, 2006). For biomass estimation, TL of single echoes was converted into fish weight by applying the vendace-specific formula weight (g) = 0.0012× TL (cm)3.58 which was derived as average from multi-mesh gillnet catches in the study lakes. For the analysis, −55 dB was used as the lower threshold for target strength and volume backscattering, excluding echoes from fish with a total length of less than about 4 cm. Catches by pelagic multi mesh gillnet surveys (see below) confirmed that vendace occurred only in the hypo- and metapelagic layers. Therefore, analysis was restricted to the layers between a depth of 12 m (thermocline) and the lake bottom. For improved comparability between the lakes, which differ in bank slope and share of shallow area, only those parts of acoustic transects with more than 20 m depth were analysed. Each transect was split into elementary sampling units (ESU) with a length of 50 m each in Lakes Breiter Luzin and Stechlin and with a length of 200 m in the bigger but sparsely populated Lake Werbellin. In Lake Werbellin,

T. Wanke et al. / Fisheries Research 186 (2017) 159–167

161

Table 1 Morphological and limnological parameters of study lakes. Epilimnetic chlorophyll a (Chl a), epilimnetic total phosphorus (TP) and secchi depth are mean values (May–October), the trophic state was assessed according to LAWA guidelines (1998). Data was collected during 2008–2013 and was provided by the Ministry of Agriculture, Environment and Consumer Protection of the Federal State of Mecklenburg-Vorpommern (Breiter Luzin), the Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Berlin (Stechlin) and the Ministry of Rural Development, Environment and Agriculture of the Federal State of Brandenburg (Werbellin). Lake

Area (ha)

Depth max (m)

Depth mean (m)

Trophic state

Chl a (␮g l−1 )

TP (␮g l−1 )

Secchi depth (m)

Breiter Luzin Stechlin Werbellin

345 452 782

58.3 69.5 55

22.3 24 22

mesotrophic oligotrophic mesotrophic

3.1 2.7 4.4

15.0 12.0 30.0

2.9 6.3 3.8

Table 2 Characteristic parameters of vendace stocks in the study lakes in 2012–2013. Commercial vendace fisheries applied 18 mm gillnets in Lake Stechlin, 24 mm gillnets in Lake Werbellin, but no gillnet fishing in Lake Breiter Luzin. Annual vendace yield (kg ha−1 ), which was mainly achieved from May to October and coregonid biomass (kg ha−1 ) as estimated by hydroacoustics in autumn (end of fishing season), both refer to the area of vendace habitat (minimum depth of 20 m) of each lake. Body mass (g), absolute fecundity (eggs fish−1 ) and relative fecundity (eggs g body weight−1 ) were measured in autumn for vendace of age 0+ (one summer old), 1+ (two summer old), 2+ (three summer old) and 3+ (four summer old). n = number, SD = standard deviation. Breiter Luzin

Annual vendace yield (kg ha−1 ) Coregonid biomass (kg ha−1 ) Body mass, wet (g) 0+ Body mass, wet (g) 1+ Body mass, wet (g) 2+ Body mass, wet (g) 3+ Fecundity (eggs fish−1 ) 1+ Fecundity (eggs fish−1 ) 2+ Fecundity (eggs fish−1 ) 3+ Relative fecundity (eggs g−1 ) 1+ Relative fecundity (eggs g−1 ) 2+ Relative fecundity (eggs g−1 ) 3+

Stechlin

Werbellin

n

mean ± SD

n

mean ± SD

n

mean ± SD

2 2 36 76 62 21 36 32 12 36 32 12

0 134.1 ± 33.0 7.6 ± 1.6 16.1 ± 3.8 33.8 ± 11.4 50.1 ± 11.2 1581 ± 621 2687 ± 994 4257 ± 1793 90.1 ± 25.5 88.4 ± 19.9 84.5 ± 21.0

2 2 61 89 34 12 40 24 10 40 24 10

16.3 ± 1.8 227.3 ± 83.4 9.4 ± 3.2 21.1 ± 7.5 39.1 ± 9.8 50.2 ± 6.5 2141 ± 874 3442 ± 1383 4710 ± 1480 92.8 ± 24.8 84.8 ± 27.0 93.3 ± 19.8

2 2 22 50 84 31 15 37 21 15 37 21

17.5 ± 8.0 7.3 ± 5.3 22.6 ± 4.5 92.4 ± 20.9 116.5 ± 22.9 139.2 ± 18.7 12,220 ± 3334 15,912 ± 3666 17,662 ± 3511 115.0 ± 20.3 121.7 ± 22.5 120.1 ± 16.5

smelt co-occur with vendace in the hypolimnetic layers, and these species could not be discriminated if the lower threshold was set to −55 dB, similar to analyses in the other two lakes. Therefore, we set the lower TS threshold to −38 dB in Lake Werbellin, which excluded fish <19.5 cm (smelt and 0+ vendace) from analysis. However, the exclusion of 0+ vendace did not bias the analysis, because in both years the abundance of 0+ vendace was extraordinarily low as a result of strong recruitment failures, which were indicated by extremely low catches with pump-driven, illuminated larvae traps (Wanke et al., 2016) and the almost entire absence of these cohorts in commercial and scientific gillnet catches in the subsequent years. Multi-mesh gillnet catches revealed that the contribution of 0+ vendace to overall vendace density and biomass was negligible and 0+ vendace accounted only for 1.1% or 0.7% of the biomass of all fish <19.5 cm caught in the hypopelagic zone in 2012 and 2013, respectively. Therefore, the estimates of vendace density and biomass in Lake Werbellin can be compared with those in the other two study lakes, despite the augmented lower TS threshold. In Lakes Breiter Luzin and Stechlin, density and biomass estimates refer to the sum of both vendace and the lake-specific endemic coregonid species (see above), whose echoes cannot be separated by hydroacoustics. However, we consider the total coregonid density as appropriate because the ecology of the coregonids is very similar (Helland et al., 2008; Scharf et al., 2008) and hence we assume that intra- and interspecific competition for zooplankton is similarly strong. Lake-wide areal density and biomass were calculated as median of all ESU values. 2.3. Fish sampling for analysis of growth, condition and fecundity To analyse growth, condition and fecundity of vendace, fish were sampled by pelagic multi-mesh gillnets modified from type NORDIC (Appelberg, 2000) shortly before the hydroacoustic surveys in 2012 and 2013. Each pelagic gillnet was 30 m long and 6 m deep and was composed of 12 mesh panels (each 2.5 m long) of different mesh size (6.25/8/10/12.5/15.5/19.5/24/29/35/43/55/70 mm

knot to knot). Sampling in each lake was conducted simultaneously close to the maximum depth and at a second location with a slightly lower depth. At each sampling location, the pelagic zone was sampled by setting the nets in a cascade design (ending of lower line of the surface net tied to ending of upper line of the next net and so on) covering the entire depth range from the surface to the bottom. Nets were set before dusk and hauled after dawn. After removal from the nets, all fish were counted, weighed (nearest 1 g body mass) and the total length was measured (nearest 1 mm). A subsample of 50–100 female vendace were packed individually and frozen at −20 ◦ C for future processing in the laboratory. At the laboratory, vendace were thawed, weighed (nearest 0.1 g body mass) and scales for age analysis were taken from the region above the lateral line between the dorsal and adipose fin. Subsequently, ovaries were dissected, weighed (nearest 0.1 g) and a subsample (composed of pieces from the anterior, posterior and middle section of the ovary) was weighed (nearest 1 mg) and stored in Gilson fluid (Lagler, 1978). Gilson fluid has a hardening effect on eggs and facilitates their separation by breaking the surrounding ovary tissue. After around two weeks of incubation (with occasional shaking), eggs were counted under a dissecting microscope and average ovarian egg density was calculated by dividing the number of eggs in the subsample by the wet weight of the subsample of the ovary. Total fecundity of each fish was calculated by multiplying its ovarian egg density with total weight of both ovaries and relative fecundity was calculated by dividing the total number of eggs by the wet weight of the thawed fish. Muscle water content was estimated in dorsal muscle tissue of thawed fish. Therefore, a dorsal piece of muscle tissue (approximately one g) was dissected and cut into small pieces, which were transferred into a pre-weighed aluminium cup. Subsequently, filled cups were weighed (nearest 1 mg), dried at 100 ◦ C for 24 h and weighed again. Muscle water content was calculated as the weight loss of muscle tissue divided by its initial wet weight and values were expressed in percent.

162

T. Wanke et al. / Fisheries Research 186 (2017) 159–167

2.4. Egg sampling For the analysis of egg dry weight and for the use in breeding experiments, eggs and semen were collected from vendace in Lakes Breiter Luzin, Stechlin and Werbellin during the spawning seasons from mid to end of December in both 2012 and 2013. Spawning vendace were sampled using pelagic gillnets (mesh size 16 and 18 mm in Lakes Breiter Luzin and Stechlin, mesh size 24 and 26 mm in Lake Werbellin). The differing mesh sizes in the lakes facilitated that age of fish was comparable, because vendace in Lake Werbellin grow faster than fish in the other lakes. Further, mesh size was chosen to target spawners of age 2+ and 3+, because in Lakes Breiter Luzin and Stechlin first spawners (age 1+) yield not enough eggs for both breeding and analysis of dry weight and in Lake Werbellin 1+ spawners were almost absent due to low reproduction in the previous years. Nets were set before dusk and hauled after dawn. In a frost-free but unheated room, female vendace were analysed for their maturity status by gently stroking the fingers along the body cavity. Only those females, which spilt eggs already at minimum pressure and whose eggs were limpid and covered with ovary liquid were chosen for analysis. For each lake and each year, 20 females were measured (nearest 1 mm total length), patted dry and eggs were stripped on a clean, dry plate. From this egg sample, 1.5 ml of eggs were carefully transferred with a measuring spoon into a dry screw cap container (40 ml) and another 1.5 ml were transferred into an Eppendorf tube and deep frozen (−20 ◦ C) for later laboratory analysis of dry weight. Milt sampling was performed simultaneously to egg sampling. Therefore, male vendace were patted dry to avoid activation of spermatozoa, then fish were gently stripped and milt was collected directly from the anal papilla by a pipette. To simulate communal spawning and to exclude individual paternal effects, milt was taken from 15 to 50 males and mixed. From the resultant semen mixture, two drops (approximately 100 ␮l) were given to each batch of eggs. Then, fertilization was initiated by adding 10 ml of lake water and each sample was gently shaken. After 30 min, the supernatant water was decanted and the screw cap containers were completely filled with lake water, closed and clipped onto a rack. The rack was put into an isolated box with lake water to guarantee a constant temperature during the transport to the laboratory, where eggs were immediately transferred into incubation jars. 2.5. Breeding Eggs were incubated in a circulation system with 100 ml round bottom flasks used as incubation jars. Eighty flasks were clipped to a board (40 each side) which was mounted on a long tank with each jar being half immersed into the water. Water entered the jar through a central tube close above the rounded bottom and flew off over the flasks rim back into the tank. Flow rate of each jar was adjusted by a valve to allow a gentle rolling movement of the eggs. The water circulated through the system via a main storage tank (600 l) with temperature regulation unit (Titan 2000, Aqua Medic GmbH, Bissendorf, Germany) and aeration system. In both years, incubation conditions were similar and breeding experiments proceeded without any disruptions. During the incubation period, water temperature (mean ± SD) was 5.33 ± 0.48 ◦ C in 2012/13 and 5.62 ± 0.26 ◦ C in 2013/14 and oxygen saturation was always close to 100%. Each batch of eggs was visually inspected at least every second day and dead (opaque) eggs were removed from the incubation jar and put into individual sample container filled with ethanol for later counting. When embryos had reached the eyed stage, the remaining live eggs in the incubation jar from each batch and the dead eggs from the respective ethanol sample were counted under a dissecting microscope. Subsequently, gauze collars were attached to the

neck of each flask to retain larvae and incubation continued. Dead eggs were removed and stored in ethanol as before. At the time of hatch, larvae were counted and removed daily. At the time of maximum hatch, 10 larvae were sampled from each batch and frozen (−20 ◦ C) for dry weight analysis. Breeding experiments were carried on until hatch or death of the last embryo. Survival rate for each batch was calculated as number of live embryos (eyed eggs or hatched larvae) divided by the initial number of eggs (live eggs at eyed stage plus dead eggs until eyed stage) and expressed in percent. Mean dry weight of eggs and larvae was measured from frozen samples in groups of ten to improve precision of weighing. Therefore, samples were thawed, 10 eggs (respectively 10 larvae) were transferred into a pre-weighed aluminium cup, dried at 60 ◦ C for 24 h and weighed to the nearest 0.1 mg again.

2.6. Potential larvae production For each lake and year we calculated the potential larvae production (larvae ha−1 ) according to our measurements of spawning stock biomass and reproductive parameters. Therefore, spawning stock biomass (kg ha−1 ) was calculated from hydroacoustics data as described before but with enhanced lower thresholds for target strength and volume backscattering to exclude immature fish in each lake (Breiter Luzin −43 dB, Stechlin −42 dB, Werbellin −38 dB). Half of the spawning stock biomass was considered as females, because multi-mesh gillnet catches revealed that the sex ratio was fairly balanced in each lake. For each lake and year, the female spawning stock biomass was divided into age groups according to the weight proportions of vendace of different age in the multi-mesh gillnet catches. Subsequently, the female spawning stock biomass of each age group was multiplied by the average relative fecundity (eggs kg body mass−1 ) measured for the respective group as described before and the resultant egg densities (eggs ha−1 ) were summarized for each lake and year. To calculate potential larvae production, egg density was multiplied by the median hatching rate measured during artificial breeding for each lake and year as described before. To illustrate the effect of variation of relative fecundity between dense and sparse stocks on potential larvae production, for each lake and year potential larvae production was alternatively calculated using the minimum and maximum average values of relative fecundity measured for each age group within the two years among the three lakes.

2.7. Statistical analyses To test for statistical difference between the stocks, we built linear-mixed-effects-models (LMMs) with year as random effect (random intercept and random slope). Model assumptions were checked by visual inspection of diagnostic plots. Even after transformation, egg survival data violated the assumptions for general linear modeling (normality), thus we built generalizedlinear-mixed-effects-models (GLMMs) with year as random effect (random intercept and random slope), binomial error distribution and logit as link function. For LMMs and GLMMs, significance of population as fixed effect was calculated by maximum likelihood ratio test between the alternative model including the effect in question and the null model without the effect in question. Correlations between the median of each variable and coregonid density (ind ha−1 ) and coregonid biomass (kg ha−1 ) in the three lakes in 2012 and 2013 were tested by Spearman‘s rank correlation coefficient. We considered correlations with p < 0.10 (␳ > 0.70) as strongly informative due to the low power of n = 6 data pairs per correlation. All statistical analyses and plotting were conducted with R (R Core

T. Wanke et al. / Fisheries Research 186 (2017) 159–167 Table 3 Differences in life history and reproductive traits of populations between the three study lakes, as estimated by maximum likelihood ratio tests for the main effect of population between the alternative model (including the effect in question) and the null model (without the effect in question). Parameters (displayed in Fig. 1a–f) were measured on female vendace caught in Lakes Breiter Luzin, Stechlin and Werbellin in the years 2012 and 2013. The factor year was included into LMMs and GLMMs (survival data) as random effect. Total number of observations (n), degrees of freedom (d.f.), chi-square (X2 ) and p-value (p) is given for each parameter. Parameter

n

d.f.

X2

p

Total length at first maturity (cm) Muscle water content (%) Relative fecundity (eggs g body mass−1 ) Egg dry weight (mg) Larval dry weight (mg) Survival eyed stage (%) Survival hatch (%)

108 189 229 102 100 100 100

2 2 2 2 2 2 2

14.5 10.4 12.7 3.81 5.94 0.45 0.67

0.0007 0.0055 0.0017 0.149 0.051 0.798 0.716

Table 4 Spearman´ıs rank correlation coefficient (␳) and significance of correlations (p) between parameters of life history and reproduction of female vendace and coregonid biomass (kg ha−1 ) in Lakes Breiter Luzin, Stechlin and Werbellin in 2012–2013. n = number of groups, d.f. = ◦ of freedom. Parameter

n

d.f.

␳

p

Total length at first maturity (cm) Muscle water content (%) Relative fecundity (eggs g body weight−1 ) Egg dry weight (mg) Larval dry weight (mg) Hatching success (%)

6 6 6 6 6 6

4 4 4 4 4 4

−0.77 0.71 −0.89 0.03 −0.09 0.37

0.07 0.11 0.02 0.96 0.87 0.47

Team, 2015) using the R-packages lme4 (Bates et al., 2015), gplots (Warnes et al., 2016) and Hmisc (Harrell et al., 2015). 3. Results Areal density of coregonids, as estimated by hydroacoustics, was substantially higher in Lakes Breiter Luzin (8998 individuals ha−1 ) and Stechlin (12,830 individuals ha−1 ) than in Lake Werbellin (122 individuals ha−1 ) in 2012 (Fig. 1). In addition to low recruitment in 2012 and 2013, strong commercial fishing (in both years, biomass of vendace culled during fishing season was higher than remaining stock biomass in autumn, Table 2) declined vendace density in Lake Werbellin even more, and hence the differences between the lakes became larger in 2013 (Table 2, Fig. 1). A similar pattern was observed in terms of areal biomass of coregonids, which was likewise substantially higher in Lakes Breiter Luzin and Stechlin than in Lake Werbellin (Table 2). If smelt and vendace densities were summed up in Lake Werbellin (as based on the lower TS threshold of −55 dB), areal density of hypopelagic fish >4 cm was 613 individuals ha−1 in 2012 and 380 individuals ha−1 in 2013. Therefore, areal density of hypopelagic fish was substantially higher in Lakes Breiter Luzin (>14 fold) and Stechlin (>20 fold) than in Lake Werbellin even if smelt was included. A similar pattern was observed in terms of biomass of hypopelagic fish. Weight-at-age and total length at first maturity were significantly lower in vendace from Lakes Breiter Luzin and Stechlin compared to Lake Werbellin (Tables 2, 3), whereas inter-annual variation of total length at first maturity was low in all lakes (Fig. 1a). Average body wet mass of first spawners (age 1+) in Lake Werbellin was more than four times higher than that of first spawners in Lakes Breiter Luzin and Stechlin (Table 2). Across all lakes and years, there was a strong negative correlation (Spearman ␳ = −0.77, p = 0.072) between areal density or biomass (Table 4) of coregonids and total length of female vendace at first maturity (Fig. 1a). Water content of muscle tissue was significantly lower in female vendace from Lake Werbellin than in females of Lakes Breiter Luzin and Stechlin (Table 3, Fig. 1b). Intra-population variation between

163

subsequent years was low (Fig. 1b). Across all lakes and years, relative water content strongly increased with coregonid stock density even though the correlation was not statistically significant (Spearman ␳ = 0.71, p = 0.111, Fig. 1b). A strong but non-significant correlation was also observed between relative water content and coregonid biomass (Table 4). Age-specific absolute fecundity differed substantially between the stocks of the three lakes (Table 2) and reflected the interstock difference in size-at-age (Table 2). First spawners in Lake Werbellin had on average 5.7 times more eggs than first spawners in Lake Stechlin and 7.7 times more eggs than first spawners in Lake Breiter Luzin. Relative fecundity also differed significantly between the stocks (Table 3) and was higher in Lake Werbellin than in Lakes Breiter Luzin and Stechlin (Table 2, Fig. 1c). In each stock, inter-annual variation of relative fecundity was low (Fig. 1c). Across all lakes and years, there was a strong negative correlation between areal density of coregonids and relative fecundity (Spearman ␳ = −0.89, p = 0.019, Fig. 1c). The same strong relationship was observed between relative fecundity and coregonid biomass (Table 4). Dry weight of eggs produced by 2+ and 3+ spawners did not differ significantly between the three stocks (Table 3). Across all lakes and years, there was no significant correlation between egg size and coregonid stock density (Spearman ␳ = 0.03, p = 0.957, Fig. 1d) or coregonid biomass (Table 4). Similarly, dry weight of larvae did not differ significantly between the three stocks (Table 3). Dry weight of larvae was strongly correlated with dry weight of eggs (Spearman ␳ = 0.89, p = 0.019). Accordingly, there was no correlation between dry weight of larvae and coregonid density across all lakes and years (Spearman ␳ = −0.09, p = 0.872, Fig. 1e). Dry weight of larvae was also not correlated with coregonid biomass (Table 4). Egg mortality was generally highest between fertilization and the eyed stage (median 86.2% of total mortality) and resulted from fertilization failure and instant embryonic mortality. Subsequently, embryonic mortality was low but then again increased around the time of hatch. At the eyed stage, egg quality measured as proportion of live eggs did not differ between stocks (Table 3), and there was also no inter-stock difference in egg quality, measured as proportion of living larvae at 100% hatch (Table 3). Across all lakes and years, embryo survival was not correlated with coregonid density (at eyed stage: Spearman ␳ = 0.14, p = 0.787; at hatch: Spearman ␳ = 0.37, p = 0.468, Fig. 1f) or coregonid biomass (Table 4). Potential larvae production calculated from the female spawning stock biomass under consideration of measured relative fecundity and observed hatching rate was lowest in Lake Werbellin in both years (Fig. 2). Across all years and lakes, potential larvae production was strongly positively correlated with coregonid density and biomass (Fig. 2). Next to the dominating effect of coregonid density and biomass, also interpopulation variation of relative fecundity, which accounted for about 30% within each age class (Table 2), contributed to the observed pattern of potential larvae production. Assuming equal coregonid biomass would consequently lead to about 30% difference in potential larvae production between Lake Werbellin and Lakes Breiter Luzin and Stechlin only due to the observed difference in relative fecundity. The potential ranges of larvae production for the given spawning stock biomasses per lake and year, as calculated using the highest average values of relative fecundity for each age class (from Lake Werbellin) and the lowest values (from Lakes Breiter Luzin and Stechlin), are displayed as arrows in Fig. 2.

4. Discussion We observed pronounced differences in stock density and biomass of coregonids between the three lakes, which induced

164

T. Wanke et al. / Fisheries Research 186 (2017) 159–167

90

ρ = −0.89 *

80 79 78 77 76 1.0

ρ = 0.03

ρ = −0.09

90 80 70 60

0.40

0.45

Hatching success (%)

0.50

f)

ρ = 0.37

50

0.35

Larval dry weight (mg)

e)

0.9

110

Egg dry weight (mg)

130

d)

70

Rel. fecundity (eggs g−1)

c)

ρ = 0.71

0.8

ρ = −0.77 *

0.7

16

18

20

22

Muscle water content (%)

24

b)

14

TL 1st maturity (cm)

a)

0

2500

5000

7500

10000

12500

−1

Stock density (fish ha )

0

2500

5000

7500

10000

12500

−1

Stock density (fish ha )

Fig. 1. Scatter plot and Spearman’s rank correlation coefficient (* p < 0.10) between coregonid stock density and total length at first maturity (cm) (a), muscle water content (%) as proxy for condition (b), relative fecundity (eggs g−1 ) (c), dry weight of eggs (mg) (d), dry weight of larvae (mg) (e) and percent successful hatch as measure of egg quality (f) of female vendace. Vendace were caught in Lakes Breiter Luzin (circle), Stechlin (square) and Werbellin (triangle) in the years 2012 (white) and 2013 (grey). Symbols display median values of respective estimates and error bars display lower quartile (25th percentile) and upper quartile (75th percentile).

slower individual growth and poorer condition of vendace in the lakes with higher densities. However, the differences in coregonid density and biomass did not result in differences in egg size, hatching success and larvae size between the stocks. Among the reproductive traits measured, only relative fecundity was significantly negatively correlated with stock density, resulting in lowest relative fecundity at the highest stock densities. However, there was no correlation between relative fecundity and egg size. Consequently, there was no indication for a strong compensatory feed-back loop between stock density and propagule and offspring quality, indicating that the early reproductive phase of vendace is affected by density-dependent processes just in terms of relative fecundity. Accordingly, prediction of the reproductive potential of vendace stocks can be performed on the basis of spawning stock

biomass and relative fecundity, but may not benefit from inclusion of spawner condition. The differences in area density and biomass of coregonids between the stocks were strong. While Lake Stechlin has been known for its exceptionally dense vendace stock since years (Mehner and Schulz, 2002), the stock density in Lake Werbellin was particularly low during the years of study as a consequence of poor reproduction combined with intensive fishing in previous periods (unpublished data). Accordingly, the study lakes included here covered the full range of densities observed in exploited vendace stocks in Northern Germany, and this range considerably exceeded the range reported from stocks in Finland (Karjalainen et al., 2016). Even inclusion of smelt into analysis hardly reduced the contrast in density and biomass of fishes inhabiting the hypolimnion between

T. Wanke et al. / Fisheries Research 186 (2017) 159–167

15 10

ρ = 1.00 *

0

5

ρ = 1.00 *

5

Pot. larvae prod. (106 larvae ha−1)

10

15

b)

0

Pot. larvae prod. (106 larvae ha−1)

a)

165

0

2500

7500

12500

Stock density (fish ha−1)

0

50 100

200

300

Stock biomass (kg ha−1)

Fig. 2. Scatter plot and Spearman´ıs rank correlation coefficient (* p < 0.05) between (a) coregonid stock density or (b) coregonid stock biomass and potential larvae production (in 106 larvae ha−1 ) of vendace in Lakes Breiter Luzin (circle), Stechlin (square) and Werbellin (triangle) in the years 2012 (white) and 2013 (grey). To illustrate the effect of variation in relative fecundity between dense and sparse stocks on potential larvae production, arrows on the right side of each symbol display the hypothetical range of potential larvae production as based on the maximum relative fecundity for each age class (values from Lake Werbellin) and the minimum relative fecundity for each age class (values from Lakes Breiter Luzin and Stechlin) which were measured during the two years of study.

the lakes. Finally, the vendace stocks in the three lakes are genetically relatively similar (Mehner et al., 2009), suggesting that the observed differences reflect phenotypic variability between the stocks as a consequence of different ratios of standing stock biomass and carrying capacity of the lakes rather than strong lineage differences. The strong contrast in stock density and biomass was reflected in a pronounced inter-stock difference of somatic growth (sizeat-age). A high plasticity in somatic growth is typical for vendace (Karjalainen et al., 2016) and intraspecific competition was often demonstrated to be the main driver for the growth rate in vendace populations (Viljanen, 1986; Helminen et al., 1993; Auvinen, 1995; Marjomäki and Kirjasniemi, 1995). In all three studied lakes, the hypopelagic fish communities were strongly dominated by vendace, suggesting a high correspondence between individual growth and vendace stock density. However, the observed inter-population differences in growth rates were probably further intensified by difference in the trophic state of the lakes (Table 1). The lakes with higher densities of vendace were characterized by lower concentrations of total phosphorus and chlorophyll a, suggesting lower overall productivity and carrying capacity, and therefore, stronger resource limitation (Schulz et al., 2004). Consequently, differences in food availability per individual between lakes were even more pronounced as suggested by differences in stock densities alone. In dense stocks, high competition for limited resources reduces the energy available for allocation into growth, but also into reproduction (Koops et al., 2004). Energy scarcity in dense populations of small individuals is further intensified when considering the dependency of metabolic rate from individual size. According to Jobling (1994), the energy required to maintain a certain stock biomass is higher, if the stock is composed of many small fish compared to few big fish, a situation, which was found in the study lakes. As a consequence, metabolic biomass and, therefore, energy available for allocation into reproduction differed even stronger than indicated by differences in stock biomass between Lake Werbellin and the other two study lakes. However, the amount of energy available for investment into reproduction can be estimated by the condition of fish several months prior to spawning when energy is initially allocated to ovaries (Eliassen and Vahl, 1982; Millner et al., 1991). Among various condition indices, the muscle water content was shown to accurately predict the energy content of muscle

and liver, thus reflecting the overall energy status of fish (Lambert and Dutil, 1997; Green, 2008). In vendace, somatic energy reserves are expected to be highest and accordingly muscle water content lowest in early autumn before gonads grow rapidly and energy is allocated to egg production (Zawisza and Backiel, 1970; Lahti and Muje, 1991; Anwand, 1998). During this period, we found significantly higher muscle water content in the dense stocks of Lakes Breiter Luzin and Stechlin, indicating reduced energy availability for gonad development with potential effects on reproductive traits, as compared to the low density stock of Lake Werbellin. Such negative effects of stock biomass on the condition of vendace were also demonstrated by Auvinen (1995). In teleost fishes, fecundity generally depends on age, size and condition of spawners (Kamler, 2005). Absolute fecundity is primarily related to the weight of spawners. This dominant effect of size explains the marked difference in age-specific absolute fecundity between the studied stocks, which reflect their difference in weight-at-age. Similar results were reported for several vendace populations (Czerniejewski and Wawrzyniak, 2008; Karjalainen et al., 2016). However, even after correcting for differences in female size between the stocks by calculating fecundity relative to individual body wet mass, the significant difference in fecundity between stocks persisted. Furthermore, the relative fecundity correlated negatively with stock density and biomass, corroborating previous studies that relative individual food shortage has a major effect on relative fecundity in fish (Bagenal, 1978; Izquierdo et al., 2001). According to life-history theory, relative fecundity is usually inversely related to egg size (Stearns, 1992; Einum and Fleming, 2000), and an increase in egg size in years with low food availability and hence lower fecundity has also been shown for vendace (Wilkonska and Zuromska, 1988; Gregersen et al., 2011). In our study, we did not observe a correspondence between egg size and relative fecundity or stock density. This is in contrast to a recent study that relative fecundity was lower but eggs bigger in low-density compared to high-density vendace populations (Karjalainen et al., 2016). However, egg size further depends on age and size of the female (Kamler, 2005), and hence it is difficult to disentangle the effects of stock density and population age structure on fecundity at conditions such as here, at which stock density also modified individual growth and size-at-age.

166

T. Wanke et al. / Fisheries Research 186 (2017) 159–167

The size of fish eggs is a strong predictor for larval size at hatch (Kamler, 2005). Larval size is generally thought to be beneficial for larval survival because it reduces initial starvation mortality of larvae (Wilkonska and Zuromska, 1988; Chambers et al., 1989), by enhancing their size-spectrum of potential prey (Knutsen and Tilseth, 1985), and reducing the vulnerability to size-selective predation (Rice et al., 1987; Miller et al., 1988). In our study, the dry mass of eggs was strongly correlated with the dry mass of larvae at hatch, as it has been shown previously for vendace (Viljanen and Koho, 1991). However, due to the missing correlation between egg size and stock density, larval size was neither correlated to the latter. This is in correspondence with results from vendace populations in Finland (Karjalainen et al., 2016). Egg quality is defined as the ability of the egg to be fertilized and to develop into a normal embryo (Bobe and Labbé, 2010). However, neither fertilization success nor embryonic survival is directly linked to egg size, but both are linked to egg matter composition (Kamler, 2005; Bobe and Labbé, 2010). Despite fast progress in the understanding of the role of different chemical compounds for early survival, the only way to accurately measure egg quality is still to conduct breeding experiments (Bobe and Labbé, 2010). In our study, egg mortality was highest in the period between artificial fertilization and the eyed stage as result of fertilization failure and instant embryonic mortality. A second mortality peak occurred at the time of hatch. Such mortality patterns were similarly observed during experimental breeding of vendace from other populations (Wilkonska and Zuromska, 1988; Kamler, 2005; Karjalainen et al., 2014). In contrast to our hypothesis, egg mortality was not correlated with the density of the coregonid stocks in the three lakes, even though stock density significantly affected maternal condition and relative fecundity. This corresponds with results from Karjalainen et al. (2016) who also found no systematic effect of maternal resource availability on fertility rates or total embryonic mortality of vendace. Apparently, vendace can maintain the quality of their propagules even at low per-capita energy availability at high fish densities. From an ecological point of view this is hardly surprising if the high energetic costs of egg production are considered. If high densities of vendace and low productivity of the lake limit the individual resource availability and hamper the condition of maturing females, vendace reduce the relative fecundity, but maintain the quality (and presumably the biochemical composition) of eggs. Similar results have recently been found for Lake Whitefish (Coregonus clupeaformis) (Muir et al., 2014) and European Whitefish (Coregonus lavaretus) (Wanzenböck et al., 2007). Whitefish spawners maintain control over the quality of their gametes through trade-offs between body condition, egg size, egg number, and egg composition to buffer the effect of changes in energy availability on recruitment potential. In line with those results, recruitment potential of the studied populations, which was calculated as the number of larvae produced per hectare under artificial incubation conditions, was largely determined by spawning stock biomass and differences in relative fecundity (Fig. 2). From our results, we conclude that stock density of vendace is the key determinant for growth, condition and relative fecundity, but has no impact on egg size and egg and larvae quality. The potential larvae production per lake and year was mainly driven by the size of the spawning stock and relative fecundity. A densitydependent impairment of reproduction in consequence of poor maternal condition at low per-capita resource availability could not be confirmed. Following Ouellet et al. (2001), maternal effects can be usefully incorporated into stock management only if there is a consistent or at least predictable relationship between attributes of the parents and offspring quality. Our study did not find evidence for such maternal effects in vendace in response to stock density and biomass. Therefore, the additional inclusion of spawner

condition into stock assessment programs will not improve the prediction accuracy of vendace fisheries management.

Acknowledgements We thank S. Zienert, F. Weichler, R. Frenzel, J. Windheuser, J. Rubin and E. Krivopuskova for their contribution to fieldwork and breeding experiments. We further thank two anonymous reviewers for their valuable comments on a former version of this article. This study received funding from the Fischereiabgabe granted by the Ministry of Rural Development, Environment and Agriculture of the Federal State of Brandenburg.

References Anwand, K., 1998. Comparisons of annual gonad cycle and fecundity between nominate and deepwater forms of vendace (Coregonus albula L.) in Lake Stechlin (State of Brandenburg, Germany). J. Appl. Ichthyol. 14, 97–100, http:// dx.doi.org/10.1111/j.1439-0426.1998.tb00621.x. Appelberg, M., 2000. Swedish standard methods for sampling freshwater fish with multi-mesh gillnets. Fisk. Inf. 2000, 1–32. Auvinen, H., 1995. Intra- and inter-specific factors in the dynamics of vendace (Coregonus albula (L.)) populations. Finn. Fish. Res 15, 87–96. Bagenal, T.B., 1978. Fecundity. In: Bagenal, T.B. (Ed.), Methods for the Assesment of Fish Production in Fresh Waters, IBP Hanbook No. 3. Oxford, UK, Blackwell Scientific Publications, pp. 166–178. Balk, H., Lindem, T., 2011. Sonar4 and Sonar5-Pro post processing systems manual version 6.0.2. Oper. Manage., 464, Version 602. Bates, D., Maechler, M., Bolker, B., Walker, S., Christensen, R.H.B., Singmann, H., Dai, B., Eigen, C., Rcpp, L., 2015. Package Lme4. Beverton, R., Holt, S., 1957. On the Dynamics of Exploited Fish Populations, Fish and Fisheries Series. Chapman & Hall, London. Bobe, J., Labbé, C., 2010. Egg and sperm quality in fish. Gen. Comp. Endocrinol. 165, 535–548, http://dx.doi.org/10.1016/j.ygcen.2009.02.011. Chambers, R.C., Leggett, W.C., Brown, J.A., 1989. Egg size, female effects, and the correlations between early life history traits of capelin: Mallotus villosus: an appraisal at the individual level. Fish. Bull. 87, 515–523. Christianus, J., 1995. Age and growth of selected vendace (Coregonus albula L.) populations in Poland. Arch. Hydrobiol. Spec. Issues Adv. Limnol. 46, 97–102. Czerniejewski, P., Wawrzyniak, W., 2008. Fecundity of vendace Coregonus albula (L.), from several lakes in western Pomerania. Arch. Pol. Fish. 16, 135–146. Diekmann, M., Brämick, U., Lemcke, R., Mehner, T., 2005. Habitat-specific fishing revealed distinct indicator species in German lowland lake fish communities: indicator fish species in German lowland lakes. J. Appl. Ecol. 42, 901–909, http://dx.doi.org/10.1111/j.1365-2664.2005.01068.x. Eckmann, R., 1995. Fish species richness in lakes of the northeastern lowlands in Germany. Ecol. Freshw. Fish 4, 62–69. Einum, S., Fleming, I.A., 2000. Highly fecund mothers sacrifice offspring survival to maximize fitness. Nature 405, 565–567. Eliassen, J.-E., Vahl, O., 1982. Seasonal variations in biochemical composition and energy content of liver gonad and muscle of mature and immature cod, Gadus morhua (L.) from Balsfjorden, northern Norway. J. Fish Biol. 20, 707–716. Graham, M., 1935. Modern theory of exploiting a fishery, and application to North Sea trawling. J. Cons. Int. Pour LExplor. Mer 10, 264–274. Green, B.S., 2008. Maternal effects in fish populations. In: Sims, D.W. (Ed.), Advances in Marine Biology. Elsevier, pp. 1–105. Gregersen, F., Vøllestad, L.A., Østbye, 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, 263–269, http://dx.doi.org/10.1111/j. 1365-2400.2010.00779.x. 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, http://dx.doi.org/10.2307/3566049. Harrell, F.E., Dupont, C., many others, 2015. Hmisc: Harrell Miscellaneous. R Package Version 316-0 URL HttpCRANR-Proj. Helland, I.P., Harrod, C., Freyhof, J., Mehner, T., 2008. Co-existence of a pair of pelagic planktivorous coregonid fishes. Evol. Ecol. Res. 10, 373–390. Helland, I.P., Vøllestad, L.A., Freyhof, J., Mehner, T., 2009. Morphological differences between two ecologically similar sympatric fishes. J. Fish Biol. 75, 2756–2767, http://dx.doi.org/10.1111/j.1095-8649.2009.02476.x. Helminen, H., Sarvala, J., 1994. Population regulation of vendace (Coregonus albula) in lake pyhäjärvi, southwest Finland. J. Fish Biol. 45, 387–400, http://dx.doi. org/10.1111/j.1095-8649.1994. tb01321.x. Helminen, H., Auvinen, H., Hirvonen, A., Sarvala, J., Toivonen, J., 1993. Year-class fluctuations of vendace (Coregonus albula) in lake pyhäjärvi, southwest Finland, during 1971-90. Can. J. Fish. Aquat. Sci. 50, 925–931, http://dx.doi.org/ 10.1139/f93-106. Hilborn, R., Walters, C.J., 1992. Quantitative Fisheries Stock Assessment. Springer, US, Boston, MA. Hjort, J., Jahn, G., Ottestad, P., 1933. The optimum catch. Hvalradets Skr. 7, 92–127.

T. Wanke et al. / Fisheries Research 186 (2017) 159–167 Huusko, A., Hyvärinen, P., 2005. A high harvest rate induces a tendency to generation cycling in a freshwater fish population. J. Anim. Ecol. 74, 525–531, http://dx.doi.org/10.1111/j.1365-2656.2005.00951.x. Illies, J., 1978. Limnofauna Europaea. Fischer Stuttgart, Germany. Izquierdo, M.S., Fernández-Palacios, H., Tacon, A.G.J., 2001. Effect of broodstock nutrition on reproductive performance of fish. Aquaculture 197, 25–42, http:// dx.doi.org/10.1016/S0044-8486(01)00581-6. ˚ Juza, T., Frouzová, J., Brämick, U., Draˇstík, V., Mrkviˇcka, T., Kubeˇcka, J., 2012. The vertical distribution of fish in the open water area of a deep temperate mesotrophic lake assessed by hydroacoustics and midwater trawling. Int. Rev. Hydrobiol. 97, 509–525, http://dx.doi.org/10.1002/iroh.201101440. Jakobsen, T. (Ed.), 2009. Fish Reproductive Biology: Implications for Assessment and Management. Wiley-Blackwell Chichester, West Sussex, U.K.; Ames, Iowa. Jobling, M., 1994. Fish bioenergetics. In: Fish and Fisheries Series, 1st ed. Chapman & Hall, London. Kamler, E., 2005. Parent-egg-progeny relationships in teleost fishes: an energetics perspective. Rev. Fish Biol. Fish. 15, 399–421, http://dx.doi.org/10.1007/ s11160-006-0002-y. Karjalainen, J., Keskinen, T., Pulkkanen, M., Marjomäki, T.J., 2014. Climate change alters the egg development dynamics in cold-water adapted coregonids. Environ. Biol. Fishes, http://dx.doi.org/10.1007/s10641-014-0331-y. Karjalainen, J., Urpanen, O., Keskinen, T., Huuskonen, H., Sarvala, J., Valkeajärvi, P., Marjomäki, T.J., 2016. Phenotypic plasticity in growth and fecundity induced by strong population fluctuations affects reproductive traits of female fish. Ecol. Evol., http://dx.doi.org/10.1002/ece3.1936, n/a-n/a. Kawasaki, T., Omori, M., 1995. Possible mechanisms underlying fluctuations in the Far Eastern sardine population inferred from time series of two biological traits. Fish. Oceanogr. 4, 238–242, http://dx.doi.org/10.1111/j.1365-2419.1995. tb00147.x. Knutsen, G.M., Tilseth, S., 1985. Growth, development, and feeding success of Atlantic cod larvae Gadus morhua related to egg size. Trans. Am. Fish. Soc. 114, 507–511, http://dx.doi.org/10.1577/1548-8659(1985)114<507:GDAFSO>2.0. CO;2. 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, http://dx.doi.org/10.1111/j.1467-2979.2004.00149.x. LAWA—Länderarbeitsgemeinschaft Wasser, 1998. Gewässerbewertung – stehende Gewässer Vorläufige Richtlinie für eine Erstbewertung von natürlich entstandenen Seen nach trophischen Kriterien. Kult.-Verl. Berl. GmbH 74. Lagler, K., 1978. Capture, sampling and examination of fishes. In: Bagenal, T.B. (Ed.), Methods for Assessment of Fish Production in Fresh Water, IPB Handbook No. 3. Blackwell Scientific Publications, Oxford, pp. 7–47. Lahti, E., Muje, P., 1991. Egg quality and female condition in vendace (Coregonus albula L.) before and during spawning. Hydrobiologia 209, 175–182, http://dx. doi.org/10.1007/BF00015340. Lambert, Y., Dutil, J.-D., 1997. Can simple condition indices be used to monitor and quantify seasonal changes in the energy reserves of cod (Gadus morhua)? Can. J. Fish. Aquat. Sci. 54, 104–112. Marjomäki, T., Kirjasniemi, J., 1995. Density dependent growth of vendace (Coregonus albula (L.)) in Lake Puulavesi: a modeling analysis. Arch. Hydrobiol. Spec. Issues Adv. Limnol. 46, 97–102. Marjomäki, T.J., 2004. Analysis of the spawning stock-recruitment relationship of vendace (Coregonus albula (L.)) with evaluation of alternative models, additional variables, biases and errors. Ecol. Freshw. Fish 13, 46–60, http://dx. doi.org/10.1111/j.0906-6691.2004.00041.x. Marshall, C.T., Kjesbu, O.S., Yaragina, N.A., Solemdal, P., Ulltang, Ø., 1998. Is spawner biomass a sensitive measure of the reproductive and recruitment potential of Northeast Arctic cod? Can. J. Fish. Aquat. Sci. 55, 1766–1783, http://dx.doi.org/10.1139/cjfas-55-7-1766. Marshall, C.T., O‘Brien, L., Tomkiewicz, J., Köster, F.W., Kraus, G., Marteinsdottir, G., Morgan, M.J., Saborido-Rey, F., Blanchard, J.L., Secor, D.H., Wright, P.J., Mukhina, N.V., Björnsson, H., 2003. Developing alternative indices of reproductive potential for use in fisheries management: case studies for stocks spanning an information gradient. JNAFS 33, 161–190. Marteinsdottir, G., Begg, G.A., 2002. Essential relationships incorporating the influence of age, size and condition on variables required for estimation of reproductive potential in Atlantic cod Gadus morhua. Mar. Ecol. Prog. Ser., 235. Marteinsdottir, G., Thorarinsson, K., 1998. Improving the stock-recruitment relationship in Icelandic cod (Gadus morhua) by including age diversity of spawners. Can. J. Fish. Aquat. Sci. 55, 1372–1377. Mehner, T., Schulz, M., 2002. Monthly variability of hydroacoustic fish stock estimates in a deep lake and its correlation to gillnet catches. J. Fish Biol. 61, 1109–1121, http://dx.doi.org/10.1006/jfbi.2002.2124. Mehner, T., Diekmann, M., Bramick, U., Lemcke, R., 2005. Composition of fish communities in German lakes as related to lake morphology, trophic state, shore structure and human-use intensity. Freshw. Biol. 50, 70–85, http://dx. doi.org/10.1111/j.1365-2427.2004.01294.x. Mehner, T., Pohlmann, K., Elkin, C., Monaghan, M.T., Freyhof, J., 2009. Genetic mixing from enhancement stocking in commercially exploited vendace populations. J. Appl. Ecol. 46, 1340–1349, http://dx.doi.org/10.1111/j.13652664.2009.01733.x. Mehner, T., 2006. Prediction of hydroacoustic target strength of vendace (Coregonus albula) from concurrent trawl catches. Fish. Res. 79, 162–169, http://dx.doi.org/10.1016/j.fishres.2006.01.014.

167

Miller, T.J., Crowder, L.B., Rice, J.A., Marschall, E.A., 1988. Larval size and recruitment mechanisms in fishes: toward a conceptual framework. Can. J. Fish. Aquat. Sci. 45, 1657–1670, http://dx.doi.org/10.1139/f88-197. Millner, R., Whiting, C., Walker, M.G., Witthames, P., 1991. Growth increment, condition and fecundity in sole (Solea solea (L.)) from the north sea and eastern english channel. Neth. J. Sea Res. 27, 433–439, http://dx.doi.org/10.1016/00777579(91)90044-2. Muir, A.M., Arts, M.T., Koops, M.A., Johnson, T.B., Krueger, C.C., Sutton, T.M., Rosenfeld, J., 2014. Reproductive life-history strategies in lake whitefish (Coregonus clupeaformis) from the Laurentian Great Lakes. Can. J. Fish. Aquat. Sci. 71, 1256–1269, http://dx.doi.org/10.1139/cjfas-2013-0254. Ouellet, P., Lambert, Y., Bérubé, I., 2001. Cod egg characteristics and viability in relation to low temperature and maternal nutritional condition. ICES J. Mar. Sci. 58, 672–686, http://dx.doi.org/10.1006/jmsc.2001.1065. R Core Team, 2015. R: A Language and Environment for Statistical Computing. R Found. Stat. Comput., Vienna Austria (URL HttpwwwR-Proj). Reznick, D.N., 1991. Maternal effects in fish life histories. In: Dudley, E.C. (Ed.), The Unity of Evolutionary Biology: Proceedings of the Fourth International Congress of Systematic and Evolutionary Biology. Discorides Press, Portland, Oregon, pp. 780–793. Rice, J.A., Crowder, L.B., Binkowski, F.P., 1987. Evaluating potential sources of mortality for larval bloater (Coregonus hoyi): starvation and vulnerability to predation. Can. J. Fish. Aquat. Sci. 44, 467–472. Ricker, W.E., 1954. Stock and recruitment. J. Fish. Board Can. 11, 559–623. Russell, E.S., 1931. Some theoretical considerations on the overfishing problem. J. Cons. Int. Pour LExplor. Mer 6, 3–20. Scharf, J., Krappe, M., Koschel, R., Waterstraat, A., 2008. Feeding of European cisco (Coregonus albula and C. lucinensis) on the glacial relict crustacean Mysis relicta in Lake Breiter Luzin (Germany). Limnol.—Ecol. Manage. Inland Waters 38, 147–158, http://dx.doi.org/10.1016/j.limno.2007.12.001. Schulz, M., Koschel, R., Reese, C., Mehner, T., 2004. Pelagic trophic transfer efficiency in an oligotrophic, dimictic deep lake (Lake Stechlin, Germany) and its relation to fisheries yield. Limnol.—Ecol. Manage. Inland Waters 34, 264–273, http://dx.doi.org/10.1016/S0075-9511(04)80050-1. Schwartzlose, R.A., Alheit, J., Bakun, A., Baumgartner, T.R., Cloete, R., Crawford, R.J.M., Fletcher, W.J., Green-Ruiz, Y., Hagen, E., Kawasaki, T., Lluch-Belda, D., Lluch-Cota, S.E., MacCall, A.D., Matsuura, Y., Nevárez-Martínez, M.O., Parrish, R.H., Roy, C., Serra, R., Shust, K.V., Ward, M.N., Zuzunaga, J.Z., 1999. Worldwide large-scale fluctuations of sardine and anchovy populations. South Afr. J. Mar. Sci. 21, 289–347, http://dx.doi.org/10.2989/025776199784125962. Solemdal, P., 1997. Maternal effects—a link between the past and the future. J. Sea Res. 37, 213–227. Stearns, S.C., 1992. The Evolution of Life Histories. Oxford University Press, New York. Trippel, E.A., 1995. Age at maturity as a stress indicator in fisheries. BioScience 45, 759–771, http://dx.doi.org/10.2307/1312628. Valtonen, T., Marjomäki, T.J., 1988. Stock-recruitment relationship in certain Finnish vendace (Coregonus albula L.) populations. Finn. Fish. Res. 9, 267–270. Venturelli, P.A., Shuter, B.J., Murphy, C.A., 2009. Evidence for harvest-induced maternal influences on the reproductive rates of fish populations. Proc. R. Soc. B: Biol. Sci. 276, 919–924, http://dx.doi.org/10.1098/rspb.2008.1507. Viljanen, M., Koho, J., 1991. The effects of egg size and incubation conditions on life history of vendace (Coregonus albula L.). Arch. Hydrobiol. Spec. Issues Adv. Limnol. 24, 2418–2423. Viljanen, M., 1986. Biology propagation, exploitation and management of vendace (Coregonus albula L.) in Finland. Arch. Hydrobiol. Spec. Issues Adv. Limnol. 22, 73–97. Viljanen, M., 1988. Relations between egg and larval abundance, spawning stock and recruitment in vendace (Coregonus albula L.). Finn. Fish. Res. 9, 271–289. Wanke, T., Brämick, U., Mehner, T., 2016. Early detection of reproduction deficits and the compensatory potential of enhancement stocking for vendace, Coregonus albula, fisheries in German lakes. Fish. Manag. Ecol. 23, 55–65, http://dx.doi.org/10.1111/fme.12162. Wanzenböck, J., Gassner, H., Hassan, Y., Lahnsteiner, B., Hauseder, G., 2007. Ecology of European Whitefish, Coregonus Lavaretus, in two Austrian lakes in relation to fisheries management and lake productivity. In: Cowx, I.G. (Ed.), Management and Ecology of Lake and Reservoir Fisheries. Blackwell Publishing Ltd, Oxford, UK, pp. 58–69, http://dx.doi.org/10.1002/9780470995679, ch6. Warnes, G.R., Bolker, B., Bonebakker, L., Gentleman, R., Huber, W., Liaw, A., Lumley, T., Maechler, M., Magnusson, A., Moeller, S., Schwartz, M., Venables, B., 2016. gplots: Various R programming tools for plotting data. R Package Version 301 URL HttpCRANR-Proj. Waterstraat, A., 1990. Ökologische Untersuchungen an Populationen der Kleinen Maräne (Coregonus albula L.) im Breiten Luzin (Bezirk Neubrandenburg). Fortschritte Fisch. 9, 93–104. Wilkonska, H., Zuromska, H., 1988. Effect of environment on Coregonus albula (L.) spawners, and influence of their sexual products on the numbers and quality of offspring. Finn. Fish. Res. 9, 81–88. Zawisza, J., Backiel, T., 1970. Gonad development, fecundity and egg survival in Coregonus albula L. Biol. Coregonid Fishes, 363–397. Zwolinski, J.P., Demer, D.A., 2014. Environmental and parental control of Pacific sardine (Sardinops sagax) recruitment. ICES J. Mar. Sci. 71, 2198–2207, http:// dx.doi.org/10.1093/icesjms/fst173. de Roos, A.M., Boukal, D.S., Persson, L., 2006. Evolutionary regime shifts in age and size at maturation of exploited fish stocks. Proc. R. Soc. B: Biol. Sci. 273, 1873–1880, http://dx.doi.org/10.1098/rspb.2006.3518.