EDP Sciences logo
Subscriber Authentication Point
Search
Menu
All issues Volume 60 (2024) Int. J. Lim., 60 (2024) 15 Full HTML
Issue
Int. J. Lim.
Volume 60, 2024
Special issue - Biology and Management of Coregonid Fishes - 2023
Article Number 15
Number of page(s) 18
DOI https://doi.org/10.1051/limn/2024015
Published online 02 September 2024

© F. Bourinet et al., Published by EDP Sciences, 2024

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1 Introduction

Inland fisheries have often been overlooked by policymakers and researchers with large landings from the marine environment absorbing the most attention. However, inland fish catches still represent a globally important, albeit unevenly distributed resource (Lynch et al., 2016; Funge-Smith and Bennett, 2019). They are locally and internationally of great socioeconomic importance, contributing to human consumption and food security (Welcomme et al., 2010; Lynch et al., 2016).

As with all fish stocks, the exploitation of freshwater species raises questions about the sustainability of fisheries and the impacts on the fish populations. While complete extirpation of fish populations due to fishing pressure alone is scarce, overfishing can lead to the collapse of fish stocks, where fish populations become so depleted that it undermines their sustainability (Jackson et al., 2001; Mullon et al., 2005). Fish population dynamics are influenced by various factors, both environmental and anthropogenic. However, fishing pressure has been widely recognized as the primary driver explaining abundance fluctuations in many fish stocks (e.g., Alcala and Russ, 1990; Kleisner et al., 2013; Lyach, 2020).

Studying and quantifying the relationship between fishing effort, fish abundance and productivity has been a key focus of fisheries science and management over the past decades. These efforts have been aimed at ensuring the sustainable management of fishing activity and the health of fish populations (Hilborn and Walters, 1992). Various stock assessment methods have been developed to quantitatively estimate variations in fish abundance under different fisheries management scenarios (Hilborn and Walters, 1992). First developed for marine fisheries, these methods were then adapted for inland fisheries and tailored to various fisheries and ecological contexts, including invasive species, stock enhancement, habitat degradation and others (Berger et al., 2012; Munyandorero, 2012; Alves et al., 2013; Pitcher, 2015; Lorenzen et al., 2016).

Except for a few fisheries in Europe or in the North American Great Lakes, inland fisheries are usually less intensively monitored than marine fisheries (Pitcher, 2015). Inland fisheries scientists frequently face data-poor situations, where adequate time series of catch-at-age, fishing effort, abundance indices, and accurate estimations of critical life-history traits are missing or only partially available. Diverse methods have been developed to work on population and fishery dynamics in a data-poor context (e.g., Pilling et al., 2008; Fitzgerald et al., 2018), including methods based on catch data and life-history traits (Chrysafi and Kuparinen, 2016), the analysis of simple indicators from Catch Per Unit of Effort (CPUE) of both commercial and recreational fisheries statistics (Gerdeaux and Janjua, 2009), population or biomass models (Fisch et al., 2019), management experiments (Lorenzen et al., 2016), or simply expert knowledge (Stokes et al., 2021).

Whitefish, salmonid species belonging to the genus Coregonus, are cold water stenothermic fish representing major resources for inland fisheries in Europe and North America (Fera et al., 2015; Winfield and Gerdeaux, 2015). In peri-alpine lakes, whitefish often constitute a substantial part of the annual landings for both commercial and recreational fishermen (41% of the total Swiss landings in 2020 [OFEV, 2021], > 40% of the French landings in lakes Geneva, Annecy and Bourget [Haute Savoie and Savoie Direction Départementale du Territoire]). However, these species exhibit large stock fluctuations over time, partly because of their high sensitivity to both environmental conditions (Straile et al., 2007; Myers et al., 2015) and fishing pressure (Lukin et al., 2006; Anneville et al., 2015; Rook et al., 2022). Moreover, whitefish are especially sensitive to climate change because of their ecology and physiology (Roberts et al., 2024). These variations highlight the need for a better understanding of the factors influencing recruitment and abundance to better quantify the current whitefish stock biomass, predict future fluctuations, and ensure sustainable fisheries management.

Various stock assessment methods have been applied to whitefish fisheries, mostly in the North American Great Lakes (Cox and Kitchell, 2004; Harford et al., 2006, 2007). These vast lakes contain multiple water units and, consequently, distinct whitefish stocks managed with separate Total Allowable Catches (TAC). Stock assessments and whitefish population dynamics studies in European lakes are scarcer and have to adapt to heterogeneous data coming from various sources (Bartley et al., 2015; Vehanen et al., 2020; Stokes et al., 2021). They often rely on trend analyses of indicators derived from basic fisheries statistics (Anneville et al., 2009; Gerdeaux and Janjua, 2009), scientific surveys with gillnets, or hydroacoustic abundance estimations (Jurvelius et al., 2006; Harrison et al., 2010).

Similar trends in whitefish landings have been observed over the last two decades in three geographically proximate peri-alpine lakes (Bourinet et al., 2023): Lake Geneva, Lake Neuchâtel, and Lake Bourget, while a contrasting trend was observed in a fourth nearby lake (Lake Annecy). These interannual variations in landings raise questions for both scientists and managers about the key determinants of the population dynamics of these whitefish stocks. Previous research has already described the sensitivity of these four lakes to various environmental or anthropogenic factors (Bourinet et al., 2023), taking into account environmental variables, fishing effort and stocking intensity as explanatory variables. However, recent knowledge of the fishery characteristics and their interactions with whitefish populations is lacking. Indeed, the last stock assessment and population dynamics study took place at the end of the previous century and was conducted only at Lake Annecy (Caranhac, 1999).

We aimed to further explore the impact of whitefish harvest on some key processes of population dynamics in the four peri-alpine lakes over the last two decades, with a particular focus on the similarities and differences among lakes. We hypothesized that the large interannual variations in fishing efforts may conceal periods of strong fishing exploitation. To make the best use of the heterogeneous available data, we developed an inter-lake Delay-Difference stock assessment model within a hierarchical state-space framework (Meyer and Millar, 1999). It estimated the time series of the main population dynamics variables, including annual recruitment, spawner biomass, fishing mortality, and exploitation rates. With important fluctuations in productivity and fishing efforts in these lakes, negative impacts from exploitation may have occurred and could still be a problem. This multi-lake model allowed us to assess the status of past and current exploitation and whether these fisheries could benefit from fishing mortality reductions in the future.

2 Materials and methods

2.1 Studied lakes and fisheries

Lakes Geneva, Neuchâtel, Bourget, and Annecy are neighboring deep peri-alpine lakes situated in France and Switzerland (Fig. 1; Tab. 1). Each lake supports both commercial and recreational fisheries targeting a common large pelagic ecomorph of European whitefish that (1) lives in the pelagic zone below the thermocline, (2) feeds mainly on zooplankton and (3) spawns in winter near the shore. European whitefish refers to the species from the genus Coregonus living in the European lakes. This genus is known for its high genetic diversity and numerous species encountered in the peri-alpine lakes (De-Kayne et al., 2022). Within this study the ecomorph is represented by two whitefish species: Coregonus lavaretus in Lake Bourget and Lake Annecy; and Coregonus palea in Lake Geneva and Lake Neuchâtel (Bernatchez and Dodson, 1994; Østbye et al., 2005; Vonlanthen et al., 2012; Alexander and Seehausen, 2021; Bourinet et al., 2023).

These fisheries are characterized by similar fishing gears, seasonality and regulations. Fishermen deploy large drifting gillnets at appropriate depths according to the season, usually targeting age-3 to age-5 whitefish, corresponding to the age of sexual maturity. Fishing season goes from January to October. Mesh sizes are defined by the local fisheries managers of each lake to target age-3 fish (age-4 fish for Lake Annecy). The total landings are dominated by commercial fishermen, although in Lake Annecy, recreational fisheries also significantly contribute to the annual landings (Tab. 1).

Over the period from 2000 to 2020, landings in these lakes have varied significantly, with a ratio between the minimum and the maximum values ranging from 3.3 to 12.6 according to the lake (Tab. 1). The fishing effort, measured as the annual total number of drifting nets deployed on Lake Geneva, Lake Bourget, and Lake Annecy, or the number of whitefish fishing days on Lake Neuchâtel, varied with a ratio between the minimum and the maximum values ranging from 2.1 to 14.3. Fisheries statistics were then gathered from the local fisheries management organization of each lake to whom commercial fishermen report their daily performance: from the Haute-Savoie and Savoie Direction Départementale du Territoire for the French territory and from each respective fishery service of the Swiss cantons (Genève, Vaud, Fribourg and Neuchâtel). These fisheries have been supplemented by traditional stocking activities, with contemporary stocking occurring on all lakes except Lake Annecy (Bourinet et al., 2023).

thumbnail Fig. 1

Location of the four studied lakes. FR = France; SW = Switzerland; IT = Italy. Purple circles represent cities with over 100,000 inhabitants. Adapted from Bourinet et al. (2023).
Table 1

Main physical, trophic, and fishery characteristics of the four studied lakes between 2000 and 2020.

2.2 Adapting a delay-difference model to the multi-peri-alpine lakes context

2.2.1 State model

A Delay-Difference model (Deriso, 1980; Schnute, 1985, 1987) was used to assess the effect of the fisheries in the four studied lakes (Fig. 2). Delay-Difference models can be seen as an intermediary that bridges the gap between simpler surplus production models or other data-limited models and the more complex, fully age-structured models that require complete and detailed age-structure of the landings (Hilborn and Walters, 1992). While not based on full age-structured dynamics and contrary to the surplus production model, the Delay-Difference model explicitly accounts for growth, recruitment, and natural mortality, three key processes of fish population productivity. Despite extensive monitoring of peri-alpine whitefish fisheries for years, data remain incomplete for the use of a fully age-structured model for the period between 2000 and 2020 (Appendix A). As such, the use of a Delay-Difference model within a hierarchical state-space Bayesian framework (Meyer and Millar, 1999) was considered the most suitable approach (Fig. 2). This choice facilitated the integration and enhancement of diverse incomplete data sources, accounting for the heterogeneous availability of information across various lakes while addressing the inherent stochasticity arising from both process and observation errors (Meyer and Millar, 1999; Parent and Rivot, 2012).

The Delay-Difference model is a stage-structured model, dividing the populations into recruits and post-recruits (i.e., the first cohort fully exploited and the older cohorts, respectively). In our case study, recruits correspond to the age-3 group, except in Lake Annecy, which is composed of the age-4 group (Goulon and Guillard, 2022). The age structures of the annual landings were determined from scale readings of annual landing samples spread over the fishing season (Appendix B) (Bourinet et al., 2023). The biomass equation used is adapted from the Deriso-Schnute Delay-Difference model (Deriso, 1980; Schnute, 1985, 1987) (Eq. (1)), starting from a von Bertalanffy weight model (Meyer and Millar, 1999), with p being the Brody growth coefficient:

(1)

(2)

where l is the lake and y corresponds to the year. Bl,y is the biomass for a lake in a given year, and Survis the survival rate, calculated from the annual natural mortality rates (M) (Tab. 2) and the annual fishing mortality rates (F) (Eq. (2)). The same F was assumed for both stages. The size selectivity of these gillnet fisheries is adapted on each lake to target the recruitment stage and the first age class of the post-recruitment stage. The Brody growth coefficient (ρ) represented the only growth parameters in this simplified set of equations. Wrecr and Wpost correspond to the mean weight of a fish from the recruitment stage and the post-recruitment stage, respectively (Appendix B). They were calculated on each lake based on the length-at-age data converted to weight-at-age through a length-weight relationship (Appendix B). As the length-weight relationships were found to be similar between lakes, the same length-weight relationship was used.

The annual recruitment Rl,y, corresponding to the abundance (in numbers) of the first cohort entering the fishery, follows a log-normal distribution (Eq. (3)) around a mean recruitment productivity value meanRl, defined for each lake and expressed as a number of individuals per hectare. The studied ecomorph exhibits the same reproductive ecology, spawning in December on shallow gravel beds close to the shore. Thus, the mean recruitment is assumed to be relative to the surface area of the lakes. tauRl is the precision parameter of the log-normal distribution, integrating the stochasticity for the recruitment process (Tab. 2). A non-informative prior was assigned for meanRl (Tab. 2), as no recent information on the recruitment productivity was available for any of the lakes.

(3)

The yields were calculated for three stages s(Eq. (4)): the yield from the recruitment stage, the yield from the post-recruitment stage, and the total yield corresponding to the sum of the two previous partial yields.

(4)

where Bl,y,s is the biomass of the respective stage s:

(5)

(6)

(7)

Values for parameters used in the main calculations were assigned from independent prior probability distributions. The Fishlife R package (Thorson et al., 2017) was employed to generate suitable priors for life-history parameters based on life-history invariants. This package has compiled life-history traits for 32, 000 fish species and applied a multivariate model to predict distribution and standard deviation of several life-history traits of any fish species, including growth and mortality. It incorporated the taxonomy of the fish species, here the Coregonus lavaretus, to calculate priors for natural mortality M1 and the Brody growth coefficient ρ1 (Tab. 2). The Fishlife package was also utilized to define the prior on tauRi (Tab. 2). For other precision parameters (tau) used in the model to account for stochasticity in each process, a non-informative prior definition was applied based on an inverse-gamma distribution (Tab. 2).

Because of the iterative process in equation (1), there was a need to initialize the first two timesteps of biomass and the first three time-steps of recruitment, using a non-informative prior proportional to each lake surface (Tab. 2). The initial recruitment was forced to be inferior to the initial biomass.

The annual fishing mortality was defined using a random-walk process, as interannual variations are supposed to be low compared to the trends and shifts over the studied period. Thus, the first time-step of F was initiated using a non-informative prior (Tab. 2).

thumbnail Fig. 2

Directed Acyclic Graphic (DAG) of the adapted Delay-Difference Model. Observed data are represented in grey ellipses and unknown data in white ellipses. Arrows represent the causal relationship between two variables. Frames represent the repetition of structure over to lakes (l) or years (y). Parameters and variables are fully described in the text.
Table 2

Definition of the prior of the model parameters. MVNormal corresponds to the multivariate normal distribution, 𝒰 to the uniform distribution, Γ to the gamma distribution, 𝒩 to the normal distribution, and Beta to the Beta distribution.

2.2.2 Observation model

The model was fitted on all available fishery-dependent and fishery-independent data. Total landings were almost always available for each lake and year. Staged-landings, corresponding to the respective landings from each stage, were scattered depending on the lakes (Appendix A). Staged landings were calculated from the respective proportion weighted by the respective mean weight of the fish from the recruitment stage (Wrecr) and from the post-recruitment stage (Wpost), respectively. To take into account stochasticity around the observation process, observed total and staged landings (YieldObsl,y,s) were defined through a log-normal distribution (Eq. (8)).

(8)

CPUE was used as an index of abundance (Ricker, 1940). While the use of CPUE in fishery modeling has been frequently debated (Richards and Schnute, 1986; Maunder et al., 2006; Campbell, 2015), it is commonly utilized in studies where fishery-independent abundance estimates are missing (Gerdeaux and Janjua, 2009; Campbell, 2015; Lappalainen et al., 2020). Moreover, no significant changes that could modify the catchability (q) of the fishery had occurred in the studied period (e.g., same boats, technology and gears were used). Some years were missing for fishing effort at the beginning of the time series (e.g., up to eight years for Lake Neuchâtel; Appendix A). CPUE was calculated only from commercial fishery statistics. While the recreational fishery can represent up to 60% of annual landings in Lake Annecy, the commercial CPUE was considered more reliable and accurate (Gerdeaux and Janjua, 2009). Still, recreational and commercial CPUE presented similar trends between 2000 and 2020. CPUE was assumed to be proportional to the true biomass (Ricker, 1940), and CPUE scaled to the mean CPUE followed a log-normal distribution (Eq. (9)).

(9)

where CPUEobsl,y represents the observed CPUE values for lake l and year y, represents the mean of CPUEobsl,y over the time period; tauCPUEl is the precision parameter of these log-normal distributions, its prior is defined in Table 2; qCPUEl is the proportionality factor linking observed and estimated CPUE, assumed constant over the period and defined in Table 2. CPUE values were standardized by their mean ) to facilitate interpretation of abundance trends with consistent scales between lakes.

Abundance index (AI) values for both the recruits stage and the total biomass on each lake were extracted from hydroacoustic surveys, with surveys spanning one to nine years (Appendix A). The echosounders used were SIMRAD (Simrad Kongsberg Maritime AS, Horten, Norway) EY500, EK60 and EK80, using 70 or 120 kHz frequencies: previous analyses showed that results are independent of the equipment and from the two frequencies (Draštík et al., 2017; Mouget et al., 2019; Rautureau et al., 2022). The same method (i.e., sampling strategy, threshold, filtering, analysis software) was deployed on each lake to estimate fish abundance per hectare during the end of summer according to European standards (CEN, 2014) and described in previous publications (Guillard et al., 2006; Girard et al., 2020). In peri-alpine lakes when the thermocline is well established, the fish community is distributed above or below the thermocline: deeper and cooler habitats support salmonids and coregonids (Mehner and Schulz, 2002; Winfield et al., 2008), while the upper pelagic area supports percidae and cyprinidae (Imbrock et al., 1996; Guillard et al., 2006; Deceliere-Vergès and Guillard, 2008; Probst et al., 2009). So, fish detected under the thermocline in open water at the end of summer were assumed to be European whitefish, as Arctic charr (Salvelinus alpinus) is scarce and mainly found close to the bottom (Lemaire et al., 2020). AI values from hydroacoustic surveys, expressed in biomass per hectare, were assumed to be proportional to the respective biomass reduced by the surface of the lake (Eqs. (10) and (11)).

(10)

(11)

where AIRobsl,y and AIBobsl,y indicate the AI in kg ha−1 taken from hydroacoustic data for the recruitment stage and total biomass, respectively; qAIhydrol represents the proportionality factor linking the observed values to the estimated values, common for both recruitment stage and total biomass, but lake dependent (Tab. 2); Surfacel is the surface area of each lake; Ml is the annual natural mortality rate (Tab. 2); tauAIhydro is the precision parameter of these lognormal distributions, common across all lakes (Tab. 2).

2.2.3 Bayesian inference

The model was run in the JAGS software using Markov Chain Monte Carlo (MCMC) sampling. For the modeling process, the runjags (Denwood, 2016), rjags (Plummer, 2023) and coda (Plummer et al., 2006) R packages were used. Three chains with different initial values were run in parallel with an adaptation phase of 50,000 iterations, a burn-in phase of 20,000 iterations and sampling one out of ten of the 10,000 following iterations. The convergence of the MCMC process was assessed using the potential scale reduction factors (Gelman and Rubin, 1992). The fitting of the model’s estimates to the observation data was also evaluated by ensuring that all observations were encompassed within the 95% Credibility Interval (CI). The code were run using the R software (R Core Team, 2023) and the figures were generated using the ggplot2 package (Wickham, 2016).

2.3 Metrics for stock assessment status

The assessment of the past and present status of the four stocks involved the use of various metrics.

2.3.1 Exploitation rate (E)

The first indicator employed was the comparison of annual fishing mortality rates to annual natural mortality, with the former ideally not exceeding the latter (Gulland, 1970; Bousseba et al., 2021). Thus, the ratio of the exploitation rate (Eq. (12)) should not exceed 0.5, representing roughly the MSY target (Pauly, 1985; Bousseba et al., 2021).

(12)

The fishing mortality rate was used as a proxy for exploitation intensity. Given the assumption of constant natural mortality rates over the study period, variations in fishing mortality served as direct indicators of total mortality, survival rates, and exploitation rates as estimated by the model.

2.3.2 Stage structure

The stage structure of the landings provides informative insights. Well-managed fisheries often target older and larger fish (Griffiths et al., 2023), and a higher proportion of post-recruitment fish in landings indicates the presence of older fish in the population. These older individuals are better spawners, contributing to higher egg production and larger larvae (Berkeley et al., 2004; Birkeland and Dayton, 2005).

2.3.3 Remaining biomass B/Bf=0

Another diagnostic was to compare biomass estimates from the complete model (B) with those in a scenario assuming no fishing exploitation between 2000 and 2020 (Bf=0). Taking the outputs estimated by the model for parameters and latent variables time series, the biomass time series can be calculated under a no-fishing scenario by replacing the fishing mortality time series with zero. The ratio of annual biomass estimated by the complete model divided by the annual biomass estimated under this hypothetical no-fishing scenario (B/Bf=0) is equivalent to a measure regularly used in fishery management: the Spawning Potential Ratio (SPR) (Goodyear et al., 1993; Slipke et al., 2002; Walters and Martell, 2004; Hordyk et al., 2015).

3 Results

The model exhibited good convergence, with potential scale reduction factors from Gelman and Rubin’s convergence diagnostic all below 1.043. Most of the observed values fall inside of the CI range, and the posterior distribution confirms a good fit of the model (Figs. 3 and 5). However, differences in trends between observed CPUE and AI from hydroacoustic seem to cause residual patterns. Moreover, high interannual variations in landings can be hard to match for the model, as is the case for staged landings at Lakes Neuchâtel and Bourget.

On Lake Geneva, a sudden increase in total biomass was noted in 2009 (Fig. 3a), corresponding to a high recruitment that year (Fig. 3b). Lake Bourget displays a similar biomass trend (Fig. 3a) but is associated with strong recruitment after 2009 (Fig. 3b). The biomass in Lake Annecy increased after 2010, as indicated by observed CPUE. This increase did not appear to correspond to substantially greater numbers of recruits (Fig. 3b). Lake Neuchâtel’s total biomass was rather stable but decreased abruptly at the end of the time series, reaching very low levels by 2017 (Fig. 3a), which is concurrent with a decline in mean recruitment after a prolonged period of strong interannual variations (Fig. 3b).

The annual values of fishing mortality rates exhibit a large CI around the median of the estimates. The trends of fishing mortality presented in Figure 3c as the annual fishing mortality estimates, divided by the mean of the respective simulation, are more accurate with a narrower CI. Fishing mortality rates on Lake Annecy continuously decreased over the period, experiencing a significant drop around 2010. Annual fishing mortality rates are the smallest on this lake, with an average median posterior estimate of 0.19 year−1. On Lake Neuchâtel, fishing mortality steadily decreased with an average median posterior estimate of 0.28 year−1. More variability is observed for Lake Geneva and Lake Bourget, with high fishing mortality rates on both lakes in the most recent years, surpassing the natural mortality rates (Tab. 3). Lake Geneva reached 0.81 year−1 in 2017, and Lake Bourget 0.55 year−1 at the end of the time series.

All four lakes exhibited an initial decrease of the remaining biomass ratio B/Bf=0 between 2000 and 2006 (Fig. 4). This finding supports the basic notion that fishing removal directly reduces the biomass of whitefish. Subsequently, two distinct trends emerge among the four lakes. In the cases of lakes Geneva and Bourget, the increase in the ratio B/Bf=0 corresponds to the increase in total biomass previously observed in 2009 (Fig. 3). Afterwards, both lakes experienced a steady decrease in this ratio until the end of the studied period, although Lake Geneva showed an increase in 2020. The Lake Bourget ratio dropped to 0.44, and the Lake Geneva ratio reached its lowest value of 0.23 in 2019. Conversely, for Lake Neuchâtel and Lake Annecy, a stable trend appears, with the ratio B/Bf=0 fluctuating around 0.75.

Comparing the mean recruitment over the period relative to the surface area of each lake provides insights into their overall productivity between 2000 and 2020 (Tab. 3). Lake Annecy emerges as the most productive of the four lakes, with Lake Neuchâtel being the least productive. Lake Geneva and Bourget exhibited similar intermediate values.

Both natural mortality rates and the mean recruitment productivity exhibited different posterior distributions between lakes (Tab. 3). Lake Annecy was the most productive lake for recruitment and displayed the lowest natural mortality rate. Conversely, Lake Neuchâtel had the highest natural mortality rates and the lowest recruitment productivity. Lake Geneva and Lake Bourget had similar outputs for both parameters. For natural mortality rates and the mean recruitment productivity, the posteriors standard deviations were lower than the prior ones, and their mean values were distinct between lakes, indicating that the priors were effectively updated by the information in the data. In contrast, the posterior distributions for the Brody growth coefficient ρ were similar across lakes and not markedly different from the prior distribution (Tab. 3).

The estimates for total landings fit well with the observed landings (Fig. 5). For lakes Geneva, Neuchâtel and Annecy, recruitment landings displayed important interannual variations (Fig. 5a), probably reflecting interannual recruitment variations. Post-recruitment landings were more stable on these lakes (Fig. 5b), with trends very similar to the ones for total landings on Lake Geneva and Lake Annecy (Fig. 5c). High interannual variations for observed values between post-recruitment and recruitment on Lake Neuchâtel were hard to match for the model, with some observed annual data out of the CI range (Fig. 5). Lake Bourget exhibited the same pattern for both staged landings and thus for total landings (Fig. 5), with first a stable low landings period, then a sudden increase in 2009 and then a steady slow decrease.

thumbnail Fig. 3

Estimated outputs from the model for (a) total biomass (tons), (b) recruitment (in number of 1000 individuals) and (c) trends in fishing mortality (corresponding to the ratio of the annual F over the mean F from the period). Lines represent the median of posterior estimates, while the shaded areas denote the 95% CI (Credibility Interval). The dots represent observed values after conversion using the model parameter outputs; for both the biomass and recruitment plots, these correspond to the AI (Abundance Index) obtained from the hydroacoustic data.
Table 3

Posterior statistics of the three parameters.

thumbnail Fig. 4

Ratio of the biomass estimated by the model (B), divided by the biomass estimated under a no-fishing scenario (Bf=0). Lines represent the median of posterior B/Bf=0 estimates; the shaded areas denote the 95% CI (credibility interval).
thumbnail Fig. 5

Estimated outputs in tons (t) of (a) recruitment landings, (b) post-recruitment landings and (c) total landings. Lines represent the median of posterior landings estimates; the shaded areas denote the 95% CI (credibility interval); the dots are the observed values of landings.

4 Discussion

Whitefish fisheries across the world have experienced significant landings fluctuations, influenced by both interannual environmental changes (Kangur et al., 2020; Bogdanov et al., 2021) and variations in stock productivity due to modifications in ecosystem functioning (Rook et al., 2022) or fishery overexploitation (Sarvala et al., 2020). Bourinet et al. (2023) identified links between whitefish abundance and environmental variables in Lakes Geneva, Neuchâtel and Bourget, revealing inter-lake differences. Conversely, Lake Annecy demonstrated a negative relationship with fishing intensity. The authors pointed out the need for a more detailed analysis of the relationship between whitefish population dynamics and fishing activity. The application of a multi-lake Bayesian Delay-Difference model provided detailed metrics related to population dynamics in each lake. It leveraged diverse data sources despite some gaps in the fisheries data.

4.1 Comparison of population dynamics between lakes

Significant synchronicity in both total landings and CPUE was described in these lakes over the 2000–2020 period by Bourinet et al. (2023). The period from 2013 to 2020 was marked by a continuous decrease in whitefish biomass in Lake Geneva, Neuchâtel, and Bourget. No common environmental driver could explain this abundance synchrony. Similarly, the present work did not reveal any similar population dynamics variable which could account for this synchrony. The estimated biomass and landings are consistent with these observations, but the time series of recruitment, fishing mortality rates and stage-structured landings were dissimilar between lakes. The parameters of the model do not show inter-lake similarities, except for the Brody growth coefficient (ρ) that exhibits close posterior distributions among lakes, similar to the prior definition. As determined from the Fishlife package, considering the life-history traits of Coregonus lavaretus, the Brody growth coefficient might be close to the actual value, given all the studied lakes harbor a common ecomorph. However, it seems more plausible that the model did not use ρ for biomass estimates, with no significant deviation from the prior distribution. Variations in biomass estimates were most probably driven by annual survival rates instead (Eq. (1)). It might be because the growth differences are already incorporated through the mean weight of the two stages. It could also be due to the informative prior definition, already close to the actual values.

Interestingly, all four lakes exhibited distinct values of recruitment, landing structure, fishing mortality dynamics, and other parameters. The natural mortality rates aligned with values reported in other studies on the same species, ranging between 0.2 and 0.5 depending on the local context and the species (Auvinen, 1987; Linløkken, 1995; Vainikka et al., 2017). As anticipated for the lake with the highest age at maturity and the slowest growth (Pauly, 1980; Beauchamp et al., 2004; Gislason et al., 2010; Swain, 2011), Lake Annecy exhibited the lowest mortality rates. In contrast, Lake Neuchâtel had higher natural mortality rates, potentially due to its whitefish population having a shorter lifespan (Beverton and Holt, 1959; Hewitt and Hoenig, 2005). This could also be influenced by greater predation, considering the large colonies of cormorants (Phalacrocorax carbo) in Lake Neuchâtel (Keller and Müller, 2012). For modeling purposes, we had to constrain natural mortality rates to be constant over time, a common assumption in fisheries stock assessment models (Lorenzen et al., 2016). However, it is recognized as not always accurate, and its variability should not be overlooked (Vetter, 1988).

4.2 Indicators of fishing pressure impact

The adapted model used in this study represents an innovative approach, providing insights into the effects of fisheries on European whitefish stocks. Such valuable information complements the qualitative analyses of landings and fishermen’s feedback currently used to manage the fisheries. A negative impact of fishing exploitation was observed in three out of four lakes over the period 2000–2020, based on three complementary indicators. Higher fishing mortality rates correspond to elevated total mortality and exploitation rates, underscoring the significant impact on whitefish stocks (Rochet and Trenkel, 2003; Keskar et al., 2017). The low contribution of older fish (assessed through post-recruitment) in total landings reveals a low proportion of post-recruitment in the total biomass, as the model assumed equal F for both stages. This may rather reflect unstable fisheries sensitive to the high instability of recruitment abundance (Griffiths et al., 2023). High exploitation impact is also revealed with low remaining biomass ratio B/Bf=0, indicating that the biomass estimated by the model is low compared to the hypothetical situation of a no-fishing scenario. This ratio is equivalent to the SPR frequently used in fisheries science, for which different thresholds have been debated with different reference points (Clark, 2002; Legault and Brooks, 2013; Larijani et al., 2024). As it depends on the management goal and of the species, we did not pick a specific limit, but opted for working qualitatively with this indicator.

Fisheries strategies vary based on the lake and population dynamics. The increase in total landings on Lake Geneva and Lake Annecy seems to be mainly supported by an increase in post-recruitment landings and lower fishing mortality rates. These observations suggest that efforts should be made to protect the recruitment stage to increase survival rates. Multi-cohort populations, coupled with the conservation of older fish, are known to enhance population stability. Older fish typically possess better spawning capacities, resulting in higher quality and quantity of larvae (Scott et al., 2006; Hixon et al., 2014). The presence of several cohorts and old fish in a population not only contributes to population growth but also increases resilience and resistance to natural and anthropic stressors (Hsieh et al., 2010; Charbonneau et al., 2022). Lower fishing mortality rates thus play a crucial role in this dynamic by directly enhancing the survival rates of recruits, leading to a higher abundance of post-recruitment stages in subsequent years (Charbonneau et al., 2022).

4.3 Fishing pressure impact and status of the fisheries

For Lake Annecy, past overexploitation has already been discussed in Bourinet et al. (2023). Significant regulation changes occurred in 2011, including reduced daily retention limits for the recreational fishery, a halving of the number of commercial fishermen between 2007 and 2010, and a reduction of the nets’ mesh size (Goulon and Guillard, 2022). This management resulted in a substantial overall reduction in fishing efforts. A subsequent steady increase was observed in both biomass and landings, suggesting that the stock was overexploited before 2011 (Rosenberg, 2003). The ratio B/Bf=0 was stable after 2010 at a high value. The Lake Annecy whitefish fishery appeared to be in good status at the end of the time series. Because fishing mortality rates were well below natural mortality rates, there was high productivity relative to its small surface area, and only a small fraction of the biomass was missing compared to the no-fishing simulation. This high productivity is particularly noteworthy considering the lake’s oligotrophic nature, where conventional wisdom might anticipate lower whitefish productivity due to bottom-up control (Downing et al., 1990; Müller et al., 2007). Lake Annecy is known as an oligotrophic lake with good trophic efficiency and notable high productivity, as reported in two Ecopath models (Janjua and Gerdeaux, 2009; Lemaire et al., 2020). It is interesting to highlight that our biomass estimates are consistent with the ones obtained through these models, which ranged between 74 t and 143 t.

On Lake Geneva, the reduction in fishing mortality estimated after 2006 was followed by an increase in both biomass and total landings, driven by increased post-recruitment biomass and landings. It shows the negative impact of exploitation at the beginning of the study period, followed by a rapid recovery. At the same time, an increase in the B/Bf=0 ratio was observed. Such an increase cannot be solely attributed to variations in exploitation. It seems closely tied to abundant annual recruitment. Thus, one or several environmental factors seem to be the most plausible drivers of high recruitment. This sudden increase in biomass starting in 2009 corresponds to the integration of the abundant 2006 cohort into the stock that year (Anneville et al., 2017). The subsequent high abundance of whitefish with a gradual decrease in the following years appears to be driven by the survival of the 2006 cohort, as total mortality rates were low for this cohort. Low B/Bf=0 ratio values on Lake Geneva after 2013 might suggest major impacts from fishing exploitation on fish biomass, removing a large part of the potential stock abundance. The three recent consecutive years of decreasing fishing mortality up to 2020 may be the consequence of a reduced fishing effort after several years of decreasing landings or an effect of the recent fishery regulation changes (Goulon et al., 2023). These hypotheses could explain the sudden increase in the ratio B/Bf=0 in 2020.

Lake Bourget’s whitefish population showed low recruitment before 2009; then, it suddenly increased to a higher level. This sudden shift suggests an important change in ecosystem functioning. Lake Bourget experienced intense eutrophication and re-oligotrophication processes over the last decades, with total phosphorus concentrations dropping constantly since 1980 and going below the 10 µg L−1 threshold around 2010 (Eckmann et al., 2007a; Jacquet et al., 2014a). This total phosphorus concentration decrease drove changes in phytoplankton and zooplankton composition and abundance (Jacquet et al., 2014b; Moiron et al., 2021). This shift may have impacted whitefish population dynamics through a bottom-up effect (Müller et al., 2007), explaining upward whitefish trends after 2009. This hypothesis is also supported by the increase in B/Bf=0 in 2009: as the only difference between B and Bf=0 is the absence of fishing removal in the latter, an increase of this ratio implies an environmental influence. It could be induced by the high spring daphnia abundance observed in these years (Perga and Lainé, 2010; Jacquet et al., 2023), as a positive relationship was found in Bourinet et al. (2023). Still, important changes in management tactics also occurred in 2008 and 2009 and the ensuing changes to fishing effort cannot be ignored (Jacquet et al., 2022). Starting in 2017, the biomass estimated by the model significantly deviated from the observed hydroacoustic AI, which displayed a more stable trend. During recent years, the catch for commercial fishermen has decreased. A sudden shift in the growth of European whitefish has been observed (Jacquet et al., 2023), with age-3+ whitefish still below the minimum required size for commercial fishermen. Thus, the overall biomass remains high, with a substantial number of small fish not integrated into the fishery. Our model did not account for variations in growth, and such sudden shifts are likely challenging to model.

In Lake Neuchâtel, after a continuous decrease in fishing mortality rates, there was an acceleration in the last three years, along with a drop in biomass, recruitment and landings. The decrease in the fishing mortality rates resulted from a decrease in fishing effort, as the fishery became unprofitable with critically low total whitefish biomass levels. However, the ratio B/Bf=0 remained stable, with only a slight increase. This means that, without any fishing removal, the abundance would have followed the same declining trend, according to other model outputs. This is particularly evident in the recruitment time series, reflecting the interannual variability of reproduction success. One or several environmental parameters have likely reached threshold values, hindering successful egg deposition, development, or larvae survival. Winter water temperature has been identified as negatively correlated with whitefish abundance on Lake Neuchâtel (Bourinet et al., 2023). The mean winter water temperatures exhibited a sudden increase in 2014, with all subsequent values (except 2017) surpassing the maximum recorded temperature between 2000 and 2013. Despite Lake Neuchâtel being the coldest of the four studied lakes, the local population of pelagic European whitefish might be facing excessively high winter temperatures, restricting reproduction success (Brooke, 1975; Cingi et al., 2010; Stewart et al., 2021). This lake is also managed with high stocking intensity, releasing millions of just-hatched larvae each spring. However, it does not seem to offset the natural reproduction failure, suggesting that other environmental parameters affecting post-hatch stages may also be at play. It is worth noting that Lake Neuchâtel has the least accurate and the most limited data among the four lakes studied. The fishing effort is recorded as the number of whitefish fishing days, whereas the other lakes have a more accurate measure of the annual number of drifting nets deployed. Furthermore, Lake Neuchâtel lacks more annual data, displaying the shortest CPUE time series and only one hydroacoustic observation. While the hierarchical framework helps transfer information from data-rich to data-poor lakes, Lake Neuchâtel remains more susceptible to bias than the others. It highlights the importance of ensuring accurate long-term monitoring of fisheries statistics and scientific surveys, as the invaluable hydroacoustic data provide a comparable AI between lakes and across time. The method applied to the four lakes was the same, with a common error process and similar stochasticity. These data ensure the absolute productivity comparisons between the lakes. Consequently, instituting an annual survey encompassing all four lakes would be highly beneficial for enhancing the accuracy and completeness of the data. Such a comprehensive survey could further refine our understanding of the population dynamics and productivity variations across the peri-alpine lakes.

To conclude, fishing exploitation has been shown as a crucial driver of European whitefish population dynamics in peri-alpine lakes, with periods of negative impacts on population abundance over the last two decades. However, it is essential to also acknowledge the significant influence of environmental factors, both direct and indirect (Lynch et al., 2015; Myers et al., 2015; Zischke et al., 2017; Stewart et al., 2021). In a study by Bourinet et al. (2023), direct sensitivity to spring daphnia abundance in Lake Geneva and Lake Bourget, as well as winter surface temperature in Lake Neuchâtel, was highlighted. However, the environmental impact on these populations is likely a complex interplay of multiple factors, including environmental and anthropic variables along with fishing exploitation (Pepin, 2016; Bai et al., 2022). The model could benefit from integrating these factors. Recruitment could be delineated by a Stock-Recruitment Relationship (SRR) that accounts for the influence of selected environmental variables (Chen and Irvine, 2001; Heikinheimo et al., 2014; Bai et al., 2022). In addition, it would be relevant to incorporate stocking effects into the SRR, given its historical significance as a management measure in three of the four lakes, Lake Annecy being the only lake without whitefish stocking since 2000. Although the inclusion of a stocking parameter in a SRR has not been extensively studied, some insights have been gleaned over the last few decades through peri-alpine lake-specific, ad hoc studies (Eckmann et al., 2007b; Champigneulle and Cachera, 2008; Wedekind et al., 2022; Baer et al., 2023). This expanded model would offer a more comprehensive understanding of the interplay between environmental factors, stocking practices, fishing exploitation and European whitefish population dynamics. It could assist peri-alpine lake whitefish fishery managers by testing fishing scenarii and making short-term predictions of recruitment and abundance.

Acknowledgements

The authors want to thank the funders of the PhD project: the Association Nationale de la Recherche et de la Technologie (ANRT), three Swiss cantons (the Geneva Canton, the Vaud Canton, the Valais Canton), the Office Fédéral de l’Environnement (OFEV), and the Direction Départementale des Territoires of Savoie and Haute-Savoie. Also, many thanks to all the people who contributed to collecting the data used in this study.

Appendix A Available observed data

Table A1 displays the number of annual data available for our study and used as observed data in the model.

Table A1

Number of annual observed data used in the model, covering the period between 2000 and 2020.

Appendix B Length-weight relationships (LWR) and data

We used available individual length and weight data on Lake Annecy and Lake Bourget from standardized scientific sampling surveys (Table B1) EN 14757 (Appelberg, 2000). This type of data allows information on fish below the legal catch size. On Lake Annecy, data were completed with data collected from professional fishermen as part of the fisheries monitoring program. We added data from another lake, Lake Aiguebelette, a smaller peri-alpine lake situated a few kilometers away from Lake Bourget, where the same type of data was available on Coregonus lavaretus. Furthermore, on Lake Annecy, the database was completed with data collected from professional fishermen as part of the fisheries monitoring program. Data required were not available for Lake Geneva and Lake Neuchâtel. The three length-weight relationships (LWR) showed very close parameter values (intercept and slope of the equation value) (Fig. B1 and Tab. B1). These relationships were also compared to the 11 relationships gathered on the Fishbase website (Froese and Pauly, 2024), with similar parameter values between this study and the value estimated as reliable on Fishbase. So, we decided to use only one length-weight relationship for all studied lakes, assuming that the relationships for these lakes are similar (Fig. B1). We picked the relationship derived from Lake Annecy data, as it was the most complete dataset.

thumbnail Fig. B1

Length-weight relationship fitted on each lake dataset. Dots are observed values; lines are the statistically fitted relationship. Yellow = Lake Aiguebelette; Blue = Lake Bourget; Purple = Lake Annecy.
Table B1

Length-weight relationship for Coregonus lavaretus collected during gillnetting standardized surveys and with additional data from fisheries for Lake Annecy, based on theequation log (W) = log a + b log(L) (a: intercept and b: slope of the equation). Sample size (N); length (L) in mm; weight (W) in g; minimum (min) and maximum (max) of L & W; Standard error (SE); coefficient of determination (r2), p is the significance of regression (p) with p significant at <0.05.

On each lake, length-at-age was determined from scale reading and back-calculation data. The scales used in this study came from the COLISA collection (DOI: https://doi.org/10.15454/D3ODJM) collected as part of the monitoring of landings by professional fishermen in Lake Geneva, with one sample taken each month. In the case of Lake Annecy, the scales came from monitoring of landings by professional fishermen, carried out directly by scientists or by fishermen as part of participatory science approaches. On Lake Neuchâtel, scales were collected monthly for growth monitoring by scientists. The data range is 2002–2020 for Lake Geneva, 2001–2020 for Lake Neuchâtel, 1989–2020 for Lake Bourget, and 2003–2020 for Lake Annecy. The steps for determining ages are detailed in Hamelet et al. (2022).

We pulled all data covering the studied period to get an average length of the age dataset. Then, it was converted in weight-at-age through the length-weight relationship. This way, we could calculate directly the weight of the two stages: Wrecr fort the mean weight of the recruitment stage, and Wpost for the mean weight of the post-recruitment stage (Tab. B2).

Table B2

Mean weight at recruitment (Wrecr) and post-recruitment (Wpost) stages in grammes for the four studied lakes.

Appendix C Fitting of biomass trends estimates with CPUE time series

The CPUE (Catch Per Unit of Effort) scaled by the mean CPUE was supposed to be proportional to the total biomass trend, i.e., the total biomass scaled by the mean of the total biomass. The fit of the biomass estimate trends is presented in Figure C1.

thumbnail Fig. C1

Estimated outputs from the model for the total biomass scaled to the mean total biomass. Lines represent the median of posterior estimates, while the shaded areas denote the 95% CI (Credibility Interval). The dots represent observed values of CPUE scaled to the mean CPUE.

References

  • Alcala AC, Russ GR. 1990. A direct test of the effects of protective management on abundance and yield of tropical marine resources. ICES J Mar Sci 47: 40–47. [CrossRef] [Google Scholar]
  • Alexander T, Seehausen O. 2021. Diversity, distribution and community composition of fish in perialpine lakes. ‘Projet Lac’ synthesis report. Eawag Wasserforschungsinstitut ETH-Bereichs, 284. [Google Scholar]
  • Alves DC, Minte-Vera CV, Agostinho AA, Okada EK, Vasconcelos LP. 2013. Hydrological attributes and rheophilic freshwater fish: stock assessment. Rev Fish Biol Fish 23: 375–394. [CrossRef] [Google Scholar]
  • Anneville O, Lasne E, Guillard J, Eckmann R, Stockwell JD, Gillet C, Yule DL. 2015. Impact of fishing and stocking practices on coregonid diversity. Food Nutr Sci 6: 1045–1055. [Google Scholar]
  • Anneville O, Souissi S, Molinero JC, Gerdeaux D. 2009. Influences of human activity and climate on the stock-recruitment dynamics of whitefish, Coregonus lavaretus, in Lake Geneva. Fish Manag Ecol 16: 492–500. [CrossRef] [Google Scholar]
  • Anneville O, Vogel C, Lobry J, Guillard J. 2017. Fish communities in the Anthropocene: detecting drivers of changes in the deep peri-alpine Lake Geneva. Inland Waters 7: 65–76. [CrossRef] [Google Scholar]
  • Appelberg M. 2000. Swedish standard methods for sampling freshwater fish with multimesh. Rep Inst Freshw Res 26. [Google Scholar]
  • Auvinen H. 1987. Growth, mortality and management of whitefish (Coregonus lavaretus L. s.l.), vendace (Coregonus albula L.), roach (Rutilus rutilus L.) and perch (Perca fluviatilis L.) in Lake Pyhäjärvi (Karelia). Finn Fish Res 8: 38–47. [Google Scholar]
  • Baer J, Kugler M, Schubert M, Schotzko N, Rösch R, Vonlanthen P, DeWeber JT. 2023. A matter of time—efficacy of whitefish stocking in a large pre-alpine lake. Fish Manag Ecol 1:12. [Google Scholar]
  • Bai X, Gao L, Choi S. 2022. Exploring the response of the Japanese sardine (Sardinops melanostictus) stock-recruitment relationship to environmental changes under different structural models. Fishes 7: 276. [CrossRef] [Google Scholar]
  • Bartley DM, De Graaf GJ, Valbo-Jørgensen J, Marmulla G. 2015. Inland capture fisheries: status and data issues. Fish Manag Ecol 22: 71–77. [CrossRef] [Google Scholar]
  • Beauchamp KC, Collins NC, Henderson BA. 2004. Covariation of growth and maturation of lake whitefish (Coregonus clupeaformis). J Gt Lakes Res 30: 451–460. [CrossRef] [Google Scholar]
  • Berger AM, Jones ML, Zhao Y, Bence JR. 2012. Accounting for spatial population structure at scales relevant to life history improves stock assessment: the case for Lake Erie walleye Sander vitreus. Fish Res 115–116: 44–59. [CrossRef] [Google Scholar]
  • Berkeley SA, Hixon MA, Larson RJ, Love MS. 2004. Fisheries sustainability via protection of age structure and spatial distribution of fish populations. Fisheries 29: 23–32. [CrossRef] [Google Scholar]
  • Bernatchez L, Dodson JJ. 1994. Phylogenetic relationships among Palearctic and Nearctic whitefish (Coregonus sp.) populations as revealed by mitochondrial DNA variation. Can J Fish Aquat Sci 51: 240–251. [CrossRef] [Google Scholar]
  • Beverton RJH, Holt SJ. 1959. A review of the lifespans and mortality rates of fish in nature, and their relation to growth and other physiological characteristics. In Ciba Foundation Symposium – The Lifespan of Animals (Colloquia on Ageing). John Wiley & Sons, Ltd, pp. 142–180. [Google Scholar]
  • Birkeland C, Dayton PK. 2005. The importance in fishery management of leaving the big ones. Trends Ecol Evol 20: 356–358. [PubMed] [Google Scholar]
  • Bogdanov DV, Sendek DS, Lajus DL. 2021. Coregonine fisheries in the eastern Gulf of Finland, Baltic Sea: history and current status. Adv Limnol 65–81. [CrossRef] [Google Scholar]
  • Bourinet F, Anneville O, Drouineau H, Goulon C, Guillard J, Richard A. 2023. Synchrony in whitefish stock dynamics: disentangling the effects of local drivers and climate. J Limnol 82: 2134. [CrossRef] [Google Scholar]
  • Bousseba M, Ferraj L, Ouahb S, Droussi M, Hasnaoui M. 2021. Study of the mortality and exploitation parameters of Sander lucioperca (Linnaeus, 1758) in the Al Massira Reservoir. E3S Web Conf 314: 03003. [CrossRef] [EDP Sciences] [Google Scholar]
  • Brooke LT. 1975. Effect of different constant incubation temperatures on egg survival and embryonic development in lake whitefish (Coregonus clupeaformis). Trans Am Fish Soc 104: 555–559. [CrossRef] [Google Scholar]
  • Campbell RA. 2015. Constructing stock abundance indices from catch and effort data: some nuts and bolts. Fish Res 161: 109–130. [CrossRef] [Google Scholar]
  • Caranhac F. 1999. Modélisation de la dynamique de populations piscicoles exploitées intégrant la variabilité individuelle de croissance: application aux corégones (Coregonus lavaretus) du lac d’Annecy. Thèse de doctorat. Lyon 1. [Google Scholar]
  • Champigneulle A, Cachera S. 2008. Evaluation de la stratégie de pacage lacustre (repeuplement en lac) pour le corégone (Coregonus lavaretus) au lac du Bourget. Rapport SHL 284-2008, INRA-Thonon. 45. [Google Scholar]
  • Charbonneau JA, Keith DM, MacNeil MA, Hutchings JA. 2022. Effects of fishing mortality on the age structure of marine fishes. Can J Fish Aquat Sci 79: 2225–2236. [CrossRef] [Google Scholar]
  • Chen DG, Irvine JR. 2001. A semiparametric model to examine stock-recruitment relationships incorporating environmental data. Can J Fish Aquat Sci 58: 1178–1186. [Google Scholar]
  • Chrysafi A, Kuparinen A. 2016. Assessing abundance of populations with limited data: lessons learned from data-poor fisheries stock assessment. Environ Rev 24: 25–38. [CrossRef] [Google Scholar]
  • Cingi S, Keinänen M, Vuorinen PJ. 2010. Elevated water temperature impairs fertilization and embryonic development of whitefish Coregonus lavaretus. J Fish Biol 76: 502–521. [CrossRef] [PubMed] [Google Scholar]
  • Clark WG. 2002. F35% revisited ten years later. North Am J Fish Manag 22: 251–257. [CrossRef] [Google Scholar]
  • Cox SP, Kitchell JF. 2004. Lake Superior ecosystem, 1929–1998: simulating alternative hypotheses for recruitment failure of lake herring (Coregonus artedi). Bull Mar Sci Bull Mar Sci 74: 671–683. [Google Scholar]
  • Deceliere-Vergès C, Guillard J. 2008. Assessment of the pelagic fish populations using CEN multi-mesh gillnets: consequences for the characterization of the fish communities. Knowl Manag Aquat Ecosyst 04. [CrossRef] [EDP Sciences] [Google Scholar]
  • De-Kayne R, Selz OM, Marques DA, Frei D, Seehausen O, Feulner PGD. 2022. Genomic architecture of adaptive radiation and hybridization in Alpine whitefish. Nat Commun 13: 4479. [CrossRef] [PubMed] [Google Scholar]
  • Denwood MJ. 2016. runjags: An R Package providing interface utilities, model templates, parallel computing methods and additional distributions for MCMC models in JAGS. J Stat Softw 71: 1–25. [CrossRef] [Google Scholar]
  • Deriso RB. 1980. Harvesting strategies and parameter estimation for an age-structured model. Can J Fish Aquat Sci 37: 268–282. [CrossRef] [Google Scholar]
  • Downing JA, Plante C, Lalonde S. 1990. Fish production correlated with primary productivity, not the morphoedaphic index. Can J Fish Aquat Sci 47: 1929–1936. [CrossRef] [Google Scholar]
  • Draštík V, Godlewska M, Balk H, Clabburn P, Kubečka J, Morrissey E, Hateley J, Winfield IJ, Mrkvička T, Guillard J. 2017. Fish hydroacoustic survey standardization: a step forward based on comparisons of methods and systems from vertical surveys of a large deep lake. Limnol Oceanogr Methods 15: 836–846. [CrossRef] [Google Scholar]
  • Eckmann R, Gerdeaux D, Müller R, Rösch R. 2007a. Re-oligotrophication and whitefish fisheries management – a workshop summary. Biol Manag Coregonid Fishes – 2005 Proc. Ninth Int. Symp. Biol. Manag. Coregonid Fishes Held Olszt. Pol. 21-27 August 2005, 353–360. [Google Scholar]
  • Eckmann R, Kugler M, Ruhlé C. 2007b. Evaluating the success of large-scale whitefish stocking at Lake Constance. Adv Limnol 60: 361–368. [Google Scholar]
  • Fera SA, Rennie MD, Dunlop ES. 2015. Cross-basin analysis of long-term trends in the growth of lake whitefish in the Laurentian Great Lakes. J Gt Lakes Res 41: 1138–1149. [CrossRef] [Google Scholar]
  • Fisch NC, Bence JR, Myers JT, Berglund EK, Yule DL. 2019. A comparison of age-and size-structured assessment models applied to a stock of cisco in Thunder Bay, Ontario. Fish Res 209: 86–100. [CrossRef] [Google Scholar]
  • Fitzgerald CJ, Delanty K, Shephard S. 2018. Inland fish stock assessment: applying data-poor methods from marine systems. Fish Manag Ecol 25: 240–252. [CrossRef] [Google Scholar]
  • Froese R, Pauly D. 2024. FishBase. www.fishbase.org. [Google Scholar]
  • Frossard V, Goulon C, Guillard J, Hamelet V, Jacquet S, Lainé L, Rautureau C, Rimet F, Tran-Khac V. 2022. Suivi de la qualité écologique du lac d’Annecy. Rapport 2021. SILA Éd INRA-Thonon, 47. [Google Scholar]
  • Funge-Smith S, Bennett A. 2019. A fresh look at inland fisheries and their role in food security and livelihoods. Fish Fish 20: 1176–1195. [CrossRef] [Google Scholar]
  • Gelman A, Rubin DB. 1992. Inference from iterative simulation using multiple sequences. Stat Sci 7: 457–472. [Google Scholar]
  • Gerdeaux D, Janjua MY. 2009. Contribution of obligatory and voluntary fisheries statistics to the knowledge of whitefish population in Lake Annecy (France). Fish Res 96: 6–10. [CrossRef] [Google Scholar]
  • Girard M, Goulon C, Tessier A, Vonlanthen P, Guillard J. 2020. Comparisons of day-time and night-time hydroacoustic surveys in temperate lakes. Aquat Living Resour 33: 9. [CrossRef] [EDP Sciences] [Google Scholar]
  • Gislason H, Daan N, Rice JC, Pope JG. 2010. Size, growth, temperature and the natural mortality of marine fish. Fish Fish 11: 149–158. [CrossRef] [Google Scholar]
  • Goodyear CP, Smith SJ, Hunt JJ, Rivard D. 1993. Spawning stock biomass per recruit in fisheries management: foundation and current use. Can Spec Publ Fish Aquat Sci 120: 67–81. [Google Scholar]
  • Goulon C, Bourinet F, Guillard J. 2023. Principaux résultats du suivi halieutique concernant la population de corégone du Léman en 2022. rapport du groupe de recherche piscicole. Convention INRAE-DDT74. 27. [Google Scholar]
  • Goulon C, Guillard J. 2022. Suivi halieutique du lac d'Annecy 2021. Convention Sila-DDT-ALP-PêcheursPro. 32. [Google Scholar]
  • Griffiths CA, Winker H, Bartolino V, Wennhage H, Orio A, Cardinale M, 2023. Including older fish in fisheries management: a new age-based indicator and reference point for exploited fish stocks. Fish Fish 00: 1–20. [Google Scholar]
  • Guillard J, Perga ME, Colon M, Angeli N. 2006. Hydroacoustic assessment of young-of-year perch, Perca fluviatilis, population dynamics in an oligotrophic lake (Lake Annecy, France). Fish Manag Ecol 13: 319–327. [Google Scholar]
  • Gulland JA. 1970. The fish resources of the ocean. West Byfleet, Surrey, Fishing News (Books), Ltd., for FAO, 255 p. Rev. ed. of FAO Fish. Tech.Pap., (97):425 p. [Google Scholar]
  • Hamelet V, Goulon C, Marchand F. 2022. Interpretation of age based on the scales of lacustrine Salmonidae, the common whitefish (Coregonus sp.) and the arctic char (Salvelinus alpinus). NOVAENuémro Spéc. 2: 4. [Google Scholar]
  • Harford W, Crawford S, Coppaway CW. 2006. 2006 Saugeen Ojibway Nations commercial harvest TACs for lake whitefish (Coregonus clupeaformis) in Lake Huron. [Google Scholar]
  • Harford W, Latremouille D, Crawford S. 2007. A Bayesian stock assessment of lake whitefish (Coregonus clupeaformis) in Lake Huron and evaluation of total allowable catch options for 2007 Saugeen Ojibway Nations commercial harvest. Chippewas Nawash Unceded First Nation Saugeen First Nation, 61. [Google Scholar]
  • Harrison AJ, Kelly FL, Rosell RS, Champ TWS, Connor L, Girvan JR. 2010. First record and initial hydroacoustic stock assessment of pollan Coregonus autumanlis pallas in Lough Alle, Ireland. Biol Environ Proc R Ir Acad 110B: 69–74. [CrossRef] [Google Scholar]
  • Heikinheimo O, Pekcan-Hekim Z, Raitaniemi J. 2014. Spawning stock-recruitment relationship in pikeperch Sander lucioperca (L.) in the Baltic Sea, with temperature as an environmental effect. Fish Res 155: 1–9. [Google Scholar]
  • Hewitt DA, Hoenig JM. 2005. Comparison of two approaches for estimating natural mortality based on longevity. Fish Bull 103: 433. [Google Scholar]
  • Hilborn R, Walters CJ. 1992. Quantitative fisheries stock assessment: choice, dynamics and uncertainty. Boston, MA: Springer US. [CrossRef] [Google Scholar]
  • Hixon MA, Johnson DW, Sogard SM. 2014. BOFFFFs: on the importance of conserving old-growth age structure in fishery populations. ICES J Mar Sci 71: 2171–2185. [CrossRef] [Google Scholar]
  • Hordyk A, Ono K, Sainsbury K, Loneragan N, Prince J. 2015. Some explorations of the life history ratios to describe length composition, spawning-per-recruit, and the spawning potential ratio. ICES J Mar Sci 72: 204–216. [CrossRef] [Google Scholar]
  • Hsieh C, Yamauchi A, Nakazawa T, Wang W-F. 2010. Fishing effects on age and spatial structures undermine population stability of fishes. Aquat Sci 216: 165–178. [CrossRef] [Google Scholar]
  • Imbrock F, Appenzeller A, Eckmann R. 1996. Diel and seasonal distribution of perch in Lake Constance: a hydroacoustic study and in situ observations. J Fish Biol 49: 1–13. [Google Scholar]
  • Jackson JBC, Kirby MX, Berger WH, Bjorndal KA, Botsford LW, Bourque BJ, Bradbury RH, Cooke R, Erlandson J, Estes JA, et al., 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science 293: 629–637. [CrossRef] [PubMed] [Google Scholar]
  • Jacquet S, Cachera S, Crépin L, Goulon C, Guillard J, Hamelet V, Hustache JC, Laine L, Perney P, Raphy J, et al., 2022. Suivi environnemental des eaux du lac du Bourget pour l’année 2021. Rapport INRAE-CISALB. 174p. [Google Scholar]
  • Jacquet S, Cachera S, Crépin L, Goulon C, Guillard J, Hamelet V, Hustache JC, Laine L, Perney P, Raphy J, et al., 2023. Suivi environnemental des eaux du lac du Bourget pour l’année 2022. 175p. [Google Scholar]
  • Jacquet S, Domaizon I, Anneville O. 2014a. The need for ecological monitoring of freshwaters in a changing world: a case study of Lakes Annecy, Bourget, and Geneva. Environ Monit Assess 186: 3455–3476. [CrossRef] [PubMed] [Google Scholar]
  • Jacquet S, Kerimoglu O, Rimet F, Paolini G, Anneville O. 2014b. Cyanobacterial bloom termination: the disappearance of Planktothrix rubescens from Lake Bourget (France) after restoration. Freshw Biol 59: 2472–2487. [CrossRef] [Google Scholar]
  • Janjua MY, Gerdeaux D. 2009. Preliminary trophic network analysis of subalpine Lake Annecy (France) using an Ecopath model. Knowl Manag Aquat Ecosyst 02. [CrossRef] [EDP Sciences] [Google Scholar]
  • Jurvelius J, Marjomäki T, Hirvonen E, Lilja J, Riikonen R. 2006. Vendace (Coregonus albula) stock assessment in winter using a mobile echo-survey under ice. Boreal Environ Res 11: 415-420. [Google Scholar]
  • Kangur K, Ginter K, Kangur A, Kangur P, Möls T. 2020. How did the late 1980s climate regime shift affect temperature-sensitive fish population dynamics: case study of vendace (Coregonus albula) in a large north-temperate lake. Water 12: 2694. [CrossRef] [Google Scholar]
  • Keller V, Müller C. 2012. Breeding colonies of Great Cormorants Phalacrocorax carbo in Switzerland 2012. Swiss Ornithol. Inst. Sempach. [Google Scholar]
  • Keskar A, Raghavan R, Kumkar P, Padhye A, Dahanukar N. 2017. Assessing the sustainability of subsistence fisheries of small indigenous fish species: fishing mortality and exploitation of hill stream loaches in India. Aquat Living Resour 30: 13. [CrossRef] [EDP Sciences] [Google Scholar]
  • Kleisner K, Zeller D, Froese R, Pauly D. 2013. Using global catch data for inferences on the world’s marine fisheries. Fish Fish 14: 293–311. [CrossRef] [Google Scholar]
  • Lappalainen A, Hyvönen J, Söderkultalahti P, Heikkinen J. 2020. Estimating annual CPUE indices for perch (Perca fluviatilis) from monthly logbook data of a gill-net fishery in the Bothnian Bay, Baltic Sea. BOREAL Environ Res 25: 91–103. [Google Scholar]
  • Larijani M, Kordjazi Z, Patimar R, Bandani G, Ansari Z, Jafarian H. 2024. The management of common carp fisheries using spawning potential ratio in the southeast of the Caspian Sea. Fish Manag Ecol 31: e12660. [CrossRef] [Google Scholar]
  • Legault CM, Brooks EN. 2013. Can stock-recruitment points determine which spawning potential ratio is the best proxy for maximum sustainable yield reference points? ICES J Mar Sci 70: 1075–1080. [CrossRef] [Google Scholar]
  • Lemaire M, Guillard J, Anneville O, Lobry J. 2020. Major biomass fluctuations in lake food webs – an example in the peri-alpine Lake Annecy. J Gt Lakes Res 46: 798–812. [CrossRef] [Google Scholar]
  • Linløkken A. 1995. Monitoring pelagic whitefish (Coregonus lavaretus) and vendace (Coregonus albula) in a hydroelectric reservoir using hydroacoustics. Regul Rivers Res Manag 10: 315–328. [CrossRef] [Google Scholar]
  • Lorenzen K, Cowx IG, Entsua-Mensah REM, Lester NP, Koehn JD, Randall RG, So N, Bonar SA, Bunnell DB, Venturelli P, et al. 2016. Stock assessment in inland fisheries: a foundation for sustainable use and conservation. Rev Fish Biol Fish 26: 405–440. [CrossRef] [Google Scholar]
  • Lukin AA, Sharova YN, Prishchepa BF. 2006. The impact of fishery on the state of populations of the European whitefish Coregonus lavaretus in Lake Imandra. J Ichthyol 46: 383–390. [CrossRef] [Google Scholar]
  • Lyach R. 2020. The effect of fishing effort, fish stocking, and population density of overwintering cormorants on the harvest and recapture rates of three rheophilic fish species in central Europe. Fish Res 223: 105440. [CrossRef] [Google Scholar]
  • Lynch AJ, Cooke SJ, Deines AM, Bower SD, Bunnell DB, Cowx IG, Nguyen VM, Nohner J, Phouthavong K, Riley B, et al. 2016. The social, economic, and environmental importance of inland fish and fisheries. Environ Rev 24: 115–121. [CrossRef] [Google Scholar]
  • Lynch AJ, Taylor WW, Beard TD, Lofgren BM. 2015. Climate change projections for lake whitefish (Coregonus clupeaformis) recruitment in the 1836 Treaty Waters of the Upper Great Lakes. J Gt Lakes Res 41: 415–422. [CrossRef] [Google Scholar]
  • Maunder MN, Sibert JR, Fonteneau A, Hampton J, Kleiber P, Harley SJ. 2006. Interpreting catch per unit effort data to assess the status of individual stocks and communities. ICES J Mar Sci 63: 1373–1385. [Google Scholar]
  • 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. [CrossRef] [Google Scholar]
  • Meyer R, Millar RB. 1999. Bayesian stock assessment using a state–space implementation of the delay difference model. Can J Fish Aquat Sci 56: 37–52. [Google Scholar]
  • Moiron M, Rimet F, Girel C, Jacquet S. 2021. Die hard in Lake Bourget! The case of Planktothrix rubescens reborn. Ann Limnol Int J Limnol 57: 19. [CrossRef] [EDP Sciences] [Google Scholar]
  • Mouget A, Goulon C, Axenrot T, Balk H, Lebourges-Dhaussy A, Godlewska M, Guillard J. 2019. Including 38 kHz in the standardization protocol for hydroacoustic fish surveys in temperate lakes. Remote Sens Ecol Conserv 5: 332–345. [CrossRef] [Google Scholar]
  • Müller R, Breitenstein M, Bia MM, Rellstab C, Kirchhofer A. 2007. Bottom-up control of whitefish populations in ultra-oligotrophic Lake Brienz. Aquat Sci 69: 271–288. [CrossRef] [Google Scholar]
  • Mullon C, Fréon P, Cury P. 2005. The dynamics of collapse in world fisheries. Fish Fish 6: 111–120. [CrossRef] [Google Scholar]
  • Munyandorero J. 2012. A recruitment-mortality model in the precautionary management toolkit of African tropical inland, single-species fisheries. Fish Res 127–128: 26–33. [CrossRef] [Google Scholar]
  • Myers JT, Yule DL, Jones ML, Ahrenstorff TD, Hrabik TR, Claramunt RM, Ebener MP, Berglund EK. 2015. Spatial synchrony in cisco recruitment. Fish Res 165: 11–21. [CrossRef] [Google Scholar]
  • OFEV. 2016. Le lac de Neuchâtel - Qualité de l'eau du lac. Off. Fédéral Environ. OFEVDiv. Eaux,. [Google Scholar]
  • OFEV. 2021. Statistiques de peche, Office Fédéral de l’Environnement. [Google Scholar]
  • Østbye K, Bernatchez L, Næsje TF, Himberg K-JM, Hindar K. 2005. Evolutionary history of the European whitefish Coregonus lavaretus (L.) species complex as inferred from mtDNA phylogeography and gill-raker numbers. Mol Ecol 14: 4371–4387. [CrossRef] [PubMed] [Google Scholar]
  • Parent E, Rivot E. 2012. Introduction to hierarchical Bayesian modeling for ecological data. CRC Press. [CrossRef] [Google Scholar]
  • Pauly D. 1980. On the interrelationships between natural mortality, growth parameters, and mean environmental temperature in 175 fish stocks. ICES J Mar Sci 39: 175–192. [CrossRef] [Google Scholar]
  • Pauly D. 1985. Quelques méthodes simples pour l’estimation des stocks de poissons tropicaux. FAO Doc Tech Pêche 234: 56p. [Google Scholar]
  • Pepin P. 2016. Reconsidering the impossible — linking environmental drivers to growth, mortality, and recruitment of fish. Can J Fish Aquat Sci 73: 205–215. [CrossRef] [Google Scholar]
  • Perga M-É, Lainé L. 2010. Changes in the zooplankton of Lake Geneva. Rapp Comm Int Prot Eaux Léman Contre Pollut, Campagne 2009, 95–102. [Google Scholar]
  • Pilling GM, Apostolaki P, Failler P, Floros C, Large PA, Morales-Nin B, Reglero P, Stergiou KI, Tsikliras AC. 2008. Assessment and management of data-poor fisheries. In Advances in Fisheries Science. John Wiley & Sons, Ltd, pp. 280–305. [Google Scholar]
  • Pitcher TJ. 2015. Assessment and modelling in freshwater fisheries. In Freshwater Fisheries Ecology. John Wiley & Sons, Ltd, 483–499. [CrossRef] [Google Scholar]
  • Plummer M. 2023. rjags: Bayesian Graphical Models using MCMC_. R package version 4–14, https://CRAN.R-project.org/package=rjags. [Google Scholar]
  • Plummer M, Best N, Cowles K, Vines K. 2006. CODA: Convergence Diagnosis and Output Analysis for MCMC, R News, vol 6, 7–11. [Google Scholar]
  • Probst WN, Thomas G, Eckmann R, 2009. Hydroacoustic observations of surface shoaling behaviour of young-of-the-year perch Perca fluviatilis (Linnaeus, 1758) with a towed upward-facing transducer. Fish Res 96: 133–138. [CrossRef] [Google Scholar]
  • Rautureau C, Goulon C, Guillard J. 2022. In situ TS detections using two generations of echo-sounder, EK60 and EK80: the continuity of fishery acoustic data in lakes. Fish Res 249: 106237. [CrossRef] [Google Scholar]
  • R Core Team. 2023. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/. [Google Scholar]
  • Richards LJ, Schnute JT. 1986. An experimental and statistical approach to the question: is CPUE an index of abundance? Can J Fish Aquat Sci 43: 1214–1227. [CrossRef] [Google Scholar]
  • Ricker WE. 1940. Relation of ‘Catch per Unit Effort’ to abundance and rate of exploitation. J Fish Res Board Can 5a: 43–70. [CrossRef] [Google Scholar]
  • Roberts BJ, Chucholl C, Brinker A. 2024. Coldwater, stenothermic fish seem bound to suffer under the spectre of future warming. J Gt Lakes Res 102351. [CrossRef] [Google Scholar]
  • Rochet M-J, Trenkel VM. 2003. Which community indicators can measure the impact of fishing? A review and proposals. Can J Fish Aquat. Sci 60: 86–99. [CrossRef] [Google Scholar]
  • Rook BJ, Lenart SJ, Caroffino DC, Muir A.M, Bronte CR. 2022. A 90-year record of lake whitefish Coregonus clupeaformis abundances in Michigan waters of the upper Laurentian Great Lakes. J Gt Lakes Res 48: 1618–1635. [CrossRef] [Google Scholar]
  • Rosenberg AA. 2003. Managing to the margins: the overexploitation of fisheries. Front Ecol Environ 1: 102–106. [CrossRef] [Google Scholar]
  • Sarvala J, Helminen H, Ventelä A-M. 2020. Overfishing of a small planktivorous freshwater fish, vendace (Coregonus albula), in the boreal lake Pyhäjärvi (SW Finland), and the recovery of the population. Fish Res 230, 105664. [CrossRef] [Google Scholar]
  • Schnute J. 1985. A general theory for analysis of catch and effort data. Can J Fish Aquat Sci 42: 414–429. [CrossRef] [Google Scholar]
  • Schnute J 1987. A general fishery model for a size-structured fish population. Can J Fish Aquat Sci 44, 924–940. [CrossRef] [Google Scholar]
  • Scott BE, Marteinsdottir G, Begg GA, Wright PJ, Kjesbu OS. 2006. Effects of population size/age structure, condition and temporal dynamics of spawning on reproductive output in Atlantic cod (Gadus morhua). Ecol Model 191: 383–415. [CrossRef] [Google Scholar]
  • Slipke JW, Martin AD, Pitlo JJr., Maceina MJ, 2002. Use of the spawning potential ratio for the upper Mississippi River channel catfish fishery. North Am J Fish Manag 22: 1295–1300. [CrossRef] [Google Scholar]
  • Stewart TR, Mäkinen M, Goulon C, Guillard J, Marjomäki TJ, Lasne E, Karjalainen J, Stockwell JD. 2021. Influence of warming temperatures on coregonine embryogenesis within and among species. Hydrobiologia 848: 4363–4385. [CrossRef] [Google Scholar]
  • Stokes GL, Lynch AJ, Funge-Smith S, Valbo-Jørgensen J, Beard TD, Lowe BS, Wong JP, Smidt SJ. 2021. A global dataset of inland fisheries expert knowledge. Sci Data 8: 182. [CrossRef] [PubMed] [Google Scholar]
  • Straile D, Eckmann R, Jüngling T, Thomas G, Löffler H. 2007. Influence of climate variability on whitefish (Coregonus lavaretus) year-class strength in a deep, warm monomictic lake. Oecologia 151: 521–529. [CrossRef] [PubMed] [Google Scholar]
  • Swain DP. 2011. Life-history evolution and elevated natural mortality in a population of Atlantic cod (Gadus morhua). Evol Appl 4: 18–29. [CrossRef] [PubMed] [Google Scholar]
  • Thorson JT, Munch SB, Cope JM, Gao J. 2017. Predicting life history parameters for all fishes worldwide. Ecol Appl Publ Ecol Soc Am 27: 2262–2276. [Google Scholar]
  • Tran-Khac V, Quetin P, Anneville O’ 2021. Evolution physico-chimique des eaux du Léman et données météorologiques. Rapp Comm Int Pour Prot Eaux Léman Camp. 2020. [Google Scholar]
  • Vainikka A, Jakubavičiūtė E, Hyvärinen P. 2017. Synchronous decline of three morphologically distinct whitefish (Coregonus lavaretus) stocks in Lake Oulujärvi with concurrent changes in the fish community. Fish Res 196: 34–46. [CrossRef] [Google Scholar]
  • Vehanen T, Piria M, Kubecka J, Skov C, Kelly F, Pokki H, Eskelinen P, Rahikainen M, Keskinen T, Artell J, et al., 2020. Data collection systems and methodologies for the inland fisheries of Europe. FAO. [Google Scholar]
  • Vetter EF. 1988. Estimation of natural mortality in fish stocks: a review. Fish Bull 86: 25–43. [Google Scholar]
  • Vonlanthen P, Bittner D, Hudson AG, Young KA, Müller R, Lundsgaard-Hansen B, Roy D, Di Piazza S, Largiader CR, Seehausen O. 2012. Eutrophication causes speciation reversal in whitefish adaptive radiations. Nature 482: 357–362. [CrossRef] [PubMed] [Google Scholar]
  • Walters C.J, Martell SJ. 2004. Fisheries ecology and management. Princeton, USA: Princeton University Press. [Google Scholar]
  • Wedekind C, Vonlanthen P, de Guttry C, Stadelmann R, Stadelmann N, Pirat A, Perroud G. 2022. Persistent high hatchery recruitment despite advanced reoligotrophication and significant natural spawning in a whitefish. Glob Ecol Conserv 38: e02219. [Google Scholar]
  • Welcomme RL, Cowx IG, Coates D, Béné C., Funge-Smith S, Halls A, Lorenzen K. 2010. Inland capture fisheries. Philos Trans R Soc B Biol Sci 365: 2881–2896. [CrossRef] [PubMed] [Google Scholar]
  • Wickham H. 2016. ggplot2: elegant graphics for data analysis. Springer-Verlag New York. [Google Scholar]
  • Winfield IJ, Fletcher JM, James JB, 2008. The Arctic charr (Salvelinus alpinus) populations of Windermere, UK: population trends associated with eutrophication, climate change and increased abundance of roach (Rutilus rutilus). Environ Biol Fishes 83: 25–35. [CrossRef] [Google Scholar]
  • Winfield IJ, Gerdeaux D. 2015. Fisheries in the densely populated landscapes of Western Europe. In Freshwater Fisheries Ecology. John Wiley & Sons, Ltd, 181–190. [CrossRef] [Google Scholar]
  • Zischke MT, Bunnell DB, Troy CD, Berglund EK, Caroffino DC, Ebener MP, He JX, Sitar SP, Höök TO. 2017. Asynchrony in the inter-annual recruitment of lake whitefish Coregonus clupeaformis in the Great Lakes region. J Gt Lakes R., 43, 359–369. [CrossRef] [Google Scholar]

Cite this article as: Bourinet F, Anneville O, Drouineau H, Goulon C, Guillard J, Richard A. 2024. Impact of fishing activities on the population dynamics of European whitefish in four peri-alpine lakes. Int. J. Lim. 60: 15:

All Tables

Table 1

Main physical, trophic, and fishery characteristics of the four studied lakes between 2000 and 2020.

In the text
Table 2

Definition of the prior of the model parameters. MVNormal corresponds to the multivariate normal distribution, 𝒰 to the uniform distribution, Γ to the gamma distribution, 𝒩 to the normal distribution, and Beta to the Beta distribution.

In the text
Table 3

Posterior statistics of the three parameters.

In the text
Table A1

Number of annual observed data used in the model, covering the period between 2000 and 2020.

In the text
Table B1

Length-weight relationship for Coregonus lavaretus collected during gillnetting standardized surveys and with additional data from fisheries for Lake Annecy, based on theequation log (W) = log a + b log(L) (a: intercept and b: slope of the equation). Sample size (N); length (L) in mm; weight (W) in g; minimum (min) and maximum (max) of L & W; Standard error (SE); coefficient of determination (r2), p is the significance of regression (p) with p significant at <0.05.

In the text
Table B2

Mean weight at recruitment (Wrecr) and post-recruitment (Wpost) stages in grammes for the four studied lakes.

In the text

All Figures

thumbnail Fig. 1

Location of the four studied lakes. FR = France; SW = Switzerland; IT = Italy. Purple circles represent cities with over 100,000 inhabitants. Adapted from Bourinet et al. (2023).
In the text
thumbnail Fig. 2

Directed Acyclic Graphic (DAG) of the adapted Delay-Difference Model. Observed data are represented in grey ellipses and unknown data in white ellipses. Arrows represent the causal relationship between two variables. Frames represent the repetition of structure over to lakes (l) or years (y). Parameters and variables are fully described in the text.
In the text
thumbnail Fig. 3

Estimated outputs from the model for (a) total biomass (tons), (b) recruitment (in number of 1000 individuals) and (c) trends in fishing mortality (corresponding to the ratio of the annual F over the mean F from the period). Lines represent the median of posterior estimates, while the shaded areas denote the 95% CI (Credibility Interval). The dots represent observed values after conversion using the model parameter outputs; for both the biomass and recruitment plots, these correspond to the AI (Abundance Index) obtained from the hydroacoustic data.
In the text
thumbnail Fig. 4

Ratio of the biomass estimated by the model (B), divided by the biomass estimated under a no-fishing scenario (Bf=0). Lines represent the median of posterior B/Bf=0 estimates; the shaded areas denote the 95% CI (credibility interval).
In the text
thumbnail Fig. 5

Estimated outputs in tons (t) of (a) recruitment landings, (b) post-recruitment landings and (c) total landings. Lines represent the median of posterior landings estimates; the shaded areas denote the 95% CI (credibility interval); the dots are the observed values of landings.
In the text
thumbnail Fig. B1

Length-weight relationship fitted on each lake dataset. Dots are observed values; lines are the statistically fitted relationship. Yellow = Lake Aiguebelette; Blue = Lake Bourget; Purple = Lake Annecy.
In the text
thumbnail Fig. C1

Estimated outputs from the model for the total biomass scaled to the mean total biomass. Lines represent the median of posterior estimates, while the shaded areas denote the 95% CI (Credibility Interval). The dots represent observed values of CPUE scaled to the mean CPUE.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.