Connectivity is an important determinant of population dynamics, genetic differentiation, and biodiversity conservation because it affects key processes such as migration and dispersal, population growth, gene flow, and ultimately population resilience (Luque et al. 2012; Kool et al. 2013). Although high connectivity can also lead to negative demographic consequences in some circumstances, such as through the rapid spread of disease (e.g., Borg et al. 2017) or the facilitation of natural predation affecting source-sink dynamics (e.g., Olin et al. 2023), in general, it has been shown to mitigate local and regional ecological perturbations and/or overexploitation, e.g., by increasing overall population resilience and allowing negatively affected areas to be repopulated in cases of localised extinction (Hilborn et al. 2003; Gido et al. 2015).
In large aquatic systems, subpopulations of fish are connected either via the passive dispersal of eggs and larvae or via the active movement of juvenile and adult individuals (Brown et al. 2016). While passive dispersal is often studied using hydrodynamic models predicting particle movement (Palmas et al. 2017), for monitoring the active movement of juveniles or adults, telemetry offers a suitable toolbox to determine exchange processes, provided that the spatial scale of study is tractable (Matley et al. 2022). The ultimate outcome of dispersal can also be inferred from genetic techniques that assess the differentiation or relatedness among subpopulations and reveal the heritable consequences of dispersal (Riginos et al. 2014).
Population genetic approaches, however, primarily give an insight into genetic population structure resulting from past gene flow. This depends on the total number of dispersers between subpopulations over intergenerational time scales (Lowe and Allendorf 2010) as well as patterns of adaptive evolution and the related selective removal or enhancement of specific genotypes (Freeland et al. 2011). Although patterns of genetic connectivity emerging from population genetic analyses are fundamental for delimiting the stocks and identifying evolutionarily significant management units (Hawkins et al. 2016), such techniques are not always well aligned to capture ecological connectivity among habitats at year-to-year scale (Lowe and Allendorf 2010; Hawkins et al. 2016). In other words, whereas genetic connectivity provides information on the degree to which gene flow affects evolutionary processes over generational scales (Lowe and Allendorf 2010), ecological connectivity is of central importance for shorter-term ecological and fishery dynamics, such as population growth and vital rates influenced by dispersal as well as local abundance (Nichols et al. 2000; Runge et al. 2006). Decrease or interruption in ecological connectivity may not have an immediate effect on population genetic structure (Marandel et al. 2018), yet, it is highly relevant to local management decisions because it affects, for instance, the risk of localised overfishing, which may be overlooked when solely long-term evolutionary outcomes are considered (Hawkins et al. 2016). That is because even very small levels of exchange may contribute enough gene flow so that the subpopulations in different habitats remain genetically homogenous (Lowe and Allendorf 2010; Hawkins et al. 2016). For example, in a metapopulation, defined as an assemblage of discrete local groups with limited dispersal between them (Hanski and Simberloff 1997), ecological connectivity may be sufficient to maintain genetic panmixia, but low enough to inhibit the rapid recovery of subpopulations when a local mortality event occurs (Hawkins et al. 2016). Furthermore, safeguarding biocomplexity in such populations, which incorporates the diversity of spawning strategies and other behavioural adjustments in animals living in complex ecological systems, has been shown to be critical for achieving long-term stability and high productivity (Hilborn et al. 2003; Schindler et al. 2010).
Only by combining methods that track ecological (on year-to-year time scales) and genetic connectivity (on long-term evolutionary time scales) can the degree of population connectivity be fully characterised and appropriate management and conservation recommendations be tailored to different objectives (Lowe and Allendorf 2010; Hawkins et al. 2016). Over time, telemetry researchers and evolutionary geneticists have independently developed increasingly fine-tuned methods (see Matley et al. (2022); Nathan et al. (2022) for telemetry; and Benestan (2020); Hohenlohe et al. (2021) for genetics) but both communities remain largely siloed (for exceptions, see, e.g., Moore et al. (2017); Hahn et al. (2019); Finlay et al. (2020)). Our work combines these two toolsets by integrating and linking behavioural and genetic data to analyse population structure and understand the complex role of dispersal and connectivity on ecological and evolutionary time scales in a coastal population of northern pike (Esox lucius, hereafter ‘pike’) population in the southern Baltic Sea. Based on an improved understanding of the behaviour and genetic population structure, we derive implications for management and conservation of pike in brackish lagoons in the Southern Baltic Sea.
The Baltic Sea is one of the world's largest brackish water bodies, characterized by a strong salinity gradient from sea salinity (30 Practical Salinity Units, PSU) in its western part, connected to the North Sea, to almost fresh water (2 PSU) in the northeast (Schubert et al. 2017). Also, on local levels salinity gradients are pronounced, especially in the coastal areas, such as, for example, the lagoon network in the southern Baltic Sea, where salinities can range from almost freshwater oligohaline (< 5 PSU) to mesohaline conditions (< 18 PSU) (Arlinghaus et al. 2023b). These regional and local ecological gradients in salinity shaped a unique species assemblage, comprising both marine and freshwater species (Wennerström et al. 2013). Pike, a large-sized predator typical of freshwaters in the northern hemisphere (Craig 2008), is widely distributed throughout the coastal waters of the Baltic Sea where salinity does not exceed 18 PSU (Dahl 1961), taking advantage of an abundant foraging environment that provides access to energy-rich marine prey like herring (Clupea harengus) (Winkler 1987).
For freshwater fishes, successful survival at these salinity levels requires either evolutionary physiological adaptations allowing them to complete their life cycle in brackish water or development of behavioural traits, such as anadromy, that allow them to forage in brackish areas while continuing to spawn in the adjacent freshwater habitats (Engstedt et al. 2010; Ferguson et al. 2019; Aguirre et al. 2022). Resulting migration during the spawning time can contribute to reproductive isolation of subpopulations, for example, by affecting the timing of spawning among different groups (isolation by time, e.g., Brannon et al. (2004)) or their spawning site preferences (isolation by location, e.g., Neville et al. (2006)) (Kawecki and Ebert 2004; Engstedt et al. 2014). Such spatiotemporal processes may lead to intraspecific differentiation along the ecological gradients, such as salinity, or various habitat patches, where in some species and particular situations, a continuum of ecotypes and/or life-history pathways will be expressed and co-exist (Clemens and Schreck 2021).
Earlier studies have shown that Baltic pike have developed three distinct reproductive strategies to successfully spawn in varying salinity levels. Part of the population has undergone local adaptation and can carry out their complete life cycle, including reproduction, in brackish conditions up to 10 PSU (Sunde et al. 2022). Other individuals undertake anadromous spawning migrations from brackish feeding grounds to freshwater tributaries and wetlands (Engstedt et al. 2010; Sunde et al. 2019; Roser et al. 2023), while some fully reside in freshwater throughout the year, making only occasional forays into brackish areas (Birnie-Gauvin et al. 2019). In addition, natal homing and spawning-site fidelity, mechanisms that contribute to reproductive isolation, are common in pike (Miller et al. 2001; Bosworth and Farrell 2006), and both have been reported in Baltic pike (Diaz-Suarez et al. 2022; Engstedt et al. 2014; Nordahl et al. 2019).
The presence of ecotypes with different reproductive strategies may have a strong influence on both genetic and ecological connectivity within pike population. As there is no wide potential for dispersal in pike during the adhesive egg and larval stages (Bry 1996), pike dispersal is based solely on the movements of juveniles and adults. However, pike is a classically described as a sedentary ambush predator that has a rather small home range outside spawning time (Diana et al. 1977; Kobler et al. 2008; Craig 2008), although some studies in freshwater lakes showed that some individuals can be quite mobile and utilize all available habitats (Haugen et al. 2006). In the Baltic Sea, mark-recapture (Karås and Lehtonen 1993) and acoustic telemetry (Flink et al. 2023; Dhellemmes et al. 2023b) studies showed relatively stationary behaviour and rather small home ranges in coastal pike. However, during the spawning period, which usually takes place from March to May, Baltic pike exhibit increased mobility as they seek to reach the spawning grounds either in freshwater tributaries (Tibblin et al. 2016) or in brackish lagoons (Flink et al. 2023). This suggests that subpopulations mix in various combinations throughout the year: brackish water residents and anadromous fish intermingle in foraging habitats but separate during spawning as anadromous fish move to freshwater habitats where they, in turn, share space with resident freshwater pike.
Thus, on the one hand, the general sedentary lifestyle of pike suggests that ecological connectivity between parts of the population may be low, potentially fostering adaptive divergence on small geographic scales of a few km due to limited exchange between groups residing in different areas (Nordahl et al. 2019). But on the other hand, the occasional bursts of movements during spawning, may connect sites that are otherwise disconnected and potentially contribute to a gene flow among the subpopulations (Möller et al. 2021).
Genetic research on pike across the Baltic Sea showed population structuring shaped by pattern of isolation by distance, where geographically close subpopulations are more similar than geographically distant ones (Laikre et al. 2005; Wennerström et al. 2017). There is also genetic evidence for differences between sympatric anadromous and brackish pike ecotypes, likely in response to physiological salinity adaptations and/or natal homing and site fidelity (Nordahl et al. 2019; Sunde et al. 2022), which was also supported by otolith microchemical analyses (Engstedt et al. 2010; Möller et al. 2019). Similarly, studies in the coastal lagoons of the southern Baltic Sea, our study area, showed the influence of salinity on genetic structure, with pike in certain oligohaline lagoons differenting from pike in nearby mesohaline lagoons, which, in the absence of physical barriers between these areas, suggests that physiological reasons, i.e., salinity difference, may be limiting gene flow between them (Möller et al. 2021; Roser et al. 2023). Furthermore, Roser et al. (2023) demonstrated the occurrence of freshwater spawning activity in our study system and showed that putative anadromous subpopulation appears to be genetically intermediate between mesohaline brackish and freshwater or oligohaline brackish stocks.
Substantial declines in pike abundances were documented in many Baltic coastal areas in recent decades (van Gemert et al. 2022; Bergström et al. 2022; Olsson et al. 2023), calling for a well-informed management actions. To contribute to the understanding and conservation of these declining stocks, our study focused on addressing the following key questions:
-
Does the sedentary lifestyle of lagoon pike cause low ecological connectivity between parts of the population, increasing the potential for local overfishing?
-
Do spawning migrations enhance both ecological and genetic connectivity, or do they rather promote reproductive isolation and population differentiation through fostering behaviourally differentiated ecotypes such as brackish residents, freshwater residents, and anadromous fish?
-
Do patterns of space use and reproductive behaviour of pike align with the overall genetic structure of the population, or do environmental factors such as salinity gradients and geographic distances have a greater influence on current genetic differentiation patterns?
We hypothesized that the pike population in the study area is (H1) composed of several subpopulations with relatively stationary space use and low ecological connectivity among them, (H2) shows spawning site fidelity and behaviourally differentiated ecotypes, and (H3) its genetic structure is driven both by limited ecological connectivity and by environmental factors such as salinity gradients and geographic distances.