Metapop

Keywords: dispersal, spatial structure, adaptation network, management, Atlantic salmon, exploitation, model, demo-genetic, portfolio effect.

Context. In a context of rapid environmental changes and anthropogenic pressures, there is an urgent need to better understand the responses of species and to provide recommendations on how we can manage and conciliate human activities with the functioning of ecosystems. However, there is often a mismatch between the scale of species functioning, threats, and management actions. In particular, even if dispersal is a ubiquitous trait in organisms, implications of connectivity and spatial structure for adaptation, persistence, and management of populations are still poorly appreciated for numerous species.

Rationale and aim. This is the case of Atlantic salmon (Salmo salar), for which eco-evolutionary dynamics, exploitation and management of populations are rarely considered within a metapopulation perspective. Yet, growing evidence shows individuals and gene flow among populations of salmon. Motivated by this context, we aimed to use an innovative multi-scale approach, from genes to metapopulation, to 1) better understand the capacities of adaptation and persistence of exploited populations of Atlantic salmon and 2) explore management practices that would meet both objectives of conservation and exploitation.

Methods. Using a spatially explicit demo-genetic agent-based model, we simulated a network of interconnected populations of Atlantic salmon based on the metapopulation of Brittany. Through several scenarios, we explored the influence of a gradient of dispersal rates, various spatial genetic structures, the spatial configuration of populations, and spatial management strategies on the demographic (e.g., stability, persistence) and eco-evolutionary (e.g., life history traits, genetic diversity) dynamics of interconnected and exploited local populations.

Main results. We highlighted a non-linear relationship between dispersal rates and the stability of the metapopulation, resulting in an optimal portfolio effect for dispersal rates around 20% in a homogeneous network. Simulations also showed adaptation of local populations was fostered by dispersal between initially diverse populations, providing general support for the adaptation network theory. But the spatial configuration of populations also played an important role in their evolutionary trajectories by modulating dispersal patterns. Ultimately, this thesis showcased no clear benefit from a spatialized management strategy protecting some populations from exploitation over others. Contrasted and complex evolutionary consequences emerged from the different management strategies in interaction with dispersal.

Conclusions. It is critical to account for complex interactions between dispersal, eco-evolutionary processes, and spatial structure of populations to better understand and manage Atlantic salmon response to environmental change and anthropogenic pressures. This work also encourages empirical knowledge about dispersal rates and metapopulation structure of this species, and advocates for conservation actions according to the portfolio and adaptation network approach, protecting connectivity and biocomplexity in the face of uncertainty. It also opens up further perspective of work and illustrates the potential of our modelling approach as a prospecting tool for theoretical and applied research on metapopulations.

Funding. This project was funded by INRAE, Region Nouvelle-Aquitaine and OFB Pole Migrateur.