Cooperation against invasions: multi-agent control of alien species in connected landscapes

Invasive alien species (IAS) increasingly threaten biodiversity, ecosystem services, and economic sustainability, particularly in fragmented landscapes where management responsibilities are decentralized. The spread of IAS is not confined within administrative borders: it follows ecological connectivity, making isolated local interventions often ineffective. This work presents a dynamic game-theoretic framework for modeling the strategic management of IAS across a network of heterogeneous areas linked by spatial diffusion. Each agent, responsible for local control of the invasion, faces a trade-off between reducing ecological damages and sustaining the economic costs of intervention. The analysis explores how the interaction between spatial structure, ecological features, and decentralized decision-making shapes outcomes under non-cooperative, coalition-based, and fully cooperative strategies. To support the cooperative behavior, fair cost-allocation mechanisms are proposed based on Nash bargaining and the Myerson value, explicitly accounting for spatial externalities. Numerical experiments on a synthetic three-node network illustrate how cooperation can substantially reduce invasion spread and management costs, while strategic defection may exacerbate both ecological and economic losses. The main findings of the analysis underline the vulnerabilities of fragmented management and the need for institutional arrangements to promote adaptive, equitable, and spatially informed strategies for IAS control.