A unified framework for prioritizing habitat and connectivity conservation through analytical sensitivity

Abstract Context Connected habitats underpin key ecological processes such as movement, gene flow, and recolonization. As land-use change accelerates habitat loss and fragmentation, there is an urgent need to identify priority areas for conserving habitat and maintaining connectivity. Objectives We develop a unified spatial prioritization framework for functional connectivity that is grounded in metapopulation theory, accommodates species-specific movement behavior, and captures the dual role of landscape units as both habitat providers and connectivity facilitators. Methods We formalize the landscape as a network in which nodes represent spatial units characterized by habitat quality and edges encode permeability for movement. Pairwise ecological proximities are defined from species-specific movement capacity and an ecological distance that integrates costs across paths, expressed as least-cost distance, effective resistance, expected cost, or survival probability depending on the movement ecology of the focal species. From these proximities and node qualities, we construct the landscape matrix. Based on the sensitivity of the landscape matrix to local perturbations in habitat quality or permeability, we develop a prioritization approach that identifies where local changes most strongly influence landscape-scale habitat connectivity. Results Our framework unifies widely used landscape connectivity metrics across the full spectrum of movement models, including least-cost paths, circuit theory, spatial absorbing Markov chains, and randomized shortest paths. We show that summation-based metrics serve as computationally efficient approximations of metapopulation capacity, enabling evaluation of functional connectivity in large, high-resolution landscapes. We derive sensitivities for these metrics with respect to habitat quality and permeability and show that they correspond to two families of centrality measures: a closeness-like centrality reflecting the influence of habitat quality, and a betweenness-like centrality reflecting the influence of habitat permeability. Applying our prioritization approach based on these sensitivities to a case study involving wild reindeer in Norway, we identify high-value habitat and key movement corridors that closely align with a reference node-removal approach while reducing computation time by several orders of magnitude. Conclusions By linking metapopulation theory with sensitivity analysis, our work provides a unified and scalable framework for spatial prioritization of functional connectivity. It supports evidence-based management of fragmented landscapes and advances both theoretical understanding and practical application in landscape ecology.