Research overview: deriving connectivity and resistance surfaces from step-selection

A major application of step-selection analysis is predicting landscape connectivity by converting estimated selection coefficients into a resistance surface, a map of how difficult each location is for an animal to traverse. This overview describes how integrated step-selection models, by jointly capturing selection and movement, can simulate realistic dispersal paths or be combined with least-cost and circuit-theory algorithms to identify corridors. For wide-ranging cats threatened by fragmentation, such empirically grounded resistance surfaces improve on expert-opinion maps by deriving permeability directly from observed movement. The overview discusses validation against independent tracks and genetic data and the risk of extrapolating beyond sampled conditions. Movement-derived connectivity has become central to planning crossing structures and conserving the corridors that link felid populations.