Estimating landscape resistance to movement: a review

Resistance surfaces are often used to fill gaps in our knowledge surrounding animal movement and are frequently the basis for modeling connectivity associated with conservation initiatives. However, the methods for quantifying resistance surfaces are varied and there is no general consensus on the appropriate choice of environmental data or analytical approaches. We provide a comprehensive review of the literature on this topic to highlight methods used and identify knowledge gaps. Our review includes 96 papers that parameterized resistance surfaces (sometimes using multiple approaches) for a variety of taxa. Data types used included expert opinion (n = 76), detection (n = 23), relocation (n = 8), pathway (n = 2), and genetic (n = 28). We organized the papers into three main analytical approaches; one-stage expert opinion, one-stage empirical, and two-stage empirical, each of which was represented by 43, 22, and 36 papers, respectively. We further organized the empirical approaches into five main resource selection functions; point (n = 16), matrix (n = 38), home range (n = 3), step (n = 1), and pathway (n = 1). We found a general lack of justification for choice of environmental variables and their thematic and spatial representation, a heavy reliance on expert opinion and detection data, and a tendency to confound movement behavior and resource use. Future research needs include comparative analyses on the choice of environmental variables and their spatial and thematic scales, and on the various biological data types used to estimate resistance. Comparative analyses amongst analytical processes is also needed, as well as transparency in reporting on uncertainty in parameter estimates and sensitivity of final resistance surfaces, especially if the resistance surfaces are to be used for conservation and planning purposes.