A summary of the landscape genetics based ranking of the 48 landscape representations, with the parameterisation of the associated accumulated-cost dispersal kernel.

a<p>Each landscape representation measured connectivity as a function of distance (D) and the traversal costs resulting from the inter-feature weights of elevation (E), plan curvature (P), tree and scrub cover (T), river order (R), highway traffic volume (H), and bridge length (B).</p>b<p>Landscape representations were ranked by their Akaike weight (<i>w</i>_i). Only the top ranked cost-surface landscape representations (Σ<i>w</i>_i≤0.95) plus the uniform landscape representation are listed.</p><p><i>r</i>² = coefficient of determination (in all cases <i>p</i>≤0.03), μ_log = the mean of the logarithm of the lognormal dispersal kernel, σ_log = the standard deviation of the logarithm of the lognormal dispersal kernel, maxDC = maximum observed dispersal connectivity value.