“…Between-site variables (elevation, Euclidean distance, water, development, forest, scrub, open, agriculture, and riparian) were transformed into resistance surfaces in CIRCUITSCAPE (McRae, 2006). We optimized the valuation of each resistance surface (see Hauser et al, 2020) for more details on optimization) using a linear-mixed effect model (R package lme4; Bates et al, 2015) with Dps as the response variable, each resistance value as the fixed effect, and site as the random effect. Only the optimized resistance values for a given variable, the value with the lowest AICc score via the univariate linear-mixed effect models, were then used in subsequent hypotheses testing.…”