“…Although SDMs are used to identify the importance of biotic interactions or other ecological mechanisms (Elith & Leathwick, ; Fraterrigo, Wagner, & Warren, ; le Roux, Lenoir, Pellissier, Wisz, & Luoto, ; Roux, Pellissier, Wisz, & Luoto, ; Sexton et al., ), there are sound statistical reasons to be skeptical of this approach. In particular, abiotic environment often indirectly influences the effects of biotic interactions (Callaway et al., ; Davis, Jenkinson, Lawton, Shorrocks, & Wood, ; Sexton et al., ; Tylianakis, Didham, Bascompte, & Wardle, ) and the resultant correlations among variables (i.e., multicollinearity) can make it difficult to infer how important each variable is for shaping species' distributions (Graham, ).…”