From dispersal constraints to landscape connectivity: lessons from species distribution modeling

necessitates knowledge of species requirements for dispersal. Yet, current connectivity modelling and management often implicitly assume that dispersal requirements of species are approximated by characteristics of their occupied habitat (Zeller et al. 2012). Th is set of environments constitutes the species occupied niche (sensu Peterson et al. 2011; Table 1). Th is assumption (i.e. that species ’ occupied habitat or niche approximates conditions suitable for successful dispersal) emerges from the frequent use of connectivity metrics that treat non-habitat regions of the landscape (i.e. the matrix) as homogenously impermeable to movement (Calabrese and Fagan 2004, Smolik et al. 2010, Sawyer et al. 2011). Even when matrix heterogeneity is considered, relative matrix resistances to movement are typically assessed based on the similarity of matrix types to habitat of species (Epps et al. 2007, Eycott et al. 2012). As a direct consequence, connectivity modelling can have poor predictive power (Fletcher et al. 2011) and interventions for connectivity conservation are largely habitat-focused, frequently implemented through habitat consolidation or corridor demarcation (Moilanen et al. 2009, Worboys et al. 2010). Ecography 38: 001–012, 2015 doi: 10.1111/ecog.01306 © 2015 Th e Authors. Ecography © 2015 Nordic Society Oikos Subject Editor: Dominique Gravel. Editor-in-Chief: Jens-Christian Svenning. Accepted 15 December 2014