Functional connectivity of an imperilled Arctic ungulate – where melting sea ice and human trails increase isolation

Sea ice loss, disturbance, and habitat modification by humans can alter functional landscape connectivity, with negative impacts on wildlife. Connectivity facilitates movement and gene flow, and contributes to genetic diversity, metapopulation dynamics, and species range-shifts under climate change. For Arctic ungulates, which disperse over large areas including sea ice, environmental change threatens further isolation. Protecting habitat and its linkages is critical and depends on identifying such areas at commensurate, broad scales. Using caribou, the world's most vagile species, we modelled and mapped the drivers of connectivity across ca. 2 million km2 of the Canadian Arctic Archipelago — where the pace of climate change is among the fastest and caribou are threatened with extinction. First, we quantified hierarchical genetic structure and identified two discrete groups. Next, using circuit theory and simultaneous multi-surface optimization, we tested whether land- and sea-scape heterogeneity or geographic distance better accounted for movement and gene flow within each group. We show that anthropogenic interference is far-reaching. High Arctic Peary caribou displayed isolation-by-resistance, where glacier cover, low sea-ice concentrations during fall, and human trails impeded connectivity. In contrast, more southerly barren-ground caribou displayed largely unrestricted gene flow. These divergent outcomes underscore that organism-landscape relationships can vary across space and highlight the importance of intra-specific structure and responses. By leveraging genetic data, our study demonstrates how critical movement pathways can be identified, even for remote and imperilled species. Such knowledge supports broad-scale conservation planning, in particular, by accounting for complex organism-landscape relationships, across vast, heterogeneous ecosystems.