Multiple impacts on multiple species: Landscape connectivity for tropical mammals

Background/Question/Methods

Dispersal is a critical life-history process for many organisms, and helps determine the structure and dynamics of metapopulations and metacommunities. Yet we still lack a general understanding of what affects the movement of individual organisms or the connectivity of landscapes, particularly for large and wide-ranging taxa. Nevertheless, conservation strategies increasingly focus on maintaining or re-creating connectivity in order to facilitate dispersal of organisms through fragmented habitats. This can be difficult without detailed understanding of how connectivity is altered by different types and spatial arrangements of habitat disturbance. Connectivity plans may be most effective when they address the habitat needs of multiple species simultaneously. But this can come at the cost of designating habitat corridors that are less effective for any single species than a corridor tailored specifically for that taxon. There have been few tests of the trade-offs between single- and multi-species connectivity approaches, particularly in the context of multiple disparate anthropogenic impacts. Here we assess habitat selection and landscape-scale connectivity for large mammals in tropical rainforests of Borneo, including measuring species-specific responses to direct exploitation, habitat disturbance, climate change, and interactions among these factors.

Results/Conclusions

Hunting reduced mammal species richness by 31%, and exploitation persisted even in chronically overhunted areas. Newly logged sites had 11% lower species richness than unlogged sites, but areas logged >10 years previously had richness levels similar to old growth. Hierarchical modeling of camera trap data demonstrates that habitat selection varies strongly among species. Carnivores responded negatively to habitat disturbance, while several large ungulates responded positively. We used these habitat selection data in conjunction with electrical circuit algorithms to assess landscape-scale connectivity for individual species and multi-species combinations. An “all species combined” connectivity scenario generated total costs for movement across the landscape that were 23% and 27% higher for clouded leopards (Neofelis diardi) and banded civets (Hemigalus derbyanus), respectively, than the connectivity scenarios for those species individually. A “carnivores” multispecies connectivity scenario, however, only increased movement cost by 2% for clouded leopards and banded civets. Likewise, an “herbivores” multispecies scenario provided more effective connectivity than the “all species combined” scenario for sambar deer (Rusa unicolor) and pig-tailed macaques (Macaca nemestrina). We suggest that habitat connectivity plans be tailored to groups of ecologically similar, disturbance-sensitive species in order to maximize their effectiveness at supporting dispersal of organisms across highly heterogeneous landscapes.