Estimating functional connectivity in fragmented landscapes using the Stochastic Movement Simulator (SMS)

Background/Question/Methods

It is now well established that species will need to shift their ranges in response to climate change, but, for many species, their ability to relocate may be prevented or substantially restricted by habitat fragmentation. Mitigating action by governmental and conservation organizations has the potential to alleviate the problem to some extent, but only if it is well informed as to which species are most affected, and how habitat connectivity should be improved for the benefit of affected species. Functional connectivity estimates, such as least cost paths (LCP), have provided some slight improvement over structural measures, but have several limitations. Here, we present an alternative individual-based modeling approach, developed to predict dispersal of animals in fragmented landscapes. We applied the model to real landscapes and species, and assessed its performance by comparing its predictions against those inferred from genetic samples.

Results/Conclusions

We show that connectivity can be highly sensitive to a species’ perceptual range and to its movement behavior. We found that SMS can be a substantially better correlate of genetic connectivity estimates than LCP, and is robust to assumptions made regarding landscape spatial grain, although certain further simple species-specific information may be required. SMS therefore has the potential to inform management of fragmented landscapes for conservation of threatened species.