Context
Functional landscape connectivity is vital for the conservation of wildlife species. Landscape connectivity models often overlook factors such as mortality and asymmetry in landscape resistance that can have a significant impact on functional connectivity. Individual-based models (IBMs) can be used to explore such factors through the implementation of mechanistic dispersal behavior. Furthermore, population-level patterns of animal dispersal and landscape connectivity resulting from the simulation of alternative landscapes or scenarios of animal behavior can be compared.
Objectives
Use an IBM to evaluate the effects of disperser mortality, asymmetrical dispersal due to landscape structure, and land-use change on the functional landscape connectivity between two populations of reintroduced American martens Martes americana.
Methods
We applied a previously calibrated IBM of marten dispersal to simulate movement between two reintroduced populations in Wisconsin and Michigan, USA. We used machine learning analyses to determine how each factor affected dispersal between populations (connectivity) across five consecutive generations.
Results
Functional landscape connectivity between populations was not always correlated with more traditional dispersal metrics, such as dispersal distance. Mortality had the greatest impact on functional connectivity. Land-use change and landscape configuration affected connectivity mostly when mortality was not incorporated into simulations. These experimental factors had a stronger effect on long-distance dispersal between populations than on more local dispersal.
Conclusions
Conservation planning for landscape connectivity may benefit from accounting for mortality risks within matrix habitat. The development of individual-based models that incorporate landscape heterogeneity and complex animal behaviors when investigating long-distance dispersal can provide unique and specific insights into both landscape connectivity and wildlife conservation.