Integrating mechanistic models of landscape change and animal behavior to measure functional connectivity

Abstract Context As land-use change and climate change transform landscapes globally, increases in habitat fragmentation and shifts in habitat composition present challenges for the conservation of wildlife. Behavioral approaches to landscape ecology can explore how animal movement across complex landscapes can drive ecological processes like functional connectivity. By integrating mechanisms that link landscape change to animal behavior, simulation models can project how individuals and populations will respond to novel landscapes. Objectives To understand how dispersal behavior and functional connectivity between populations of American martens respond to changes to the landscape due to land-use and climate. We also seek to demonstrate the explanatory power of integrating mechanistic models of landscape transformation and animal dispersal. Methods We present a novel investigation of behavioral responses to land-use and climate change by linking a landscape-level simulation of forest succession (LANDIS-II) with an individual-based model of animal behavior (SEARCH). Results Climate change caused an increase in suitable cover types for martens, but forest diversity is likely to decline. Both land-use and climate change impacted dispersal behavior of martens and functional connectivity between populations. These effects were not consistent across simulations, as dispersal behavior and success were affected by the direction of dispersal (i.e., source population) due to asymmetry in landscape configuration and its interaction with climate and land-use change. Conclusions This study demonstrates how the integration of behavioral and landscape models can inform conservation in ways that classical modeling cannot and how the use of mechanistic simulation models can produce robust projections about species responses to novel conservation challenges in an uncertain future.