Integrating Remote Sensing and Individual Based Models to Prioritize Rewilding Sites for Significant Carnivore Recovery

Abstract

Trophic rewilding is based on the restoration of large carnivores, but locating possible sites where can
be reintroduced is becoming more and more complex due to habitat fragmentation and human
anthropogenic pressures. Conventional conservation planning applications are usually based on fixed
suitability maps that do not recognize the dynamic, behavioral patterns of movement of predators in
humanized landscapes. This study fills this gap by suggesting the combined use of high-resolution
Remote Sensing (RS) and Individual-Based Models (IBMs) to rank rewilding locations by functional
connectedness. Designed structural habitat features, including canopy architecture and heterogeneity
of woody covers, using the multi-spectral satellite data to produce fine-scale landscape resistance
surfaces. It was these surfaces that parameterized an IBM, which was used to simulate individual
animal trajectories and dispersal success at varying environmental conditions. Find that the high
resolution edge detection significantly enhances the determination of bottlenecks of movement, which
are often ignored by traditional models and discovered that the shape of individual vegetation corridors
and the behavioral response to landscape resistance by individuals are more important than the area
of the total habitat in the persistence of the population. This methodology is helpful as it successfully
fills the gap between spatial ecology and management in that it can be used to identify locations that
not only accommodate occupancy, but also long-term dispersal. This combined RS-IBM solution can
be a powerful, scalable conservationist tool, whereby rewilding can be done in landscape areas with
the highest likelihood of ecological success and coexistence between humans and wildlife.