Predicting space use patterns of a territorial top predator: from individual movement decisions to Arctic fox space use

Predicting animal space use could greatly improve our understanding and forecasting of ecological processes. Despite growing interest, the development of predictive space use models amenable to the integration of spatial processes into ecological frameworks have yet to reach their full potential.

Using high-resolution tracking data collected at 4-minute intervals from 26 Arctic foxes over five years, we developed a predictive space use model based on a step-selection approach. We assessed fine-scale habitat selection in relation to prey distribution, landscape features, and ecological constraints such as central place foraging and territoriality. We then used these results to build an agent-based model simulating fox space use and evaluated its ability to reproduce observed space use patterns.

Step-selection analyses confirmed that fox movements were driven by habitat type, goose nest density, distance to den, and avoidance of distance to the home range boundary. Agent-based simulations closely matched empirical tracking data and accurately forecasted fox space use, even for individuals excluded from model parameterization.

By developing a predictive model of predator space-use, our study provides a foundation for incorporating additional components of the predation sequence and contributes to more spatially informed approaches in predator-prey ecology.