Behavioral states in space and time: understanding landscape use by an invasive mammal

Animal movement models can be used to understand species behavior and assist with implementation of management activities. We explored behavioral states of an invasive wild pig (Sus scrofa) population that recently colonized central Michigan, USA, 2014-2018. To quantify environmental factors related to wild pig movement ecology and spatio-temporal landscape use, we predicted wild pig behavioral states relative to land cover type, landscape structure (i.e., edge and patch cohesion), and weather conditions. We used global positioning system (GPS)-collars and monitored 8 wild pigs from 2014-2018. We fit local convex hulls and calculated movement metrics revealing 3 wild pig behavioral states (resting, exploratory, and relocating) and constructed a 3-level model to predict behavioral state probabilities relative to biotic and abiotic conditions. Probabilities of exploratory and resting behaviors were higher nearer to riparian and open herbaceous cover types (oftentimes emergent marsh), indicating that these cover types provided security cover during activity and bedding. Hard mast cover types had a strong positive association with relocating behaviors. More cohesive patches of agriculture and shrub cover types were associated with higher probabilities of exploratory behaviors, while resting was more likely in continuous patches of agriculture (mostly mid-summer corn). The probability of exploratory behaviors increased exponentially with warming ambient temperature. Our results may be used by managers to develop control strategies conducive to landscape and environmental conditions where the likelihood of encountering wild pigs is highest or targeting wild pigs when in a behavioral state most vulnerable to a particular removal technique.

Automated summary

Animal movement models can help understand species behavior and guide management activities. This study investigated the behavioral states of an invasive wild pig population in central Michigan from 2014 to 2018. By analyzing environmental factors and using GPS collars, the researchers predicted the probabilities of different pig behavioral states in relation to land cover type, landscape structure, and weather conditions. The results provide valuable information for managers to develop control strategies based on the landscape and environmental conditions where encounters with wild pigs are most likely, or when pigs are in a behavioral state most vulnerable to removal techniques.