Drivers of Wolf Activity in a Human-Dominated Landscape and Its Individual Variability Toward Anthropogenic Disturbance

Abstract

Wolves (Canis lupus) exhibit contrasted activity patterns along their distribution range. The shift from diurnal to nocturnal habits within and among populations appears to be primarily driven by localized levels of human activity, with ambivalent responses toward such disturbance reported among populations. Yet, the drivers and the underlying individual variability of temporal avoidance patterns toward human remains unexplored. We equipped 26 wolves with GPS–GSM collars, obtaining 54,721 locations. We used step lengths, turning angles, and accelerometer data from recorded locations to infer activity through hidden Markov models (Conners, M. G., T. Michelot, E. I. Heywood, et al. 2021. “Hidden Markov Models Identify Major Movement Modes in Accelerometer and Magnetometer Data From Four Albatross Species.” Movement Ecology 9, no. 1: 1–16.). We further explored the probability of activity as a function of a set of proxies of anthropogenic disturbance at different spatial scales and its interaction with different periods of the day by fitting population-level and individual-based hidden Markov models. Wolves were predominantly active during dusk and night, yet variations in activity emerged among individuals across day periods. We did not find clear population-level effects of anthropogenic disturbance predictors, as these were masked by a wide range of individual-specific responses, which varied from positive to negative, with inter-individual variability in responses changing according to different predictors and periods of the day. Our results suggest a non-uniform strategy of wolves in adapting their behavior to human-dominated environments, further underscoring the role of vegetation patches acting as functional refuge cover for buffering the effects of anthropogenic disturbance and boosting the persistence of the species in human-dominated landscapes. This study, for the first time, reveals the individual variability in wolf responses to human disturbance. By fitting hidden Markov models to data from GPS–GSM collars deployed on 26 wolves, we found significant variation between individuals in their responses to different levels of anthropogenic pressure and across different times of day, highlighting a non-uniform strategy for coping with perturbations in human-dominated landscapes. Our findings underscore the diverse behavioral adjustments employed by wolves to persist in these environments and highlight the critical importance of vegetation patches serving as refuge cover.