Abstract Background Understanding the spatial ecology of marine predators is essential for informing conservation in coastal ecosystems experiencing rapid change from ocean-based industries and climate-driven environmental shifts. During the non-breeding season, many Brandt’s Cormorants ( Urile penicillatus ) occupy the Salish Sea, on Canada’s West Coast, a region characterized by intense human use, including commercial shipping, fisheries, and coastal development. As responsive marine predators, shifts in cormorant distribution and behaviour may reflect changes in prey availability and habitat conditions associated with human-induced ecosystem change, making them valuable indicators of ecosystem responses in dynamic nearshore environments. However, the environmental drivers of non-breeding foraging behaviour and the location of key foraging habitats in the Salish Sea remain poorly understood. We applied hidden Markov models to infer behavioural states of Brandt’s Cormorants tracked in the Salish Sea during the non-breeding season. High-resolution GPS data (step lengths and turning angles) were integrated with dive metrics and environmental covariates to classify behavioural states (resting, foraging, transiting) and to evaluate how oceanographic, bathymetric, and diel factors influenced behavioural transitions. Results Three behavioural states were identified: a low-movement, low-dive resting state and two high-movement states representing foraging (frequent diving) and transiting (little or no diving). State classification improved with the inclusion of environmental covariates, with time of day, bathymetry, and current speed emerging as the most influential predictors of movement and behavioural transitions. Foraging behaviour was concentrated during daylight hours and was most likely in shallow areas with strong currents, including the mouth of the Fraser River, Porlier Pass, and Active Pass. Resting primarily occurred at roost sites, with foraging activity clustered nearby. Conclusions This study provides the first fine-scale assessment of non-breeding movement and foraging behaviour of Brandt’s Cormorants in the Salish Sea. By integrating multi-sensor biologging data with state-space modelling, we identify key non-breeding foraging and resting habitats and demonstrate how environmental features structure habitat use in a heavily used and rapidly changing marine system. These findings support conservation planning and emergency response efforts where important cormorant habitats overlap with major marine transportation corridors and other ocean-based activities.
Lalach et al. (Mon,) studied this question.