Behavioural state‐dependent habitat selection and implications for animal translocations

Abstract Post‐release monitoring of translocated animals is often used to inform future translocation protocols. Quantifying habitat selection of translocated individuals may help identify features that characterize good settlement habitat and thus inform the choice of future release sites. However, translocated animals often undergo post‐release behavioural modification, and their habitat selection may vary depending on the underlying behavioural state. To investigate this, we analysed behavioural state‐dependent habitat selection in female greater sage‐grouse Centrocercus urophasianus translocated from Wyoming to North Dakota, USA, using Hidden Markov Models combined with Integrated Step Selection Analysis. We segmented individual trajectories into behavioural phases corresponding to an exploratory state, characterized by broad and directed movements, and a restricted state, characterized by short and tortuous movements. Then, we quantified habitat selection in each state while accounting for seasonality and individual reproductive status. While in the exploratory state, sage‐grouse exhibited natal habitat preference induction by selecting for high sagebrush cover, which is typical of their natal area in Wyoming but not of the release area in North Dakota. In the restricted state, sage‐grouse selected for gentle topography and also adjusted their habitat selection to constraints imposed by seasonality and reproductive needs by selecting for high herbaceous cover during brood rearing. Synthesis and applications . Habitat selection of translocated sage‐grouse differed between the post‐release exploration and the settlement phase. Features selected after settling, not during exploration, are likely indicative of suitable settlement habitat. Our results suggest that areas characterized by gentle topography and high herbaceous cover are well‐suited as release sites for sage‐grouse translocated during the summer, especially brood‐rearing females, and that sagebrush cover may not be a critical factor in determining the appropriateness of release sites for sage‐grouse in North Dakota. Our findings highlight the need to consider behaviour when using habitat selection estimates to inform the choice of future release sites.