Avoidance of anthropogenic sounds has been measured in many species. The results, which are typically based on observations in limited exposure contexts, are frequently used to inform policy and the regulation of industrial activities. However, the occurrence and magnitude of avoidance may be a consequence of complex risk-balancing decisions made by animals. The importance of the factors in decision-making, such as perceived risks associated with the sounds or prey quantity and quality during sound exposure, is unknown.

Here we address this knowledge gap by measuring the relative influence of perceived risk of a sound (silence, pile driving, and a tidal turbine) and prey patch quality on decision making and foraging success in grey seals (Halichoerus grypus). Seals were given access to two underwater ‘prey patches’ in an experimental pool where fish were delivered at controlled rates to simulate a low density (LD) and a high density (HD) prey patch. Acoustic playbacks were made using an underwater speaker above one of the prey patches (randomised during the study), and three decision and foraging metrics (foraging duration, foraging effort allocation between the prey patches, and foraging success) were measured.

Foraging success was highest during silent controls and was similar regardless of speaker location (LD/HD). Under the tidal turbine and pile driving treatments, foraging success was similar to the controls when the speaker was located at the HD prey patch but was significantly reduced (~16-28% lower) when the speaker was located at the LD prey patch.  Foraging decisions by the seals were consistent with a risk/profit balancing approach.  Avoidance rates depend on the quality of the prey patch as well as the perceived risk.

Policy implications: The results suggest that foraging context is important when interpreting avoidance behaviour and should be considered when predicting the effects of anthropogenic activities. For example, sound exposure in different prey patch qualities may result in markedly different avoidance behaviour, potentially leading to contrasting predictions of impact in Environmental Assessments. We recommend future studies explicitly consider foraging context, and other contextual factors such as behavioural state (e.g. foraging or travelling) and habitat quality.