Inferring leader-follower dynamics in three shark species using acoustic telemetry data.

Background: Understanding collective behaviours and interactions in sharks is still in its infancy. Although recent studies have revealed some social structures in several shark species, little is known about complex interactions and social processes such as leader-follower dynamics. Recognising the dynamics in shark populations can help to further understand population structure and the influence of specific individuals.

Methods: We developed a methodological approach to detect and analyse leader-follower behavioural patterns using acoustic telemetry data. By utilising lag-time distributions from acoustic telemetry detections for pairs of individuals we infer directed relationships based on temporal patterns. We applied this method to existing datasets from grey reef sharks (Carcharhinus amblyrhynchos), blacktip reef sharks (Carcharhinus melanopterus), and tiger sharks (Galeocerdo cuvier).

Results: We found evidence of leader-follower behaviour in both reef-associated species, with half of the tagged grey reef sharks forming leader-follower networks at eight locations. Size was a significant influence on female grey reef sharks leading-following behaviour. We found similar behaviours at three locations for blacktip reef sharks, with one-third of the tagged individuals forming separate and non-overlapping networks. Size was a significant influence on male blacktip reef sharks leading-following behaviour. No species showed a significant effect of sex alone on leader-follower behaviours. Aggregating networks did not show an overall hierarchy for either species but showed that grey reef shark coordinated in smaller networks than expected with strong influences from more dominant individuals. We found no leader-follower networks for tiger sharks.

Conclusions: Our methodology reveals leader-follower behaviours in blacktip reef sharks and grey reef sharks and corroborates findings from the literature which have previously either been described using visual observation or using a different analytical approach. We demonstrate how existing acoustic telemetry datasets are a valuable source which can be used to detect social interactions associated with leader-follower behaviours in sharks, especially when visual observations are not feasible. Our approach provides new insights into understanding the social dynamics in sharks and offers a way to be applied to many more species already acoustically tagged.