Collective Escape Dynamics of Blackbuck Herds
Theoretical Ecology and Evolution Laboratory.
Indian Institute of Science, Bengaluru.
2015 - 2021
In this project, we aim to understand how individual behaviour affects the collective response of animal groups to predation-like events in blackbuck herds. : "We overcome many of the challenges associated with such a study via a combination of : "(i) design of a novel and relatively non-invasive simulated threat, : "(ii) a non-invasive UAV-based high-resolution video recording of the collective motion and escape response of the herds, : "(iii) an implementation of the state-of-the-art animal tracking techniques from such high- resolution videos and finally, : "(iv) analyses of the tracked trajectories to investigate the patterns of collective escape using techniques from the physics and social network science literature. : "We expect that in blackbuck herds, which exhibit ephemeral groups due to fission-fusion dynamics, decision-making is egalitarian. More specifically, we address our questions in the following order: We first look at how do group patterns change during the escape response. We then investigate how do these group patterns relate to individual-level behaviours. There are two reasons to focus first on group-level patterns: : "First, there are standard and relatively easy-to-use metrics to measure group dynamical properties. : " Secondly, but importantly, measurement of group-level patterns can provide insights on individual level behaviours. : "Our analysis of group dynamics revealed signatures of four different phases in the group activities: a baseline phase, an escape phase, coordinated motion phase and post-response phase. We demonstrate that the blackbuck herds respond to threats by first increasing group polarisation alignment and group speed (both of these actions happen simultaneously). With a small time-lag, groups then become cohesive while moving away from the threat. We identify the response initiators and the most influential individuals in this collective escape response. We reveal that the most influential individuals are rarely the same as the response initiators. : " Furthermore, we find that when leader-follower relationships exist, they are short-lived and not persistent. We infer that collective escape dynamics in fission-fusion grouping animals is likely driven by egalitarian decision-making processes. : "We investigated the mechanism of group response by studying relative movement patterns (speed correlations) of individuals and learnt that leader-follower relationships emerge during the escape response. We also observed that physical associations between individuals are short-term and are not maintained during the escape response. : "This finding is expected for fission-fusion societies as individuals do not preserve spatial fidelity; here, frequent merge-split events should do lead to a fast turnover of neighbouring individuals. Indeed, blackbuck herds have a relatively high fission-fusion rate and smaller groups may merge and split even within hours. Moreover, these groups do not have a stable hierarchical structure regarding dominance or access to information. Hence, we do not expect stable leader-follower patterns. It would be interesting to investigate such dynamics in other fission-fusion populations having smaller group sizes. We quantitatively characterised individuals who identify the threat and initiate the movement. We expected these individuals, called response initiators, to guide group movement during the response. However, we observe that the most influential individuals are rarely the same as the response initiators. This result suggests that anyone can emerge as a leader in these groups. The probable reason for this would be that an individual’s influence is guided by other factors (intrinsic motivation or habitat cues) rather than access to the information of predatory threat. : "In summary, we argue that individuals exhibit a typical set of collective decision rules in case of threat-like perturbations in the fission-fusion blackbuck society. Despite variation in the group composition (age, sex structure and behaviour) of blackbuck herds, there are various characteristic features of collective escape dynamics across all groups we studied. However, a vital factor which may influence the group dynamics is habitat structure. So far, we have analysed only the videos which were recorded in open grassland habitat, and we may expect that the decision rules and group dynamics to be different for a patchy habitat i. e. shrubby/woody areas. It would be exciting to look into the dynamics of these groups in other types of habitat that could affect visibility and information flow. It is also interesting to investigate the generality of our results to other fission-fusion societies.
Fig: We conduct a natural experiment by approaching the blackbuck herds and record their response behaviour using drones.
Fig: Blackbuck exhibit a diverse array of response patterns, some of those are illustrated here.
Publication:
Rathore A., Vadavalli B. D., Jadhav V., Isvaran K., Torney C., Guttal V. Leadership and information transfer in groups escaping a (simulated) threat in the wild. (Ready to be submitted)