Collective Behaviour of Insect Swarms under External Perturbations
Final Report Abstract
The goal of the project “Collective Behaviour of Insect Swarms under External Perturbations” was to investigate collective mating swarms of the non-biting midge Chironomus riparius in a laboratory setting to quantitatively disentangle the effects of external perturbations, intrinsic collectivity and induced correlations. To do this, we extracted three-dimensional, time-resolved trajectories of all swarming midges, including their velocities and accelerations and performed a variety of experiments with a wide range of well-controlled external perturbations ranging from external flow to lighting signals and visual cues to sound signals. While thermally driven, convective flows do not seem to influence the swarm’s behaviour in any noticeable way, wind, optical and acoustic signals tend to induce strong, collective responses. One previously unsolved discrepancy between laboratory experiments and field measurements has been that natural swarms typically exhibit long-range correlations in the swarm’s velocity fluctuation correlations, while laboratory midge swarms are typically uncorrelated. As a result of this study, we were able to reconcile these findings by showing that any kind of perturbation leads to an increase of correlation lengths within Chironomus riparius swarms. The correlations do not intrinsically arise, but are caused by behavioural changes due external perturbations. In the wild, these perturbations are ubiquitous, resulting in long correlation lengths within swarms. We further developed a framework to quantitatively capture the collective behaviour of these swarms, in the spirit of classical thermodynamics. We were able to define a full set of state variables, namely number, volume, pressure, entropy and temperature and showed that equipartition holds for swarming Chironomus riparius, with each midge contributing nine degrees of freedom. These state variables allowed us to formulate an equation of state, a generalized power law relation between state variables that efficiently reproduces time series of pressure within the swarm. Performing perturbation experiments using interleaved light and sound cues, we were able to move the swarm along a cycle in a pressure-density phase diagram much akin to a thermodynamic engine in classical thermodynamics. Using the equation of state obtained from unperturbed swarms, we were able to reproduce the pressure-density phase diagram for swarms under perturbations, showing that the collective response of the swarm is encoded within the equation of state and that the behaviour of this complex system consisting of many individual living beings can efficiently be described in a statistical manner, using a simple set of state variables.
Publications
- "Response of insect swarms to dynamic illumination perturbations," Journal of the Royal Society Interface 16, 20180739 (2019)
M. Sinhuber, K. van der Vaart, and N. T. Ouellette
(See online at https://doi.org/10.1098/rsif.2018.0739) - "Three-dimensional time-resolved trajectories from laboratory insect swarms," Scientific Data 6, 190036 (2019)
M. Sinhuber, K. van der Vaart, R. Ni, J. G. Puckett, D. H. Kelley, and N. T. Ouellette
(See online at https://doi.org/10.1038/sdata.2019.36)
