Midge swarms behave “like clusters of stars”

Species of small flies collectively known as midges routinely experience stronger acceleration-related forces than those endured by human test pilots, new research shows.

Calculations made by UK physicist Andy Reynolds from agricultural institute Rothamsted Research found that midges in flight change direction so rapidly that they encounter a g-force of 10 – the same as the maximum permitted during a sharp turn by a Red Bull Air Race stunt plane.

And while all humans except highly trained and expertly secured pilots would lose consciousness at such levels of acceleration-induced pressure, the insects suffer no ill effects at all.

“Fortunately for midges, insect brains do not move around in the skull,” says Reynolds.

The finding emerged from Reynolds’ modelling of midge swarming behaviour – which is significantly different to the emergent properties of other large groups of animals, such as schooling fish, flocking birds or herd mammals.

In these instances, large groups of animals exhibit forms of global order, wherein most members of the group maintain a roughly uniform distance from each other and show a commonality of velocity and direction change.

Midge swarms behave very differently. 

This was first noted by US researcher Akira Okuba from the State University of New York in a 1986 paper. 

Using stereoscopic photographic recordings, Okuba discovered that in three dimensional flight midges within a swarm move randomly in terms of both direction and acceleration. Nevertheless, the swarm remain swarm, rather than dissipating because of overwhelming randomness. He also noted that inward acceleration for each insect increased in inverse proportion to its distance from the centre.

Okubo described midge swarms as “self-gravitating systems”. Reynolds clarifies, describing them as entities that “maintain cohesion but do not possess global order.”

In his paper, Reynolds explores and quantifies another of Okubo’s insights – that the actions of midge swarms can be described mathematically. In doing so, he entertains more than curiosity. If midge swarm behaviour can be described, it can also be, within limits, predicted – and this has implications for forecasting insect infestation in agricultural areas.

To describe the movements of the swarms, Reynolds used what he describes as “old school physics”, a framework known as the Langevin equation (named after mathematician Paul Langevin, who first formulated it in 1908).

The equation will be familiar to anyone who did high school science, because it is used to describe Brownian motion – the seemingly random movements of particles colliding in fluid.

The exercise produced some surprising results. 

“The equation shows that midge swarms are effectively bound together by gravitational-like forces and so are behaving a lot like clusters of stars,” he says.

Reynolds describes modelled midge swarms as a “coexistence of core condensed phases surrounded by dilute vapour phases”.

The analysis revealed why swarms aren’t destroyed by weather conditions.

“Displace a swarm with a gust of wind and it behaves as a solid despite all that empty space,” explains Reynolds. 

“The swarm consists of a dense inner core and outer vapour phase with strange thermodynamic properties.”

The resilience of swarms to phenomena such as wind gusts, he suggests, explains why midges bred in captivity exhibit different swarm characteristics and are not useful for predicting wild behaviours.

Using the findings, Reynolds now hopes to refine the models to predict swarm movements down to “postcode-level” accuracy.

The study is published in the Journal of the Royal Society Interface.