Cannibal locusts may cause swarms

Locust swarms are particularly hard to defend against because they occur unexpectedly, and scientists have not had much luck predicting when or where they will happen. According to new evidence uncovered in the University study, though, the key to understanding and predicting this swarming behavior may be the locusts’ hunger — for each other.

The study shows that swarming locusts bite at each others’ abdomens, and this cannibalistic chase in turn spurs on the massive group swarms, with each insect simultaneously fleeing from predators and chasing its own prey. Since a locust can survive even with half of its abdomen gone, this cannibalism does little to diminish the size of the swarms.

The researchers were motivated by curiosity about what causes the typically isolated insects to form these enormous herds, said ecology and evolutionary biology professor Iain Couzin, who led the research team.

“Contrary to popular opinion, locusts actually hate to be near each other, but if the population density increases, then they switch their behavior and become gregarious,” Couzin said. “Once they start exhibiting this gregariousness, we wanted to understand why they start forming these swarms.”

Previous studies had hypothesized that overcrowding might cause the swarming behavior, but the recent study suggests that the population growth is merely responsible for a resource shortage, specifically a food shortage, which in turn causes the insects to turn on each other for nutrients. This discovery initially stemmed from the mysterious disappearance of several locusts in Couzin’s lab.

“When we were doing our previous studies we noticed that some of the locusts were disappearing, and this was very confusing to us,” Couzin said. “So we went back to the videos of the experiments, and we saw some instances of really violent cannibalism in the videos.”

The cannibalism meant, in some cases, that the insects would vanish from their lab habitats without leaving behind a corpse or any clue about what might have happened to them.

“If you leave them overnight, you could come back in the morning and find just a tiny leg in the locust arena,” said Sepideh Bazazi GS, who also worked on the study.

These discoveries will help scientists analyze when and where swarms are likely to form and allow agricultural workers to better protect their crops from the locusts, several of the researchers said.

“Our results will allow better prediction of when and under which environmental circumstances solitary locust populations will ... start to aggregate, at what densities they will begin to march, and in which direction and at what speed,” Stephen Simpson, a biology professor at the University of Sydney, said in an e-mail. “In this way we will be able to manage locust outbreaks more effectively and with minimal environmental impact, for example by aerial monitoring followed by application of targeted strips of insecticidal treatment, rather than having to blanket-spray large areas of habitat.”

The results may also have important applications outside of pest control, Couzin said. The research sheds light on the field of group dynamics, which looks at collective behavior and interactions within groups.

“[Group dynamics] is becoming a really interesting area because a fundamental problem in biology is scaling up from microscopic to macroscopic properties,” Couzin said. “We’re trying to understand things like how the genome scales up to developments of the body and more generally how we can look at functional complexity at a higher scale.”

Get the best of the ‘Prince’ delivered straight to your inbox. Subscribe now »

Studying animal groups, like locust swarms, is helpful in addressing these types of questions because scientists can manipulate and track these groups, Couzin said. The recent study is particularly important because it provides concrete data about group behavior that cannot be as easily obtained in some of the other fields where group dynamics is relevant, such as physics, economics and robotics.

“Because the field [of group dynamics] is dominated by models and theoretical speculations, empirical data is more than ever needed to validate these models,” Jerome Buhl, a biology professor at the University of Sydney who worked on the study, said in an e-mail. “Locusts are becoming a model system in the emerging field of collective behaviour because a lot is known about this animal at many levels from their physiology to their ecology and their migrations at continental scale.”

Perhaps the most surprising implication of these results for group dynamics is the hostile nature of the swarming activity, Couzin said. Instead of working together as a group, the locusts are instead eating each other and forming swarms only because they are trying to simultaneously flee from and pursue the others.

“I think there was some perception that [swarming] was a cooperative behavior because it helps all of them escape from these nutrient-deprived areas and find new resources,” Couzin said. “We’ve really been able to turn that theory on its head. Instead of being cooperative, the swarms are cannibalistic.”