Read our COVID-19 research and news.

New tracking system could show—at last—how pesticides are harming bee colonies

Neonicotinoids, the world’s most commonly used insecticides, effectively thwart many crop pests but they also have insidious effects on vital pollinators: bees. At high doses, these neurotoxins—which wind up in the pollen and nectar the bees collect—harm their memory and ability to gather food. Now, using an innovative tracking technique, researchers have shown that neonicotinoids broadly reduce activity in bumble bee colonies, making bees less likely to care for their young, and making it hard for the colony to regulate nest temperature. The findings could help unravel a long-standing mystery: how the pesticides harm bee colonies.

For years, laboratory studies have shown the damage that neonicotinoids can inflict on individual bees. But it’s much harder to conclusively demonstrate how the pesticides damage entire colonies, which contain hundreds or even thousands of bees, all interacting as one complex “superorganism.” Part of the difficulty is the variability of conditions in nature, where weather, disease, the floral richness of the landscape, and other factors that influence colony health can interact and skew results in unknown ways.

To figure out how the pesticides were affecting colonies, James Crall, an animal behavior biologist at Harvard University, decided to examine the bees’ collective behavior after exposure to the chemicals. But doing so was far from simple. Past efforts to track bees involve dotting them with paint, taping the footage for short periods of time, and then carefully examining and annotating their actions. “It’s hard to track them even for a 5-minute video,” Crall says. “It’s unimaginable to do that for many days for multiple colonies.”

Crall and his team found a solution by turning to tracking software that he had written as a Ph.D. student studying insect flight biomechanics at Harvard. He and colleagues glued uniquely patterned 3-by-4-millimeter tags onto the backs of hundreds of bumble bees. Finally, by adapting robotic equipment from a fruit fly lab, they assembled a moveable platform with two high-resolution cameras. Those cameras can regularly spy on up to a dozen bumble bee colonies, picking up the movement of the tags, and passing them onto computers for analysis.

The group chose bumble bees because they are much easier to work with than the iconic honey bee for two reasons: Their colonies contain hundreds, rather than tens of thousands, of individuals; and they are relatively content to forage in a confined space, whereas honey bees want to fly free outdoors.

The team then gave nine colonies sugar syrup laced with six parts per billion of a common neonicotinoid called imidacloprid, allowing them to feed on it whenever they wanted. Over the 12-day experiment, the overall level of activity of the bees and their social interactions decreased. Whereas bees in control colonies spent about 25% of the night caring for the brood, for example, the pesticide-consuming bees spent less than 20%, the researchers report today in Science. The team discovered that the lethargy was, inexplicably, stronger at night. In a further experiment, Crall and his colleagues showed that imidacloprid can hinder the ability of colonies to regulate their temperature, which they normally do by flexing their muscles and fanning their wings.

It’s important that a hive stay at a constant temperature for the colony’s larvae to develop properly. “That brood is their future. If they don’t take care of them, then there’s a likelihood of an effect on the colony,” says Richard Gill, a bee ecologist at Imperial College London. More broadly, he says, it’s important that the many workers communicate and interact. “All the cogs need to be turning at the right time for the machine to be functioning well,” Gill says. It’s possible that the various pesticide-induced effects could stunt the growth of the colony.

Now that Crall has shown these effects, he plans to develop tools for tracking and manipulating temperature in colonies to learn more about how pesticides and temperature interact. Ultimately, he hopes, the system of automated video surveillance could be used to make pesticide testing faster, cheaper, and more sophisticated. Entomologist Reed Johnson of The Ohio State University in Wooster, who was not involved in the research, thinks that likely. “It’s the future of how we’re going to be looking at pesticide effects.”