New research is settling the debate over whether drone congregation areas really exist

A common honey bee party fact is that drones fly to congregation areas to find a queen. But this nugget of knowledge is debated among scientists, since most of the data that led researchers to this conclusion used lures to attract the drones. Drones, therefore, may have ostensibly formed congregation areas even in the absence of an existing gathering, and for some scientists, the very existence of drone congregation areas is still an unsettled question.

But new research from Queen Mary University of London, England, is providing some answers. Dr. Joe Woodgate and his colleagues used a technique called harmonic radar tracking to analyze the flight patterns of individual drones during mating flights. This let them watch where the drones ended up, without using a lure to bias the bees’ natural behavior.

“Many beekeepers and scientists have long suspected that drones gather in large numbers in places that remain stable year after year, but it’s incredibly difficult to study such behavior when the drones are so small and fly so high above us,” says Woodgate. “The vast majority of evidence for congregation areas comes from people using caged queens or queen pheromone lures, which they raise into the air on long poles or dangle from balloons.”

Woodgate explains that since drones will return to a location at which they have smelled a queen, scientists might even have unwittingly created the congregation areas they were studying. To avoid this problem, he and his colleagues glued small radar transponders, which resemble sewing pins, to the drones’ backs, enabling the researchers to track the drones’ flights and see if congregation areas really exist.

A drone with a radar transponder attached to his thorax. Photo by Joe Woodgate.

How to track a bee

Regular radar technology works a lot like sonar, but uses beams of radio waves instead of sound. However, honey bees are too small, and their environment too complex, for this technique to enable tracking of individual bees. With harmonic radar, though, the radio wave doesn’t exactly bounce: The special transponder attached to the bee absorbs the signal, then emits another one at a higher frequency (a harmonic signal), making it easier to detect in an otherwise noisy landscape. 

When a drone equipped with a transponder exits its hive, the researchers can detect the harmonic signal using a large radar receiver parked out in the field. The receiver station has what look like large satellite dishes attached to a rotating turret, which pick up the drone’s signal as it flies around, exploring the outside world. Dr. James Makinson, one of the study’s coauthors, tweets that he “used to live in the back of a radar van,” to help make sure they didn’t lose the drone’s signal.

Harmonic radar receiver equipment parked out in the field. Photo by Joe Woodgate.

Frisky flights

The researchers’ diligence paid off, and they made some surprising discoveries. “The first thing we noticed was that drones seemed to switch between two distinctive modes of flight,” says Woodgate. “They get around from place to place with fast, efficient, straight line flights, but occasionally make an abrupt switch to a very different behavior in which they make tight loops, staying within a small area of space and with a lot of speeding up and slowing down.”

Even drones from different hives make these small, loopy flight patterns at the same places, which appear to be the ever-elusive congregation areas. Whether these are actually mating hot-spots is yet to be proven, but they have similar properties to other species’ mating swarms.

As if tethered by elastic bands, drones venturing farther from the core of these tight congregations would usually reverse course and accelerate back toward the center. “This creates the apparent effect of a physical force, binding the drones to the congregation and allowing a swarm of bees to remain stable over long periods, even though each individual drone only stayed for a few minutes,” Woodgate explains. “This pattern is shared by mating swarms of male midges and other insects.”

However, approximately one in five drones will change its flight path completely from this boomerang pattern, and instead make a beeline to a neighboring congregation. A single drone appears to shop around for virgin queens at different congregation areas, should his initial gallivants be unsuccessful.

My way is the flyway

Despite being interested in the topic for decades, scientists have not yet been able to define consistent attributes that make a good congregation area. As Drs. Cyprian Zmarlicki and Roger Morse describe in 1963, with what sounds like a hint of frustration, “It has so far been almost impossible to define a drone congregation area physically.”

But Woodgate and his team found that drones from different hives appear to navigate the landscape, including to and from congregation areas, using aerial highways, or “flyways,” which are defined by the terrain ― a finding which agrees with previous research. A location’s potential as a congregation area, therefore, probably depends as much on its specific landscape as its convenience, like whether it’s on a flyway. 

The researchers infer that flyways help define congregations because drones were able to efficiently find congregation areas despite being apparently naïve to their location. Radar tracking showed that the drones took short orientation flights, averaging at about 13 minutes and flying around 100 m from their hive — just long enough to recognize their home, but not broad enough to discover new landscape features or congregation areas.

“On subsequent flights, drones were able to go straight to a congregation area without showing any evidence of extensive searching,” Woodgate says, “so whatever they use to guide them must be something they can observe from pretty much anywhere.” Moreover, drones congregated in the same location year after year, so the congregations weren’t discovered by following or learning from their brothers. The drones probably find the party, no invitation needed, by following the lay of the land.

These data explain why, as Drs. Hans and Friedrich Ruttner describe in the 1970s, drones from colonies newly relocated to unfamiliar terrain quickly find and participate in the established congregation areas. At the time, the authors postulated that the drones likely found the congregations using optic clues, and now the data more firmly back that claim up. Pheromonal cues from other drones may also play a role, but this is yet to be investigated in detail.

Woodgate and his colleagues say that visual reconstruction of the drone’s perspective, using path, trajectory, and altitude data from the radar tracks, will further clarify the visual cues that drones might use to find the congregation. “We are using virtual reality to recreate what a drone sees as it approaches a congregation area with the aim of discovering exactly what they are using to guide them,” he says.

Dr. Keith Delaplane, professor and director of the University of Georgia honey bee program, and who was not involved in the research, says that Woodgate and his colleagues skillfully use technology to resolve behaviors down to the individual drone. “Although this study informs emergent processes that sustain the drone congregations, we still don’t know how they achieve permanence year after year,” he adds. “I am confident that this too will be shown to be outcomes of landscape features and emergent properties of the drones themselves.”

How big are congregations, really?

These findings are so fundamental to honey bee reproduction, it is hard to believe the data hadn’t been gathered already. It certainly wasn’t for lack of trying: Scientists have been interested in honey bee mating flights since at least the 1960s. Early investigations into congregation areas involved tying helium balloons to queens, floating them in the air, and recording which locations attracted drones and which didn’t.

From these early studies, the data suggested that congregations were huge — up to a kilometer wide, with tens of thousands of drones swarming at a single location. But other work, also using radar tracking, showed that the congregations can be much smaller, with a diameter of just 100 m and fewer than 100 drones participating at any given time.

It is likely that the lure-based sampling technique attracted drones from not only the nearest congregation, but also more distant ones, blending multiple congregations together into a single sample. Woodgate’s method, however, shows that four distinct congregation areas can be found within just a 500 m transect. Moreover, drones cycle in and out of perusing the congregations and resting back at home, so lure-based sampling over a long period of time likely caught drones which would not have normally all been flying simultaneously, inflating the number of congregation participants, too.

Drone on drone

While using lures to study drone congregation areas has its caveats, it can still be a powerful technique. Julia Mahood, master beekeeper and former president of the Metro Atlanta Beekeepers Association, used mechanical drones (the kind you fly with a remote control) to suspend queen lures for her research on congregation areas. Mahood says that in her experience, having a queen lure isn’t sufficient to spontaneously create a congregation area.

“I’ve had many, many [mechanical drone] flights in areas where the landscape is conducive to hosting congregation areas and found zero bees,” Mahood says. Using radar is superior to lure sampling in some ways, because individual bees can be followed to and from their hives while measuring their position and velocity, but it also has its limits: Radar techniques work best in open fields, but that is not necessarily where drones like to gather.

“I’ve found congregation areas over tree canopy more often than in textbook congregation areas, like where open land has a wind break, like a tree line,” Mahood says. She recalls one apparent congregation which was located above the treetops, when a few hundred meters away there was a soccer field that looked like it should be the perfect location for a congregation. While the influence of the lure Mahood used is difficult to assess, these observations do reveal the uncomfortable notion that what is “textbook” may need to be redefined.

Both old (tying lures to balloons or poles hoisted in the air) and new (radar tracking) techniques work best in open fields, and not noisy spaces like forests. This may have incorrectly led researchers to believe that drones prefer to congregate in open spaces, when really, those were just the easiest congregations to observe. Using mechanical drones can sample more diverse landscapes, which Mahood says “opens a whole new world.”

Uncooperative queens

Drones are not the only bees who need to find a congregation: Queens, too, must efficiently navigate to them, but queen mating behavior is even more difficult to study. Drones appear not to be bothered by having a radar transponder attached to them, despite its long antenna that sticks straight in the air, but queens have trouble with the unusual backpacks.

Using observation hives, Woodgate and his colleagues could see that the queens were not moving normally when equipped with transponders, so the researchers had to come up with a new technique to tag the queens: They glued a small magnet to the queen’s thorax ahead of time, then clipped the transponder to the magnet as the queen exited the hive.

But because they were attached with magnets, the transponders could become dislodged. Indeed, that was the fate of one virgin queen that Woodgate tried to track. Queens were, for some reason, more difficult to track than drones, and may require smaller or more specialized equipment.

The researchers attempted to track the flight paths of almost 100 virgin queens, but only obtained data from three mating flights, which is too few to draw reliable conclusions. “They [queens] did not seem to have any difficulty flying with the transponders so it was not that their ability was affected, it simply seemed that they lost their motivation to take nuptial flights,” Woodgate says, adding that they don’t yet know why this might be.

An unmated queen with a magnet and transponder, ready for flight. Photo by Joe Woodgate.

More strategic mating yards

If scientists can eventually decipher the exact visual cues drones and queens use to follow flyways and what leads them to form a mating swarm, this might let queen producers do a better job of choosing locations for their mating yards. “Many matings with a diverse drone population is key to colony health,” Mahood says. “Queen breeders are very interested in knowing the behavior of drones so they can strategically place their mating nucs.”

Ruttner and Ruttner’s research from decades ago showed that drones can fly up to 4.3 miles (7 km), even traversing mountains, to find a congregation area. Likewise, they discovered that queens could fly up to 3.1 miles (5 km). But these are extreme scenarios: Given the chance, drones would prefer to stay closer to their hives, usually less than 800 m. They are likely only pushed to make extremely long mating flights if there are no suitable congregation areas closer to home.

And with increased distance comes increased risk to both the queens and the drones. The farther they must fly, the more resources they consume, and chances are higher that they might encounter a predator, sudden change in weather, or fatigue. Strategically placing mating nucs and drone source colonies near flyways and convenient congregation locations may increase the chances of queens having successful mating flights.

According to Dr. Lars Chittka, professor of sensory and behavioral ecology and Woodgate’s advisor, the researchers’ efforts to reconstruct drones’ three-dimensional visual experience during flights are well underway. It is possible that, once navigation rules are defined, we may be able to better predict where the flyways and congregations will occur in uncharted landscapes. The secret lives of drones might not remain a mystery for long.

This article originally appeared in the August 2021 issue of American Bee Journal.