Bees delivered less pigment — a proxy for pollen — to flowers across roads than to flowers on the same side, suggesting that streets and highways may act as barriers to pollination.

In a study, published May 10 in the Journal of Applied Ecology, researchers found that even small streets with little traffic seemed to impede pollination by the fuzzy insects, and the effect was worse for smaller bees.

"Obviously, there are implications for the bees, but I think the issue might actually be more important for plants that rely on these insects for pollination, particularly in urban areas or other places where there's a lot of roads, and habitats tend to be fragmented into small areas," said study co-author Gordon Fitch, a Ph.D. candidate in ecology and evolutionary biology at the University of Michigan. "I think this really has the potential to meaningfully limit pollination."

As bees move about landscapes, they shuttle pollen from one plant to another. These pollination services are essential for seed and fruit production in many crops and wild flora. The European honeybee is often the poster child for its kind, but there are thousands of native bee species in the U.S. alone.

Like many insects, bees are facing global declines from anthropogenic drivers such as pesticide use, climate change and habitat loss. Although urban development can destroy or degrade insect habitat, research shows that cities can still support bees and other pollinators. But little is known about how human structures, such as roads, influence insect movement.

"If you have two plants on either side of the road, are bees going to cross the road?" said Chatura Vaidya, a co-author of the study and a Ph.D. candidate in ecology and evolutionary biology at the University of Michigan. "If there's a lot of traffic, and whether the roads are big or small, what sort of impact is that going to have on [the bees'] movement?"

To answer these questions, Fitch and Vaidya established 47 study sites near a small city in Michigan, including roads with between one and five lanes, as well as sites with bike paths or pedestrian paths. 

While mammals and bird movements can be tracked with GPS devices, this technology is still too bulky to use on small-bodied insects; instead, the researchers used fluorescent pigment as a pollen analog to monitor bee movement indirectly.

"We first tested to see that the bees weren't particularly bothered by [the pigment]," Fitch said. "They readily came to flowers that had pigment on them, and then when they visit one of those flowers, they get totally covered in the pigment. And then the next flower they go to, they drop a little bit off."

The team compared two native plant species, which are pollinated by different-sized bees. Wild bergamot, or Monarda fistulosa, has clusters of purple flowers that are visited by larger bees, mainly bumble and honey bees. In contrast, the bright-yellow daisy-like blooms of threadleaf coreopsis, or Coreopsis verticillata, are mostly graced by smaller bees, such as sweat bees.

At each of their study sites, the researchers sprinkled pigment onto the flowers of a potted bergamot or coreopsis, which was put near the roadside. They also placed two plants of the corresponding species without pigment: one across the road and another at an equal distance, but on the same side of the road.

After leaving the plants outside for one day, the researchers took them to a dark spot and used a black light to count the number of flowers with glowing specks. Because wind and nonpollinating insects could also transfer pigment, the researchers only considered flowers with pigment on their reproductive parts to have been transferred by bees.

To corroborate that bees — rather than other pollinating insects such as butterflies or flies — were responsible for moving pigment between flowers, the team recorded floral visitors at each of the study sites and found that 97% were bees. 

According to Fitch, the main finding was that plants across roads received less pigment than those on the same side of the road, suggesting that bees were deterred by these asphalt expanses.

"So these were really clear effects: that bees were not traveling across roads," Fitch said. "One of the things that most surprised us was that even on these really quiet residential roads, where maybe a few cars go by an hour, we still saw an effect on bee movement." 

To look at why roads might deter the pollinators, Fitch and Vaidya analyzed features such as traffic volume, road width and speed limit. They found that wider roads were associated with reduced pigment transfer, while traffic speed and volume were less important.

"Once you got about three lanes of traffic or more, we very rarely saw any pigment getting moved across the road," Fitch explained.

According to Vaidya, more research is needed to tease out the precise reasons why bees don't tend to cross large roads, but she suspects it's a combination of factors.

"A road in general is this paved surface where there are no flowering resources, so there's no reason for bees to fly over," she said. "From having plants, you suddenly have this entire stretch of land which has nothing — it's bare. So we think that could be one of the reasons why they are not traveling over roads." 

Another finding was that in bergamot, which attracts big bees, pigment transfer was reduced by about one-third when a road was between plants, while in coreopsis, which is mostly visited by small bees, the presence of a road reduced pigment transfer by about half.

These findings indicate that roads pose more of a barrier to little bees, something that the researchers expected because of other research that found bee body size was related to dispersal ability and flight distance. As cars whoosh down roads, the wind that they generate may also disrupt smaller bees' flight.

Roads and highways fragment landscapes and obstruct the movement of mammals, birds and other animals. This study suggests that roads also obstruct bee movement and the pollination services that they provide, which could restrict gene flow among plants. According to Fitch, this could have implications for plant reproduction and negatively affect populations, something he would like to investigate in future work. 

"Similarly, if bees are not able to go from one habitat to another because there are many roads cutting through habitats, then you could also find bottlenecks of populations within these species, as well," Vaidya said. "That could be especially true for smaller bees than for larger bees." 

The researchers say their findings could inform more bee-friendly urban planning.

"Cities are not all bad for bees, and how we plan our cities, and particularly plan roadways, can really have a big impact on the trajectory of bee populations," Fitch said. "Having some space [along roads or sidewalks] that's dedicated to flowers or even planter boxes can be a way to provide habitat that reduces the degree to which these roads are barriers."

The study, "Roads pose a significant barrier to bee movement, mediated by road size, traffic and bee identity," published May 10 in the Journal of Applied Ecology, was authored by Gordon Fitch and Chatura Vaidya, University of Michigan.