citizen science - free-living bees

I’ve been poking around the internet for the next interesting journal article. This month, I typed citizen science + honeybee into Google Scholar, and one paper really stood out: “Monitoring free-living honeybee colonies in Germany: Insights into habitat preferences, survival rates, and citizen science reliability.”  The study compares research-grade observations of free-living honeybee colonies around Munich (N=107), with observations from a much larger citizen science project across Germany (N=423). Where possible, the authors aim to combine high-quality research data with citizen science observations to expand the geographic scope of the study. 

Free-living honeybee takeaways

• 63% of colonies nest in large-diameter, mature trees
• 31% nest in buildings

Why survival was overestimated
Colonies monitored directly by researchers in Munich showed only about 12% overwinter survival. Citizen scientists reported much higher survival (around 29%), but this turned out to be biased due to fewer follow-up reports and poorly timed observations.
Many “surviving” colonies reported by volunteers were actually new swarms discovered early in the season (46% overlapped with the swarming period).
Abandoned nest sites were often not reported, making colonies appear longer-lived than they really were.

Cavity entrance direction: buildings vs. trees
Radar plots show that:

• Tree cavities have a strong preference for south-facing entrances (20%)
• Building cavities show a preference for west (22%) and south (16%)

Cavity entrance height: buildings vs. trees

• Median entrance height for tree cavities: 4 meters (13.1 ft)
• Median entrance height for building cavities: 6 meters (19.7 ft)

Big-picture takeaway
What are your swarm trap plans for 2026? Will they look like unused bee equipment stacked on blocks, or purpose-built swarm traps raised and baited with pheromones? Does this study change how you’ll deploy your traps?  When building and tree cavity entrance direction are combined, even an unlikely north-facing entrance appeared about 10% of the time. In other words, it’s surprisingly hard to get this wrong. From a next-door homeowner’s perspective, having a swarm-trap plan—any plan—is far better than having none at all.

Feral Honey Bees in Urban Environments

 

Above: Exterior view showing the window shutter from outside.
Below: Interior view of the window shutter box with the access panel removed.

Introduction to the Citizen Science Article Earlier this year, I uploaded a photo of a local Atlanta bee nest entrance—found in a cut-off sycamore limb—to the Free Living Bees website. Not long after, I came across a fascinating citizen science journal article featured on the same site. As someone obsessed with swarms, I found it absolutely captivating.

Urban Population Density The research focuses on the city of Belgrade, Serbia—a large and densely populated urban environment. Residents reported both honey bee nests and swarms, often seeking help with their removal. Interestingly, the number of reports increased exponentially with population density (a relationship shown clearly in the paper’s data).  To verify and expand on these reports, the authors conducted phone interviews with the residents and followed up with on-site field studies.

Reported Swarms vs. Nests At first, I assumed that reports of swarms and nests would occur in roughly equal numbers—but not so. The study found 261 swarms compared to only 90 nests, meaning swarms were reported nearly three times more often.  Most swarms (about 78%) were seen at heights between 1–9 meters, remarkably similar to what Seeley describes in Honeybee Democracy (p. 52). Reported nest heights, on the other hand, were concentrated between 3–15 meters (about 74% of cases).

Interior–Exterior Shutter System Jovana Bila Dubaić kindly sent me photos to help clarify the distinction between (1) the wooden window shutter and (2) the wooden window shutter box. Many buildings in Belgrade feature a combined interior–exterior shutter system.
When closed, pedestrians see the wooden window shutter covering the exterior. When retracted, however, the shutter rolls upward into a wooden box located inside the building, above the window. Astonishingly, 30.7% of reported nests were found inside these shutter boxes.

Exploring with Google Street View Curious, I used Google Street View to explore some of Belgrade’s most populated neighborhoods—imagining myself as a scout bee. What I “saw” was extraordinary: street after street lined with windows that offer easy access to those wooden shutter boxes. From a bee’s perspective, it’s a smorgasbord of indoor nesting opportunities.

In summary Don’t be intimidated by the lengthy title: “Unprecedented Density and Persistence of Feral Honey Bees in Urban Environments of a Large SE-European City.” The paper is well worth the read, and if you reach out with questions, you’ll likely find that the lead author, Jovana Bila Dubaić, is remarkably generous with her responses.

hive preparation for winter

 

Refer to Derek’s 2024 paper in the Journal of Thermal Biology, “Are man-made hives valid thermal surrogates for natural honey bee nests?” Derek reports that “bee space above combs increases heat loss by up to ∼70%; hives, compared to tree nests, require at least 150% the density of honey bees to arrest convection across the brood area.”

In a hive, brood heat typically circulates by (1) rising between frames, (2) striking the inner cover, (3) spreading sideways, and (4) descending through the cold bee space.

To improve insulation, I added an external Bee Cozy (R-value 8) and took internal measures to limit brood heat circulation by blocking the bee space above the combs. I used letter-sized plastic sheets (the type made for overhead projectors), placing one spliced and trimmed sheet directly on the top bars beneath each hive’s inner cover.

learning from natural honey bee nests

It’s winter and I’m thinking about my bees in these recent cool temperatures; my curiosity turned to Google Scholar and searching for the beekeeper and engineer - Derek Mitchell. Here I found Derek’s 2024 paper in the Journal of Thermal Biology - Are man-made hives valid thermal surrogates for natural honey bee nests.  This is a highly technical read with an exotic mix of scientific units, but I found take-home nuggets useful to my hobby beekeeping. My goal is not to keep bees in a tree, but to modify my wintering of bees in a thin-walled wood box by learning what bees have, on their own, achieved living in tree cavities.

	Man-made hive	Tree nest
Top ventilation	Beekeeper’s  choice	None
Wall properties	Thin low R-value	Thick high R-value
Comb to wall surface	Continuous bee space separation	Comb attached to the wall with a few small gaps
Comb to roof surface	Continuous bee space separation	Comb maximally attached to tree cavity roof
Comb to bottom surface	Continuous bee space separation made more complex with multiple boxes	One continuous separation
Vertical comb length	Single frame height whether using one or multiple boxes	Continuous

Made with HTML Tables

Using 16 variables adjusted to experimental observation, Derek used computational fluid dynamics (CFD) modeling and compared the convective air flow and metabolic impact to bees in a standard British National Hive (one box of 35 Liters) versus a synthetic tree nest (45 liters).

Derek writes, “bee space above combs increases heat loss by up to ∼70%; hives, compared to tree nests, require at least 150% the density of honey bees to arrest convection across the brood area.”

Derek recommends blocking the bee space above the combs in a man-made hive, which I interpreted as adding a clear plastic sheet resting directly on the top bar frame. I have a box of letter-sized plastic sheets designed for overhead projector presentations - it seems perfect for the task. Stay tuned for how this works out for me as I am over-wintering two hives with a different number of boxes. One hive has two medium boxes, and the other hive has one deep box. I spliced sheets together with tape and trimmed the extended sheet to fit inside my eight-frame box. Each hive received one trimmed plastic sheet beneath its inner cover resting directly on the top bar frame. No doubt the bees will glue the plastic to the top bar frame and this will require some vigorous spring cleanup effort. 🤣

Perhaps you want to give this small modification a try and use different materials (e.g., wood, thick acrylic, or coroplast).

Swarm Trap - May 5th 2024

Swarm summary: It was an unusual swarm trap season.🤣 My neighbors helped me see a swarm select a sycamore tree cavity, and 43 days later this bee nest swarms and selected my swarm trap 7 houses away. Flight vigor at the bee nest has since increased - the bee nest is a gift that keeps giving!

 

I was hoping for something else! 

• 9th March - I installed 4 swarm traps in my neighborhood
• active scouting began at two 40-liter swarm traps hanging from crepe myrtle limbs. Each swarm trap contained:
• 5x deep frames with plastic starter strips
• UV-reactive blue swarm trap entrance
• synthetically produced Nasonov pheromone
• 23rd March in the early afternoon - my traps are passed over as the bees selected a cavity in the sycamore tree trunk using a cut-off limb entrance and active scouting of swarm traps subsided
  
• 29th April - sycamore bee nest entrance was obscured by a huge bee beard - see the 2nd photo.
  

In a roundabout way, the moment I've been waiting for!   

• active scouting restarted at the two 40-liter swarm traps
• 5th May at 5:30PM - bees selected my backyard 40-liter swarm trap and the sycamore bee nest entrance is no longer obscured by a huge bee beard - a flight journey of 290 feet (88 meters) - see the 1st photo.
  
• over the next 6 days after dark, I moved the occupied swarm trap in 5 steps towards my empty backyard hive stand.  The after-dark move was made easier by leveling the next location during daylight hours.
  
• 13th May - I transferred 5 deep swarm trap frames to 8-frame equipment and moved the swarm lure to the 8-frame landing board. No smoking or brushing of bees was required. - see the 5th photo.