How Honeybees Can Help Us Monitor Pollution
The Observation
After a year of beekeeping, I started noticing something odd: the honey from my urban Manchester hives had a slightly different taste and colour than honey from rural hives I’d sampled.
Not bad. Just different. Darker, more complex, hints of something metallic.
Which raised a question: what exactly is in urban honey?
Bees as Environmental Samplers
It turns out honeybees are extraordinary environmental sampling machines:
Foraging Range
A single colony:
- Forages within 3km radius (~28 sq km)
- Individual bees visit 500-1000 flowers per trip
- Colony makes thousands of foraging trips daily
- Collectively samples millions of data points
No human monitoring network approaches this density.
What They Collect
Bees bring back:
- Nectar (becomes honey)
- Pollen (protein source)
- Water
- Propolis (tree resins)
Each carries traces of whatever’s in the environment:
- Airborne particles
- Surface residues
- Atmospheric deposition
- Soil contaminants
Concentration Effect
Bees don’t just sample – they concentrate:
- Many flowers → one honey cell
- Environmental traces accumulate
- Honey becomes archive of local environment
- Detectable signals from diffuse pollution
The Literature
This isn’t a new idea. Research shows honey contains:
Heavy Metals
- Lead (from traffic, industry)
- Cadmium (from combustion)
- Copper and zinc (various sources)
- Mercury (atmospheric deposition)
Organic Pollutants
- PAHs (polycyclic aromatic hydrocarbons)
- Pesticides and herbicides
- Industrial chemicals
- Particulate matter residues
Radioactive Isotopes
- Chernobyl contamination detected in European honey decades later
- Atmospheric weapons testing signatures
- Industrial sources
Honey essentially archives environmental exposure.
The Question
If bees are sampling the environment anyway, and I’m keeping bees anyway, can I:
- Analyse honey for pollutants?
- Map pollution patterns across Manchester?
- Identify pollution hotspots?
- Track changes over time?
This feels like a perfect Friday night experiment:
- Tangential to main work
- Uses existing bees
- Probably too ambitious
- Might reveal something interesting
- High chance of instructive failure
The Pilot Experiment
Sample Collection (Summer 2019)
Harvested honey from three locations:
- My garden (Didsbury, suburban South Manchester)
- Allotment hives (Levenshulme, more urban)
- Rural comparison (Peak District, control site)
From each:
- 100g honey sample
- Carefully cleaned jars
- Detailed location data
- Collection date recorded
The Analysis Challenge
This is where it got complicated.
Analysing honey for trace metals requires:
- ICP-MS (Inductively Coupled Plasma Mass Spectrometry)
- Or AAS (Atomic Absorption Spectroscopy)
- Laboratory facilities
- Technical expertise
- Expensive
Cost per sample: £150-300
My budget: £0 (self-funded)
Problem: Can’t afford proper analysis.
The Budget Solution
Found several partial solutions:
University Collaboration
Approached chemistry colleagues:
- Some interested in principle
- No available equipment time
- Not their research area
- Would need ethical approval
- Too many barriers
Result: Not feasible
Citizen Science Labs
Investigated community labs:
- Some have basic equipment
- Not heavy metals analysis
- Insufficient sensitivity
- Training required
Result: Not adequate
Commercial Testing
Food safety labs offer analysis:
- Designed for contamination detection
- Limited pollutant range
- Still expensive (£80-150/sample)
- But more accessible
Result: Possible but limited
DIY Approaches
Could test for some parameters:
- pH (simple)
- Conductivity (indicates dissolved ions)
- Colour (correlates with some pollutants)
- Microscopy (particulate matter)
Result: Suggestive but not definitive
What I Actually Did
Phase 1: Visual and Basic Analysis
Compared three honey samples:
Colour:
- Rural: Light amber
- Suburban (mine): Medium amber with grey tint
- Urban: Dark amber, slightly cloudy
Crystallisation:
- Rural: Slow, fine crystals
- Suburban: Moderate, mixed crystal size
- Urban: Fast, coarse crystals (suggests different sugar composition)
Taste:
- Rural: Floral, light, clean
- Suburban: Complex, slightly metallic
- Urban: Strong, mineral notes, hint of bitterness
Particulate: Microscopy showed visible particles in urban samples, less in rural.
Conclusion: Visible differences, consistent with pollution hypothesis, but not conclusive.
Phase 2: Limited Chemical Testing
Sent one sample (urban) to commercial lab:
Tested for:
- Lead
- Cadmium
- Copper
- Zinc
- Arsenic
Results:
- All detected (trace levels)
- Below food safety limits
- Higher than expected for rural honey
- Copper and zinc elevated (consistent with urban sources)
Cost: £120
Usefulness: Proof of concept, but need more samples for pattern
Phase 3: Reality Check
Realised this experiment needs:
- Multiple sample locations (minimum 10-20)
- Multiple time points (seasonal variation)
- Proper controls
- Statistical analysis
- Validated methods
Estimated cost: £2,000-5,000
My budget: Still £0
Reality: This exceeds Friday night experiment scope
What I Learned (Besides Pollution Levels)
Analytical Chemistry Is Expensive
“Just test the honey” assumes:
- Equipment access
- Technical expertise
- Validation protocols
- Quality control
- Interpretation skill
None of this is free or simple.
Citizen Science Has Limits
Great for:
- Data collection
- Observation
- Simple measurements
- Engagement
Difficult for:
- Sophisticated analysis
- Trace detection
- Method validation
- Publication-quality data
Gap between ambition and resources is large.
Collaboration Is Essential
Should have:
- Found chemistry collaborator first
- Designed experiment together
- Secured resources before starting
- Aligned on objectives
Lesson: Even Friday night experiments need planning.
Questions Outpace Answers
Each finding raises new questions:
- Which pollutants matter most?
- How do levels vary seasonally?
- What’s causing spatial patterns?
- Are bees affected by what they’re sampling?
- What about pollen vs. honey?
Complexity exponential.
What Actually Worked
Despite analytical limitations, learned:
Bees Do Sample Widely
Foraging observations confirmed:
- 3km range accurate
- Visit diverse locations
- Bring back environmental traces
- Concentrate materials in hive
The sampling mechanism works as hypothesised.
Spatial Variation Exists
Even crude methods show:
- Urban ≠ rural
- Visual and taste differences
- Consistent patterns
- Detectable signals
There’s real information there.
Honey Archives Environment
Limited testing confirmed:
- Metals present
- Reflects local sources
- Different from control samples
- Temporally stable
Honey does record exposure.
Interest Is High
Discussed with:
- Local beekeepers (very interested)
- Environmental groups (keen to collaborate)
- School teachers (educational potential)
- Residents (care about local pollution)
Network for future work exists.
The Pivot
Can’t do sophisticated analysis (too expensive).
Can do:
- Collect samples systematically
- Build sample library
- Document locations and timing
- Network with others doing similar work
- Wait for cheaper analysis options
New approach: Banking samples for future analysis.
Started collecting:
- 50g samples from each harvest
- Dated and location-tagged
- Frozen for preservation
- Documented thoroughly
Cost: Essentially free (using existing harvest)
Utility: Creates resource for future work
Unexpected Findings
Urban Honey Quality
Despite pollution concerns:
- Urban honey is safe (below safety limits)
- Often more interesting flavor
- Greater floral diversity (parks, gardens)
- Comparable or better than some rural honey
Urban isn’t automatically worse.
Pollution Awareness
Project raised awareness:
- Neighbours asking about air quality
- Local groups interested in monitoring
- Schools wanting education projects
- Policy makers receptive to data
Even partial data stimulates discussion.
Beekeeping Motivation
Knowing bees sample environment:
- Increases appreciation for their role
- Highlights urban ecology complexity
- Connects to broader environmental questions
- Makes beekeeping feel more purposeful
Reframes hobby as research participation.
Current Status
Sample bank: 45 samples (2019-present)
Analysis: Limited (1 sample fully tested)
Cost: ~£150 total
Publications: 0
Learning: Substantial
Is this successful? Depends on definition:
- ✗ Didn’t map Manchester pollution
- ✗ Didn’t publish findings
- ✗ Didn’t secure funding
- ✓ Proved concept
- ✓ Built sample archive
- ✓ Learned methods and limitations
- ✓ Created collaborator network
- ✓ Generated ideas for future work
For a Friday night experiment: success.
What Would Make This “Real” Research?
To do this properly would require:
Resources
- £5-10k for analysis
- Lab access for sample prep
- Collaborators with chemistry expertise
- Ethics approval for environmental work
Design
- Systematic sampling strategy
- Control sites
- Temporal replication
- Validated methods
- Statistical power analysis
Time
- Not Friday nights
- Dedicated project time
- Student or postdoc involvement
- Multi-year commitment
Justification
- Novelty (it’s been done before)
- Impact (how would findings be used?)
- Feasibility (demonstrate capability)
- Funding (convince reviewers)
Moving from “interesting question” to “funded research” is substantial leap.
The Friday Night Experiment Sweet Spot
This project occupies perfect space for Friday night experiment:
Too ambitious for casual hobby
Too unfunded for serious research
Too interesting to abandon
Too educational to regret
It’s:
- Beyond my expertise (learning chemistry, ecology)
- Outside my field (no publication pressure)
- Unfunded (no deliverables)
- Tangential (no career impact)
- Compelling (genuinely curious)
Exactly what Friday night experiments should be.
Broader Implications
Citizen Science Potential
If analytical costs decrease:
- Beekeepers could monitor local pollution
- Create distributed sensor network
- Track temporal and spatial patterns
- Contribute to environmental monitoring
Thousands of beekeepers × systematic sampling = valuable data
Bee Health Questions
If bees concentrate pollutants:
- What’s the exposure to bees themselves?
- Does urban pollution affect colony health?
- Are some locations more risky than others?
- Should beekeeping in some areas be discouraged?
Monitoring might reveal concerning patterns.
Urban Planning
Pollution mapping via bees could inform:
- Green space placement
- Traffic management
- Industrial siting
- Air quality improvement
Novel data source for policy.
What I’d Tell Someone Starting This
Do
- Start small (a few samples)
- Document everything
- Network with others
- Be prepared for high costs
- Treat it as exploration
Don’t
- Expect easy answers
- Assume cheap testing exists
- Underestimate complexity
- Work alone
- Seek publication-quality data (unless funded)
Consider
- Partnering with university chemistry dept
- Crowd-funding analysis costs
- Focusing on one pollutant
- Banking samples for future
- Sharing methods with others
The Meta-Learning
Beyond bees and pollution, this project taught:
Sophistication vs. Accessibility
Interesting questions often require:
- Expensive equipment
- Technical expertise
- Institutional access
- Significant resources
Citizen science faces real barriers.
Collaboration Is Essential
Solo Friday night experiments have limits:
- Can’t do everything yourself
- Need diverse expertise
- Resources require partnerships
- Scale needs networks
Even “independent” work needs collaborators.
Partial Results Have Value
Don’t need complete answers to:
- Learn valuable skills
- Generate interesting questions
- Build capacity for future work
- Contribute to conversations
Perfection isn’t requirement.
Process Beats Outcome
The value here isn’t publications or pollution maps. It’s:
- Hands-on learning
- Connecting domains
- Building networks
- Experiencing research process
- Maintaining curiosity
Journey beats destination.
Future Possibilities
Could this evolve beyond Friday night experiment?
Possible paths:
- Secure funding for proper analysis (requires grant writing, institutional support)
- Find collaborator with chemistry resources (requires networking, shared interest)
- Join existing project (contribute samples to larger study)
- Pivot to education (use samples for teaching, citizen engagement)
- Keep banking samples (wait for cheaper analysis methods)
- Accept limitation (remain interesting failure)
Most likely: Some combination.
Conclusion
Can honeybees monitor pollution? Yes, in principle.
Can I, with limited resources, create a pollution monitoring network? Not really.
Was this worth doing anyway? Absolutely.
The experiment taught me:
- Chemistry is hard and expensive
- Bees are remarkable samplers
- Urban ecology is complex
- Citizen science has real limitations
- Collaboration is essential
- Partial data has value
- Curiosity justifies effort
And created:
- Sample archive (potential future resource)
- Collaborator network
- Public interest
- Educational opportunities
- Ideas for future work
For £150 and some Friday nights (and weekends, and occasional weekdays), that’s good return on investment.
Perfect Friday night experiment. Ambitious enough to be interesting, limited enough to be realistic, instructive enough to be worthwhile, and open-ended enough to potentially lead somewhere unexpected.
Just like separating graphite with tape or levitating frogs, you don’t know where these tangential investigations will go.
That’s the point.