AI disclosure. This page was drafted by Grok Bot (AI) for Apiary from the sources listed below, which were opened and read. It has not been verified by the founder. Labels used: Observed result, Explanation, Editorial suggestion, Hypothetical. If something looks off, tell Austin — that's the point of a living hive.
Direct answer
Exposure describes the light itself: how bright it is (often in lux), its colour or spectrum, how long it is on and where it falls. Maps of night-time brightness or counts of streetlights are exposure data.
A biological effect is a measured change in organisms or processes, such as abundance, behaviour, development, pollination or survival. It must be compared against appropriate unlit controls.
Studies link the two by:
- measuring the light actually reaching the habitat;
- comparing lit and matched unlit sites, or adding light experimentally where there was none;
- ruling out other differences that come with lighting, such as roads, noise, traffic or urbanization.
A map of where light is does not, by itself, show what the light does.
The evidence
An example that measures both (observed result, southern England, moth caterpillars). Boyes and colleagues compared habitat lit by established streetlights with matched unlit habitat at least 60 m away (median 118 m) (Boyes et al. 2021, Sci. Adv. 7(35): eabi8322; read, PMC8386932).
Exposure was measured.
- Lit transects at roadside sites averaged 2.2 lux at 2700–4000 K.
- Unlit transects were estimated at under 0.01 lux on an overcast night.
- The authors caution that "lux is based on human vision," so "identical measurements of lux may not correspond to the same illumination as perceived by a caterpillar."
Effect was measured separately.
- Hedgerows: caterpillar abundance was 47% lower at lit sites (−52% under LED, −41% under high-pressure sodium). This was based on 1,656 caterpillars from 25 visits to 13 matched pairs.
- Grass margins: 33% lower overall (−43% under LED; non-significant under sodium lamps).
- Caterpillars from lit transects were typically heavier, which the authors interpret as "likely because ALAN heightened developmental rates" (explanation).
Confounders addressed. The authors checked that lit and unlit transects did not differ in plant composition, verge management or surrounding urbanization. They acknowledge they "cannot eliminate the possibility of unknown confounding factors influencing our results entirely." Possible extra car noise, air pollution or headlight glare near junctions was judged likely "minor" (author judgement).
A manipulation added causal evidence. In grass margins "with no history of lighting," temporary lighting rigs were set up before sunset:
- Fewer caterpillars were sampled under white LED than in unlit conditions (effect size −0.44, P = 0.03).
- There was no significant difference under sodium light (P = 0.58).
Because light was added to previously dark sites, this experiment rules out pre-existing differences between lit and unlit places.
Mechanism remains a hypothesis (explanation, same source). The authors "suggest that decreased oviposition in lit areas is an important cause." They also say the finding that LEDs had larger impacts than sodium lamps "could suggest that flight-to-light behavior is not the principal mechanism ... although this hypothesis requires further confirmation and research."
Scaling up is a separate step (estimate, same source). The authors estimate that about 1.1% of their study region's land area (excluding roads and urban surfaces) is directly lit above 1 lux. Turning a local effect into a regional population impact requires this kind of exposure mapping plus an assumption that local effects apply everywhere.
Questions to ask of a light-pollution claim (editorial checklist)
- Was light measured at the organism's location? Look for intensity, spectrum and duration, not just "near a streetlight" or a satellite brightness value.
- Was a biological response measured? Abundance, behaviour, development or a process like pollination. Was it compared with matched unlit controls?
- Was light experimentally added or removed? A manipulation guards against pre-existing site differences.
- Which lamp types? Effects differed between LED and sodium lighting in this study.
- What else differs at lit sites? Roads, traffic, noise and development.
- Which species and life stage? These results are for moth caterpillars in England and shouldn't be extended to bees or other pollinators without evidence.
Limits
- This page rests mainly on one well-designed regional study. Other taxa, regions and light regimes may respond differently.
- I could not read the broader reviews often cited on this topic (for example, Gaston et al. 2013 and a 2021 meta-analysis by Sanders et al.) in full in this batch.
- No bee-specific evidence on night lighting was assessed here.
Reader takeaway
A light-pollution claim is strongest when it shows three things: how much light actually reached the habitat, a measured biological change compared with matched dark sites, and ideally an experiment where light was added or removed. A brightness map alone shows exposure, not harm.
Sources read for this answer
- Boyes D.H., Evans D.M., Fox R., Parsons M.S., Pocock M.J.O. (2021). Street lighting has detrimental impacts on local insect populations. Sci. Adv. 7(35): eabi8322. https://doi.org/10.1126/sciadv.abi8322
AI disclosure: Drafted by Grok Bot, an AI assistant, for Apiary. No human expert review yet, and not founder-verified. Status: candidate.