Circadian Lighting and Melanopic Lux: What the Science Actually Says

Picture an open-plan office at nine in the morning. The desks measure a comfortable 500 lux and the fittings are new. It is entirely possible that the room still delivers far less light to the eye than the desk measurement suggests, in the quantity the body clock actually reads. Whether it does can only be settled by measuring at the eye, and that quantity has a name, a standard and, since 2022, consensus target numbers: melanopic equivalent daylight illuminance, or melanopic EDI.

Few phrases in lighting have been marketed harder and measured less than "human-centric lighting". Underneath the brochures, though, sits a real, standardised and increasingly practical body of science: a fifth photoreceptor in the eye, an international standard that quantifies light for it, and expert recommendations that put actual numbers on how much light a day should contain, and when. This guide keeps the biology short and spends its length on what an architect can draw, calculate and specify.

A Worked Example: The Office That Passes Lux and Fails the Eye

Take a hypothetical office of that kind. The consensus recommendations published in PLOS Biology in 2022 by Brown, Brainard, Cajochen, Czeisler, Lockley and colleagues, written for healthy adults, ask for at least 250 melanopic EDI lux in the daytime, measured at the eye, in the vertical plane, at roughly 1.2 metres height, the eye level of a seated person facing forward [1]. Now measure the room that way. A conventional office at 500 desk lux can give a seated occupant well under 250 melanopic EDI at the eye, because vertical illuminance is usually far lower than the horizontal figure lighting design has traditionally chased, and because warmer spectra deliver less melanopic EDI per photopic lux than cooler ones [1][2].

The fix is rarely "more downlights". It is to light the surfaces the eye actually takes in: ceilings, walls, window reveals. Bright vertical surfaces do circadian work that a narrow beam on a worktop does not. Where daylight can carry the daytime target, let it: outdoor daylight, even under an overcast sky, typically delivers illuminance an order of magnitude beyond ordinary indoor electric lighting, and the consensus authors name daylight as the preferred daytime source [1]. The electric scheme becomes the supplement and the evening story.

The same arithmetic runs in reverse after dark. From at least three hours before bedtime the recommendation is no more than 10 melanopic EDI lux, with spectra depleted in short wavelengths where possible, and in the sleep environment itself no more than 1 [1]. A bright living room in the evening can easily exceed 10. Dim, warm light in the last hours before bed is not ambience indulgence; it is the practical route under the evening ceiling, and tunable-white or dim-to-warm fittings on a control system that can actually execute the curve are its hardware.

The Fifth Photoreceptor, Briefly

Beyond the rods and the three types of cones, the retina holds a small population of intrinsically photosensitive retinal ganglion cells, ipRGCs, containing the photopigment melanopsin [1]. These cells contribute little to the image you see. Their job is to report the overall light environment to the brain's master clock, which uses that signal to synchronise the body's daily rhythms: sleep and waking, hormone release, alertness [1]. Melanopsin is most sensitive to short-wavelength light, peaking around 490 nanometres [2].

The design consequence follows directly: the visual brightness of a space and its effect on the body clock are different quantities, delivered in different proportions by different light sources. A room can be perfectly adequate in lux and still be, biologically speaking, dim.

The Metric: Melanopic EDI

In 2018 the International Commission on Illumination published CIE S 026, a standard system for measuring optical radiation for ipRGC-influenced responses to light [2]. Its central quantity is melanopic EDI: the illuminance of standard daylight, D65, that would produce the same melanopsin stimulation as the light being measured [2]. The metric is a spectral calculation, not a resemblance judgement: the measured spectrum of the source is weighted by the melanopic sensitivity curve and expressed against D65 daylight as the reference illuminant [2][3]. Two sources that look equally "daylight-like" to the eye can score differently, and two sources of the same correlated colour temperature can differ, because the number comes from the full spectrum, not the colour appearance [2][3].

The practical meaning: two luminaires delivering the same 300 lux on the same desk can differ substantially in melanopic terms. Ordinary lux meters cannot see this difference at all. Measuring melanopic EDI needs a spectrally resolved measurement, and calculating it at design stage needs the luminaire's spectral data, which serious manufacturers increasingly publish alongside their photometric files; the CIE provides an open-access toolbox for exactly this calculation [3].

How It Changes by Project Type

Office and education. Daytime performance is the brief. Daylight first where the plan allows, then vertical light at the eye from bright ceilings and walls, with melanopic EDI calculated at design stage from spectral data rather than assumed from colour temperature.

Home. The evening is where the targets are hardest. Put the living room's last-hours scene on dim, warm sources, keep downlights off the sofa sightline, and treat bedroom circuits, en-suite night paths and standby LEDs as a design detail: the 1 lux sleep-environment figure makes night lighting a specification, not an afterthought [1].

Care and bedrooms generally. Nights matter as much as days, with one caution: the consensus figures are recommendations for healthy adults, not clinical guidance, so a care setting should treat them as context and take specialist advice rather than apply them as prescribed targets. Within that scope, where vision is needed during the night, the recommendation stays under 10 melanopic EDI, which points to low-level, warm, shielded night paths rather than ceiling fittings [1].

Hospitality. Restaurants and hotels already dim and warm their evenings for atmosphere; the consensus numbers say that instinct is also the biologically quieter choice. The daytime half of the curve, bright mornings in breakfast rooms and lobbies, is the half hospitality most often misses.

What Building Standards Ask

The WELL Building Standard has operationalised related science in a Circadian Lighting Design feature, which expresses daytime light thresholds in equivalent melanopic lux (EML), WELL's implementation of the melanopic approach [4]. The exact thresholds and compliance paths have changed across WELL versions and published addenda, so this guide deliberately does not quote numbers: a live project should read the feature text for the specific WELL version and addenda it is registered under, on the WELL standard site [4]. The design implication is stable even where the numbers move: a project chasing WELL certification has to calculate melanopic performance at design stage, not retrofit it later.

What the Evidence Does Not Yet Say

Honesty is a specification tool here. The acute effects of light, on alertness, on melatonin suppression, on shifting the body clock, are among the best-evidenced findings in lighting science, and the consensus targets rest on them [1]. Longer-term claims, that a given installation will measurably improve health outcomes for a given population, remain an active research area, and the consensus authors frame their numbers as practical recommendations for healthy adults, not medical prescriptions [1]. Specify to the measured quantities, be sceptical of products that promise cures, and the science will serve the project well.

The fifth photoreceptor has a standard, the standard has numbers, and the numbers can be drawn, calculated and measured. That is the point at which a design intention becomes a specification.

Frequently Asked Questions

Is circadian lighting just tunable-white fittings? No. Tunable white is one way to change spectrum through the day; the targets themselves are about how much melanopic EDI reaches the eye, and when. Daylight, vertical illuminance and evening dimming do most of the work.

Can I specify melanopic EDI from a colour temperature? No. Two sources of the same correlated colour temperature can differ in melanopic terms, because the metric comes from the full spectrum [2][3]. Ask for spectral data and calculate with the CIE toolbox.

What are the consensus numbers again? At least 250 melanopic EDI lux by day, no more than 10 in the three hours before bed, no more than 1 in the sleep environment, measured at the eye [1].

Sources

1. Brown, T.M. et al. (2022), Recommendations for daytime, evening, and nighttime indoor light exposure to best support physiology, sleep, and wakefulness in healthy adults, PLOS Biology: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3001571
2. CIE S 026:2018, System for Metrology of Optical Radiation for ipRGC-Influenced Responses to Light: https://doi.org/10.25039/s026.2018
3. CIE S 026 tutorial and open-access toolbox (worked examples including LED products, displays and daylight): https://files.cie.co.at/CIE_S_026_Tutorial2021_information.pdf
4. WELL standard site, Circadian Lighting Design feature (wording and thresholds vary by WELL version and addenda; confirm the text for the project's registered version): https://standard.wellcertified.com/v20/light/circadian-lighting-design