Broad Spectrum Stops at 400 Nanometres. Your Pigment Does Not.

Here is a sentence from the FDA’s own consumer explainer on sunscreen labelling, paraphrased only because I am not going to make you read regulatory prose: “broad spectrum” means the product’s UVA protection is proportional to its UVB protection.

That is the entire claim. UVA and UVB. Two bands, both ultraviolet, both stopping at 400 nanometres, because that is where ultraviolet stops and visible light begins.

Visible light is not mentioned. Not in the definition, not in the testing requirement, not anywhere in the framework. There is no US standard, no mandated test, and no label claim that would tell a shopper whether a sunscreen does anything at all about the wavelengths between 400 and 700 nanometres.

Which would be a footnote, except that for a very large number of people those are the wavelengths that matter most.

The blue-light mechanism, and the part where red light gets exonerated

Melanocytes contain a photoreceptor called opsin-3. It responds to blue light. In 2018, Regazzetti and colleagues published the mechanism: blue light hits OPN3, which triggers a calcium signalling cascade — CAMKII, then CREB, ERK and p38, then upregulation of MITF, the master switch for pigment production. The end of that chain is melanin.

The dose-response has since been mapped, and the detail is where it gets useful. Blue light at 450 to 495 nanometres and green light at 530 to 595 induce melanogenesis specifically in Fitzpatrick IV through VI skin, at doses from 8 to 480 J/cm². A single exposure is enough to produce immediate and persistent darkening in skin types IV to VI. Higher doses, in the 80 to 120 J/cm² range, produce delayed tanning. Shorter wavelengths do more: 415 nm at 50 J/cm² activated opsin-3 and produced what the review calls potent and sustained hyperpigmentation, while 450 nm at 200 mJ/cm² did not activate it at all.

And 630 nm red light — the wavelength in every LED mask on the market — showed no melanogenic effect and was, if anything, inhibitory.

So “visible light causes pigmentation” is too coarse. Blue and green do, at dose, in deeper skin. Red does not. The LED mask is not your problem. The window is, and so is the screen, and so is the overcast afternoon that you correctly judged did not warrant sunscreen.

One more finding from that literature, and it is the one that reframes everything: in individuals with dark skin, visible-light-induced hyperpigmentation was found to be more potent and longer-lasting than UVA1-induced hyperpigmentation. The thing your sunscreen is not tested against outperforms the thing it is.

Iron oxide is the answer, and nobody will tell you how much is in there

Mineral filters do not solve this on their own. Dumbuya and colleagues tested this directly in Fitzpatrick IV skin: an iron-oxide-containing formulation significantly protected against visible-light-induced pigmentation, while a mineral SPF 50+ sunscreen without iron oxide did not. The zinc was not enough. The pigment was what worked.

This is why every dermatologist with an interest in melasma has spent five years telling people to buy tinted sunscreen. The tint is not cosmetic. The tint is the active.

Then you go to buy one.

A 2026 product analysis in the Journal of Drugs in Dermatology went through marketed tinted sunscreens and found disclosed iron oxide concentrations ranging from under 1.4 percent to 10.4 percent. Nearly eightfold, across products sitting on the same shelf, all of them making the same tinted-broad-spectrum noises. There is no regulatory floor. There is no required disclosure. Separate work found that the overwhelming majority of tinted sunscreens surveyed did not list iron oxide on the label at all.

And the threshold everyone repeats — that you need at least 3 percent, or 3.6 percent — I could not trace to a primary trial figure. It circulates constantly in industry and consumer writing. It may well be right. I am not printing it as a fact when the papers it is attributed to do not obviously contain it.

What this leaves you holding

A shopper with melasma, or with the kind of skin where a spot leaves a shadow for four months, is being asked to select for an ingredient at a concentration nobody will disclose, against a threshold nobody can source, to block a wavelength range no label mentions.

That is not a consumer problem. That is a labelling failure, and it lands hardest on exactly the people for whom the ordinary advice — wear SPF 30 — is the least complete.

So: buy tinted, check that iron oxides appear in the ingredient list rather than assuming the colour came from them, and treat the shade range as a rough proxy for how much is in there, since a brand offering four depths is at least having the conversation. Then stop thinking of it as sun protection and start thinking of it as a pigment problem, because it is one. Shade, a hat, the side of the street with the awning, and the window you sit next to every day at work are all doing work that the bottle is not being tested to do.

I would like to be able to tell you to read the label. The label does not say.

This article is educational and is not a substitute for care from someone who can look at your skin. Anything new, changing, painful or spreading should be assessed rather than researched.