The UV Index is one of the great successes of public health communication. It compresses a messy piece of atmospheric physics into a single integer, tells you roughly how fast you'll burn, and does it well enough that people actually change their behaviour. If your goal is "don't get scorched today," it's a genuinely good number.
The problem starts when that number gets borrowed for a different job. Ask almost any app how much vitamin D you made this afternoon and, underneath, it will be doing some version of the same calculation: take the forecast UV Index, multiply by something for your skin type, multiply by something for how much skin was showing, multiply by minutes. It looks reasonable. It is also built on an assumption that quietly falls apart in exactly the conditions people care most about — early spring, late autumn, high latitudes, early mornings.
Here's the assumption, stated plainly: that the UV Index is proportional to your skin's ability to make vitamin D. It isn't, and the size of the error moves around.
Two different questions about the same sunlight
Ultraviolet light isn't one thing. It's a band of wavelengths, and biology responds to each wavelength differently. Physicists and photobiologists describe that response with an action spectrum — a weighting curve that says "at this wavelength, this effect is this strong."
Sunburn has one. The erythemal action spectrum (standardised by the CIE) describes how efficiently each wavelength reddens human skin. Take the actual spectrum of sunlight hitting the ground, weight it by that curve, add it up, scale it, and you get the UV Index. That's all the UV Index is: sunlight, weighted by how well it burns you.
Vitamin D has its own. Previtamin D3 forms when UVB photons strike 7-dehydrocholesterol in your skin, and that reaction has a separate action spectrum — one that sits further toward the short-wavelength end of the UVB, and drops off faster as wavelengths get longer. The CIE standardised this one too, and researchers have since worked on refining it from in-vivo human data.
Two curves, two different shapes, applied to the same incoming sunlight. That means two different numbers — and no fixed conversion between them.
Why the gap moves
If the two curves were merely different but the ratio between them were constant, none of this would matter much — you could apply one correction factor and move on. The reason it matters is that the ratio isn't constant. It depends on two things that change every single day.
Sun angle
When the sun is low, its light travels a long, slanting path through the atmosphere. Short wavelengths get absorbed and scattered most aggressively on that journey, so the surviving light is disproportionately stripped of exactly the photons vitamin D needs. The UV Index falls too — but not as fast, because the erythemal curve draws more of its total from longer wavelengths that pass through more easily. A low sun therefore hurts vitamin D more than it hurts your burn risk.
This is the real mechanism behind the familiar "vitamin D winter": it isn't that winter sun is weak in general, it's that the specific slice of the spectrum you need is gone while the rest is still arriving.
Ozone
Stratospheric ozone is a strong absorber of short-wavelength UVB, and the total column of it above your head is not fixed. It varies with latitude, season and weather systems — swings of tens of percent through the year are normal, and day-to-day variation of ten percent or more is unremarkable. Because vitamin D synthesis is weighted toward precisely the wavelengths ozone eats, a thicker ozone column suppresses vitamin-D-effective UV more than it suppresses the UV Index.
Kazantzidis and colleagues examined this relationship directly in 2009, and the conclusion is the one that matters here: the ratio between erythemally weighted and vitamin-D-weighted UV is a function of solar zenith angle and total ozone, not a constant you can hard-code.
The practical version: two afternoons can both report "UV Index 4" and carry meaningfully different vitamin D potential — because one had the sun higher and less ozone overhead than the other. Any calculator that treats UV Index 4 as a fixed quantity of vitamin D is making an error that grows toward the edges of the season, and shrinks near midsummer noon.
What that means for the "UV 3" rule
You'll often see the guidance that you need a UV Index of about 3 before your skin makes meaningful vitamin D. As a rule of thumb it's fine, and it's far better than no rule at all — below roughly that level, the sun is usually too low for enough UVB to survive the trip.
But it's worth seeing it for what it is: a threshold on the wrong axis. UV Index 3 in March at 50° north, with a thick spring ozone column and the sun barely clearing 35°, is a different proposition from UV Index 3 at 6pm in July. Both will tell you the same thing. Your skin will not agree.
The same caution applies to the "you'll burn in X minutes" style of reasoning when it's repurposed for vitamin D. Burn time and vitamin D time are governed by different curves, so they don't scale together — which is why the honest framing has always been "get some sun without burning" rather than any precise minute count. (We wrote about that spread in detail in How much sun do you actually need for vitamin D?)
Why almost nobody corrects for this
Not out of carelessness. The UV Index is simply the number that's available. It's forecast by meteorological agencies worldwide, it's free, it comes through weather APIs, and it arrives as a tidy scalar. The vitamin-D-weighted equivalent — usually called D-UV — is not something your weather provider hands you. To get it you need the sun's position for your exact location and moment, and a source of total-column ozone data, and then you have to do the re-weighting yourself.
So the pragmatic move, industry-wide, has been to use the erythemal number and absorb the error. For a summer afternoon in a temperate city, the error is modest. For March, for October, for anywhere north of about 45°, it is not.
What we did about it in SolGlow
SolGlow 1.7 introduced what we call True-D: instead of feeding the sunburn-weighted UV Index into the vitamin D math, it re-weights the incoming UV into a vitamin-D-action-spectrum value using the sun's computed angle for your location and time, plus total-column ozone. The estimate you see is built on the spectrum your skin actually responds to.
Two deliberate design decisions came with it, and both are about honesty rather than marketing:
First, the free tier only ever sees the conservative branch of the correction. When the physics is ambiguous, the number gets clamped downward. An estimate in SolGlow can become more honest as the model improves; it cannot become more flattering.
Second, the correction is visible. SolGlow's "Show the Work" view walks through how today's estimate was built, and when the ozone step actually moved the number, it appears there as its own line with its citation attached. When it didn't move anything meaningful, it stays out of your way. A correction you can't inspect is just a different black box.
See the number your skin actually responds to
SolGlow re-weights UV into the vitamin D action spectrum using sun angle and ozone, then shows its working — including the ozone step. Free on the App Store, with SolGlow Pro a one-time $19.99.
Explore SolGlowFrequently asked questions
Does the UV Index tell you how much vitamin D you're making?
Only loosely. It's weighted for sunburn, and vitamin D synthesis follows a different action spectrum concentrated further into the UVB. The two correlate, but no fixed multiplier converts one into the other.
What UV Index do you need for vitamin D?
Roughly 3 or above is the usual rule of thumb, because below that the sun is generally too low for meaningful UVB to reach the ground. Treat it as a rough threshold — the same UV Index carries different vitamin D potential depending on sun angle and ozone.
How does ozone affect vitamin D synthesis?
Ozone absorbs short-wavelength UVB strongly, and those are the wavelengths vitamin D depends on most. A thicker ozone column suppresses vitamin-D-effective UV faster than it suppresses the UV Index.
Why do vitamin D apps disagree with each other?
Because most start from the same erythemally weighted UV Index and apply different multipliers on top of it. The shared starting error, plus different assumptions about skin and clothing, is enough to produce very different answers for the same afternoon.
References & further reading
- Kazantzidis, A., et al. (2009). On the relationship between erythemal and vitamin D action spectrum weighted ultraviolet radiation. Journal of Photochemistry and Photobiology B: Biology.
- CIE (2006). Action spectrum for the production of previtamin D3 in human skin. CIE 174:2006.
- Bouillon, R., et al. (2021). A revised action spectrum for vitamin D synthesis by suberythemal UV radiation exposure in humans in vivo. PNAS, 118(40).
- Khanna, T., et al. (2022). Comprehensive Analysis of Seasonal and Geographical Variation in UVB Radiation Relevant for Vitamin D Production in Europe. Nutrients, 14(23):5189.
- Webb, A.R., Kline, L., & Holick, M.F. (1988). Influence of season and latitude on the cutaneous synthesis of vitamin D3. Journal of Clinical Endocrinology & Metabolism.
Related reading: How SolGlow estimates vitamin D from sunlight · Vitamin D winter: when your skin stops making it entirely
This article is for general education and is not medical advice. SolGlow provides vitamin D estimates based on published research and is not a medical device; its output is an estimate, not a diagnosis. Talk to a healthcare professional about your vitamin D status, supplementation, or sun safety.