How long outside today?

• Sun position from the standard Spencer/NOAA series. Accurate to about a minute of clock time.
• The clear-sky ultraviolet curve is fitted to 7,022 clear-sky readings from 16 places between 68°N and 55°S, and agrees with them to within about 15%. The textbook formula it replaced ran roughly a third low at the shallow sun angles that decide a Nordic autumn.
• The threshold below which nothing happens comes from Engelsen and colleagues (2005): production stops once the sun drops below 15.7° above the horizon. That one number produces every grey band here. Their paper also finds nowhere poleward of 51° escapes a dead period — which this model reproduces independently, a useful check that it behaves.
• How much vitamin D per unit of sun is anchored to Webb (2006), after Holick: a quarter of your personal sunburn dose, on a quarter of your skin, makes roughly 1,000 IU.
• Skin colour follows Lo and colleagues (1986): every skin type makes the same vitamin D from the same dose measured against its own sunburn threshold. Darker skin needs proportionally longer — three to five times.
• Age thins the raw material in skin. A straight line to about 40% of a twenty-year-old's capacity by 70. The real curve is not this tidy.
• Saturation. Production plateaus, and further sun destroys what has already formed, so a long sunbathe can yield less than a short one. A smooth ceiling near 20,000 IU whole-body. First-order stand-in.
• Clothing is guessed from the temperature rather than assumed, because a page that quietly puts you in shorts in October is lying. The year chart uses your location's own daily temperatures for the last twelve months, smoothed — so the bare-skin fraction falls through autumn the way clothing actually does. You can override it.
• The ozone layer overhead is not published free as a live figure anywhere this page can reach. Instead it is solved for: the clear-sky ultraviolet index the forecast gives for your coordinates, read back through the fitted curve, yields an effective column in Dobson units. That pins today exactly. Other dates move off it along the seasonal shape of Van Heuklon's (1979) empirical model, which tracked Stockholm's satellite record to about 16 DU once pinned this way, against 63 unpinned. Its spring peak arrives a few weeks late, so late-winter days here are slightly optimistic.
• Today's sky is Open-Meteo's forecast ultraviolet index for your exact coordinates, hour by hour, divided by their own clear-sky value. That ratio is one measured number absorbing cloud and haze at once. Other dates fall back to a clear modelled sky, which is far cruder.
• Altitude comes from the forecast's own elevation for your coordinates. Ultraviolet gains roughly 10% per 1,000 metres.
• Not modelled: reflection off snow or water, shade and buildings, lying down versus standing, pollution beyond a standard rural haze, and the several-hour delay between sun on skin and vitamin D in blood.

• 400 IU is a floor, not an aim. Livsmedelsverket — the Swedish Food Agency, not Folkhälsomyndigheten — sets 10 micrograms (400 IU) a day for adults 18–74 and 20 (800 IU) from 75. That is the population dose that stops deficiency disease: soft bone, muscle weakness. The US figure is 600 IU, Europe's reference intake 600. None of them is an optimum.
• There is no agreed optimum, so the honest way to answer "how much" is to name a blood level and work back. Guidelines aim at 50 nmol/L of 25-hydroxyvitamin D. Through a winter with no sun, most adults need roughly 800–1,000 IU a day to hold that; the Endocrine Society's clinical guideline, which aims higher at 75 nmol/L, says 1,500–2,000. Bigger bodies need more — it is stored in fat, so the same pill goes further in a smaller person. This page defaults to 1,000.
• Past that, the evidence stops supporting you. VITAL — 25,871 adults on 2,000 IU a day for 5.3 years — found no drop in cancer or in major heart and circulatory events, and Australia's D-Health agreed. Those people were mostly already sufficient. More is not better; not being short is the whole benefit.
• Europe's safe ceiling (European Food Safety Authority, 2023) is 100 micrograms — 4,000 IU — a day from age 11, pregnancy included. So 1,000 sits comfortably inside it, at four times the Swedish floor.
• Being deficient does real damage: soft bone, muscle weakness, bone pain, and in older people more fractures and falls. In a Nordic winter a shortfall is the default, because there is no sun route to top up from.
• The blood test measures 25-hydroxyvitamin D. Below 30 nmol/L is deficient, 30–50 is short, 50+ is what guidelines aim for. No good evidence that pushing past 75 buys anything. It is the only way to know your own number — everything above is a population average.
• Practical shape: D3 (cholecalciferol) rather than D2, with a meal containing fat. Daily beats one big weekly dose.

Sources: Engelsen, Brustad, Aksnes & Lund, Daily duration of vitamin D synthesis in human skin, Photochemistry and Photobiology 81 (2005) 1287. Webb, Who, what, where and when, Progress in Biophysics and Molecular Biology 92 (2006) 17. Shih, Farrar, Cooke et al., Fractional sunburn threshold UVR doses generate equivalent vitamin D and DNA damage in skin types I–VI, Journal of Investigative Dermatology 138 (2018) 2244. Manson et al. (VITAL), New England Journal of Medicine 380 (2019) 33. EFSA, tolerable upper intake level for vitamin D, EFSA Journal (2023) 8145. Van Heuklon, Estimating atmospheric ozone for solar radiation models, Solar Energy 22 (1979) 63. Livsmedelsverket's recommendations and LIVSFS 2018:5. On daylight and the body clock: the retinal cells that set it respond to bright visible light, not ultraviolet — see Lucas et al., Measuring and using light in the melanopsin age, Trends in Neurosciences 37 (2014) 1. Weather, ultraviolet and temperature data from Open-Meteo, CC BY 4.0.