Sunburn, Tanning and What UV Actually Does to Your Skin
A sunburn is not a heat burn. Nothing on your skin got hot enough to cook. What actually happened is that a particle of ultraviolet light was absorbed directly by the DNA inside your skin cells, welded two neighbouring letters of the genetic code together, and set off an emergency response that takes hours to become visible. Below, watch it happen one photon at a time — and watch the two things that follow: a tan, which is your skin building a shield out of damage already done, and a mutation counter that never goes back down.
Try this: start on Fair · midday and let the redness build. Then switch to Tanning bed — watch the burn warning almost vanish while the deep blue UVA arrows pour into the dermis and the collagen bar keeps falling. Then hit 🥣 Sunscreen SPF 30 and watch the photon rain thin out.
Live skin readout
What's happening
Real, standard numbers: UVB is 280–320 nm and UVA is 320–400 nm; one unit of UV index equals 25 mW/m² of erythemally-weighted irradiance; one minimal erythema dose (MED) for fair, type-II skin is about 250 J/m², which is why UV index 9 burns fair skin in roughly 18–20 minutes; SPF 30 filters about 97% of UVB and SPF 50 about 98%; a tan is worth roughly SPF 2–4; redness peaks 12–24 hours after exposure; UV rises about 10% per 1,000 m of altitude and fresh snow reflects up to about 80% of it; strong sun can generate on the order of 100,000 UV lesions per exposed cell per hour; one whole-body MED makes roughly 10,000–20,000 IU of vitamin D. Illustrative model output: the repair-capacity ceiling, the mutation-load scale, the collagen-integrity percentage, the exact vitamin D curve, the UVA figure assumed for a broad-spectrum SPF 30 (~85%), and the time compression — one real second is about four minutes of sun, and the redness lag is squeezed from hours into seconds. Photon counts on screen are a visual sample, not a physical count. This page explains a mechanism; it is not a personal sun-exposure prescription.
The Science in Plain Language
1. A sunburn is a DNA-damage response, not a heat burn
Put your hand on a hot pan and it hurts instantly, because protein is denaturing right now. Lie on a beach for an hour and nothing hurts at all — and then, somewhere around bedtime, your shoulders start to feel tight, and by the next afternoon they are scarlet and radiating heat. That gap is the whole story.
Ultraviolet light is not warm. The infrared in sunlight is what feels hot; the ultraviolet is invisible and imperceptible. What UV does is chemical. A UVB photon carries enough energy that when a DNA base absorbs it, the molecule can react. Two adjacent pyrimidine bases — two Ts, two Cs, or a T and a C sitting next to each other on the same strand — can fuse into a single fused ring called a cyclobutane pyrimidine dimer, or CPD. A related and less common lesion, the 6-4 photoproduct, forms the same way. The DNA helix physically kinks at that point. The machinery that reads and copies DNA runs into the kink and stalls.
Stalled transcription is an alarm. The cell activates p53, the tumour-suppressor protein sometimes called the guardian of the genome. p53 halts the cell cycle so nothing gets copied while it is broken, calls in repair, and if the damage is beyond repair, orders the cell to kill itself. At the same time, damaged keratinocytes pour out inflammatory signals — prostaglandins, particularly prostaglandin E2, plus interleukin-1, interleukin-6 and TNF-alpha. Those signals dilate the dermal capillaries. More blood in the skin is what redness is. The nerve endings, bathed in prostaglandins, become sensitised, which is why a sunburn hurts to touch.
All of that takes time. The lesions form during the exposure, in real time, silently. The inflammatory cascade takes 2–6 hours to become visible and peaks at 12 to 24 hours. That is why you cannot use pain as a warning signal for sun exposure: by the time your skin tells you, the exposure ended half a day ago. And it is why an NSAID like ibuprofen genuinely takes the edge off a sunburn — it blocks the prostaglandin arm. It does nothing whatsoever about the DNA.
2. UVA and UVB — two different photons, two different problems
Sunlight's UV comes in two bands that behave so differently they are almost separate topics. UVC (100–280 nm) is the most damaging of all and is absorbed entirely by the ozone layer; it never reaches you outdoors.
UVB, 280–320 nm. Shorter wavelength, higher energy per photon, and poor penetration — it is largely absorbed in the stratum corneum and the epidermis, with only a small fraction reaching the upper dermis. UVB is the burner: it is the band the UV index is weighted around, and it is the direct cause of CPDs, because DNA's absorption peak sits at about 260 nm, close enough to the UVB band for direct absorption. UVB is also the only band that makes vitamin D, because the same photochemistry that welds DNA bases together also cracks open the B-ring of 7-dehydrocholesterol in your skin. UVB is strongly filtered by the atmosphere, which is why it collapses at low sun angles — early morning, late afternoon, winter, high latitude — and why ordinary window glass blocks essentially all of it.
UVA, 320–400 nm. Longer wavelength, lower energy per photon, but roughly 95% of the UV that reaches the ground, and it penetrates far deeper — through the epidermis and well into the dermis. UVA is much less efficient at causing redness (it takes something like 500 to 1,000 times more UVA than UVB to produce the same erythema), which sounds reassuring and is exactly the problem: there is no warning signal. UVA does its damage mostly indirectly, generating reactive oxygen species that oxidise lipids, proteins and DNA bases, and it is the dominant driver of collagen breakdown. It also produces CPDs by a stranger route — UVA energy absorbed by melanin can be transferred to DNA in the dark, generating so-called "dark CPDs" for hours after you have come inside. Unlike UVB, UVA is barely filtered by the atmosphere, is nearly constant across the day and the seasons, passes through thin cloud, and passes straight through window glass.
3. How a tan actually works — and why it is a scar, not a shield
There are two tans, and they are different processes.
Immediate pigment darkening is driven by UVA and appears within minutes. It is not new pigment at all — it is existing melanin being oxidised and redistributed. It fades within hours to a day and provides essentially no protection. This is the "glow" a tanning bed sells you.
Delayed tanning is real new pigment, and it takes two to three days to appear because it is a manufacturing process. Here is the chain, and notice where it starts. A keratinocyte gets its DNA damaged. p53 rises. One of the genes p53 turns on is POMC, which is cleaved into α-melanocyte-stimulating hormone. That hormone diffuses to the melanocytes in the basal layer and binds MC1R, a receptor on their surface. MC1R raises cyclic AMP, which activates the transcription factor MITF, which switches on tyrosinase and the rest of the pigment factory. The melanocyte fills organelles called melanosomes with eumelanin and pushes them out along its dendrites into the 30-odd keratinocytes it services. Inside each keratinocyte, the melanosomes migrate to the sun-facing side of the nucleus and park there as a supranuclear cap — a literal parasol held over the DNA.
Read that chain again. The first step is DNA damage. A tan is not your skin getting healthier; it is your skin responding to injury by building armour, and the armour only gets built because the injury already happened. There is no mechanism by which you get the tan without the damage — which is precisely why the "safe tan" does not exist and why tanning beds are classified by the International Agency for Research on Cancer as a Group 1 human carcinogen, the same category as tobacco and asbestos.
And the armour is thin. A full delayed tan on previously fair skin is worth roughly SPF 2 to 4. That is not a rounding error away from a real sunscreen; it is one or two doublings of your time-to-burn, against the 30-fold a labelled SPF 30 is tested at.
4. Melanin, skin tone, and the risk that does not disappear
Constitutive melanin — the pigment you have without any sun — is genuinely protective, and the difference is large. Deeply pigmented skin has an intrinsic sun protection factor somewhere in the range of 10 to 13, and eumelanin is a broadband absorber that also quenches free radicals. Melanoma incidence in Black Americans is roughly 20 to 30 times lower than in white Americans. That is a real biological advantage and it should be stated plainly.
It is also not immunity, and the way it fails is specific and important. Melanoma in darker skin disproportionately occurs at acral sites — palms, soles, nail beds — and on mucous membranes. Those sites are barely sun-exposed, which means their melanoma is largely not UV-driven and carries a different mutational signature. Because it is rare, because it appears where nobody thinks to look, and because both patients and clinicians are less likely to suspect it, acral melanoma is caught later and stage-for-stage survival is worse. Bob Marley died at 36 of an acral melanoma that began under a toenail and was initially dismissed as a football injury. The lesson is not "darker skin is at high risk of sunburn"; it is "check the soles of your feet and under your nails, and take a new, changing, or non-healing pigmented spot there seriously regardless of your skin tone."
The other end of the spectrum is the MC1R gene. Certain common MC1R variants — the ones associated with red hair, freckling and very fair skin — shift melanin production away from brown-black eumelanin toward reddish-yellow pheomelanin. Pheomelanin is a poor sunscreen, and worse, it appears to be a pro-oxidant that can generate oxidative DNA damage even without further UV. That is one reason red-haired individuals carry elevated melanoma risk that is not fully explained by sun exposure alone.
5. Repair, p53, and how decades of small damage becomes cancer
Your cells are extremely good at this. The pathway that removes CPDs is nucleotide excision repair. XPC (or, on actively transcribed genes, a stalled RNA polymerase) recognises the distortion; the TFIIH complex unwinds the helix around it; XPF–ERCC1 and XPG cut the damaged strand on either side, excising a patch of roughly 24–32 nucleotides; DNA polymerase fills the gap using the undamaged strand as a template; ligase seals it. Done correctly, the lesion leaves no trace.
We know exactly how much this matters because of what happens when it is broken. People with xeroderma pigmentosum inherit defects in the nucleotide excision repair genes. Their skin cancer risk is increased on the order of a thousandfold, and children with XP develop skin cancers in the first decade of life. Repair is not a background detail; it is the only reason ordinary sun exposure is survivable.
But repair is a rate, and rates saturate. Under strong sun the lesions arrive faster than the machinery clears them, and a backlog builds. Three things can happen to a backlogged lesion. Most are eventually repaired correctly. Some belong to cells so damaged that p53 gives up and triggers apoptosis — those are the shrunken, orange-pink sunburn cells a pathologist can see in a biopsy of burned skin, and they are the reason a bad burn peels: a whole cohort of the epidermis deleted itself. And a few are copied over before repair finishes, or repaired wrongly. Because a CPD between two Ts is preferentially misread, the classic UV mutation is a C→T transition at a dipyrimidine site, and tandem CC→TT changes. That specific pattern is a fingerprint: find it in a tumour genome and you know the sun did it.
Now stack that over decades. Sequencing of normal, healthy, sun-exposed skin from middle-aged adults has found something genuinely startling — that ordinary-looking skin is already a patchwork of thousands of competing mutant clones, many carrying the same driver mutations found in skin cancers, with NOTCH1 and TP53 among the most common. The mutation burden of healthy sun-exposed skin is comparable to that of many tumours. What separates a clone from a cancer is the accumulation of the additional hits, and that is what "cumulative dose" means. This is why the risk profiles differ: basal cell carcinoma tracks intermittent, intense, blistering exposure and is by far the most common human cancer but rarely metastasises; squamous cell carcinoma tracks total lifetime dose, which is why it appears on the face, ears, scalp and forearms of outdoor workers, and why it is preceded by the rough, scaly precursor lesions called actinic keratoses; and melanoma is associated most strongly with intermittent severe burns, particularly in childhood, and with the number of moles.
6. Photoaging — most of "looking older" is UVA
Compare the skin on the inside of your upper arm with the skin on the back of your hand. Same age, same genes, same nutrition. The difference is photons.
Estimates commonly attribute the large majority of visible facial ageing — wrinkles, leathery texture, sagging, mottled pigmentation, broken capillaries — to UV rather than to the passage of time. The mechanism is mostly UVA, because UVA is what reaches the dermis, and the dermis is where the structure lives. UVA generates reactive oxygen species there, which activate the AP-1 transcription factor, which upregulates matrix metalloproteinases — principally MMP-1 (collagenase), plus MMP-3 and MMP-9. Those enzymes chop up type I and type III collagen. At the same time AP-1 suppresses new procollagen synthesis. So you get faster demolition and slower construction at once.
The elastin story is the visible one. Rather than simply degrading, elastic fibres in chronically sun-exposed dermis accumulate as tangled, non-functional masses — a state called solar elastosis, which under the microscope looks like blue-grey amorphous material replacing normal pink collagen. That is what gives long-exposed skin its thickened, yellowed, deeply furrowed appearance. Add the dilated, permanently visible capillaries (telangiectasias) and the uneven pigment of sun spots (solar lentigines), and you have essentially the entire visual vocabulary of "sun-damaged skin".
Two practical implications. First, a sunscreen that only carries a high SPF is protecting you from the burn and doing much less about the ageing — SPF is a UVB measure. Second, this is the arm of UV damage that a driver accumulates through a car window and an office worker accumulates through an unfilmed office window, on the side of the face nearer the glass, over years.
7. How to actually read a sunscreen label
SPF is a UVB number, and it is not linear. SPF is defined as the ratio of the UV dose needed to redden protected skin versus unprotected skin. Because redness is overwhelmingly a UVB phenomenon, SPF is essentially a UVB rating. And the arithmetic of the numbers surprises people: SPF 15 filters about 93% of UVB, SPF 30 about 97%, SPF 50 about 98%, and SPF 100 about 99%. The step from 30 to 50 buys you roughly one extra percentage point of blocked photons. The number is a ratio of tolerable dose, not a percentage.
"Broad spectrum" is the UVA claim, and you need it. A product can carry SPF 50 and offer weak UVA protection. In the US the phrase "broad spectrum" indicates the product passed a critical-wavelength test; in the EU the UVA seal indicates UVA protection of at least one third of the labelled SPF. Ingredients that do meaningful UVA work include zinc oxide, titanium dioxide, avobenzone (usually stabilised with octocrylene), and in many countries Tinosorb S and M, Mexoryl and bemotrizinol. If the label has an impressive SPF and no broad-spectrum statement, it is half a product.
The tested dose is more than you apply. SPF is measured at 2 mg per square centimetre of skin. Real-world application is typically a quarter to a half of that, and protection falls off faster than proportionally — applying half the dose does not give you half the SPF, it gives you considerably less. In practical terms, a labelled SPF 30 applied the way most people apply it may be behaving like an SPF 5 to 10. Which is why the reliable advice is boring and mechanical: use more than feels necessary, cover the ears, the back of the neck, the tops of the feet and the hairline, and reapply every two hours and after swimming or towelling. Reapplication beats a bigger number on the bottle. And shade, a wide-brimmed hat, a long-sleeved shirt and UV-blocking sunglasses are not the consolation prize — woven fabric is a perfect broad-spectrum filter that never needs reapplying.
8. The vitamin D trade-off, honestly stated
This page would be dishonest if it stopped at "UV bad". The same UVB photons that make CPDs also convert 7-dehydrocholesterol in your skin into previtamin D3, and skin synthesis is, for most of human history and most of the world, the dominant source of vitamin D. A single whole-body minimal-erythema dose generates roughly 10,000–20,000 IU — far more than any ordinary supplement.
Several details make the trade-off less brutal than it first looks. Synthesis is fast and self-limiting: most of it happens in the first fraction of an MED, well below the dose that reddens you, and continued exposure photodegrades previtamin D3 into inert products, which is why you cannot reach vitamin D toxicity from the sun no matter how long you stay out. Deep skin pigmentation slows synthesis substantially, so darker-skinned people at high latitudes need materially more exposure or dietary intake for the same result. Above roughly 35–37 degrees latitude, the winter sun angle is too shallow for meaningful UVB at all, which is the reason for the term "vitamin D winter". Sunscreen, in laboratory conditions, cuts synthesis sharply — but field studies have generally not found sunscreen users to be vitamin D deficient, most likely because real-world application is thin and incomplete.
The honest summary is that the two effects have different dose-response curves. Vitamin D synthesis saturates early and low; DNA damage keeps accumulating linearly with dose. That asymmetry is the entire argument, and it means the trade-off is much less painful than the tanning industry likes to imply — and that a supplement is available while a repaired genome is not. For the full pathway from skin to liver to kidney, see the companion animation on how your skin makes vitamin D.
9. Three myths worth retiring
Myth: "I'm getting a base tan before the holiday so I don't burn." A base tan is worth roughly SPF 3. It extends your time-to-burn from about 20 minutes to about an hour at UV index 9 — on a beach day that is nothing. Worse, the tan is not a preparation for damage, it is damage: to acquire it you already absorbed the UV dose, already formed the dimers, and already banked whatever mutations came out of them. You have paid the cost up front for a discount that does not cover it. A pre-holiday sunbed session buys a few percent of protection at the price of a Group 1 carcinogen exposure.
Myth: "SPF 100 is twice as good as SPF 50." SPF 50 blocks roughly 98% of UVB; SPF 100 blocks roughly 99%. The difference between them is about one percent of the incoming photons. The two ways very high SPF numbers actually mislead are subtler: they encourage people to apply less and stay out longer, and a very high SPF says nothing at all about UVA, so the collagen damage and the photoaging can proceed at nearly full speed behind an impressive-looking number. A well-formulated broad-spectrum SPF 30, applied thickly and reapplied, outperforms an SPF 100 applied once and thinly.
Myth: "It's cloudy and cool, so I'm fine." Temperature is infrared; burning is ultraviolet; the two are only loosely correlated. Thin or broken cloud transmits a large fraction of UV — and scattered cloud can briefly push ground-level UV above the clear-sky value through edge-of-cloud reflection. Some of the worst burns happen on cool, breezy, overcast days precisely because the heat cue that normally drives people into the shade is missing. Add the reflectors: fresh snow bounces back up to about 80% of UV, dry sand around 15–20%, sea foam and water surfaces around 10–25% — which is why you can burn under a beach umbrella and why skiers burn the underside of the chin and the inside of the nostrils. And add altitude: UV climbs roughly 10% per 1,000 metres, so a snowfield at 2,500 m is a fundamentally different exposure from the same sun at sea level. Finally, ordinary window glass blocks UVB but transmits most UVA, so the commute and the desk by the window are contributing to photoaging all year, in every weather.
Connections
- All Interactive Visualizations
- Animation: How Your Skin Makes Vitamin D
- Animation: DNA Repair
- Animation: Apoptosis — Programmed Cell Death
- Animation: The Cell Cycle and Cancer
- Animation: Oxidative Stress and Antioxidants
- Melanoma
- Basal Cell Carcinoma
- Squamous Cell Carcinoma
- Actinic Keratosis
- Melasma
- Dermatology
- Vitamin D3
- Antioxidants