Build a virtual light filter, uncover a missing patch, explore real skin microscopy and inspect a DNA structure. Discover why visible light and warmth cannot tell the whole story.
Enable JavaScript to change the conditions and run the interactive experiment.
Your eyes tell you about visible light. They do not measure all the ultraviolet arriving alongside it. A material can change one part of a spectrum while leaving another almost unchanged.
Can two covers give the same total reading while passing different amounts of UV?
Our two 2.2 W/m² examples pass 0.2 and 2.0 W/m² of UV respectively.
Light spans many wavelengths. Ultraviolet, or UV, lies beyond the violet end of ordinary vision. Our bench uses four separate channels. Colored moving markers help you follow them; UV and infrared markers are explanatory colors, not what those invisible wavelengths look like.
Keep the visible channel at 100%, then change the two UV settings to 10%. The visible detector stays unchanged while the UV reading falls. A real cover needs its own measured transmission spectrum; these settings do not describe all glass, fabric or sunscreen.
An uncovered cell bypasses the ideal cover. Each open cell transmits the full uniform source; covered cells transmit the selected fraction. Add their contributions before averaging. An average of 28% does not mean every location receives 28%.
Absorbed light can change molecules. A real skin section shows tissue organization; a separately measured DNA structure shows one kind of chemical lesion. Neither is a personal damage meter. Wavelength, amount, material and biological response all matter.
Using the WHO convention, UVA spans 315–400 nm, UVB 280–315 nm and UVC 100–280 nm. These are naming boundaries, not biological cliffs. Some sources use a 320 nm UVA boundary. Our 305 and 365 nm examples are unambiguous under either convention. We do not model an unattenuated UVC beam as ordinary ground-level sunlight.
The authored source supplies 1 W/m² in each of four monochromatic channels: 305, 365, 550 and 1000 nm. Each value is channel-integrated irradiance, not W/m²/nm. Their sum is 4 W/m². This artificial four-line source is not a solar or screen spectrum. Irradiance is arriving power per area; it is not temperature or a visible-brightness score.
For one photon, E = hc/λ. NIST’s defining constants give about 4.0651 eV at 305 nm and 3.3968 eV at 365 nm. At 1 W/m², arrival rates are about 1.5354 × 10¹⁸ and 1.8375 × 10¹⁸ photons per second per square meter. Equal irradiance does not mean equal photon counts per second per area. Both travel at the same speed in vacuum. Photon energy alone does not determine a biological outcome.
For cover transmission T and uncovered fraction g, the mean outgoing fraction is g + (1 − g)T under uniform illumination. With T = 0.10 and g = 0.20, the mean is 0.28. Uncovered regions supply 0.20/0.28 ≈ 71.43% of the transmitted channel. Moving the same gaps changes the local map but leaves this area average unchanged.
Each covered cell transmits T and absorbs 1 − T; uncovered cells transmit everything. Reflection, scattering, fluorescence and angle dependence are omitted. Absorbed plus transmitted irradiance equals the incident value per channel. Absorption can contribute to heating, but the bench has no heat capacity, cooling law or temperature calculation.
The optional technical table uses the erythema-reference weighting printed in NOAA’s NEUBrew report. Its weight is about 0.219786 at 305 nm and 0.000407380 at 365 nm: a ratio of about 539.51. This endpoint-specific convention is not an absorption spectrum, universal DNA-damage factor or individual forecast. Our four-line weighted sum is not a measured outdoor UV Index.
A cis-syn cyclobutane pyrimidine dimer can connect neighboring thymine units on the same DNA strand. In 1TTD, the highlighted extra links are C5–C5T and C6–C6T inside component TTD. Normal opposite-strand pairing is different. Whole-skin experiments reported these lesions after UVA as well as UVB exposure; “UVA only ages, UVB alone damages DNA” is misleading.
McAteer and colleagues used solution NMR constraints and structural calculations. We render one deposited conformer each from 1TTD and parent duplex 1COC, preserving their coordinates. Each has 761 modeled atoms, 486 excluding hydrogen. The damaged structure has 23 residue records because one TTD record represents two linked thymine units; the parent has 24. This is not a deleted nucleotide. Switching structures compares them; it does not replay a reaction.
A chemical lesion, a lasting sequence mutation and cancer are not interchangeable. Cells have repair and other responses; not every lesion becomes a mutation. A structure establishes geometry, not a repair rate or a clinical probability. This lesson collects no skin photos and predicts no individual safe exposure time.
The Human Protein Atlas section is normal thin skin stained with hematoxylin and eosin. Staining reveals features. It does not represent natural skin color, UV penetration or damage after exposure. No spatial calibration is supplied for this derivative, so we do not invent a micrometer scale or reconstruct a measured 3D tissue volume.
Lawrence and colleagues measured finite-width spectra for experimental 385 and 405 nm sources. Figure 7 is normalized: a peak of 1 does not establish equal absolute irradiance. Reported widths are 7 and 10 nm. The 405 nm source includes shorter wavelengths. Their high-dose experimental setup is not evidence that an ordinary phone produces the same exposure.
FDA describes SPF in terms of relative sunburn-producing energy, not a guaranteed multiplier for time in the sun. A single SPF number is not a transmission spectrum. Broad-spectrum labeling and actual instructions matter. Our imaginary cover assigns no product SPF. Use the linked WHO and FDA guidance for real-world protection.
A camera or laboratory instrument can select wavelengths that eyes alone cannot separate. Performance needs a spectrum and stated conditions. A material’s name or apparent darkness does not supply that evidence.
Clothing, shade, environmental UV information and product instructions work together in public-health guidance. Our coverage puzzle explains geometry; it does not test sunscreen or recommend an exposure schedule.
Histology locates tissue features. NMR-derived structures locate atoms. Spectroradiometry characterizes a source. Their questions connect, but one measurement cannot silently supply all the others.
Draw a 5×5 grid. Mark one row uncovered and four rows covered. Each square represents equal area under uniform illumination.
Write “10 same-wavelength tokens arrive” by each square. They represent equal portions of light, not lamps or photon observations.
An open square passes 10 tokens; a covered square passes 1. Five open squares pass 50. Twenty covered squares pass 20. Together, 70 pass out of 250.
What fraction of all arriving tokens passes? 70/250 = 28%. What fraction of the passing tokens came through gaps? 50/70 ≈ 71%. Explain the different denominators.
Move the five uncovered squares to different places. Keep their number the same. The total stays unchanged, but the locations receiving most tokens move.
Write “Visible: passes 10/10” and “UV: passes 1/10.” Explain how the visible result can stay the same while the UV changes. These are imaginary properties.
Finish “This paper model cannot tell me…” with a product’s SPF, skin temperature, DNA damage count, or safe time outdoors. Explain what evidence is missing.
How can 20% uncovered area supply about 71% of the light that passes?
Indoor paper only. No sunlight, UV lamp, laser, magnifying lens, skin exposure, sunscreen test or indicator is needed. Follow actual product instructions and public-health guidance for real sunshine.
You need UV information as well; cool conditions do not determine the UV level. Sun position, atmosphere and other conditions affect UV independently of how warm you feel.
They carry equal radiant power per area but different photon counts and reference erythema weights. Our independent fixtures calculate both distinctions; neither predicts an individual injury.
No; transmission can depend on wavelength. This is an ideal authored cover. A real material would need its own measured spectrum.
28%. 0.20×1 + 0.80×0.10 = 0.28. This average does not make every location equally covered.
Two neighboring thymine units on the same DNA strand. The cyclobutane dimer creates additional covalent connections between adjacent bases.
No; cells have repair and other responses, and not every lesion becomes a mutation. The structure demonstrates one kind of damage; it is not a prediction of a person's future.
It is a tested relative sunburn-protection measure; follow the actual label and other protection measures. Broad-spectrum coverage is relevant too. Our ideal-filter arithmetic does not assign product SPF.
The full spectrum, absolute irradiance and experimental conditions. Normalizing a curve to a peak of 1 does not make two sources equal in total power or isolate one wavelength.
General environmental/protection context, not a personal timer.
WHO · ultraviolet radiationNaming conventions are not abrupt changes in every biological response.
WHO · UV band definitionsApplied to the authored monochromatic channels.
NIST · SI defining constantsDomain starts strictly above 250 nm. The 328 nm piecewise discontinuity is retained. Our sum is not a measured UV Index.
NOAA–EPA NEUBrew · UV Index calculationsCC BY 4.0; HPA and Uhlén et al., Science 2015, doi:10.1126/science.1260419. Tiles assembled/cropped without recoloring.
Human Protein Atlas · skin image 1788The evidence download preserves exact image/version, publication and license information.
HPA · reuse licenseJ. Mol. Biol. 282, 1013–1032 (1998), doi:10.1006/jmbi.1998.2062. One isolated-duplex conformer, not a photograph or reaction movie.
McAteer et al. · 1TTDChemical records distinguish the parent despite potentially confusing legacy titles.
Parent duplex · 1COCCCD topology supplies component bonds. Source coordinates remain unchanged.
wwPDB · data usage policyNarrow claim supported by the official abstract; not an individual forecast.
Mouret et al. · UVA and DNA lesionsScientific Reports 8, 12722 (2018), CC BY 4.0. Full axes/legend retained; relative peaks do not mean equal absolute irradiance.
Lawrence et al. · measured spectra, Figure 7Use actual labels and broad-spectrum guidance. The bench assigns no product rating.
FDA · Sun Protection FactorIndependent subject review is pending.
Read the sources and model assumptions
Interactive explorationSlide a texture beneath a finger. See the contact change, compare a moving touch with a steady one, and zoom from skin to a real sensing protein. Discover how touch differs from pain.
Interactive explorationOpen a connected follicle, follow its normal oil route, and compare structures hidden beneath the surface. Read real human microscopy and discover why acne is more than a story about dirt.
Interactive explorationOpen a sourced 3D eye, move a target nearer, and catch the moment its light comes together. Then discover why a smaller pupil is different from better focus.