Why the Sky Is Blue: Scattering, Sunlight and the Human Eye

Air molecules scatter blue light across the sky far more strongly than red, a process called Rayleigh scattering. Why the sky is not violet, why sunsets are red and why Mars has blue sunsets all follow from the same physics, with some open questions about the details.

The daytime sky looks blue because air molecules, mostly nitrogen and oxygen, scatter sunlight in every direction, and they redirect short wavelengths much more strongly than long ones. Light arriving at the eye from all parts of the sky is therefore weighted toward the blue end of the spectrum. The process is called Rayleigh scattering, and many independent sources agree on it, among them NOAA, the Royal Museums Greenwich, Georgia State University's HyperPhysics and a 2023 peer-reviewed paper.

That tidy answer leaves a puzzle. Violet light scatters even more strongly than blue, yet the sky does not look violet. Resolving that gap takes the Sun's output, the way human eyes respond to color and, at twilight, absorption by ozone.

The same physics also accounts for red sunsets, white clouds and the pale, washed-out look of the sky near the horizon. It even explains why Mars, with a very different atmosphere, produces blue sunsets under a tan daytime sky.

What air does to sunlight

Scattering, in the NOAA Storm Prediction Center's usage, is the redirection of light by small particles. Which kind of scattering occurs depends on the size of the particle compared with the wavelength of the light. When particles are much smaller than the wavelength, below about a tenth of it according to HyperPhysics, the result is Rayleigh scattering. Individual air molecules fall comfortably in that category.

The strength of Rayleigh scattering varies as 1/λ⁴, where λ is the wavelength. An arXiv preprint on atmospheric extinction (not peer reviewed) notes that light at 400 nanometers scatters more than nine times as strongly as light at 700 nanometers. A trade-magazine explainer in Spectroscopy gives a smaller figure, saying blue light scatters about six times more than red, but it uses a blue wavelength longer than 400 nanometers, so the two figures are consistent.

Because the scattered light comes from every direction around the observer, blue seems to fill the whole dome. Spectroscopy also notes that, anywhere except directly at the Sun, the light seen is partially polarized, a fingerprint of scattering rather than direct sunlight.

What the sky is not

One popular explanation holds that the sky is blue because it mirrors the ocean. Both the Royal Museums Greenwich and Live Science identify that as a myth. The color comes from the atmosphere scattering sunlight, not from reflection off the sea.

Another claim in circulation says nitrogen absorbs violet light. The research behind this article found it only in a patent document, which is not a credible science source, and no reputable source supporting it turned up. It is left out here.

Why the sky is not violet

If shorter wavelengths scatter more, violet should win. The table below shows relative Rayleigh scattering computed from the 1/λ⁴ law alone, normalized so that 700 nm red equals 1. Violet at 400 nm comes out at about 9.4 and blue at 450 nm at about 5.9, so violet scatters roughly 1.6 times as strongly as blue.

These figures describe only the scattering law. They ignore how much light the Sun emits at each wavelength, how much ozone absorbs and how the eye responds, which is exactly where the explanation for the missing violet lies.

Two widely cited reasons

According to Spectroscopy and Live Science, the Sun emits less violet than blue, and human eyes respond less to violet. Together these push the sky's apparent color away from violet and toward blue. Sources disagree on where the Sun's spectrum peaks, with weaker sites variously naming blue, green or yellow, so the safe statement is the narrower one: the Sun puts out less violet than blue.

A third reason: how the eye reads the mixture

A 2005 report from LiveScience carried by NBC News describes a study by physicist Glenn Smith in the American Journal of Physics. As the report tells it, skylight contains roughly equal parts violet and blue by spectrum, and the eye's cones respond to that mixture much as they would to pure blue plus white light. The sky therefore appears as an unsaturated, or pale, blue rather than a deep violet.

That account is a news report of the study, and the original paper was not available for this article. The qualitative explanation is consistent across sources, but how much each factor contributes remains unquantified here. Websites that put numbers on the eye's reduced sensitivity to violet give figures ranging from about 3 times to about 100 times, without reliable sourcing, so no figure is offered.

Ozone, twilight and an incomplete textbook story

A 2023 paper by Christian von Savigny in Atmospheric Chemistry and Physics revisits the question. It reports that Edward Hulburt showed in 1953 that Rayleigh scattering accounts for only about one third of the blue of the zenith sky at sunset, while ozone absorption accounts for about two thirds. The paper calls the common Rayleigh-only explanation "not entirely correct."

That finding applies to the sky overhead at twilight, not to midday. The research available for this article did not settle how large ozone's role is for the midday sky. Weaker sites disagree, one saying ozone absorbs some violet and another saying Hulburt found little daytime effect, and neither is reliable enough to rest a claim on.

Von Savigny's point is an extension of the standard account, not a refutation. Scattering by air molecules remains the foundation, and ozone adds to it under certain lighting conditions.

A fringe challenge

A 2021 arXiv preprint argues that Rayleigh's and Thomson's theories conflict with quantum mechanics and proposes a single-photon theory in their place. The preprint has not been peer reviewed, no corroboration was found, and it is treated here as unverified and outside the mainstream. It does not change the established picture.

White clouds, pale horizons and red sunsets

Particles comparable to or larger than the wavelength of light scatter differently. The National Weather Service's JetStream site explains that cloud droplets are about the size of visible wavelengths, producing what is called Mie scattering. HyperPhysics adds that Mie scattering depends only weakly on color, so it scatters all colors about equally, and the result looks white.

The same logic explains why the sky is whiter near the Sun and near the horizon, according to HyperPhysics. In those directions, larger particles contribute more of the scattered light and dilute the blue.

Sunsets

Near sunrise and sunset, sunlight crosses a much longer stretch of air before reaching an observer. NOAA's Storm Prediction Center explains that blue is scattered out of the beam along that path, leaving red and orange. The blue removed from a sunset beam does not vanish; it can contribute to the blue afternoon sky somewhere else.

That link led NOAA's Stephen Corfidi to state that sunsets are red because the daytime sky is blue. The two colors are products of one process seen from different positions.

Air quality changes the show. Corfidi notes that pollutants and large particles soften and mute sky colors, and that vivid sunsets need clouds high enough to catch relatively clean sunlight above the dirty boundary layer near the ground.

How scientists worked it out

According to Wikipedia's summary of the history, Leonardo da Vinci conjectured that scattering of sunlight by small particles gives the sky its color. John Tyndall later found that bright light scattered off nanoscopic particulates is faintly blue-tinted and proposed that this explained the sky. Per the same summary, he could not explain why blue was favored, and atmospheric dust could not account for the intensity of the color. This is a tertiary source, and a fuller history by Lilienfeld was not retrieved.

In 1871, Lord Rayleigh (John William Strutt) gave a theory of the polarization and wavelength dependence of light scattered by particles small compared with the wavelength, according to von Savigny's 2023 paper. What those particles were remained unsettled. In 1899, Rayleigh showed that scattering by air molecules is enough to explain the observed effects, in a paper in the Philosophical Magazine.

A statistical-physics footing

An Indian Academy of Sciences Resonance article describes how Marian Smoluchowski linked density fluctuations in a gas to light scattering in 1908, and how Albert Einstein completed the quantitative step in 1910. A European Journal of Physics teaching article by Potenza and colleagues says Einstein's result is formally identical to Rayleigh's, as Einstein himself noted. A historical essay on arXiv adds that his formula gave a quantitative link between Rayleigh scattering and critical opalescence, the cloudiness of a fluid near its critical point.

The fluctuation view points to something subtle. As Potenza and colleagues explain, the randomness of molecular positions in air matters, because molecules arranged perfectly and scattering coherently would cancel the sideways scattering. This description is simplified, and the primary papers were not reviewed for this article.

A comparison: the sky of Mars

Mars shows what changes when the atmosphere does. The Royal Museums Greenwich notes that Mars's thin atmosphere makes Rayleigh scattering a very small effect there. Reports from Space Daily and Science News Today describe a dusty daytime sky ranging from tan to butterscotch.

Sunsets there are different. NASA scientist Mark Lemmon of Texas A&M describes Martian dust as the right size to let blue light penetrate more efficiently, giving bluish sunsets. NASA's caption for a 2005 image from the Spirit rover, as reproduced by CBC, says the dust scatters blue light forward, creating a bluish glow near the setting Sun. NASA's Visualization Studio records that the Curiosity rover captured its first color sunset on April 15, 2015.

Popular articles sometimes say Mars reverses Earth's colors, which oversimplifies. The blue comes from dust scattering, not Rayleigh scattering, and it is concentrated around the Sun while the rest of the sky stays tan or reddish.

What the blue sky teaches

The blue sky is a lesson in how a simple-sounding physical law meets a messy world. Scattering by air molecules supplies the core mechanism, but the color people actually see is also shaped by the Sun's output, by the quirks of human vision and, at dusk, by ozone. Each added factor refines the textbook answer rather than overturning it.

The open items are quantitative ones: how much each factor contributes, and how those contributions shift with angle, haze and time of day. Careful measurements of real skies and the primary papers that remain to be reviewed would answer them. Until then, the best-supported summary is that the blue comes from air, sunlight and the way an eye reads the result.

Frequently asked questions

Why is the sky blue and not violet?

Violet scatters more than blue, but the Sun emits less violet and human eyes respond less to it, according to Spectroscopy and Live Science. A 2005 study by Glenn Smith, as reported by LiveScience, adds that the eye reads the blue-violet mix as a pale blue. The size of each factor has not been firmly quantified in the sources reviewed.

Is the sky blue because it reflects the ocean?

No. The Royal Museums Greenwich and Live Science both identify the ocean-reflection idea as a myth. The color arises from air molecules scattering sunlight.

Why are sunsets red?

At sunrise and sunset, sunlight travels a much longer path through the atmosphere. NOAA's Storm Prediction Center explains that blue light is scattered out of the beam along the way, leaving red and orange.

Why are clouds white if the sky is blue?

Cloud droplets are comparable in size to the wavelengths of visible light, which produces Mie scattering, according to the National Weather Service. HyperPhysics notes that this kind of scattering depends only weakly on color, so all colors are scattered about equally and the cloud looks white.

Is Rayleigh scattering the whole explanation?

Not entirely. A 2023 peer-reviewed paper reports that ozone absorption accounts for about two thirds of the blue of the zenith sky at sunset, per Hulburt's 1953 work. The role of ozone in the midday sky was not settled in the sources reviewed.

Sources

1. Why is the sky blue? , Live Science
2. Why is the sky blue? , Royal Museums Greenwich
3. Blue Sky and Rayleigh Scattering , HyperPhysics, Georgia State University
4. Why Is the Sky Blue? , Spectroscopy
5. Why skies are blue instead of purple , NBC News (LiveScience) (2005-01-01)
6. NWS JetStream - The Color of Clouds , NOAA National Weather Service
7. The Science of Sunsets , NOAA Storm Prediction Center
8. Revisiting the question "Why is the sky blue?" , Atmospheric Chemistry and Physics 23, 14829 (2023-01-01)
9. The atmospheric extinction of light , arXiv 1511.03399 (preprint)
10. On the transmission of light through an atmosphere containing small particles in suspension, and on the origin of the blue of the sky , Philosophical Magazine (Taylor & Francis) (1899-01-01)
11. Critical opalescence (Resonance) , Indian Academy of Sciences
12. The daylight sky and Avogadro's number , European Journal of Physics (2015-01-01)
13. Einstein's impact on the physics of the twentieth century , arXiv physics/0507107 (preprint)
14. Rayleigh scattering , Wikipedia
15. Different Perspective on Blue Sky Theory: Theory of Single-Photon Scattering on Bound and Free Electrons , arXiv 2107.06143 (preprint, not peer reviewed) (2021-01-01)
16. Brazil - Por que o céu é azul? Um experimento sobre espalhamento Rayleigh , Revista Brasileira de Ensino de Física (SciELO)
17. NASA On Air: NASA's Curiosity Sees Blue Sunset On Mars , NASA Scientific Visualization Studio (2015-05-21)
18. Blue Mars sunset captured in NASA video , CBC (2015-01-01)
19. Perseverance Captures a Beautifully Strange 'Blue Sunset' on Mars , ScienceAlert (2023-01-01)

This article was drafted with the help of AI and reviewed by Jaroha editors before publication.