Skip to content

Why Is the Sky Blue?

It has nothing to do with the ocean. The answer is a scattering law — and the same law explains why sunsets are red.

The short answer

The sky is blue because of Rayleigh scattering. Air molecules are far smaller than the wavelength of visible light, so they scatter short wavelengths much more strongly than long ones — blue roughly five times more than red. That scattered blue light reaches your eye from every direction at once, so the whole sky glows blue. It is not the ocean reflecting upward.

0:001:06

Transcript

Okay, why is the sky blue? Not because it reflects the ocean. That is the myth everybody repeats, and it is the wrong answer on every test.

Sunlight looks white, but it is every color mixed together. When it enters the atmosphere it hits nitrogen and oxygen molecules far smaller than the light's own wavelength, and they scatter it in all directions — but not every color equally. Short wavelengths scatter far more than long ones. That is Rayleigh scattering, and it goes as one over wavelength to the fourth power, so blue scatters about five times more strongly than red. Blue arrives at your eye from every direction at once, and the whole dome glows.

Then why not violet, which scatters even more? The sun emits less violet, and your eyes are much less sensitive to it.

And sunsets? Same physics, opposite result. Near the horizon the light crosses far more atmosphere, so the blue is scattered away before it reaches you — and what survives the trip is red and orange.

Test yourself on Physics

5 questions, easy to hard. No account needed to try it.

Question 1 of 5Easy

What is the name of the effect that makes the sky blue?

Question 2 of 5Easy

Which colors of sunlight are scattered most strongly by air molecules?

Question 3 of 5Medium

Rayleigh scattering varies with wavelength as one over lambda to the fourth power. Roughly how much more is blue light scattered than red light?

Question 4 of 5Medium

Why does the sky look red near the horizon at sunset?

Question 5 of 5Hard

Violet light has a shorter wavelength than blue and scatters even more. Why does the sky not look violet?

The longer answer

The short answer is Rayleigh scattering. The answer worth having is why the wavelength dependence is so steep, because that single fact explains the blue sky, the red sunset and the blue-grey haze over a distant mountain range — and it is the part exams actually test.

Sunlight arrives as a mixture of every visible wavelength, running from about 400 nanometres at the violet end to about 700 at the red end. The molecules it meets on the way down — overwhelmingly nitrogen and oxygen — are roughly 0.3 nanometres across. That is not slightly smaller than the light hitting them; it is hundreds of times smaller. This size relationship is the whole story, because it decides which scattering regime applies.

When a particle is much smaller than the wavelength striking it, the light's oscillating electric field drives the particle's electrons into oscillation, and the particle re-radiates as a tiny antenna. The efficiency of that re-radiation is extremely sensitive to frequency: intensity scales as one over wavelength to the fourth power. Put real numbers through it. Red at 700 nanometres against blue at 450 gives a wavelength ratio of about 1.55, and 1.55 raised to the fourth power is close to six. The same air scatters blue several times more strongly than red, and nothing about the air needs to change for that to happen.

That is why the sky has a color at all. Light coming straight from the sun's disc is mostly unscattered, which is why the sun looks white-ish rather than blue. Light reaching you from any other part of the sky got there by being redirected, and redirected light is disproportionately blue. When you look up at an empty patch of sky you are not looking at a blue object. You are looking at scattered sunlight arriving off-axis.

Two refinements matter, and both tend to be where exam questions hide the trick.

The first is violet. Violet sits at a shorter wavelength than blue, so the formula says it should dominate even harder. It does not, and the reason has nothing to do with scattering physics. The solar spectrum peaks in the blue-green and falls away toward the violet end, so there is less violet in the beam to begin with. On top of that, the human eye's cone response to violet is weak compared with its response to blue. The atmosphere really is scattering plenty of violet; our detectors are simply poor at registering it, and the mixture we perceive reads as blue. Any question that asks "why not violet" is checking whether you know the difference between what is physically scattered and what is physiologically seen.

The second is the sunset, which is not a separate phenomenon but the same one measured along a different path. With the sun overhead, sunlight takes the shortest possible route through the atmosphere. With the sun on the horizon, that path is many times longer — it cuts across the atmospheric shell rather than dropping through it. Along a long enough path, so much blue has been scattered sideways out of the beam that almost none is left travelling toward you, and what survives is the red and orange that scattered least. A red sunset is the leftovers of a blue sky. The same logic explains why distant mountains look hazy and blue-tinged: you are seeing scattered light added along the line of sight, which is exactly what atmospheric perspective means in a landscape painting.

It is also worth being precise about what this is not, because the wrong answers are popular enough to be worth naming.

The sky is not blue because it reflects the ocean. The effect is just as strong over the Sahara and over the middle of a continent, and it works looking down from an aircraft at any surface below. The causation in fact runs the other way for the ocean: sea water looks blue partly because water absorbs red wavelengths over a few metres of depth, and partly because the surface reflects the sky that is already blue.

The sky is also not blue because of water vapour, dust or pollution. Those particles are comparable to or larger than the wavelength of visible light, which puts them in a different regime — Mie scattering — where the strong wavelength dependence disappears and all colors scatter roughly alike. That is precisely why a humid or smoggy sky looks washed-out and white rather than deep blue, and why clouds, made of droplets far larger than a wavelength, are white rather than colored. If dust made the sky blue, dustier skies would be bluer. They are the opposite.

Finally, the reason the sky is black from the Moon or from orbit: no atmosphere means no molecules to scatter the light. The sunlight is still there, arriving with the same spectrum, but there is nothing in the path to redirect any of it toward your eye. The blue you see from the ground is not a property of sunlight or of space. It is a property of the air.