As far as I know, the sky is blue because as light from the sun enters the atmosphere composed primarily of wavelengths of visible light. However, as the light happens across molecules of oxygen and nitrogen in the air, they have a small chance of scattering off, and the shorter wavelength photons such as blue light have much more chance of scattering. Hence, the reds and oranges sort of pass through the atmosphere in a straight beam, whereas the blues and greens get scattered around the place randomly, meaning that the sky appears pale blue. enter image description here

If, however, the scattering is random, then shouldn't just as many blue photons be sent flying back out to space as make it to our eyes on the ground? Why aren't all pictures we take from space of the earth blue? I can understand if the effect is not very prominent, but in all of these pictures the effect seems almost non-existent. Is there a flaw in my understanding, or is the effect just a lot smaller than I thought it would be?

  • $\begingroup$ Hm, good question. I would have thought it would be because the scattering takes place mostly in the forward direction, but in fact it has a $(1+\cos^2\theta)$ dependence which means light is equally likely to be scattered backwards. Hopefully someone else can follow this up into an actual answer. $\endgroup$ – David Z Nov 3 '15 at 7:41
  • $\begingroup$ I think wikipedia let this very clear. Yes, the intensity of light scattered in one direction, $\theta$, is proportional to $(1+\cos^2\theta)$, but what is important is the average of the scattering in all direction, which gives the $\textit{ Rayleigh scattering cross-section}$. Then, the fraction of light scattered by a group of "scattering particles" is the number of particles per unit volume times the cross-section. It also brings an example. $\endgroup$ – raul Nov 3 '15 at 8:56
  • $\begingroup$ "the major constituent of the atmosphere, nitrogen, has a Rayleigh cross section of $5.1\times 10^{−31} m^2$ at a wavelength of 532 nm (green light). This means that at atmospheric pressure, where there are about $2×10^25$ molecules per cubic meter, about a fraction $10^{-5}$ of the light will be scattered for every meter of travel." $\endgroup$ – raul Nov 3 '15 at 8:56
  • $\begingroup$ Standing on Earth, you look up through the glowing blue sky at the absolute blackness of space. Up in Earth orbit, you look down through the glowing blue sky at clouds, oceans, and land forms that are all illuminated by blazing sunlight. $\endgroup$ – Solomon Slow Nov 3 '15 at 22:02

Well, the Earth is kind of blue when viewed from far away:

Pale Blue Dot

In the picture above, you see that the "edge" of the Earth is more bluish, which is due to the reason you give in your answer; here we see through a deeper layer of the atmosphere, so we see more blue light.

But the light that does reach Earth's surface is reflected back through the atmosphere, so this is seen as well. And since after all non-blue photons outnumber the blue ones, we see what's underneath the surface.


  • The Earth does look blue, even continents, from satellite. (Still, in normal direction the amount of atmosphere is not that large).
  • Almost all the satellite photographs that you see are retreated to improve the contrast and restore the white balance. (and many other issues such as band-variations of illumination, hot spot, etc).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.