enter image description here
Sun's light appear to travel as parallel beams towards earth $_1$. Sun produces electromagnetic radiations through pp chain and other reactions in Photosphere $_2$.

I don't see whether these reactions send photons in that neatly arranged parallel rays, or else any other effects make the rays have such parallel beautiful motion.

So what makes Sun's light travel as parallel beams towards earth?

Links: A very good news for those who wants to desperately know the answer. I have replaced the word "ray" with "beam", as both have difference in their meaning. I have extracted a small passage from the webpage Light beam - Wikipedia :

A light beam or beam of light is a directional projection of light energy radiating from a light source. Sunlight forms a light beam (a sunbeam) when filtered through media such as clouds, foliage, or windows.

If the passage is saying truth, we are only left with the question, how clouds (the question is concerned with this) form beams?

Reference: $_1$ Crepuscular rays-Wikipedia $_2$ Sun-Wikipedia

  • 4
    The image above is stolen from Nathaniel's answer to other question. – Immortal Player Dec 26 '14 at 10:29
  • 3
    Quick answer: Rays are going in all directions, but the further a planet is from the Sun, the smaller the arc of the Sun's rays that intercept the planet. The earth is far enough away for the rays to be considered parallel in this case. – Phonon Dec 26 '14 at 10:37
  • @Phonon: I have seen your quick answer in other website! Just joking:) Let me be clear, I want to know how these photons are sent in that parallel order. As I understand or imagine, photons should be thrown randomly from Sun's surface. – Immortal Player Dec 26 '14 at 10:41
  • They are not sent in any parallel order, they're sent in every possible direction! The ones that are intercepted on earth's surface, correspond to a very small solid angle, for which all rays can be considered roughly as parallel. – Phonon Dec 26 '14 at 10:44
  • @Phonon: they're sent in every possible direction- I hope your "they're" is nothing but photons, isn't it? – Immortal Player Dec 26 '14 at 10:53

Close to the surface of the Sun, the light it emits is not in parallel beams at all. It sends photons out in pretty much every direction at random, like this:

enter image description here

But the Sun is far away from Earth. Most of those photons don't reach the Earth because they're heading in the wrong direction. Only those photons that happen to be heading toward the Earth actually get there - and the photons that are heading to the Earth are all heading in more or less the same direction. So the photons that actually reach the Earth are all travelling in more or less parallel directions.

Below I've drawn the photons that reach the Earth in red, and the other photons in grey. You can see that the red ones are more or less parallel. (In reality, the Earth is further away than that, so the photons are more parallel than in the diagram.)

enter image description here

To answer your last question: sunbeams are formed when a gap in the clouds lets some light through, like this:

enter image description here

Photons from the Sun are always travelling in more or less the same direction, because they all came from the Sun, but it's easier to notice this when the light is concentrated in one particular place like this.

Finally, we have to explain why you're able to see the beam at all. This is just because some of the photons in the beam collide with particles in the atmosphere, which scatter them in all directions, and some of these scattered photons happen to hit your eye.

  • +1 Thank you for the answer. Diameter of earth is 12,742 km. The width of the photon cloud from the Sun is a small fraction of a millimeter [Extracted from the book "God does play dice with the universe" by Shan Gao]. Seeing these numbers in mm and km, I think, the second diagram seems to conclude wrong conclusions (though it is not to scale, scale might make a difference here). – Immortal Player Dec 30 '14 at 9:37
  • If we consider scale, earth will be bigger and bigger than photon cloud (which seems to be opposite in the diagram including earth and sun), then the angles between photons (which seems to be very very less in the diagram) makes more difference w.r.t earth, so, even now I have trouble in understanding your view (that photons will be parallel as seen in the clouds diagram), what I think is photons should come in criss-cross form. Finally, once again, thank you for the effort. – Immortal Player Dec 30 '14 at 9:45
  • I'm not familiar with the reference you cite, and the term "photon cloud" is not standard terminology, so I can't say anything about its width. You're right that the incoming photons criss-cross each other, but the point is that they only do it a bit. (If they didn't criss-cross each other, you wouldn't be able to form an image of the sun using a pin-hole camera.) The photons' paths differ from each other by at most the Sun's apparent diameter in the sky, which is about half a degree, so they're pretty close to parallel, which is why you can see sunbeams as distinct lines. – Nathaniel Dec 30 '14 at 9:53
  • The scattering of photons is not the correct explanation. That would not lead to well defined beams of light as in the picture. – JezuzStardust May 17 '16 at 11:04

The Sun rays are not exactly parallel. They seem parallel as "Phonon" says, see her comments, but at a precise examination they are not so. Are you aware of the Hanbury Brown Twiss experiment? Look in Wikipedia at the site


and see the lower picture in that article. I copied the picture below. The experiment was done by observing the star Sirius but with our Sun it can be done also. In short, rays coming from a points $a$ on the star's disk, reach on the Earth two points, A and B, and so do rays from another point $b$ from the star's disk.

enter image description here

  • Thank you for the answer. Are those Red and green arrow composed of photons coming back to back? – Immortal Player Dec 26 '14 at 12:00
  • My mother-tongue is not English, s.t. what is "back-to-back? At the place A, the red ray and the green ray reach exactly the same point. Similarly at the place B. (The Hanbury Brown Twiss experiment was a great discovery. It showed that there is a correlation between what we get at the point A, and what we get at the point B, but this goes beyond the scope of your question.) – Sofia Dec 26 '14 at 12:13
  • Let me say in other way, I hope the ray is composed of photons all moving in the same direction, one behind the other, did you also mean this? I have started understanding the complexity in finding the solution to the question, like photon clouds, path of photons, etc. I feel uncomfortable now whether ray is actually a straight line with photons being lined up one behind the other. – Immortal Player Dec 26 '14 at 12:35
  • No, it's not exact. By Huygens' principle, any point of the solar surface emits a spherical wave (well, solar surface is an imprecise term, but for our purpose it is good enough). Part of the wave enters the Sun and part goes outside the Sun, from which, further, an extremely small part comes to us. What I try to say is that the photons emitted by each such point are, each photon, a spherical wave. They are not small wave-packets, many, many of them along each straight-line ray from the sun to us. No, the photons are big creatures growing wider and wider as the spherical wave expands. – Sofia Dec 26 '14 at 13:13
  • But, when touching an apparatus of ours, such a fat creature delivers it energy to one molecule on a photographic plate. This is why we draw a straight line between our apparatus and the point a (or b ) on the Sun surface. (To be rigorous, the energy is delivered by some more complicated process than I said, but again, for our example, the delivery to a molecule is a good enough expression.) – Sofia Dec 26 '14 at 13:13

Reading some of your earlier comments, it seems that there is a slight confusion at play here. From a comment on the original question:

I want to know how these photons are sent in that parallel order. As I understand or imagine, photons should be thrown randomly from Sun's surface.

Technically, you are correct: photons - including streams of photons following the same path, or "back-to-back" as you have said - are emitted in every direction from the surface of the sun. The sun is not a big flashlight, pointing only at the Earth; it's a big flame, shining all around it.

However, just because the photons leave the sun in all directions, does NOT mean that they strike the Earth from all directions. In order for that to happen, the sun would have to be all around us.

If you briefly imagine the Sun and Earth as points, and draw a quick sketch of the sun emitting straight rays of light, you will see that all the rays that actually strike the Earth (rather than missing it) are going the same direction - are parallel, in other words.

Now, the Earth is larger than a point, and the Sun is larger than the Earth, so the average of all the rays that leave the Sun and hit the Earth actually converge, but because the Earth is so small and so far away (i.e. it subtends a very small angle from the Sun) the rays are almost completely parallel, coming uniformly from the direction of the sun.

Concerning the question of how the clouds form sunbeams, the short answer is that they simply cast shadows that provide enough contrast so you can see the sunbeams. The long answer is that under a clear conditions the entire sky looks like sunbeams; sunlight scattering off of dust and gases in the atmosphere: and with the clouds in the way, only certain portions of the sky allow light through to the ground, so you see distinct lit beams along the path the light is taking from the sun to the ground, against a dark background.

  • Thank you for the answer. Do you agree that photons are thrown randomly? If you agree that, I see no way how photons can be sent back to back? – Immortal Player Dec 27 '14 at 2:01
  • 1
    What do you mean by "sent back to back?" – Asher Dec 27 '14 at 5:09
  • Let me cut and paste what I said to Carlwithoft above in a comment: "What is bothering me is, by reactions in sun, (I think) each photon should be ejected in different directions, then there will be no chance in the formation of a ray (which I think to be line formed by photons one behind the other), so no question of whether rays are parallel or not. Though, I see some uncertainty going on here with rays and feel myself to be not accurate. I wish the discussion will become accurate as it goes." I hope this makes me clear. I look forward for your comments. Thank you. – Immortal Player Dec 27 '14 at 10:00
  • I have edited the question. – Immortal Player Dec 27 '14 at 10:36
  • I've also added a paragraph at the end of my answer that answers your question edit, I hope. – Asher Dec 28 '14 at 1:48

Suns rays are isotropic, but if we assume a small area $dA$, then the rays would be perpendicular to that crossection, and parallel with eachother.

Light from the sun actually converges on the earth - the sun is about 220 Earths across, so light from one edge and light from the other edge must converge to reach the same point on Earth. The angle is small enough (about 0.5 degrees) that for most practical purposes we can consider the sun to be either a point source or a uniform flat source as needed.

  • 3
    The rays don't really "converge". Instead they both converge and diverge - depending on which pair of rays you take. Two rays form the opposite edges indeed converge on the Earth if we can see them. But those two emitted from a single point will actually diverge. Yet some of the rays can be parallel. – Ruslan Dec 26 '14 at 12:18
  • @paul: I have edited the question. – Immortal Player Dec 27 '14 at 10:37

protected by Qmechanic Aug 17 '16 at 6:29

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

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