Recently, it was a fine sunny morning (around 100 minutes after sunrise) and the sun was briefly behind a small cloud. Near the sun was one of those wispy clouds one sometimes sees (possibly Cirrus clouds?), at the very top of which I suddenly saw a very bright 'band' of rainbow (wider in the red to violet direction than a normal rainbow, and spanning the width of the cloud) and perhaps 15 or so degrees away from the sun. I unfortunately did not have a camera so my appalling paint drawing will have to suffice (the lines are the cloud, and the band is the approximate size of the band of rainbow; the omelette is the sun behind a cloud). It was very bright at first and surprisingly vivid (all colours red through violet), but faded within about 20 seconds from when I saw it; it was one of the most remarkable things I have ever seen: a bright rainbow band in a small cloud near the sun. My question is: I know about iridescent clouds, and circumhorizontal arcs, but does anyone know what this would be since it was a band of wide rainbow spanning the width of the cloud, so near the sun?

enter image description here

Edit: From what's been said, I think what I saw was likely a sun dog, something I've been waiting years to see, and didn't even know it when I saw one! The interesting thing about this one was that it had distinct blue and violet bands in addition to the rest of the spectrum, something I haven't seen in any pictures (although I am aware that cameras don't necessarily capture violet well), and which some sites e.g. here and here appear to imply is not a feature of most sundogs, with a blue-white tail instead being typical. I wonder if anyone could explain what might make one more distinct. Perhaps increased stability of the ice crystals might be the answer, or is it more likely that the violet band was not actually as vivid as I remember it to have been?

  • I just came across the answer yesterday, strangely enough:… – Xcheckr Sep 20 '16 at 22:49
  • That link explains the colour formation at that particular angle, which I was aware of. What was actually interesting me was how the violet might appear brighter than usual. The first site mentioned in my edit explains the blue-whiteness as being due to the ice's birefringence, which would mean that should be a constant, so I don't see how that would change; thus I thought stability might be a factor. – Anon Sep 20 '16 at 23:06

There is a full display of these that can occur. The picture here illustrates one of these enter image description here

The names given to the parts of this are seen below. This is a St. Petersburg display. I first saw one 25 years ago. I shut the hatchback on my car and the circumzenithal arc (see image below), the one at the top arcing away from the sun, was reflected in the hatchback window. I looked up and was rather stunned. I have seen two other such displays and photographed one of them as I was home at the time with a camera. The last one I saw was maybe 7 years ago.

These occur with ice crystals, such as cirrus clouds. The different arcs are due to different paths through hexagonal prism ice crystals. If you are interested you can visit this web page to get more of the atmospheric physics on this. Sundogs are the most common feature to appear. They are common in the early morning. A full display happens on a day with an extremely cold upper atmospheric layer. The three times I have seen these it was actually not that cold near the ground. enter image description here

  • Thanks for the answer; I will wait a bit before nominating a correct answer. That image is truly a feast of halos! Although I'd seen pictures of sundogs like the one in your image, I had no idea that they could look like a piece of rainbow. Frankly I'm still surprised, since none of the pictures I have seen (e.g. Chris's answer) have been as vivid and saturated as the one I saw. It was almost like the circumzenithal arc in your picture, only much wider in the direction of scattering, and horizontal and parallel with the sun. – Anon Sep 20 '16 at 0:33
  • 3
    Wow, what an amazing photo! – rob Sep 20 '16 at 2:51
  • Is it possible for this kind of thing to happen with the moon? – Dawood ibn Kareem Sep 20 '16 at 9:01
  • 1
    The desert southwest US is ideal for creating these. Sundogs are an almost everyday occurrence from late fall to early spring. You can also get optical effects around the edges of cirrus clouds that produce fushia type colors or rainbow effect in the cloud. – Lawrence B. Crowell Sep 20 '16 at 10:46
  • @DavidWallace, you can indeed get these with the moon. See moon halo and moon dog. I understand they are fainter, like a moonbow. – Anon Sep 20 '16 at 22:22

That will be a sun dog. I saw one last week. (The sun is over on the left in this picture.)

enter image description here

  • 1
    That is indeed much what it looked like. The one I saw was very vivid, and lucky because it was just at the top of a cloud, so I almost missed it. As to whether it was a sundog though, I wasn't aware that sundogs usually had monochromatic splitting in them. – Anon Sep 19 '16 at 23:04
  • 1
    @Anon - the rainbow colours were much more vivid though my sunglasses, and still looked more vivid without sunglasses than the picture shows. – Chris Degnen Sep 20 '16 at 10:08
  • I see, I guess your eyes are better at filtering out the glare than the camera is, and capture the colours more truly. – Anon Sep 20 '16 at 22:23

Was it a cold day? In that case, I think you saw an halo, a 22° halo to be more precise.

enter image description here

Halos are formed when the sunlight is reflected in ice crystals in the atmosphere, often present in the cirrus clouds that you mentioned.

Quoting from the Wikipedia page:

As light passes through the 60° apex angle of the hexagonal ice prisms it is deflected twice resulting in deviation angles ranging from 22° to 50°. The angle of minimum deviation is almost 22° (or more specifically 21.84° on average; 21.54° for red light and 22.37° for blue light). This wavelength-dependent variation in refraction causes the inner edge of the circle to be reddish while the outer edge is bluish.

Sun dogs are a related but different phenomenon. They are

two subtly colored patches of light to the left and right of the Sun, approximately 22° distant and at the same elevation above the horizon as the Sun

(from here). In fact, they are often accompanied by a 22° halo, but the two phenomena are independent and you can see one without the other.

  • It wasn't a cold day. I have been wondering about the 22 degree halo, but what makes me doubt it is the rainbow splitting which I don't think is so clear with halos (e.g. your diagram). There was only a narrow horizontal band (although my diagram looks like there was an arc - that's the cloud) which is typical of sundogs, but the cloud was small so it could have been a lucky view of halo. At that stage, I think the distinction between the two breaks down: see here which says the distinction between tall sundogs and partial halos is not important. – Anon Sep 21 '16 at 6:16
  • 1
    @Anon Yes, at some point the distinction becomes indeed quite arbitrary...but I think we can agree that the origin of the phenomenon must be the presence of ice crystals (and therefore that the cloud was a cirrus). Actually, even if it was not a cold day, the temperature must have been pretty low at 5000 m, where cirrus clouds usually form. – valerio Sep 21 '16 at 8:21
  • I agree that the phenomenon was most likely ice crystals, and that the cloud was probably cirrus. – Anon Sep 22 '16 at 4:21

rainbow clouds

My google searches for what I saw today led me here. I've been seeing yellow and orange streaks in the sky near the sun recently but today was like nothing I've seen before. The clouds were highlighted with all the colors of the rainbow. Reminded me of an oil spill.... doesn't seem natural and it wasn't ice because it was hot today. Wish people would notice/talk about these things more. Seems to go unnoticed by the majority of people.

  • Welcome to Physics.SE! If you want to post something relevant but not answer-worthy, please post it as a comment. This answer does not answer the OP's question. – Fine Man Mar 10 '17 at 4:40
  • Interesting picture though. It sounds like cloud iridescence. – Anon Mar 11 '17 at 6:04

Your Answer


By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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