Especially on a sunny day in the winter or spring, I often notice faint rainbows when I look down at the edge of the road by the curb. I am not sure what causes these rainbows, but I believe that they are caused by light reflecting off tiny salt crystals that have accumulated on the ground.

Why do rainbows form under such conditions? The standard mathematical explanation for rainbows assumes spherical raindrops for simplicity, and for water droplets suspended in air, that seems like a reasonable approximation. But salt crystals (or whatever particles on the ground that are causing the rainbows) have irregular shapes and are dispersed randomly on the ground. Wouldn't such random configurations preclude a reflection pattern as coherent as a rainbow?

EDIT: To help clarify what I'm talking about, here is a short video of the type of rainbow effect that I have observed. (You might want to turn off the sound; unfortunately I don't have a video editing program to remove the sound.)

EDIT: The accepted answer below mentions retroreflector glass beads. I took a sample from the ground, which a friend of mine took a close-up photo of as well as a flash photo of. Both photos support the glass bead theory.

  • $\begingroup$ Or thin films of oil make nice rainbow effects. You have assumed a cause that doesn't seem to necessarily apply. $\endgroup$
    – Jon Custer
    Commented Apr 20, 2022 at 15:52
  • $\begingroup$ @JonCuster It's true that my assumption may be incorrect. On the other hand, when I observe this phenomenon, it's when the ground is bone dry and the weather is bright and sunny. (Oil is certainly not the cause; those color patterns are swirly and not band-shaped or bow-shaped, as these patterns are). If I move my head side to side, it creates an illusion that the rainbow is a few inches above the ground, and moving as I move. $\endgroup$ Commented Apr 20, 2022 at 17:19

2 Answers 2


I suspect the rainbow is from retroreflector glass beads rather than salt. Tiny glass beads are placed on the surface of road markings, halfway embedded, making them reflect light from vehicles back to them with total internal reflection. But when they get loose they act as cloud droplets instead, making rainbows.

  • 1
    $\begingroup$ I think you are right. See the photos in this blog atoptics.wordpress.com/tag/glass-bead-rainbow and epod.usra.edu/blog/2010/07/glass-bead-bow.html. A simple test is to measure the radius of the rainbow, which should be somewhere between 20 and 30 degrees. $\endgroup$
    – jkien
    Commented May 1, 2022 at 10:42
  • $\begingroup$ @jkien - Yes, the rainbow in the video does look like it is tighter than a rain rainbow given the distance to the shadow of the camera, although it is hard to tell exactly. $\endgroup$ Commented May 1, 2022 at 23:50
  • $\begingroup$ I like this proposal. I will see if I can collect a sample of the particles on the ground to examine more closely. $\endgroup$ Commented May 2, 2022 at 13:29
  • $\begingroup$ There do seem to be some tiny glass beads on the ground where I have observed the effect I described. I don't have a camera that is good enough to photograph these beads in a meaningful way, since they're so small, but I am going to accept this answer as being most likely correct. $\endgroup$ Commented May 3, 2022 at 23:10

Looks like, e.g., ice crystals can cause some rainbow-like effects. So maybe salt crystals or some other crystals can cause the effect in your video. It seems that in your video the rainbow is present only on the white(ish) areas of the road. Maybe those are some transparent crystals.

  • $\begingroup$ In the video the rainbow is in the back-scattering direction (i.e. >90°), while ice halos are usually in the forward direction (e.g. 22° and 46° halos), and they don't look as colorful. $\endgroup$
    – Ruslan
    Commented May 1, 2022 at 13:59
  • $\begingroup$ @Ruslan : "Halos can have many forms, ranging from colored or white rings to arcs and spots in the sky. Many of these appear near the Sun or Moon, but others occur elsewhere or even in the opposite part of the sky." (en.wikipedia.org/wiki/Halo_(optical_phenomenon) ). I guess it depends on the shape of the crystals. $\endgroup$
    – akhmeteli
    Commented May 1, 2022 at 14:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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