I noticed that the skylight where I work looked white to me, but its reflection from the tiles below showed a colorful pattern. See the images below, and also a video that shows this effect from various angles.

How do these patterns form, and why can't I see them directly?


The skylight, its reflection on the tiles, an enhanced version of the second image.

Some observations that might be useful:

  1. The change in color of the reflection near the beams may suggest that this has something to do with the skylight glass and not an interference pattern from the tiles.

  2. The colors only appear when you look it at from an angle.

  3. The birefringence patterns mentioned in this answer look very similar.

  4. That answer also mentions that light from the sky can be polarised.

  5. But you need a polariser to see this effect. Where is the polariser here? I am not using any polarisers on my phone camera.

  • $\begingroup$ Are the floor tiles glazed? $\endgroup$
    – PM 2Ring
    Commented Dec 27, 2023 at 12:24
  • 1
    $\begingroup$ @PM2Ring Yes, they are glazed vitrified tiles I guess. $\endgroup$
    – AlphaLife
    Commented Dec 27, 2023 at 15:38

1 Answer 1


I think the effect is the same as in the question you linked. The glass is birefringent, causing different wavelengths to get polarized differently (note that in the video, the color patterns appear to move with the windows and not with the tiles). While in the question you linked a polarizing filter is responsible for the appearance of colors, here the tiles act like the polarizer. Reflection from planar surfaces is polarization-dependent: transverse electric (TE, also called s-polarized) and transverse magnetic (TM, also called p-polarized) waves are reflected by different amounts. As an example, the figure below shows reflectances and transmittances as a function of angle of incidence on glass. (Image credit, CC BY-SA 4.0)

While the reflectances are the same for normal incidence (in which case we cannot meaningfully speak of TE and TM polarization anyway), reflectance tends to be higher for TE polarization at off-normal incidence. Famously, if the angle of incidence is equal to Brewster's angle, the reflected wave is entirely TE polarized, so the floor acts essentially like a linear polarizer. The fact that the floor acts like a polarizer only for off-normal incidence would explain why you only see colors when you look at the floor at an angle.

  • 1
    $\begingroup$ it's likely the stressed polymer films embedded in or applied to the windows rather than the glass itself that are polarizing. $\endgroup$
    – uhoh
    Commented Dec 27, 2023 at 1:51
  • $\begingroup$ Don't forget that scattered light from the sky is also polarised. The skylight is birefringenant and the reflection Is the analysing polarisor. en.m.wikipedia.org/wiki/Rayleigh_sky_model $\endgroup$
    – D Duck
    Commented Dec 27, 2023 at 18:47

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.