
On a rough reflective surface like the ocean or a dark wet road, reflections from bright lights like city lights, car brake lights, or the Moon appear stretched vertically. Why is this?
When a surface is perfectly flat, like a regular mirror, the image we see in the reflection isn’t distorted at all. Even if we put a mirror flat on the ground, we wouldn’t see a vertically stretched reflection like this.
Neither the road nor the ocean are perfectly flat though. Their surfaces are rough, and this rough surface scatters light when it’s reflected. If we imagine that each piece of the surface was a little flat mirror, with each piece facing in a random direction, some of these would be at the right angle to reflect light from a source (like the Sun) directly into our eyes, and most would not. We’d only see a reflection in those pieces that are at the correct angle to reflect the light into our eyes.
The further these little mirrors are from the area where we’d see the reflection in a flat mirror, the more extreme an angle they will need in order to still reflect the light at us. If every one of these little surfaces was really really tiny, what we’d expect to see is a blurry reflection. The smaller the pieces get, the less blurry the reflection would get.
We can actually see this in effect when we compare pictures of the Sun reflected off the ocean. When you’re quite near the ocean, all the different reflecting surfaces are relatively large so the reflection is quite blurry and broken (especially if there are lots of waves):

In comparison, if we look at a reflection of the Sun on the ocean that was taken from space, all the waves and ripples that distort the reflection are far too tiny to see, and as a result the reflection is quite clear and crisp:

Another difference that’s quite apparent between these photos is the vertical stretching that I’ve been wondering about. From up close, it’s very stretched. From a distance? Not so much. This gives me a thought, one that actually hadn’t occurred to me until I got to this point in writing this post and saw those images one after another:
What if it’s important that there’s a significant relative distance between the closest and furthest parts of the surface that are reflecting the light source?
From a long way away, these distances appear quite small. For example, if I’m 1 km away from a surface, then a 1 m distance between two points on that surface is really quite small. If I’m only a metre away myself though, then that’s a very significant distance.
As we just saw, reflections on non-flat surfaces are more blurry when they’re closer to you, so what if this vertical stretching is actually just the reflection getting more blurry towards the bottom, because that part of the road or ocean is closer? As it’s more blurry, this would let the edge of the reflection creep out further, and could look like stretching.
If I’m right, then I should be able to see the same type of stretching if I look at a reflection on a vertical surface, except the stretching would be horizontal in that case. I should also be able to replicate the same stretching effect if I can get a reflecting surface that is smooth on top and gets rougher towards the bottom, and look at a reflection of a light in it like I would a normal mirror (i.e. with the reflecting surface vertical and the light source behind me).
Let me know what you think of this idea in the comments, and if you have any ideas of your own for why we see these stretched reflections. Any ideas about how I could try to disprove my idea would be welcome too! In the meantime, I’ll try to do these experiments, and see if I can find an expert to talk to about this question.
I’ve written more on this topic in another article: Natural Curiosity: Stretching Reflections 2