Edit: oops, this should have gone on the Refraction page.
Sorry, but that is patently false (says me, a stranger on the internet). But seriously: the different angles observed in the photo are merely a geometrical effect of perspective. All rays can be followed back to the position of the sun in the photograph.
This is a terrific image, however I do not think that what is happening is the result of refraction. The obvious, the light rays can be seen because of dust in the atmosphere. The source of the light is the sun. We can determine where the sun would be in this image by extending the lines of the rays to the point where they meet--just like single point perspective. If the clouds disperse enough to reveal the sun that's where it would be. There must be some degree of refraction of sunlight caused by the vacuum/atmosphere transition, but I don't think we visually perceive it, (except maybe at sun rise and set) because we can't see the rays outside our atmosphere to make a comparison. Refraction of sunlight could be perceived and photographed in an air/water transition.
The reason the light rays spread out is the same as when we perceive (parallel) railroad tracks converging to a point in the distance. WillN2 (talk) 18:06, 5 September 2011 (UTC)
