Why Do Polarizers Make The Sky Bluer and The Trees Greener? I've been reading a lot about polarization and I have pretty much understood the physics behind the phenomenon. However, until now I couldn't find a scientific reference that explains clearly why the sky, for instance, appears bluer than usual if it were to be photographed with a polarization filter on top of a camera? The same for grass and tree leaves and haze I suppose.
The closest I've read so far is, that is due to multiple reflections occurring in the atmosphere that causes a specific wavelength, say the blue, to lose its polarity when reaching our eyes. While putting a polarizer in front of a camera it will eliminate all these randomness of polarity caused by those reflections.
PS: if you could provide a scientific reference to that, that would be appreciated.
 A: It's not true in general that a polarizing filter makes colors appear more saturated or changes their hue in some universal way. If you take shot A with the polarizer in a certain orientation, then rotate the filter 90 degrees and take shot B, the sum of the two pictures is identical to what you would have gotten without the filter. Therefore if A has more saturated greens for the leaves on the trees, then B has less saturated greens. If the sky is more blue in A, it's less blue in B.
When light is scattered by air molecules or by aerosols such as fog or smog, the scattering is stronger for shorter wavelengths and the scattered light is polarized. The polarization depends on the angle of scattering.
If you photograph the sky through a polarizer, you will see changes that vary depending on the orientation of the polarizer, and that also vary from one part of the sky to another.
If you photograph a distant mountain, then you're seeing both light from the mountain and light scattered by the haze between you and the mountain. Here is a picture from Wikipedia that shows the effect:

As you go to more distant parts of the landscape, the colors become bluer and less saturated. The less saturated color is basically the same effect you get when you're standing in a thick fog: the light is completely scrambled, so the color in any given direction is just an average of all the colors around you. If you were to put a polarizing filter in, you could manipulate this photo to make the distant mountains look "better," i.e., less influenced by the haze. Light that has only been reflected from the mountains is partially polarized by reflection (especially if the sun is overhead-ish at close to Brewster's angle). The multiply scattered light is nearly unpolarized. If you can enhance the amount of directly reflected light in proportion to the multiply scattered light, you'll get a more saturated and less washed out image of the mountains.
Something similar will happen any time you're photographing an object that acts like a dielectric, so it works with leaves, water, snow, and glass, for example. There's an angle of reflection that is defined by the sun, the object, and the camera. The closer this angle is to Brewster's angle, the stronger the polarization will be. You can then use the polarizer to either enhance or reduce this light. For example, people often want to reduce the glare from snow or water.
One thing that confused me when reading about this topic was that people talk about using linear polarizing filters and circular polarizing filters for photography. Actually both of these act like linearly polarizing filters in terms of what they do to the image. The things photographers call circularly polarizing filters are not what a physicist would call a circularly polarizing filter. They're linearly polarizing filters that contain an additional quarter-wave plate. Modern light sensors, etc., do not work well with linearly polarized light.
