Hands down, my favorite book for browsing is M. Minneart’s The Nature of Light & Color in the Open Air. This small, easy to carry (and cheap if you get the Dover edition) book has 233 short sections on things one can see 'in the open air'; for example, twinkling of stars, the color of lakes, and the rainbow. Section 93 is on black snow, in which he notes how the apparent whiteness of a falling snowflake seems to change from blackish to whitish when the flake’s background changes from the grey sky above to a darker background below (such as dark-green trees). It is a simple enough observation, but I didn’t bother looking for it until a few years ago. Indeed, the illusion of whiteness change is strong. Look for it next time you go out when it's snowing.

On a walk on Jan. 2, I took the picture below because I thought the sun poking through the cloud was interesting, and then later realized that I could easily compare the whiteness of the flakes. Minneart's next section describes a related contrast illusion that also can be tested in the picture - the feeling that the snow on the ground is brighter (or whiter) than the sky above. See if you can shake this illusion next time you go out in the snow.

The sky in the picture is not uniformly grey, as Minneart assumes in his discussion, but is a yellowish-red at and below the level of the sun. So, for the comparison of the sky brightness to the snow brightness, I chose a patch of sky of sky at the place within the green box marked 'A' in the upper left, where the sky is darker. (To see this and other boxes, you can click on the picture to enlarge it.)

For the snow patch, I chose the green-box region marked 'B' in the lower left. Then I enlarged both patches, removed the obvious flakes of snow that had introduced darker spots into the areas, and put the regions side-by-side at right. One can see that indeed the sky is brighter than the snow. The tints are also different, with the sky having a pinkish tinge and the snowpack having a bluish tinge. This may be because the snowpack was partly shaded from the redder parts of the sky, and received most of its illumination from the bluish-grey regions away from the sun. Minneart discusses blue shadows under a blue sky, and this seems to be an example of the same phenomenon, except the sky is a cloudy grey-blue.

To compare the whiteness of the snowflakes, I chose the flakes contained with the red boxes labeled 'C', 'D', and 'E' in the picture. C has a sky background, D has a dark forest background, and E is hard to see, being a snowflake against a snow background. In appearance, C seems darkest, the 'black snow' discussed by Minneart, E is hard to see, and D seems the brightest and whitest. To see if D really is whitest, I cut out copies of the flakes, enlarged them, and put them on the right. Then I took a central region, marked by the red rectangles, and enlarged them again. When compared on the far right, flake C, the one against the sky, actually is a little brighter than the other two, and E, the lowest one, is darkest. The flake in the sky has a pinkish tinge, whereas the other two are bluish. This is strange because all three flakes were relatively close to me - their only difference is their background. The explanation may be that some of the background light penetrates or diffracts around the flakes, so the color of the illumination directly behind the flake can have a relatively strong influence on the color.

One problem with this test is that the camera may record a different color than the color one sees directly. I notice this when I photograph a rising sun. Here in Japan, with all of the haze and particulates, the rising sun is often a deep red to my eye but a whitish-yellow to my camera. However, it would be an odd coincidence if the camera made one flake brighter and pinker while making another darker and bluer.  By the way, Minneart gives a very nice, concise explanation of the 'black snow' illusion:

One must bear in mind that the only quality in which black, grey, and white differ is their brightness, for which the surrounding background provides the standard of comparison. In this case, we refer all brightnesses to that of the sky, and this sky is much brighter than one would have thought, much brighter, anyhow, than falling snowflakes.

You can make the obvious application of this explanation to the bright snowpack illusion.

- JN