In an earlier post I shared a photo of these same palm trees with a nearly full moon behind them. Now that the Moon is gone (until much later tonight), I can compose a shot with a much darker object behind the trees, the Orion constellation (most of it). The large bright spot in the lower center is the Orion Nebula, and you can just barely see signs of Barnard’s Loop, a large, but very faint semi-circle of red around the left side of Orion.
The trees are lit by artificial light. I don’t know for sure what type of light bulbs are being used in the area, but they certainly have a very different color than starlight. And this after I did a fair amount of color adjustment on them!
Which leads to the interesting topic of “white balance” in astrophotography. Although human vision is remarkable, it has many flaws. One of these flaws is that color perception is quite relative. Our eyes (or more precisely, the part of the brain that processes signals from the eyes) adapt to the light source so that white looks more or less white regardless of the color of the light illuminating it. So what is really white? Most people seem to think that our Sun is yellow, but it is (by definition) actually white. That is, sunlight is what human vision is designed for. When it falls on a sheet of white paper, the light reflected off the paper is our defacto definition of true white. The Sun only looks yellow sometimes because it is surrounded by blue sky (there’s that flaw in human vision again).
Stars do vary in color, of course, and the color is a big clue to the classification of a star. Our star is of the type G2V, and there are many other G2V stars in our galaxy, and in other galaxies. If we care about color accuracy in an astrophoto it is usually a fairly simple matter to identify a G2V star in the image and adjust the relative brightness of the red, green, and blue channels so that they are equal in the G2V star(s). When this adjustment applies to the entire image, it will be properly color balanced.
I almost never do this. A simpler (although less precise) method of adjusting color balance is that in most astrophotos the average color of the stars will be white. So as long as the image does not appear to be “tinted” in the stars, it is probably pretty close to the correct color balance. This would not suffice for scientific imaging, but in the “pretty picture” type of astrophotography that I enjoy, the visual appearance is actually more important than mathematical correctness.