Tuesday, May 8, 2012

A Horsehead Nebula of a Different Color

Having recently undertaken to summarize some astrophysical topics for color scientists, I am confronting another language barrier between neighboring fields.

About ten years ago, astrophysicists Karl Glazebrook and Ivan Baldry introduced the world to “the color of the cosmos”---an average color of all the stars, corrected for red shift. I reviewed that work as it evolved in interaction with ISCC feedback [ISCC News # 397, May/June 2002]. The astrophysicists’ first answer, “turquoise,” changed quite a bit as the data conversion to color was refined. The final accepted color, obtained via a color-appearance model, was declared to be “beige” or “cosmic latté.” Some of us dissented: The answer “salmon” was consistent with the correct chromaticity, and the answer “black” could not be ruled out because, relative to a screen white, the average night sky is---after all---black.

The dialogue about cosmic color was instructive to everyone, partly because it revealed a difference of language and thinking between astrophysics and color science. Within the ISCC community, we already know some peculiarities of language---e.g., colors that are warm to artists have low color temperatures for scientists; and contrast in the display industry is defined to be greater than 1, whereas contrast in the paint industry is defined to be less than 1. These are familiar enough. But since we of the ISCC don’t often talk with astrophysicists, we may be unaware of some language stumbling blocks between these communities.

One example is “luminosity,” which to us means integrating a spectral power distribution weighted by a bell-shaped function that peaks at 555 nm.  However, in astrophysics, “luminosity” means the power integrated over the whole electromagnetic spectrum. Sometimes, astrophysicists prefix their “luminosity” with “bolometric,” and that removes the confusion. However, we still should watch out, as such niceties as a prefix are often lost in a technical discussion.

Another term, “brown dwarf,” refers to an infrared-radiating object that has too little mass to burn as a star (i.e., to sustain nuclear fusion). To color scientists, it is an oxymoron to call a self-luminous object brown. Some astrophysicists are aware of this problem. For example, Kenneth Brecher of Boston University, in a talk called “How Now, Brown Dwarfs,” referred to Joseph Silk’s objection that “brown is not a color.” (Well, actually brown is a color, but Silk had a point.) Brecher concluded that an isolated brown dwarf would look similar to a neon gas-discharge light.

Still another astrophysical term that will bemuse color scientists is the “green valley” of a galaxy color-magnitude diagram.  A color-magnitude diagram is a plot of galaxy color (actually difference between logarithms of light received through a blue and a violet filter) and luminosity (the bolometric kind, if you please). At the top of the diagram is the “red sequence” of galaxies, at the bottom is the “blue cloud” of galaxies, and in between is the “green valley”. In one sense the term should not confuse, because the “green valley” is a place with a conspicuous lack of galaxies, that lack being because there are no green black-body radiators. However, to name a thing for an absent attribute is a bit of a brain-boggler.

The galaxy color-magnitude diagram is an example of the use of color versus brightness to sleuth out the evolution of distant astronomical objects. An earlier such diagram, designed for individual stars rather than galaxies, is the Hertzsprung-Russell diagram, which has existed in various forms for about 100 years. The sleuthing process is quite intricate, given our limited perspective on the universe, and I am trying to learn more about it. For readers who want a reasonable introduction, I recommend the Wikipedia articles on the Hertzsprung-Russell diagram and the galaxy color-magnitude diagram.

As a final example of a word whose meaning becomes less certain in the astrophysical arena, consider how to render a “true” color. Together with such familiar issues as camera-to-tristimulus transformation, one wonders whether to correct the red shift of a distant receding object. If we do, then the object has the color we would have seen in its vicinity long ago; if not, then it’s the color we see now from afar. Fortunately that issue doesn’t affect near objects such as the Horsehead Nebula (a mere 1500 light-years away). Even so, Internet search reveals a variety of colors for the Horsehead Nebula. The colors may not be “true,” but are surely “different”---hence the title of this essay.

Michael H. Brill
Datacolor