Russell Croman Astrophotography  



Astrophotography: Art or Science?

I often receive questions as to how I produce the images that appear on my site. Most of these are from people who want to learn some of the techniques in order to improve their own images. Occasionally there is an interest in how much an image has been modified from the original exposure data, and if it is truly representative of the actual astronomical object.

In general, astronomical images must all be enhanced to one degree or another to display all of the information present in them. Indeed, many beginning astrophotographers are very surprised when they start to work with their first image, for it can appear almost totally black except for the brightest stars in it. The reason for this is partly because of the limited ability of our current display devices (computer monitors, printers, etc.) to replicate the range of brightness levels in the image. The brightness of objects in astronomical images can vary over a huge range, from the brightest super-giant stars, to the faintest tendrils of gas in a nebula. There is often interesting detail present across this whole range, and it is the job of the astrophotographer, in processing the image, to bring out this data and make it all visible simultaneously in the final result.

The techniques that are used to bring out this detail are quite powerful, and so naturally there is an element of care that must be taken not to overdo it. Modern image enhancement techniques can bring out details that would not otherwise be visible, but they can also be overused. Sometimes this just results in an over-emphasis of details that isn't what would be considered a good representation of the true nature of the object. Other times, "details" can appear in an image that really aren't there, or at best are misrepresentations of reality.

By now, there is one point that should be fairly plain: virtually all astronomical images that are prepared for general viewing have been significantly modified from their original, scientifically useful form. Because of this, and because the degree of modification is largely determined by the judgment and taste of the person doing the processing, we must quickly dispense with hard-lined notions of what is right and wrong in image processing, for these also become the province of various individual opinions and personal preferences. In other words, preparing an image for general viewing is largely an art. It may be an art that is guided by scientific notions and attention to detail, but it is still an art.

There may be purist attitudes and ideals which state that any modification of the image from its original form is somehow "wrong" or "unethical." About this, there is a key point that can be made. The original, scientifically useful data in general just aren't that fun to look at. They may yield answers to enormously important questions, such as the precise distance to a faraway galaxy or the exact orbit of a near-Earth asteroid, but in general they lack the aesthetic appeal that has the power to inspire the spirit and the imagination, to fill us with the desire to know more about what we are looking at. Where would the next generation of professional astronomers come from, if it weren't for these images of celestial sights that conjure up an interest in the subject in the first place? Indeed, how much public support, and therefore funding, would a pivotal program such as the Hubble Space Telescope receive if it were not for the extraordinary images produced by artful enhancement of some of its scientific data as a side project?

Considerations such as these demonstrate that the art form known as astrophotography has a very important, if indirect role to play in science.

- Russell Croman
December 2003