The UtahAstronomy e-mail discussion group recently hosted an exchange about the accuracy of astrophotos.
Let’s begin by ignoring the deeper issues, such as whether color actually exists in space. When you consider that light is electromagnetic radiation moving in all directions, both particle and wave, at 186,000 miles per second, it becomes apparent that shape and color of distant objects do not exist except as they are perceived; that is, focused by eye or lens. But the colors and shapes are real in that cameras and telescopes operated anywhere can produce similar images of given targets. This subject was kicked around by the UA group in the past, and it’s an interesting topic, but the latest talk isn’t about that.
We have been noodling over the legitimacy of the type of image processing used by nearly all astrophotographers, whether NASA or some of our talented Utah astronomers. That involves manipulation to bring out subtle details that would not be visible without the manipulation.
An analogy I used in the discussion is this: suppose you took a night photo with dimly-lit people on the sidewalk and a car in the road with its headlights blazing at you. Imagine you’re a purist who refuses to do anything to one part of the photo that you don’t do with all of it. You can adjust it on your computer so that the people on the sidewalk are well-lighted, but then the headlights will be so bright that they will wash out a good part of the picture. Or you can adjust the scene so that the headlights are small white orbs in the darkness, but then the sidewalk activity is lost in the black background.
The alternative is to manipulate parts of the photo. You can “burn in” the headlamps so that they aren’t blown out, while at the same time allowing the people on the sidewalk to be relatively well-exposed. The result is a view you could not have seen with your eyes at the time.
The same dilemma comes up when making photos of many celestial objects. A galaxy’s core usually is orders of magnitude brighter than its distant spiral arms; dimmer yet are filmy sprawling material flung off many galaxies during collisions. How do you handle the image? One that shows a lot of detail in the central bulge –- such as swirling dust lanes — will be so dark in the remaining areas that it barely looks like a galaxy photo. A view displaying the arms in their best light, let alone the material around the galaxy, will render the center washed out, nearly a featureless blank.
Almost all astrophotographers will manipulate the picture to show detail in both the bright, dense inner region and the arms, and some take such long exposures that the filmy material shows up too. I argued that this verges on cheating because it treats parts of the image differently from other parts. After all, the center really much brighter, yet that’s not how it looks in the finished product; instead, all parts of the galaxy will appear a lot closer in brightness than they really are.
Is that cheating?
No, is the resounding answer from nearly every astroimager. They contend that the details are present in the center of the galaxy, are detected by the camera, and that there is nothing wrong with showing them as well as the details in the dimmer parts. It’s like scientists taking measurements of natural objects and explaining the results with pie charts and graphs and drawings. They show what is there, even if it doesn’t look like the object itself.
Thinking this over, I have to concede that the purist attitude I was promoting is flawed. It’s better not to hide information in a blaze of a washed-out galaxy core. Or in the center of a starburst nebula, or in other objects with vastly differing contrast.
The problem is not that we can’t capture details; it’s that we don’t yet have a medium that will reproduce a wide enough range of them.
A challenge for the future is to design a display to show it all. Maybe a computerized image will have available dozens of photos of the same object, taken at different exposures. Imagine a screen that the viewer can easily brighten or darken, and adjust the contrast and tonal values. As the parameters change, photos with different exposures pop up seamlessly. Turn a knob in one direction and all the glorious details of the galaxy center are visible. Twist it the other direction and study the knobs of nebulosity in the finest, darkest sections of the arms.
Until that’s possible, and as long as the artifice –- or should we say art? –- is acknowledged, manipulation seems a valid way to represent cosmic reality through our limited photographic palette.