Adventures in Image Reduction and Composition

Today I have an image of the Sculptor galaxy, NGC 253 (also known as the Silver Coin or Silver Dollar galaxy for its appearance in small telescopes). This dusty galaxy is fairly nearby by cosmic standards, only about 11 million light years away in the direction of the constellation Sculptor. Unlike most of the images I put up here, this one was not taken at the Vis. Instead, it was taken by the 2-meter Faulkes North telescope on Haleakalā. Last week my group in our Observational Astronomy lab were using the Faulkes telescope to get data for our project, and it wound up that we had a free hour, so on the advice of our teacher we used it to image this galaxy.

The Sculptor galaxy, NGC 253.

This image is by far the hardest one I've ever put together. To begin with, when I got the data from the telescope it had no easy identifying information as to which picture was what, so by trial and error I had to work out which pictures had been taken through the red, green, blue, and H\(\alpha\) channels. Once I'd done that, I had to figure out how to get the images to stack, because Deep Sky Stacker couldn't find enough stars (don't ask me why) for it automatically stack them. Once that was accomplished, I had to figure out how to play with the stretch of the final image to capture a good dynamic range -- no easy feat. All of that meant that I ended up re-reducing the data at least five or six times, and that was before I even began assembling the various pictures into the final image. I put together at least three versions of this picture, playing with the light curves of the different colors to try to get something that captured the beauty of the galaxy without either blowing out the bright parts or losing detail in the darker regions of the arms. This version is fairly close to true color, whatever that means in astronomy. I only took one (conscious) liberty: in addition to the standard RGB filters, we got some images with a hydrogen-alpha (H\(\alpha\)) filter, which is a filter that lets only one particular wavelength of light through (in this case the light emitted when an electron drops from the third to the second orbital in a hydrogen atom, which appears red). But whenever I tried mapping the information from that filter to red, I ended up with a galaxy that looked much too red, so I left it as simple luminance data. It's not "correct", but I think it gives a more aesthetically pleasing picture, so I left it. Perhaps in the future I'll go back and see if I can't get it to look better and at the same time more chromatically correct, but for now I'm moderately satisfied with it.