Astronomical Observatory: Cool Images

Astr384 Photography Projects, Spring 2004

M51 Whirlpool Galaxy, Phil Ammar

M51, also known as the Whirlpool Galaxy, has been of particular interest to me over the last few years. I can’t describe why I am so drawn to it, possibly because of its beauty and size. I have made several other attempts at reproducing its majesty using our telescope and CCD cameras, but until now those attempts have been limited by poor skies here in Grand Rapids, and out of date equipment. With the new equipment, some of those problems have been overcome. Rehoboth offers clearer, and darker skies that also give better resolution due to its altitude. The OGS telescope there also gives much better tracking, allowing exposures of 5 minutes, compared to the maximum of 20 seconds before.

For this attempt at the Whirlpool, images were taken on two different nights; all the color data was collected on the 25th of April 2004 and the open images were taken on the 28th. Twelve five minute exposures of green and red, and fourteen five minute exposures of blue and open went into this final image. All the data was reduced using the following formula procedure; all pictures were dark subtracted and bias subtracted and flat fielded. I did not bother trying to subtract the ghosts that occur with the ST-10 because I knew that I would be cropping them before I made my final image.

After the data was reduced I aligned it using the ‘two star align’ which I discovered did a marvelous job. At this point I was ready for combining all the data. Up until then, all the steps were fairly straightforward, without much room for error. Now some decisions were to be made which would affect the final outcome significantly. There were many hot pixels and I tried removing them but was unable to do a perfect job. Because of this, when I tried doing just a straightforward average combine, I found that they left streaks across the combined image. I had to redo the combine, the next time choosing median combine, which worked much better. At this point I had 4 raw combined images with all different background levels and varying intensities. I set the all to the same background at ~20000 counts, but didn’t try to set their upper levels equal. I just took note of the relative intensities so that I could scale them when I did the color combine. I just ran a mild unsharp mask on the colors with a mild FFT filter, and a cutoff of 5% with a mask of 50. I didn’t bother doing too much to the colors since my luminance data was very good and I was going to give the luminance a large weighting in the color combine. I spent much more time on the open image though, running the digital development filter and unsharp mask a bit less mild than on the colors. The digital development used the FFT low pass filter with a medium hardness, and used a cutoff of 4%. I let it pick the background and midlevel counts automatically. The unsharp mask used a medium FFT filter with a cutoff of 2%. This gave me a nice crisp looking open which I then used to do the color combine. I used the LRGB four color combine, setting the color levels at:
Red = 1
Green = 1.2
Blue = 5
That accounted for the scaling difference that was around after using pixel math to set the backgrounds level. After doing the combine I applied a gamma function to it to let the faint arms still shine while at the same time not allowing the core to overexpose. For the gamma function I set the min and max levels appropriately and used a gamma value of 0.8 which I found did a quite nice job. After that I saved it as a jpg file.

There is quite a bit of science that can be extracted from this image. I tried to make the colors as realistic as possible which took some experimenting but I think I did a decent job of it. With the colors accurate we can locate different regions in the galaxy based solely on their color. One immediately notices that the arms are much bluer than the core, which shows that younger hotter stars inhabit those regions and that can lead us by comparing with other galaxies to the relative age of the Whirlpool. In my image there are some areas that are slightly redder than others and those point out the areas that have lots of hydrogen gas and where star formation is currently taking place. NGC 5195, the Whirlpool’s merging companion is a much redder, irregular galaxy that seems to be connected to the Whirlpool. It is probably approaching its first pass because M51 doesn’t look very perturbed. Taking red shifts, one could see how the galaxies are moving relative to each other and try to find the mass of them, that would give us an idea of how much dark matter there is in the galaxies because we can see much of the light (but still very heavy!) matter. There are also dust lanes running through the galaxy, and those in conjunction with the star formation clues can tell us more about galaxy formation.