Digital cameras are more sensitive than the human eye or photographic film when it comes to detecting faint astronomical objects. As an astro-imager, I sometimes see very dim “smudges” in the background of some of my images. I never investigated them until recently. I got the idea as I was working on an image of NGC 7331, the so-called Deer Lick galaxy group. Looking carefully, I noticed lots of small faint objects all over the background. I wanted to know if they were stars or galaxies and I wanted to know their magnitudes and any other information I could find.
The image I evaluated was a composite of 42 separate exposures of 500 seconds each, for a total of about 5.8 hours of exposure time. It was a monochrome image taken through a filter that admits most of the visible spectrum (a luminance filter). The images were taken at Indian Trail Spring (ITS) in the Ochoco National Forest under dark skies. Figure 1 shows the stars and galaxies with the magnitudes annotated.
My first instinct was to see if I could find any of the objects in any of the usual catalogs like the Millennium Star Atlas. I found that most of the printed atlases for visual observers only show stars brighter than about 11th magnitude and deep sky objects brighter than about 14th or 15th magnitude or so. I checked planetarium programs like SkyMap Pro and none of my objects of interest could be found. As I was to find out, the dimmest object I could detect was about magnitude 23.5, or about 100,000 times dimmer than a magnitude 11 star.
I was finally able to find the magnitudes and types of objects by using the Aladin interactive sky atlas. Users can download it for free. I opened my image of NGC 7331 in Aladin and then overlaid the Sloan Digital Sky Survey (SDSS) catalog. All my faint objects had small icons overlaid that I could click to find out more information. The information gave the object type and the magnitude. So I had some answers. Unfortunately, I couldn’t find any other information about the objects like distance, for example. I think the objects are too dim to have been studied extensively.
Another question came up: how was it that I could I detect objects dimmer than sky quality meter (SQM) readings. ITS has fairly dark skies with readings of about 21.5 magnitudes per square arc second or better (Bortle scale 2 or better).
A single short-exposure digital image usually will not show really dim objects because the sky background is so “noisy” that it obscures the objects. However, when we combine multiple images (called stacking) the background noise gets reduced by the factor 1/square root of the number of exposures. So by combining enough exposures we can reduce the noise until dim objects are revealed. This allows us to go deeper in magnitudes than the magnitudes given by the SQM. In the figure below, the ADU values are a measurement of the strength of the signal. The left-hand image shows the background noise as +/- 32 ADU and the right-hand image shows how the 24-ADU galaxy signal is revealed by stacking 42 images to reduce the noise to +/- 4 ADU. The object is magnitude 22.
The equipment I used was a Celestron 11” EdgeHD scope with a QSI 583wsg cooled CCD camera. With this combination I was barely able to detect magnitude 23.5 stars. In order to get to magnitude 25 I would need a telescope in the 24-inch range. A more sensitive camera and longer exposures would help, too. My guess is that magnitude 25 is probably the limit for amateurs unless one can afford a meter-size scope. This was an interesting exercise and helped me explore the limits of my equipment.