Containing what appear to be ghosts rising out of the cosmic dust to beckon a passing bat, the Ghost Nebula (designated Sh2-136, VdB 141) is a reflection nebula located at the edge of the Cepheus Flare molecular cloud complex at a distance of 1,470 ly.  The dust at the tip of the conical pillar at the lower right side is illuminated by a newly formed star system (BD+67 1300) at its apex.  This structure is a Bok globule (designated CB230), which are dark clouds of dense cosmic dust and gas where star formation takes place.  Light transmitted by several stars imbedded in the nebula illuminates the brown-ish color and delicate structure of the surrounding dust cloud.  The light from the background stars is filtered by the vast quantities of dust in this region so they appear as red/yellow stars.  The blue stars are likely to lie in front of this dusty zone.  There are a lot of papers about the role that interstellar dust in the Cepheus Flare region plays in star formation but most are locked behind limited access sites. If you want to learn more, here is one that is open and accessible:  https://iopscience.iop.org/article/10.1086/313076/fulltext/36173.text.html

I started working on this image three years ago back in 2019.  At the time, I gathered some pretty good subs but I knew that I definitely didn't have enough to produce a very clean image so I put it aside until then next year.  In 2020, I managed to grab more data but I still doubted that I had quite enough.  Frankly, I mostly forgot about it until the ghost rose again this year.  After a long period of time with the roof closed due to really crummy monsoon weather this year, we finally had some good conditions and I was delighted to see really good seeing conditions.  I had a number of nights where I gathered images with FWHM numbers around 1.5 arc-seconds and I laughed thinking that that NEVER happens when I'm trying to image some really small galaxy!  By the time I was finished and I sat down to asses how much data I had, I was surprised to find that over the three years that I've been trying, I had gathered around 80 hours of useable data.

Because of the length of time that I had spent gathering this data, I had to spend a fair amount of time sorting through old calibration files to get all the subs calibrated properly.  The 16803 sensor is a fantastic detector but it tends to have a time constant for stable calibration files of about 6 months.  Cosmic rays strikes kill columns and I've found that even the bias signal level tends to drift a little in form (X/Y) over time.  None of this stuff is fatal, but you do have to be aware that calibration files need to be refreshed periodically.  My sensor even has a few columns that appear to be unstable from exposure to exposure so it's also necessary to use cosmetic correction to fix those columns.  Again it's nothing serious, but it's something that may be required with the 16803.  So, when I process an image using data gathered over years, I have to spend quite a lot of time making sure that I haven't missed anything and sure enough, I had to recalibrate all of this data more than a couple of times before I was happy with it.

Processing was pretty straightforward; however, when you process dark nebula, there are a lot of choices--particularly when you start with pretty clean data.  I realized right away that there was a major choice to make about how to display the background dust in this region because there is a LOT of it.  I tend to prefer images with higher contrast so I chose to dial back the level of the almost overwhelming amount of background signal.  It was a balancing act to avoid completely losing the background dust.

Another difficulty when processing an image of a dark object is that you need a fairly strong stretch to pull out the really low level stuff and of course, that creates star bloat--particularly on over-exposed stars.  Since I use 20 minute exposures exclusively, all of the bright stars in this image are over-exposed and that leads to another problem with star color.  The over-exposed regions will always be white with wings that show the real color of the star.  Multiple exposures using HDR methods are the cleanest way to fix this but I did something different to cheat my way around the white cores in processing.  I simply created an image layer in PS and used a blue filter to change the color balance of the whole image.  I then gently layered the core of the over-exposed blue stars onto the final result. The key to this approach is to realize that less is more!  It only take a tiny hue change to make the stars look slightly more natural and it's important not to touch anything beyond the core in the star halo.

Anyway, I've messed with it about as much as can so it's time to set it free.  As always, feel free to let me know what you think and if I've screwed it up, I'll have to give it another shot.  Hopefully it's not too far off.

CS
John