One of the most important lessons to learn early in astrophotography is patience. Once a certain level of competence is achieved, our first impulse is to try and shoot as many objects in one night as we can. After all, clear skies are precious, and we must maximize the use of our time. After a short time of being told by all our family and friends that our work is AMAZING, we start to realize… our work really isn’t that great outside of the circles of people who are easily impressed by just about anything that looks like it’s from outer space. We realize that to take things to the next level, we need more than 16 x 1 minute sub-exposures to drive down our noise and get to the deeper fainter details. Now we shoot the same object all night, or for as long as we can before it goes behind the neighbors trees. This is a great leap forward in the quality of our work, albeit we aren’t cranking out four or five images per night anymore.
Our next great leap forward, also due to growing discipline and patience, is when we decide to spend more than one night on a target. Perhaps two… which depending on the weather might take a week or more. My first published image I spent a month on, getting a total of about 20 hours of narrowband data from my backyard a few hours at a time. Then there are those, of whom I am not worthy, who will spend many months or even a couple of years gathering data for extremely deep images or large mosaics. I am not only not there yet, I may well never be. I’m pretty sure I’ve got the patience thing down though for multi-night imaging… well, until I slipped recently, and that’s what I’m going to tell you about here.
I had the opportunity to attend the Texas Star Party this year (2017) as a vendor for Software Bisque, and instead of flying/shipping everything out, I drove. There are advantages to traveling light, but being prepared is not one of them, so I decided to drive so I could bring more of my own gear (last year I borrowed quite a bit). I brought my current favorite imaging setup, a Paramount MYT and and Esprit 150 refractor with an FLI Microline 16200 camera. A MoonLite NiteCrawler has been giving me exceptional focus with this combo and I was ready for the week with a list of images I planned to shoot with this and a shorter focal length DSLR system.
Most nights were good, but only three of them were good in the early morning when I wanted to shoot the Triffid Nebula (M20). I shot the region with the DSLR setup the first morning, and then gathered as much as I could on two additional mornings. Here’s where I got greedy. M20 really wasn’t high enough to shoot to gather sufficient data in only two nights. I tried getting the red subs while it’s elevation was around 30 degrees or less, and the blue when it was higher around 35 degrees (it never quite gets even to 36 degrees above the horizon here). This was foolhardy I know and I rarely will shoot even lower than 40 degrees, but I have had some good success exceptionally low to the horizon from Florida under “perfect” conditions (usually shooting out over the ocean) so why not give it a shot.
I have a core processing philosophy that if I have to work more than a couple of hours on an image, then the data probably wasn’t good enough to begin with. I simply will not spend 40 hours doing all kinds of different deconvolutions, magic stretches, and witchcraft noise reductions. Don’t get me wrong… I WILL use these, but if I have to iterate that long then I’m just trying to turn straw into gold, and I’ll simply start over or get more/better data for the next time. One of the things I love about the ML16200 from FLI is that the read noise is very low, and it’s a grade 1 sensor. I can do amazing work with that camera with shockingly little data, even with an f/7 refractor (and that Esprit 150 produces amazing contrast and that doesn’t hurt either).
I looked at the data when I got home, and decided not to even try processing the M20 images because the stars were so bloated. A few days later when cleaning up, I looked again and some where pretty good. I used the blink tool and tossed out some big fat star images, and decided to give it a go. I calibrated, aligned and stacked in PixInsight and got excited because the data was better than I was expecting. At least zoomed out, and after a few initial stretches it looked promising.
An axiom of astrophotography is that just about anything looks great if you shrink it down enough. Zoom in though and a serious problem becomes apparent. The root cause of all the reddish blotch around the stars in the image at right is that the stars in the red subs are larger than the stars in the blue and green exposures. I just tried shooting too low and the turbulence bloated the stars and finer details in the image. I’ve gotten this before, but only on a few of the brightest stars, and I will confess to fixing them with the blur brush in Photoshop. This time the problem is all over the place on perhaps a hundred stars or more in the image. A seemingly easy fix was the morphological tool in PixInsight on just the red channel. This helped a lot, but I still had many many stars with pink blocks around them. What I really need are some red subs shot higher up in elevation. Good clean data. Of course, I have good friends who will take me to task immediately… deconvolve, use the minimize, tweak
rejection, blah blah. Yes, I know and do use these tools, but I think here it’s over using them… with 10 hours I could make an APOD worthy image I’m sure, but this just violates my own personal code of imaging. Data is king, not processing. I did this once before on a Thor’s Helmet image where the red was slightly out of focus. The resulting image was good, and even made the cover of a magazine… but deep down, I have always been unsatisfied with the image and feel the obsessive compulsive need to shoot it again (or at least the red channel) and do it over. I think I killed the red channel trying to “fix it”, and the overall tone will be different the next time.
I loved this framing on M20 and M21 and I’m not giving up yet. I’m a few degrees further south at my dark sky site, and the seeing is better in Florida to boot, so before summer is over, I may yet combine the best of the data from Texas with some new good data from Florida. I will just have to be patient with this one.