Working under difficult skies!
Last night was the first night where stars were generally visible here for the first time in a LONG time – since about May 13th… Ouch! Overall seeing conditions were generally poor, but I thought I’d go out and see what I could do. On top of the generally poor seeing conditions the sky offered atmospherically, a big, bright full disc moon complicated matters more. However, I thought I’d get out there and take advantage of it while I had an opportunity and see what I could get done. Stars and bright globulars should show up okay, right? I hit up a variety of objects which I’ll be blabbering about over the next few days as I finish processing them.
Understanding that I wouldn’t be making an exciting ground breaking discoveries (again, poor skies), I thought instead I would try some test and do some comparisons. Since part of my interests lie around building a solid Altaz imaging platform, I could have wasted the night away continuing to work on that code. However, instead I thought I’d try a few other ideas.
Most people are away that long exposures on altaz platforms are troubled due to field rotation. There are many ways of dealing with this – a wedge, a field derotator, farstar, or even some clever cameras and software that can try to compensate for it. Yet at the same time, there are some pretty big issues with taking long exposures with any of these platforms: wasted pictures and time. For example, you’re out there, having a grand time with your five or ten minute exposures only to find that each one has an airplane pass through, tracking or guiding errors creep in, a slight gust of wind or other perturbation which casues the entire sub to be ruined or damaged in some way. Suddenly that past hour of imaging only gives you a few scant images worth trying to use. That’s not very fun!
Another approach to this problem which I haven’t seen much written about is to take a large number of shorter images, build intermediate subs by stacking them incrementally, and then take those intermediate files and build out the final picture. In this way, one can have “checkpoints” during exposures. Sure, you will need to process a lot more data, but one can easily deal with this as it’s just computer time.
To test this theory out, I went over the the ever-rewarding M13 and took 16x 10 second exposures on the CPC 1100, using a StarShoot Pro and f/6.3 focal reducer. In my case, four of the raw images were not good enough, leaving me with 12 workable images – or 120 seconds, a healthy amount of time for M13, even under cruddy skies. However, in the processing phase, I tried two different techniques. The first technique is pretty brain dead: grab the 12 images, align and derotate them (in software, MaximDL in this case), then stack them together using the ‘Sum’ method. This method is really simple: once the images are aligned and properly rotated, the images are simply added together. This has the effect of enhancing the image as if it were one giant and long exposure of 120 seconds.
The second method is to build intermediate files and then use a gently mechanism for combining those intermediates. In this case, I broke the set of 12 images into 4 sets of 3. I then used the ‘Sum’ method on each set of three images and saved the intermediate file. These intermediate files were now representative of 30 seconds of exposure time each. Then I went back and used the softer ‘sigma clip’ mechanism, which helps average out pixel brightness and kill hot pixels. While the overall color balancing and stretching was not quite the same for both these images, I think you’ll agree that while the second picture is a bit darker, it is more contrasty and offers a richer feel to the core of M13, while the ‘Sum only’ image has stars which feel a little bit ‘blown out’ or too bright.
It’s not yet clear to me this always will have a positive outcome, but it certainly gives an impression that it could be helpful. Next it’d be nice to try it on a fainter object that might not attract quite as much light to register enough in those shorter exposures!