I originally photographed M31 in August of 2020 as my first serious attempt in photographing a deep space object (DSO). I did not make many other attempts in DSO photography until the past couple of months where Miguel and I have had hard times thinking about anything else. There are multiple sides to this type of photography and so many ways to improve and learn. This is definitely the most technically challenging photography I have ever done. I’ll say this new attempt at M31 is a significant improvement over my first, mainly due to increased integration time and learning better processing techniques.
The Andromeda Galaxy (M31)
You can learn a little about this section of sky by visiting my first post. In addition, here are a couple other factoids about this galaxy. One of the coolest things I’ve learned since getting into astronomy and DSO imaging is the size of a lot of these objects relative to other things that we all routinely see in the night sky. Sure, stars are small in our vision and there are a lot of very small objects that need very large focal lengths to see. But, many DSOs are actually very large. We don’t notice them due to their low magnitude of brightness. Many nebulas cover large parts of our sky, for example. For M31, it’s apparent size on the long axis is ~ 3.167 degrees. The size of the full moon is ~ 1/2 a degree, so M31 covers an area a little more than six full moons!
Andromeda was first formally described by Persian astronomer Abd al-Rahman al-Sufi in the year 964. Did you know that many of the first astronomers were living in the middle east? It’s true – many of the stars still carry their original Arabic names. Andromeda and other similar galaxies were originally thought to be groups of gas and stars within our Milky Way. The discovery and proof that Andromeda was its own “island universe,” like our own, did not occur until the 1920s. Over the last century, M31 has been extensively studied and is now thought to contain ~ one trillion stars.
Collecting the data
The new moon period, which allows for the best low-light conditions for astrophotography, in October is troublesome. Miguel and I checked and double checked the forecast and found the best potential night was on October 22/23. We are finally getting some rain in eastern Missouri and most of this period is forecast for significant clouds. The night we chose was mostly cloudless but was not perfect. Winds were 10-13 mph with regular gusts up to 20 mph. In addition, seeing and transparency were on the poor side due to the winds and high humidity. This was not optimal, but it was still the best apparent night to try, so we did. To aid with the winds, we setup on the downwind side of a couple of hay bales and they did a pretty good job of acting as wind breaks. We imaged at my favorite site – Danville Glades C.A.
I had originally planned to shoot with my Canon R5 and the Canon ef 400 mm f/4 do ii lens. This framed M31 very nicely in Telescopius and I was eager to see how this lens performed in astrophotography purposes. However, with the winds forecasted, I decided to use a smaller lens that would be less likely to be affected. I wound up using the Canon 7D mk ii and the Canon 200 mm f/2.8 lens instead. This put M31 slightly smaller in the frame, but I thought it would still stand out well enough after a marginal crop.
I used the Skywatcher Star Adventurer without guiding and used my new William Optics vixen style base to mount the tracker on. This mount in combination of using a green laser pointer allowed for very good and easily obtained polar alignment. I’m happy to say that I need not dread getting PA any longer and I can make this step just part of the routine.
We had some clouds early on in the night. This wasn’t too much of a problem because they cleared out by about 9:30 pm and Andromeda was still in low latitude sky glow until about 10:00 pm. Imaging went pretty well. I did not have to throw out many lights due to wind or tracking errors. One problem that did become apparent was the quality of the stars. When wide open, this lens produces fat stars with pretty bad chromatic aberration. This was probably exasperated by the poor seeing and transparency. I knew this could potentially be an issue but wasn’t too concerned as I was mostly concerned about the galaxy. Anyway, next time I will stop this lens down by 2/3rds of a stop to try and improve this. The settings I used were f/2.8, 30 second exposures at ISO 1600.
Lights: 492 light images taken (manually removed obvious bad subs and used 447 subs for a total of 223.5 minutes of integration).
Flats: Since flats are a pain to take and since I am using a camera lens that can be corrected for vignetting when processing the raw light files to tif format, I did not take or use flats for calibration. This was the first time I tried this and it seemed to work very well. This will be my new strategy going forward.
Deep Sky Stacker worked! This was the first time I had good enough quality subs that DSS would register and process everything. After about one and a half hours of processing, DSO had processed my linear image. Processing was the biggest learning revelation I had from this project. Pieces finally came together. After stacking in DSS, I used various manual techniques in Photoshop along with StarNet for star removal, GraXpert for gradient removal and Astronomy Tools action set. I finally have a big picture of my step-wise work flow and this should get easier and better going forward.
Including driving time, setup time, imaging time and processing time, I estimate it took about 20 hours of concentrated work to produce this one image seen here. When compared to my previous attempt of this object, I am quite satisfied by the results and the time spent was well worth it to me. The increase in integration time along with my improvements in post processing really paid dividends. Maybe I’ll try this target again in a couple of years if I have made improvements in equipment, techniques and processing.