I recently took a photo of the Andromeda Galaxy that went viral, popping up everywhere from the front page of Reddit to Newsweek. The photo was taken with a four inch telescope over several nights from my backyard near Charlottesville, Virginia, USA.
In this article, I’ll share how the image came together and offer some tips for photographing one of the most photographed deep sky objects.
Indeed, Andromeda was one of the first targets I attempted when I began my astrophotography journey in earnest two years ago. As a beginner, my results were satisfactory in my beginner’s eyes, but far from what I’ve been able to capture recently. This came both from refining my techniques and adjusting my equipment to be more specialized to achieve my astrophotography goals.
So what have I learned in the past two years since the first attempts to film Andromeda? Here are a few lessons, in no particular order:
The time of year is critical
September to December are the best months to photograph Andromeda from north-northern latitudes. Although present in the sky for much of the year, shooting targets above 30° is imperative for mitigating atmospheric distortion, improving tracking, and achieving clean results. In October, Andromeda spends the night in these higher elevations.
Location, location, location
While many narrowband nebula targets can be photographed beautifully in the middle of the city, for galaxies it’s best to look for a dark sky location – at least Bortle 4. I’m lucky to have a Bortle 3 backyard where I shoot the most deep sky targets.
The Bortle Scale is a measure of sky darkness, with 1 being the darkest and 9 being downtown skies. To find dark skies near you, check out a light pollution map.
The most dramatic improvements in the quality of my images are due in large part to equipment upgrades. My first deep sky endeavors were with a Sony a7 III mirrorless and a Sony 200-600mm zoom, which together cost more than my current telescope and astro camera.
However, the optics of my Skywatcher Esprit100ED telescope are tailor-made for stars, whereas a camera lens is designed to serve many photographic purposes and is not specialized in astrophotography.
I’ve also learned that it’s critical to make sure your camera, whether it’s a DSLR/mirrorless or a specialized cooled astro camera, is suitable for your lens/telescope to avoid oversampling or undersampling of stars. I paired my Esprit100 with a ZWO ASI2600MM-Pro which, with a pixel size of 3.76µm, is in the resolution sweet spot for the 550mm focal length.
I shot Andromeda for several nights and had hundreds of subs (two minute single exposures) to eliminate. Throwing out all the subs that weren’t of the highest quality was key to getting the result I got. If a sub had tracking error, clouds, or focus issues, it went straight to the delete folder.
I was left with only the best shots to stack, and quickly scrolling through my final selections, the shots were almost indistinguishable in star shape and size, which helps ensure a better final image.
Post-processing astrophotographs is time consuming and the methods can be difficult to learn. There is no shortcut around this step because post-processing for astrophotography is just as or more important than your equipment or capture. There is a wealth of information out there, and you could spend countless hours studying post-processing techniques available in books, online tutorials, and community forums.
I found famed astrophotographer Adam Block’s teaching approach very helpful and patiently worked through his many tutorials until I was comfortable with the powerful tools available with today’s processing software. , including Pixinsight. Adam emphasizes “paying attention to your data,” and that mantra guides my approach.
I also learned that shortcuts like global adjustments, for example with level sliders in Lightroom, may seem useful but can easily compound processing errors down the road. Instead, I would recommend a careful and deliberate approach to a processing workflow that will improve your end results. My own workflow for the Andromeda photo was mostly in Pixinsight with some final global touches in Lightroom and Photoshop.
Overcome dynamic range issues
While Andromeda is one of the brightest targets in the night sky, its brightness can be tricky as it’s easy to overexpose the galactic core (see the center of my first attempt).
I approached this by shooting shorter subs, 120 seconds versus the 300 seconds or even 600 seconds I usually expose for deep sky targets. I wanted the galactic core to be as small and clear as possible in my unstretched subs. Then, during processing, I used iterative stretching, masks, and the HDR multi-scale transform tool in Pixinsight to ensure that when I brought out the fainter outer regions of the galaxy, the core remained correctly dynamic.
Astrophotography is frustrating but rewarding
If this all sounds overwhelming, that’s because it’s — at the start. Fortunately, climbing the learning curve has been extremely fun and rewarding both artistically and intellectually, if frustrating at times.
If you want to learn astrophotography, now is the perfect time. The combination of advances in optical technology and processing technology makes the hobby increasingly accessible, and resources and practitioners abound. The astrophotography community is very knowledgeable, and there’s probably an accomplished astrophotographer near you who wants nothing more than to pass on their knowledge.
Final photo and details
Here is the viral photo I recently took again and the details of how it was made:
Imaging telescope: SkyWatcher Spirit 100 ED f/5.5 APO
Imaging camera: ZWO ASI2600MM Pro (Gain 100)
To go up: SkyWatcher EQ6R Pro
Filters: Chroma RGB, Chroma Ha (3nm)
Accessories: Pegasus Focus Cube2 Pegasus Astro Pocket Powerbox Advanced
Software: Sequence generator Pro, PHD2, Pixinsight, Lightroom, Photoshop
guidance: ZWO OAG, ZWO ASI120MM-S Camera
Frames: 90x120s R, 90x120s G, 90x120s B, 20x300s Ha
Flat blacks: 15
The Andromeda Galaxy lies 2.5 million light-years from our solar system and is the closest spiral galaxy to us. It probably looks a lot like our own Milky Way. There’s probably also some jerk taking a picture of us at the same time I was taking this picture of them. Of course, the same time means 2.5 million years in our future.
Also, fun fact: we are heading towards each other at 113 km/s and we are due to collide in 5 billion years. (Although the distance between the stars is so great, there will be little or no real interaction).
Another fun thing about this photo is that there are 24 million pixels (approximately) in this image and about 1 trillion stars in the Andromeda Galaxy which takes up half of the image. That means there are somewhere in the neighborhood of 21,000 stars per pixel in the galactic glow.
You can find more of my work on my Instagram, @brennangilmorephoto, and my website, brennangilmorephoto.com.
About the Author: Brennan Gilmore is a deep sky and landscape astrophotographer based in Charlottesville, Virginia.