Sometimes the happiest results happen by accident. This is the story of how I unintentionally managed to image a small, distant asteroid with my Nikon camera and a wide-angle lens!
Shooting Orion with a Guided Camera
Saturday night, January 5, 2019 brought North Carolina the first clear skies in what seemed like ages. I planned to shoot some nebulas with my telescope and specialized camera. But tonight I also wanted to try out a new wide-angle lens I recently purchased, a Rokinon 24mm f/1.4 lens for my Nikon D5500. My goal was to point the Nikon at Orion and see how much detail I could capture if I kept shooting exposures as long as I was in the field.
I set up the Nikon on my iOptron SkyTracker Pro mount and a sturdy tripod. The SkyTracker is a motorized one-axis equatorial mount. Simply put, once you’ve pointed it at the celestial pole, it will follow the Earth’s rotation all night long. This prevents stars from forming trails in time exposures. The two cables you see coming from the camera in the photo are a data cable that connects to a laptop (running BackYard Nikon [BYN], an astrophotography app) and a “dummy battery,” which provides power to the camera from a 120V source rather than its internal battery.
I aligned the SkyTracker on Polaris, used BYN to get pinpoint focus on a star, and then framed Orion. An exposure of 20 seconds was the most I could manage without fogging from light pollution–a problem even in rural Summerfield NC. I then put BYN on loop to keep capturing 20 second exposures all night. I eventually changed this to 30 second exposures after Orion climbed higher in the sky and light pollution was less of an issue. All told, I shot 407 images between 7:51 pm and when the clouds rolled in just before midnight.
Disappointment and Curiosity
I was disappointed when I calibrated and stacked the images the next day, a process that took more than six hours. The stars were not tight circles; they showed a lot of smearing due to coma, a flaw in the camera lens. There was also so much background noise in the image that I could barely make out anything interesting.
I played with the image for several hours, trying to reduce background noise and at least make the image aesthetically pleasing, but without anything I considered as success. I was pleasantly surprised that I could clearly see the Horsehead Nebula hanging down from the left side of Orion’s belt. And there was a hint of Barnard’s Loop, a semicircle of glowing gas off to Orion’s left side.
I pulled up Cartes du Ciel (CdC), an astronomy app, to see what else I might try to discern in the image. I noticed that it showed asteroid 354 Eleonora was in Orion that evening. Was the asteroid bright enough that I might be able to see it in the photo? That would be worth further investigation.
Interstellar Detective Work Phase I: Star Hopping
My first step was to try to figure out which of the 25 million pixels in the photo might be the asteroid. I would need to use CdC to help me narrow my search. First, I made sure that CdC’s asteroid database was up to date. Then, I set the program to display Orion and the location of the asteroid for 10:30 on the night of January 5. I brought up the chart of Orion and put my image next to it, trying to match the scale and orientation as much as possible.
Amateur astronomers learn a technique called “star-hopping” to find dim objects in the sky. You begin with a bright star that’s easily identified, then you look for nearby star patterns that gradually lead you toward the dim object. My challenge with the image I took was that it showed many stars that were dimmer than the ones on the sky chart, and it was difficult to distinguish actual stars from the noise.
I worked carefully, striking out toward the west from the rightmost star on Orion’s belt. Four nearby stars formed a parallelogram that I could readily find in the image. Then it became harder. I looked for pairs of stars on the chart that I might find on the image. The photo below shows the patterns of stars I was able to match up while star-hopping to the asteroid’s predicted location.
It took several minutes, but soon I was relatively confident I had found the right spot, as there was a small triangle of stars just to the left of where the asteroid should be. Knowing that there was likely to be some possible error in the charted location, I narrowed my search for the asteroid to the area in the red circle.
There was nothing in the area that I could unambiguously identify as an asteroid. I wasn’t sure how bright it would be relative to the reference stars. I also wasn’t sure if the stack of 4 hours of exposures would leave the asteroid as an elongated streak, or if it would have just been subtracted out as background noise. Time for another technique.
Interstellar Detective Work Phase II: Blink
Asteroid discovery is typically done by comparing two photos of a star field separated by some arbitrary time period and seeing if any of the “stars” have moved in the two images. That’s how Clyde Tombaugh discovered Pluto in 1930, by using a “blink comparator” that flashed back and forth between two images of the same star field.
Would the asteroid have moved far enough in the space of the four hours between my first and last images for me to detect it? Now that I had narrowed down the approximate location, I could attempt to find it on the individual (raw) images from the camera.
The Pixinsight image processing app has a “Blink” process, which is primarily used to inspect a lot of images quickly and weed out the poor ones. However, it can also create movies by cycling through the images with little delay between each frame. It’s like a “flip book” animation. This would be a quick way to see if any of the stars in my images were moving.
My computer choked while trying to flip through the 407 images rapidly. So I chose 10 images that were evenly spaced over the course of the four hour photo session and just compared those. I zoomed in on the area to which I had narrowed my search. By Jove, something was indeed moving!
I then selected images that were taken every four minutes during the course of the night. I aligned them using Pixinsight’s StarAlign process. This minimized any motion in the stars due to the camera’s aim moving slightly during the course of the evening. Then I loaded the aligned images into Blink and ran them in rapid sequence.
The animation below clearly shows the motion of 354 Eleonora during the course of four hours. It’s the very dim dot in the center of the GIF.
Loop showing the motion of Asteroid 354 Eleonora (center) over 4 hours
So What?
If you’ve read this far, “so what?” is probably a question you don’t need me to answer. However, here’s why I’m astonished at this little “discovery”:
- Asteroids are usually found using dedicated telescopes and special cameras. This was done with a mid-level Nikon camera and a (bad!) kit wide-angle lens. The only special equipment I used in the field was the tracking mount that allowed me to take 30-second exposures.
- At the time I took the images, the 46-mile-wide asteroid 354 Eleonora was 168 million miles away, in the asteroid belt between Mars and Jupiter. At magnitude 10.3, it was about 250 times fainter than the dimmest stars I could see with my unaided eyes that evening.
- Even amateurs can find exciting things in the sky other than bright planets and nebulas.
This exercise goes to show that even what appears to be “bad” data might have some surprises lurking inside. I hate to say, “Never throw out your data,” because it eats up a lot of disk space. But you never know what you might be able to do with it.
You could even snag an asteroid!