May ends with a new moon (moonless sky), which deep-sky astrophotographers always look forward to. This month brings a potential bonus of an intense, perfectly timed meteor outburst for North American astrophotographers. In fact, it’s so perfectly timed (for US residents) that it falls on a Monday night of a three-day US holiday weekend.
The normal annual meteor showers (e.g. Perseid shower in summer) that we are used to tend to bring more drizzle than a shower. Rates of 70 to 100 meteors per hour are typical and look good on paper, but this is an optimistic estimate called the zenith hourly rate (ZHR) based on what an observer would see if the rain radiant of meteors was directly overhead. Factors such as lower radiation in the sky and light pollution reduce this rate.
In the case of the predicted meteor burst (hope), we could see a ZHR of hundreds or thousands per hour. This normally imperceptible meteor shower, called Tau Herculids, is potentially increasing its activity due to the burst of comet 73P/Schwassmann-Wachmann. The “P” designation indicates a periodic comet, which in this case has an orbital period of 5.44 years.
Normally, comets are responsible for meteor showers because each pass near the Sun accentuates their loose conglomeration of dust and boils some ice, but comet 73P was observed in 1995 to completely collapse, probably due to additional stresses related passing through Jupiter.
Although predicting the behavior of comets and associated meteor showers involves considerable uncertainty, there is the possibility of high activity. The American Meteor Society lists three patterns indicating a peak around 05:05 UTC on May 31, which translates to 22:05 PDT on May 30 for the west coast of the United States. The ZHR cannot be predicted reliably, so if the model inputs and assumptions are not correct, a fizzle could be a possibility! More information can be found in this article from earthsky.org.
Photographing a meteor shower
Given the convergence of timelines, even with such uncertainty in meteor shower activity, this is an opportunity not to be missed, especially since photographing a meteor shower is one of the most easier to do. It just takes perseverance!
In the simplest form, all you need is a decent DSLR or mirrorless camera on a tripod and a wide-angle lens. The camera should be used in manual mode with the lens wide open, fast ISO and long exposure. Typically, I use my Nikon D600 or D850 with a Sigma 15mm fisheye lens, taking 20 second images at ISO 1600. Both cameras are full-frame cameras, which maximizes sky coverage. These cameras also have internal intervalometer functions, so no additional equipment is required. If your camera doesn’t have internal intervalometer capability, you’ll need an external one.
At the start of the evening, be sure to take a few images manually to check focus as well as to judge the brightness of the sky. The sky image should not be so bright that it washes out potential meteors, but not so dark that the sky background is cut into shadows. If the background sky is not optimal (use your camera’s histogram), adjust the ISO or exposure times to achieve an acceptable image level.
If you plan to shoot all night like I do, it’s good to have a battery eliminator powered by a large power bank or AC power source. Otherwise, you’ll need to have a handful of charged batteries and swap them out every few hours.
An additional consideration to remember is to use a memory card with as large a capacity as possible. An eight-hour session at three frames per minute will produce 1,440 images. At 50MB per frame for high-end cameras, you’ll end up with 72GB! Additionally, a fast memory card is desirable to minimize dead time between frames. Pausing two seconds between frames to save an image results in a 10% timeout, and Murphy’s Law dictates that the best meteors will appear in that interval!
An optional piece of equipment to use is a star tracker or star mount. In this way, trailing stars can be eliminated and complications such as lens distortion can be largely avoided if the images are later stacked.
Consequences of the next day
Luckily for US residents, it’s Memorial Day weekend, so maybe we can take another day to sort our frames! My standard procedure is to import the entire batch of images into Lightroom and adjust a representative shot for color and contrast, then produce a set of jpeg images in a subdirectory with the output images highest quality at my monitor’s highest resolution (2K). I then use the Windows native JPEG image viewer in fullscreen mode to cycle through the frames, noting the frames with meteors. The reason for reducing the resolution in this set of JPEG images is to be able to browse through them with minimal delay between frames. Although dark meteors may be missed at this point, the best and brightest will always be evident.
While it may seem easy to identify meteors, doing a good job requires some care, as satellites and planes (even fireflies!) can be a problem. Typically, I’m looking for these features:
- Color streaks: Meteors usually ionize gases in the atmosphere into red/yellow/green colors.
- Colorless (white) streaks may mean you are looking at a satellite streak.
- A trail on two or more consecutive frames indicates a satellite or an aircraft. Meteors are usually too fast to capture in multiple frames.
- The red, green, or white dots in a path come from the aircraft’s navigation lights.
- Curved trajectories (regardless of lens distortion) are planes.
- Notwithstanding lens distortion, meteor trails from a specific meteor shower should converge on the radiant. It is also possible to capture a meteor or random meteors (deemed sporadic) from two radiants if two meteor showers are in progress.
Once the frames with meteors have been identified, I go back to Lightroom and mark them (for example with a red colored marker).
Unless this meteor shower is very intense, it is likely that only one meteor will be captured in a frame, so it will be desirable to composite some or all of the meteors into a single image. If the images were taken on a tracker, compositing multiple images should not be a problem, as the background stars should be largely aligned correctly. The exception may be stars that were close to the horizon in some frames. Atmospheric refraction will tend to come into play at low altitudes (~30 degrees above the horizon).
If untracked images were taken, star alignment between images is a much bigger issue. The distortion of a wide-angle lens makes it impossible to simply use translation and rotation to match stars in images distant in time. In this case, I use a specialized program called Registar, which can recognize and correct misalignment even in very distorted images. The original use for this program was to align scanned film images that were prone to distortion due to film shrinking or stretching. This program solves the problem of atmospheric refraction at low altitude as well. It’s not free, but works great for astrophotos where star alignment is needed and doesn’t require manually assigning reference points.
The alternative to using Registar is to manually align the frames in Photoshop, focusing only on the areas immediately around a meteor trail. All but the longest meteor trails can be reasonably composed when everything else in the frame is blanked out.
When composing frames, one of the frames should be used as the base frame of reference. Ideally, this frame would be halfway between the first and last frame with meteors, but you might want to consider the composition of the photo, placing the Milky Way in a nice position, for example.
Once the base image has been determined, Photoshop can be used to load all images into layers and hide everything but the meteor trail in all but the base image. Note that the sky background may need to be adjusted for each of the meteor images to match the background color of the base image.
Accelerated post-processing option
Another option for post-processing the collection of images from one night is to assemble them into a time-lapse movie. This is especially effective if meteor activity is high or if sporadic clouds have plagued your evening. To start putting together a movie, once the images have been batch-processed in Lightroom for color balance and contrast, they can be output to sequentially numbered JPEG files in a subdirectory. If the file frame numbers are not looped, the camera file names can be used directly. Alternatively, use Lightroom’s ability to rename the output set with sequentially numbered images.
Before outputting the footage from Lightroom, use Lightroom to crop the footage to an appropriate video aspect ratio (e.g. 16:9) and output the footage at a resolution that matches the final video file (e.g. 1920×1080 for HD video). This will minimize the further processing that Photoshop will have to do for the video conversion.
The next step is to open the first file in the set with Photoshop, making sure to check the box indicating that the file is the first in a sequence. Photoshop will then automatically assemble the frames into a video sequence which can be enriched with titles or other labels before being exported into standard video files such as H.264 .mp4 files.
Go out and get the data!
Whether your goal is to get a single frame with a meteor, create a composite, or stitch together a time-lapse video, the important step is to get out there and get the data. If the meteor storm happens, you don’t want to blame yourself for not making the effort. Good luck to all of us!