The Dark Energy Camera: a powerfully capable instrument for the modern era of massive cosmological surveys

by Kevin Reil

Author Kevin Reil. (Photo courtesy K. Reil.)


The Dark Energy Camera (DECam) is a 570-megapixel camera installed on the 4-meter Victor Blanco telescope atop Cerro Tololo, a mountain in the Chilean Andes. The science mission for the Dark Energy Survey, of which I’m a member, is nothing less than to use this camera to understand Dark Energy. Which is a tall order, since the phrase “Dark Energy” itself is, as some cosmologists say, simply words we use to describe our profound ignorance about the current-day accelerating expansion of the universe. Though this accelerating expansion was a theoretical possibility and predicted by a minority of astrophysicists in the later part of the last century, its actual confirmation through observations of a certain type of supernova (called a Type Ia supernova) in 1998 came as a major surprise to the majority of the community (and larger world). And has continued to puzzle the entire community of astronomers and physicists ever since, with mysteries such as how we square the paradigm of an ever-expanding Universe with a specific start time for it in the past. Thus, we forge ever onward, observing more of the Universe steadily, in the hope that probing further will elucidate the why and wherefore of this most recent profound cosmological mystery.

DECam is one of the tools we use in this grand endeavor, and it first opened its camera shutter in September 2012 and has thus far completed three of its planned five seasons of observation. During the five years of the survey, the instrument will collect information on millions of galaxies and look for and study thousands of supernovae.

DECam with focal plane (blue circle) visible. (Credit Brenna Flaugher.)
Above: This is a picture of the silicon focal plane (in blue, towards the left of the image) of DECam with a representative thoughtful human shown for scale. (Credit: Brenna Flaugher).


We published a paper last year about the details of the camera’s construction and operation which was rather aptly titled, “The Dark Energy Camera.” My favorite part of the paper is really the author list. It summarizes a rule called STP that I know best from the volunteer world. It stands for the “Same Ten People” who show up every time and make sure things happen. For an instrument like DECam that number is actually slightly over 100 people (so we’ll alter the rule to “SHP”) but the idea remains the same: If you want to do great science, one way is to surround yourself with people like those listed as authors on this paper, then do the best work you can as part of an excellent team.

The work done by KIPAC faculty member Aaron Roodman and myself is covered in section 7.3.3 of the paper, “Active Optics System (AOS).” The DECam’s AOS is not to be confused with the adaptive optics system the Gemini Planet Imager, another instrument with a lot of KIPAC involvement, uses. The DECam’s active optics corrects focus and alignment between exposures while the Gemini Planet Imager’s adaptive optics makes correction while exposures are being taken.

The AOS is just one small piece of a very complex system, but a rather critical one. It keeps the camera in focus and optically aligned automatically, without any manual intervention. This allows observers to focus their attention on taking data. The AOS works by using eight wavefront sensors to continuously correct the camera’s focus and alignment and needs no human intervention. With it, instead of the camera drifting out of focus by hundreds of microns through the night, the system quickly and automatically moves to better than 30 microns defocus, and the shapes of the images remain very stable. (See the image below.)

Demonstration of improvements to DECam due to active optics system. (Credit: DES Collaboration.
The illustration above shows the improvement in focus that the AOS can provide for DECam images. The AOS uses out-of-focus stars, which we call "donut images", to focus and align DECam. In the figure we see 8 out-of-focus star donuts (in the inner ring) taken from 8 special sensors (in yellow) in the focal plane. Also shown are our automatic analyses of each donut (in the outer ring). If you look closely here, you will see that the donuts labelled FN1, FN3, FS2 and FS4 are larger than the other four, which is because the former are 1.7 mm out of focus in this instance, where the others are 1.3mm out of focus (in the opposite direction). The critical point is that the size of the out-of-focus donuts is fed back into the system so that it can auto-correct the focus level quickly without outside intervention. (Credit: DES Collaboration.)

[For those who would like to know a bit more technical detail: the partitioned circle in the center is DECam’s focal plane, with the 62 data CCDs shown in white, and the eight wavefront sensors which detect aberrations in the light reaching the camera shown in yellow. Half of the wavefront sensors are normally 1.5 mm above, and the other half 1.5mm below, the focal plane when it is properly adjusted to be in focus. Surrounding the focal plane picture are donut images from an image taken during DECam commissioning with the camera roughly 200 microns out of focus at the time (which is what leads to the 1.7mm and 1.3mm out of focus images mentioned above). The inner ring of donuts are examples of out-of-focus stars from the data, and the outer ring shows the fits to the data. The graying of the inner ring shows the level of signal to noise for that image. Each pair of donuts is tagged with the name of the sensor that detected the star.]


The most immediate benefit of the AOS is survey efficiency: no fiddling with the focus saves time, and time = money in the calculus of research, as in so many other areas. DES should save several weeks of time over its five-year mission due to this system.

All of DECam’s science benefits from sharp, well-focused images, but of the four pillars of dark energy science being investigated by DES—Type Ia supernovae, the distribution of galaxy clusters, baryon acoustic oscillations, and weak gravitational lensing—weak lensing science likely benefits the most because weak lensing effects are very subtle. In weak lensing, we study tiny alterations in the shapes of distant galaxies caused by the paths of their light rays being deflected by all the mass in the Universe lying between them and us. Correcting during actual exposure minimizes the systematic changes in shapes introduced by the camera, keeping them consistent and allowing us to measure the weak lensing effect.

The system is working well; a few years of effort and several months on the mountain (Cerro Tololo) are captured in the summary sentence of our section: “Closed loop operation [of the automatic focus and alignment] was made the default condition for all observers at the end of DES SV [Science Verification time, in early 2013], and it has remained in stable problem-free operation since that time.” Phew! A huge lot of hard work is summarized in those few words, and for us it meant mission accomplished—at least for that stage of the experiment!

The DECam mounted on the 4-meter Blanco telescope. (Photo courtesy K. Reil.)
Above: DECam at the Blanco telescope—I am one of the small humans on the left, DECam is the large black cylinder next to us, and on the right is not a crashed cousin of the Millenium Falcon but the actual telescope mechanism that holds the mirrors and focuses the light that ends up being imaged in DECam. (Photo courtesy K. Reil.)


But the AOS is just one piece of DECam; a collection of really smart, dedicated people, each contributing a little piece like this, is necessary for science to work. That’s what I love about it. So you don’t have to read our entire paper, but I encourage you to look at the list of authors and think about all the other contributions that went into building a state-of-the-art scientific camera.

Star trails above the dome of the 4-m Blanco telescope on Cerro Tololo. (Credit: Tim Abbott.)
Above: time-lapse photo of star trails as the sky 'spins' around Polaris, with the Blanco Telescope (with DECam inside!) in the foreground. (Credit: Tim Abbott).