What observers really DO at the telescope -- Part 1: Wide field survey mode

Jan 31, 2018

by Mandeep Gill

I recently returned from an observing run at a telescope in Chile, and I thought our readers might wonder what astronomers do when they’re observing. After all, it can’t all be sitting around romantically staring up at the stars, right?

So here’s a fairly detailed description of what I did when I was observing for those who have wondered what actual observing is like.

Where and when

To give some context, for about two weeks in November, 2017 I was an observer (for the second year in a row) for the Dark Energy Survey (DES) project, which gets one third of the time on the Victor Blanco telescope (at the Cerro Tololo Inter-American Observatory (CTIO) in the Southern Atacama Desert of Chile) over five years (some details of the camera that was built to take the DES images, the Dark Energy Camera or DECam, were discussed in this previous KIPAC blogpost). 

CTIO is located about 50km east of the coastal city of La Serena, which is where most observers will fly into to get to the Observatory.

Map showing CTIO in relation to La Serena, Chile. (Image courtesy DES.)
Location of Cerro Tololo relative to the coastal town of La Serena. (Image: courtesy DES.)


What we observe

The Blanco telescope against a rich starfield. (Credit: Martin Murphy.)
Yes—we are seeing a lot of new objects, which are mostly stars—pick any of the pinpoints above, and most likely, you’re finding a “new” star yourself that no one has named before, one of the 300 billion or so in our galaxy! (Credit: Martin Murphy.)


Often we are asked: are you finding new things in the sky?  The basic answer is: oh, yes, we are finding plenty of new pinpricks of light, but most analyses of the images happen long after the observations are done, and the majority of the time that's when new objects are found. In general, the analyses break down into two types: cosmological analyses, and actual searches for new objects and phenomena.

Cosmological goals

We can start by discussing how the cosmological analyses work.  These analyses are largely statistical in the sense that we look at massive numbers of objects, primarily galaxies, that trace the large-scale structure (LSS) of the Universe—we generally call these objects "tracers" of the LSS, not surprisingly. After we have collected huge amounts of these objects and organized them into catalogs, we analyze these catalogs to see if they can tell us things like the concentrations of the dark matter.  As has been discussed in this prior blogpost, the dark matter makes up most of the Universe’s structural framework, in the filaments that criss-cross throughout space, and in the nodes they cross at. There is six times as much dark matter out in the Universe as there is of the ordinary matter we interact and are familiar with in our daily lives, but we can only see the normal matter (e.g., when it lights up as stars) which traces where the dark matter is distributed in aforementioned long strings we call filaments and big clumps called haloes (i.e. the nodes) as can be seen in the video clip below.

Some dark matter filaments showing the large scale structure of our Universe. (Visualization credit: Ralf Kähler and Tom Abel. Simulation credit: Oliver Hahn and Tom Abel.)

Once we know the overall structure, we can do multiple types of analyses on the objects that make it up to get more info on multiple aspects of the dark matter and dark energy that we are probing. This is because once we can figure out how the structure is distributed in space and time, which gives us a handle on what we generally term the “expansion history of the Universe,” (which we see in a schematic form below),  we can gain further insight into the nature of dark energy. 

Graphical depiction of the expansion history of the universe. (Credit: ESA Euclid Assessment Study Report.)
The Universe evolves from a homogeneous state after the big bang through cooling and expansion.  We initially see a slowing down of the expansion, then a speeding up about halfway through the its history due to the mysterious 'force' of dark energy. (Credit: ESA Assessment Study Report.)


In fact, here is an article from Sept 2017 that talks about some of the impressive results DES has obtained recently: DES clinches the most precise cosmological results ever extracted from gravitational lensing.

Wide field survey mode

Because the cosmological analyses require primarily looking at large scale distributions of the galaxies, what we want to do to collect data for these analyses is simply take large images of the sky to as much depth as possible, to collect the maximum number of images of tracer galaxies. To collect these images, we take exposures of about 90 seconds duration somewhere inside the DES "footprint," or the part of the sky we want to map out with DES, which is some 5000 square degrees, or about 1/8 of the entire sky (which amounts to about 41,000 square degrees). So we take a 90 second image, and then move to another area within the footprint (usually, one immediately adjacent to the previous image), take another picture, then move, take another and keep on doing this during this type of data-taking, which we call the “wide field survey” observing mode.


DES observing footprint. (Credit: Lahav et al., 2016.)
DES observing footprint -- the different colors indicate the coverage of the field at different times the survey images are taken.  [This image is taken from a previous post at the DES which covers some of the same ground is this piece, though from a less 'personal' perspective.] (Credit: Lahav et al., 2016.)


This mode of observation is primarily what the DES survey is about and spends time on.

Over five years, the plan is to observe most of the footprint about 10 times. As we collect data, we "stack" the images, which adds together the light from each imaged object. Once this is done, we can pick out the faint objects more easily above the background in the stacked images. We are also then able to better measure their properties (like shape, light profile, and colors) more accurately.

Single image of Orion nebula (left) vs. stacked image (right). (Credit: Eric Teske.)
Picture of a single frame (left) vs. stacked (right) image of the Orion Nebula, showing how much more one can see in the stacked images. (Credit: Eric Teske, Stellar Neophyte Astronomy Blog.)


Supernova finding and tracking mode

(As a side note, about one-seventh of the main survey time is dedicated to observing a certain set of much smaller fields in more depth, meaning longer exposures and many more visits. Our goal is to see supernovae (SNe) going off in these fields. We want to map out their "lightcurves" (their brightness as a function of time) well by getting multiple views of these SNe as they fade, so that we can get their distances precisely, as a separate type of tracer for cosmological analyses.)

Light curve of an SN1a-type supernova.
A Type Ia supernova lightcurve (made from applying a specific scaling relation to put multiple SN lightcurves, which are the differently colored points, on the same plot).  (Credit: Lawrence Berkeley National Laboratory.)


This sums up what observers do when taking data in the survey mode, collecting the data that will be used to ultimately reveal the dark ‘secrets of the Universe’ (i.e. give us insights into dark matter, dark energy, the Milky Way, other bodies in our own Solar System, etc.)—in Part 2 we will discuss the other mode observers can operate the telescope in—"Hunting for objects" mode!



Many more pictures (with captions) taken during this observing run (Nov 2017)