*** Wide and Time Domain surveys
The DES uses four separate techniques to measure dark energy
parameters, each of which makes different demands on the data. For
three of the four probes, a uniform survey of a large area of the sky
can meet these demands. For example, the weak lensing probe requires
shape measurements on as many galaxies as possible, and the
large-scale structure probe requires identification of galaxies on a
large, contiguous volume of sky. Images covering a wide area of sky
with good image quality can provide both of these.
On the other hand, the supernova probe need to measure the brightness
of supernovae at many points in time: the supernava probe
requires a time-domian survey, which repeats observations of a small
area of sky at regular intervals. The variation in the light from the
supernova in that field can then be measured.
The DES therefore consists of two separate data sets: a wide survey
and a time-domain survey. The wide survey covers a large, contiguous
area of sky, each part of which is only observed a few times over the
course of the survey. The time-domain survey covers only a small area
on the sky, but exposures are repeated a large number of times at
regular intervals over the course of the survey.
Two factors determine which survey the telescope works on at any given
time: the atmospheric conditions and the time since the previous
time-domain observations. The wide-survey is being used for
weak-lensing measurements, and exposures taken under conditions which
are acceptable for the supernova probe may be of little use for
weak-lensing measurements. Therefore, observing time with less than
ideal conditions is used primarily for time-domain exposures.
The supernove survey requires measurements of the brightness of
supernova at regular intervals, without gaps in time between
successive measurements. Therefore, if a portion of the time-domain
survey has not been observed within the past week, then observing time
will be used to obtain the needed exposures.
*** The wide-survey footprint
The dark energy science programs use images from DECam to make
catalogs of galaxies. The DES is scheduled enough time to survey
roughly one eighth of the entire sky, and not all parts of the sky are
equally useful for these measurements: the choice of which area of sky
to survey must be made carefully.
When the telecope is pointing near the Milky Way, the galaxies we are
trying to catalog are obscured by stars and dust within our own
galaxy: area far from the Milky Way is more valuable for dark energy
science.
There are also advantages to collecting data on the areas of sky that
have already been observed by complementary instruments. For example,
the South Pole Telescope is making a map of the sky in microwaves,
which is useful for measuring the masses of galaxy clusters. By
combining DES catalogs of galaxies with catalogs of galaxy clusters
measured by the SPT, science not possible with either instrument alone
becomes possible.
Several of these complementary data sets are located along the
celestial equator, parts of the sky that pass directly overhead of
someone standing at the equator. Observatories at different latitudes
are better positioned to observe different parts of the sky. Cerro
Tololo is in the southern hemisphere, so it can see the southern sky
well, but much of the northern sky (closer to Polaris) cannot be seen
at all. In contrast, observatories in the northern hemisphere cannot
see the southern sky well. Surveys of area along the equator, however,
are visible from observaties in both hemispheres, so it is the only
possible location for projects in which data from instruments in both
hemispheres need to be combined.
These different requirements result in the irregular shape of the DES
footprint: there are rectangular regions along the equator, another
region covering an area further south (observed by the South Pole
Telescope), connected to each other by an area optimal for observing
from Cerro Tololo, all bounded by the edges of the Milky Way.
*** obstac
While observing, many factors influence which exposure needs to be
taken when. Different exposures can be taken under different
conditions, and at different times in the night or times of
year. Areas of the sky which have been completed should not be done
again, while areas of the sky which have not already been attempted,
or completed under poor condition, need to be taken again. While
observing, the DES takes a new image every roughtly two minutes. To
avoid wasting time, the time between the end of one exposure and the
beginning of the next must be kept to a minimum, and it is not
practical for human observers to perform all of the bookkeeping and
conditions necessary over the course of each night: efficient
operations requires automation of the selection and execution of
science observations during the night.
The observing tactician software, obstac, automates exposure
selection, takeing into account the many factors that need to
influence the choice of exposure: the time and date, the weather
conditions, the phase and location of the moon, and records of what
has been completed, and what remains to be completed. When observing,
the DES astronomers just need to get the observing system set up, and
then turn on obstac; obstac takes care of choosing exposures though
the night.