| |
 |
| Paper: |
Cure-WISE: HETDEX Data Reduction with Astro-WISE |
| Volume: |
485, Astronomical Data Analysis Software and Systems XXIII |
| Page: |
447 |
| Authors: |
Snigula, J. M.; Drory, N.; Fabricius, M.; Landriau, M.; Montesano, F.; Hill, G. J.; Gebhardt, K.; Cornell, M. E. |
| Abstract: |
The Hobby-Eberly Telescope Dark Energy Experiment (HETDEX, Hill et al. 2012b) is a blind spectroscopic
survey to map the evolution of dark energy using Lyman-alpha
emitting galaxies at redshifts 1.9< ɀ <3.5 as tracers. The survey
will use an array of 75 integral field spectrographs called the
Visible Integral field Replicable Unit (IFU) Spectrograph (VIRUS, Hill et al. 2012c). The 10m HET (Ramsey et al. 1998) currently receives a wide-field upgrade (Hill et al. 2012a) to accomodate the spectrographs and to
provide the needed field of view. Over the projected five year run
of the survey we expect to obtain approximately 170 GB of data each
night. For the data reduction we developed the Cure pipeline, to
automatically find and calibrate the observed spectra, subtract the
sky background, and detect and classify different types of
sources. Cure employs rigorous statistical methods and complete
pixel-level error propagation throughout the reduction process to
ensure Poisson-limited performance and meaningful significance
values. To automate the reduction of the whole dataset we
implemented the Cure pipeline in the Astro-WISE framework. This
integration provides for HETDEX a database backend with complete
dependency tracking of the various reduction steps, automated
checks, and a searchable interface to the detected sources and user
management. It can be used to create various web interfaces for data
access and quality control. Astro-WISE allows us to reduce the data
from all the IFUs in parallel on a compute cluster. This cluster
allows us to reduce the observed data in quasi real time and still
have excess capacity for rerunning parts of the reduction. Finally,
the Astro-WISE interface will be used to provide access to reduced
data products to the general community. |
|
|
|
|
|
