|
UVES NGSL Project
"If I could start my life over again, I would undertake spectral classification at resolution no worse than 1 A,
and as much better than that as possible."
P.C. Keenan 1984
|
This is the home page of the UVES NGSL project. Investigators are David R. Silva (http://www.noao.edu/noao/staff/dsilva/), Reinhard W. Hanuschik (rhanusch@eso.org), Francesca Primas (fprimas@eso.org), and Michael D. Gregg (Univ. Calif.).
The UVES NGSL project is a spectroscopic survey to provide flux-calibrated, high-resolution spectra in
the optical and near-IR of stars with known effective temperature, surface
gravity and metallicity. The program stars have been carefully selected to
cover this three-dimensional parameter space. Learn
here
more about the
project.
The UVES
NGSL project complements the HST/STIS NGSL (Next Generation Spectral Library)
project.
See here for their web site.
There is a
workshop paper presenting their project. Check also
here for a compilation.
The UVES data have been collected
at the UVES high-resolution echelle spectrograph on the 8.2m UT2/Kueyen telescope at ESO's VLT on Paranal/Chile.
They have been acquired over six periods (P71-P74, P76, P79 which correspond to April 2003 - March 2005, September 2005 - March 2006, April - July 2007).
[Left: the HR diagram, from wikipedia]
|
Data processing. The data have been processed in two steps.
Step 1 involved the standard UVES pipeline and the usage of certified and archived
master calibration data. Most data have been extracted in AVERAGE mode (due to
problems with high S/N). All data from P76 have been extracted with OPTIMUM extraction.
Data are flux-calibrated. Find more about the reduction by the UVES pipeline here.
Step 2 processing ('NGSL pipeline') has been:
- organize per target (usually OB twins with a dichroic#1 and a dichroic#2
part)
- average multiple exposures from same OBs into one final product per setup --> files with suffix _avg.fits
- do a fine-correction of flux scale to minimize steps between setups (by
applying fudge factors) --> files with suffix _scl.fits
- determine S/N ratio
- reduce wavelength scale to *common* (stellar) rest system --> files with
suffix _rvc.fits
- mark telluric features
All fluxes are in 1e-13 erg/s/cm**2/A.
Telluric stars have been measured occasionally. They are marked by their
target name prefixed by a string containing the year and the month of their
observation. For instance, T0306_HD_142919 stands for "telluric standard
HD_142919 measured in 2003-06".
Spectra of telluric standards have been
treated exacly the same way as the NGSL target stars. They have been used to
build the table of telluric absorption lines which is the basis for the
telluric markers in the spectral plots red of 5800 A.
Known issues:
- Precision of flux calibration: the master response curves are collected
typically for a year and cannot reflect faster changes of instrument
response.
- There is often a step in the flux-calibrated spectra between the 580L and 580U settings.
- A slit width of about 1 arcs has been used, there have been varying slit losses under varying seeing conditions.
This has been a
'filler' programme harvesting telescope time with mediocre ambient conditions.
Send comments to <Reinhard.Hanuschik@eso.org>
Last update: 2009-10-05 17:52:37
|