III/218 ELODIE archive (Prugniel+, 2001)
A database of high and medium-resolution stellar spectra
Prugniel P., Soubiran, C.
<Astron. Astrophys. 369, 1048 (2001)>
=2001A&A...369.1048P 2001A&A...369.1048P
ADC_Keywords: Spectrophotometry ; Atlases ; Spectroscopy ; Populations, stellar
Keywords: atlases - stars: abundances - stars: atmospheres -
stars: fundamental parameters - galaxies: stellar content
Abstract:
We present a database of 908 stellar spectra of 709 stars in the
wavelength range 410 to 680 nm obtained with the ELODIE spectrograph.
The archive provides a large coverage of the space of stellar
parameters: Spectral types O to M, luminosity classes V to I and
metallicities [Fe/H] from -3.0 to +0.8. At the nominal resolution,
R=42000, the mean signal-to-noise ratio is 150 per pixel. The spectra
given at this resolution are normalized to their pseudo-continuum and
are intended to serve for abundance studies, spectral classification
and tests of stellar atmosphere models. A lower resolution version of
the archive, R=10000, is calibrated in physical flux with a broad-band
photometric precision of 2.5% and narrow-band precision of 0.5%. It is
scoped for stellar population synthesis of galaxies and clusters, and
for kinematical investigations of stellar systems. The archive is also
distributed in FITS format through the HYPERCAT database
http://www-obs.univ-lyon1.fr/hypercat/spectrophotometry.html .
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 284 908 *Spectrophotometry and atmospheric parameters
table2.dat 52 21 Comparison of Lick indices
ll/* . 919 Sub-directory: low-resolution archive (R=10000)
lh/* . 908 Sub-directory: high-resolution archive (R=42000)
--------------------------------------------------------------------------------
Note on table1.dat: For the definition of the Lick indices see Tragger et al.
(1998ApJS..116....1T 1998ApJS..116....1T) and Worthey & Ottaviani (1997ApJS..111..377W 1997ApJS..111..377W)
--------------------------------------------------------------------------------
See also:
http://www-obs.univ-lyon1.fr/hypercat/ : the HYPERCAT database
http://www.obs-hp.fr/www/archive/archive.html : the ELODIE Archive
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 5 I5 --- Num Running number in the archive
7- 16 A10 --- Name *Object identifier
17 A1 --- u_Name *[?] Uncertain name
19- 20 I2 h RAh *? Right Ascension J2000 (hours)
22- 23 I2 min RAm *? Right Ascension J2000 (minutes)
25- 28 F4.1 s RAs *? Right Ascension J2000 (seconds)
30 A1 --- DE- *? Declination J2000 (sign)
31- 32 I2 deg DEd *? Declination J2000 (degrees)
34- 35 I2 arcmin DEm *? Declination J2000 (minutes)
37- 38 I2 arcsec DEs *? Declination J2000 (seconds)
41- 50 A10 --- SpType Spectral type (from INCA)
52- 58 F7.2 km/s HRV Heliocentric radial velocity
60 I1 --- q_Flux [0,3] Quality of the flux calibration
62- 65 F4.2 --- Airm Airmass at mid-time of observation
66- 72 F7.0 K Teff Effective temperature
74- 75 I2 --- q_Teff *[-1,4] Quality flag on Teff
78- 82 F5.2 [cm/s+2] logg decimal log of surface gravity
84- 85 I2 --- q_logg *[-1,1] Quality flag on logg
86- 92 F7.2 [Sun] [Fe/H] Iron abundance relative to solar
94- 95 I2 --- q_[Fe/H] *[-1,4] Quality flag on [Fe/H]
96-102 F7.0 K TeffM Estimated effective temperature
105-109 F5.2 [cm/s+2] loggM Estimated log of surface gravity
110-116 F7.2 [Sun] [Fe/H]M Estimated [Fe/H]
117-124 F8.4 0.1nm Ca4227 Ca4227 index
125-132 F8.4 0.1nm G4300 G4300 index
133-140 F8.4 0.1nm HgamA H gamma A index
141-148 F8.4 0.1nm HgamF H gamma F index
149-156 F8.4 0.1nm Fe4383 Fe4383 index
157-164 F8.4 0.1nm Ca4455 Ca4455 index
165-172 F8.4 0.1nm Fe4531 Fe4531 index
173-180 F8.4 0.1nm Fe4668 Fe4668 index
181-188 F8.4 0.1nm Hbeta Hbeta index
189-196 F8.4 0.1nm Fe5015 Fe5015 index
197-204 F8.4 mag Mg1 Mg1 index
205-212 F8.4 mag Mg2 Mg2 index
213-220 F8.4 0.1nm Mgb Mgb index
221-228 F8.4 0.1nm Fe5270 Fe5270 index
229-236 F8.4 0.1nm Fe5335 Fe5335 index
237-244 F8.4 0.1nm Fe5406 Fe5406 index
245-252 F8.4 0.1nm Fe5709 Fe5709 index
253-260 F8.4 0.1nm Fe5782 Fe5782 index
261-268 F8.4 0.1nm NaD NaD index
269-276 F8.4 mag TiO1 TiO1 index
277-284 F8.4 mag TiO2 TiO2 index
--------------------------------------------------------------------------------
Note on Name, u_Name:
The names are generally from the HD catalogue (III/135),
BD numbers (I/122) are also used. Uncertain identification are noted
with a "?" in column u_Name.
Note on RAh, RAm, RAs, DE-, DEd, DEm, DEs:
The position was extracted from SIMBAD (when the Name is accurately
known), or from the FITS headers.
Note on q_Teff:
q_Teff=1: Poor determination -> Q_Teff=4: Excellent (see Sect. 2)
q_Teff=0: means that we have estimated Teff from the B-V colour (Tycho2
I/259 catalogue), assuming the empirical colour-temperature relation
for a main sequence star and neglecting the interstellar extinction
q_Teff=-1: Internal determination of the effective temperature
Note on q_logg:
q_logg=1 means that the determination of LOGG is taken from the
literature (see Sect. 2).
q_logg=0 means that LOGG was converted from the V absolute magnitude
from Hiparcos and Teff, using a bolometric correction valid for
a main sequence start and an empirical mass-to-light relation.
q_logg=-1 Internal determination of the surface gravity
Note on q_[Fe/H]:
See the definition in Sect. 2.
q_[Fe/H]=1: Poor determination -> q_[Fe/H]=4: Excellent
q_[Fe/H]=-1: Internal determination of [Fe/H]
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table2.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 9 A9 --- Band *Designation of the spectrophotometric band
10- 12 I3 --- Nco *Number of comparisons
17- 22 F6.3 --- Ave *Average value of the index
27- 32 F6.3 --- Diff *Mean difference
37- 42 F6.3 --- rms rms of the difference after 3σ clipping
46- 52 F7.4 --- error *Majorant of relative error
--------------------------------------------------------------------------------
Note on Band:
The designation is as in table1.dat
Note on Ave:
Average value of the measurements used for the comparison.
Note on Nco:
A (non-iterated) 3-sigma clipping was done. Nco is the number of
comparisons left afterward.
Note on Diff:
The units are in the "natural" units of the corresponding index, 0.1nm
or mag. "Molecular" indices: Mg1, Mg2, TiO1, TiO2 and in mag,
the others in 0.1nm.
Note on error:
The mean difference Diff is converted into relative error by assuming that
the error is concentrated on the central band of the index
--------------------------------------------------------------------------------
Description of files: ll/0*.fits
--------------------------------------------------------------------------------
The low-resolution archive (R=10000)
The FITS files have one axis which is the wavelength (CTYPE1='WAVE-WAV'),
scaled in units of 0.1nm (CUNIT='10**(-10)m'), with a step of 0.02nm.
They contain 2 image-type extensions. The primary extension gives the
fluxes (BTYPE='FLUX-PHY') above the atmosphere and normalized to continuum.
The other image-type extension contains the corresponding detector
noise, its EXTAME is 'NOISE'.
Two other extensions contain the flux calibration relations (FCR) needed
to convert the instrumental flux (BTYPE='FLUX-INS') into physical
flux (ie. above atmosphere) (BTYPE='FLUX-PHY') and into continuum-
normalized flux (BTYPE='FLUX-NOR'). The EXTNAME of these FCR are
respectively 'FCANOR' and 'FCAPHY'. Explanations about the content
of the FCR can be found in the documentation of Pleinpot available
from Hypercat. The processing pipeline in the FITS archive of Hypercat
allows the different conversions.
Description of some particular keywords:
TEFF [K] Teff from literature
Q_TEFF Quality flag for Teff: 4=excellent
LOGG [log cm/s**2)] log g from literature
Q_LOGG Quality flag for LOGG
[Fe/H] [Sun] [Fe/H] from literature
Q_[Fe/H] Quality flag for [Fe/H]: 4=excellent
MV [mag] Mv from Hipparcos
Q_MV Quality flag for Mv: 4=excellent
S/N S/N-per-pixel of original spectrum at 555 nm
VR [km/s] heliocentric radial velocity
FWHM [km/s] FWHM line broadening
Q_VR comment on Vr determination
SPTYPE spectral type from INCA
VMAG [mag] V-magnitude
E_VMAG [mag] error on V-magnitude
B-V [mag] B-V Johnson
R_PHOT Source of photometry: TYCHO2, INCA
H_WRESOL [0.1nm] FWHM wavelength resolution at 550 nm
Description of files: lh/0*.fits
--------------------------------------------------------------------------------
High-resolution archive (R=42000)
The structure of the files is similar to ll/0*.fits. The main
differences are that the sampling is 0.005nm and that the fluxes are
normalized to the pseudo-continuum (BTYPE='FLUX-NOR')
Description of files: ll/99*.fits
--------------------------------------------------------------------------------
These files are primarly intended to be used by the specialized reduction
programs and by the pipeline processing. Their description may not be
accurate enough for a standalone usage, and for such applications the users
are invited to contact the authors additional help.
ll/99001
Blaze function used as reference
This is a "s2d" file produced by the ELODIE reduction program (TACOS).
"s2d" means: spectra extracted by TACOS, each line is one order.
The Blaze function is obtained from a Tungsten lamp spectrum. Each
order is smoothed using a degree 17 polynomial and is normalized to
its maximum value.
ll/99002
Correction to the TACOS blaze (ll/99001)
Dividing a spectrum by a TACOS blaze produces a stair case result
(consecutive orders do not connect because of the independent
normalizations). In order to connect the orders and to reduce the
instrumental response to a "typical" stellar SED, we computed
this correction using 3 metal-deficient stars observed during
the same observing run. The original "s2d" spectra are divided
by ll/99001 and the continuum of each order is modelled
with a degree 2 polynomial.
ll/99003
Pseudo-continuum mask
It is computed after the whole archive has been reduced. In
each continuum-normalized spectrum, a simple threshold is used
to diagnostic if a given wavelength point belong to the
pseudo-continuum. We chose 0.97 as threshold. The corresponding
wavelength point is given the value 1 or 0 if it belongs
respectively to the continuum or not. ll/99003 is the
average of this mask over the whole archive. In some sense it
indicates the probability for each wavelength point to be a
continuum point. It is used as a weight for iterating the
computation of ll/99002 and of the whole archive.
(determining the continuum is required to fit and correct
variations of the blaze, see the text of the article).
ll/99004
Primary flux calibration.
The first axis is the wavelength direction
The first line is the instrumental response, The second the
airmass dependence and the third the "haze" dependence (see text).
ll/99004 is obtained after a comparison between the
instrumental fluxes and templates from the datasets L1985BURN L1998GLUS
and L1997ALEK
ll/99005
Smoothed flux calibration file
Each line of ll/99004 is modeled by a degree 17 polynomial
in order to discard the wavy residuals due to an approximate
modeling of the spectral resolution of the templates
ll/99006
Secondary flux calibration.
ll/99005 with correction after pairwyse comparisons
ll/99007
Corrected flux calibration file
The spectra calibrated in physical flux with ll/99005 are
compared with spectra from L1984JACO L1996SERO L96111KP1. The
details of these comparison are given in ll/99007. The
mean comparison is smoothed and used to compute a correction
to the instrumental response. ll/99006 serves to compute
the final flux calibration relations associated to each spectrum in
the archive.
ll/99008
Comparison with other templates.
The first axis is the wavelength direction.
The 9 first lines contains respectively the comparisons with
(1) L96111KP1 (Jone's library, red band),
(2) L96111KP2 (Jone's library, blue band),
(3) L1996SERO (OHP observations),
(4) L1996SERO (CFH observations),
(5) L1984JACO (Jacoby's library)
(6) L1987KIEH (Kiehling, 1987, Cat. III/124)
(7) L1994DANK (Danks & Dennefeld, 1994PASP..106..382D 1994PASP..106..382D)
(8) L1983GUNN (Gunn & Stryker, 1983, Cat. III/88 ) and
(9) L1984NSOA (Solar flux atlas).
The 10th line combines the first 9 lines. And the 11th is the
final correction obtained by smoothing the 10th.
ll/99010
Absolute TGM model
The first axis is the wavelength direction.
The second axis contains the 20 parameters of the TGM model fitted
to the spectra for which the atmospheric parameters are known.
The functions of the models are:
1 Cst
2 T
3 Z
4 G
5 T**2
6 T**3
7 T**4
8 T*Z
9 T*G
10 G**2
11 Z**2
12 T**2*G
13 T**2*Z
14 G**3
15 Z**3
16 T*G**2
17 T*Z**2
18 10**T
19 10**(T*2)
20 G*Z
If Teff is the effective temperature, log(G) the decimal
logarithm of the surface gravity, T and G are defined as:
T=log(Teff)-3.7 and G=log(G)-4.44. Z is [Fe/H].
Each wavelength point is fitted to the 20 functions above and
each line in 99010 are the coefficients of these functions. This
is a preliminary model and the final version will be described in the
paper presenting TGMET.
ll/99011
Internal TGM model
Same format as ll/99009 but the coefficient are fitted on
the whole archive using the atmospheric parameters obtained
from a fit to ll/99010.
Acknowledgements: Philippe Prugniel
(End) Philippe Prugniel, Patricia Bauer [CDS] 19-Jan-2001