J/MNRAS/505/4496 The extended MILES library (Garcia Perez+, 2021)
An extension of the MILES library with derived Teff, log g, [Fe/H],
and [α/Fe]
Garcia Perez A.E., Sanchez-Blazquez P., Vazdekis A., Allende Prieto C.,
Milone A.D.C., Sansom A.E., Gorgas J., Falcon-Barroso J., Martin-Navarro I.,
Cacho R.
<Mon. Not. R. Astron. Soc., 505, 4496-4514 (2021)>
=2021MNRAS.505.4496G 2021MNRAS.505.4496G (SIMBAD/NED BibCode)
ADC_Keywords: Clusters, open ; Milky Way ; Stars, normal ; Stars, nearby ;
Stars, giant ; Stars, dwarfs ; Stars, atmospheres ;
Stars, G-type ; Stars, K-type ; Stars, A-type ; Stars, M-type ;
Spectroscopy ; Photometry, UBVRI ; Models, atmosphere ;
Abundances, [Fe/H]
Keywords: stars: abundances - stars: atmospheres -
stars: fundamental parameters - galaxies: stellar content
Abstract:
Extragalactic astronomy and stellar astrophysics are intrinsically
related. In fact, the determination of important galaxy properties
such as stellar masses, star formation histories, or chemical
abundances relies on the ability to model their stellar populations.
One important ingredient of these models is stellar libraries.
Empirical libraries must have a good coverage of Teff, [Z/H], and
surface gravity, and have these parameters reliably determined. MILES
is one of the most widely used empirical libraries. Here, we present
an extension of this library with 205 new stars especially selected to
cover important regions of the parameter space, including metal-poor
stars down to [Fe/H] ∼ -1.0. We describe the observations and data
reductions as well as a new determination of the stellar parameters,
including [α/Fe] ratio. The new MILES library contains 1070
stars with homogeneous and reliable determination of [Fe/H], Teff,
logg, and [α/Fe] ratio.
Description:
In this work, we present an accurate characterization of the MILES
stellar parameters using a similar approach to Allende Prieto
(2011MNRAS.411..807A 2011MNRAS.411..807A) but including five different parameters,
logg, [Fe/H], Teff, α-element abundance, and microturbulent
velocity. We apply this method to the original MILES library and to an
extension of 205 new stars.
The original library included 985 stellar spectra taken with the
Intermediate Dispersion Spectrograph (IDS), at the 2.5-m Isaac Newton
telescope, in La Palma, between 2000 and 2001 (Sanchez-Blazquez et al.
2006MNRAS.371..703S 2006MNRAS.371..703S, Cat. J/MNRAS/371/703). The current library
(Cenarro et al. 2007MNRAS.374..664C 2007MNRAS.374..664C, Cat. J/MNRAS/374/664) contains
896 field stars and 89 stars in 17 different clusters spanning a wide
metallicity range.
We have extended the MILES library with 205 new stars selected from
severals catalogues (see Note on table10.dat for more details). The
observations were carried out in two different runs (A and B) during
2011, using the same instrument (IDS) and telescope (INT) as used in
the original MILES library. As in the original library, each star was
observed three times; two of them were to cover the blue and red part
of the spectra and the third one, obtained with a wide slit oriented
along the paralactic angle.
The spectra have been reduced with REDUCME (Cardiel
1999PhDT........12C 1999PhDT........12C), following the same steps as in the original
MILES library. Error spectra were obtained with a parallel treatment
of data and error frames through the whole reduction process (referred
to Sanchez-Blazquez et al. 2006MNRAS.371..703S 2006MNRAS.371..703S, Cat. J/MNRAS/371/703
for further details). As in the original MILES library, the spectra
were corrected for interstellar reddening using the Galactic
extinction curve of Fizpatrick (1999PASP..111...63F 1999PASP..111...63F) with RV = 3.1.
The spectral resolution of the extension stars was estimated adopting
the same methodology employed in Falcon-Barroso et al.
(2011A&A...532A..95F 2011A&A...532A..95F) for the original MILES library, with the minor
difference that the PHOENIX models (Husser et al. 2013A&A...553A...6H 2013A&A...553A...6H)
were used as templates.
We do spectroscopic analysis on MILES spectra to derive the new five
physical properties as described in the section 3. Then, we ensure
optimization on spectral analysis due to large set of spectral types
inducing big spectral shape differences (see section 3.1
optimization). After that, we encode data quality via Χ2 value
extract from FERRE algorithm (Allende et al. 2006ApJ...636..804A 2006ApJ...636..804A)
delivering parameter and abundance uncertainties (see section 3.2).
Finally, we present results of our work in three tables. The
table1.dat contains the results for the 895 fields stars contained in
the MILES catalogue. The table2.dat regroups results for the 89
cluster stars already catalogued. Then, the table3.dat lists computed
properties for the 205 stars which extend the current MILES catalogue
to 1190 stars.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table6.dat 96 895 *Photospheric parameters and abundances estimates
for field stars in MILES catalogue
table8.dat 100 89 *Photospheric parameters and abundance estimates
for cluster stars in MILES catalogue
table10.dat 69 205 *Photospheric parameters and abundance estimates
for the extra stars sample of MILES catalogue
--------------------------------------------------------------------------------
Note on table6.dat and table8.dat: We derive these physical properties directly
from the MILES (Cenarro et al. 2007MNRAS.374..664C 2007MNRAS.374..664C, Cat. J/MNRAS/374/664)
spectra using an automated spectral synthesis analysis. We generate synthetic
spectra with the ASSET
(Koesterke et al. 2008ApJ...680..764K 2008ApJ...680..764K; Koesterke 2009AIPC.1171...73K 2009AIPC.1171...73K)
code on a grid of ATLAS9 model atmospheres (plane-parallel geometry,
Local Thermodynamic Equilibrium (LTE) conditions, and a distribution of
opacities; Meszaros et al. 2012AJ....144..120M 2012AJ....144..120M).
We adopted solar abundances from Asplund et al. (2005ASPC..336...25A 2005ASPC..336...25A).
Please see section 3 for more details on spectroscopic analysis.
Note on table10.dat: 205 new stars selected from the PASTEL catalogue
(Soubiran et al. 2010A&A...515A.111S 2010A&A...515A.111S ,Cat. B/pastel) to have relatively low
[Mg/Fe] values at -1 ~< [Fe/H] ~< 0, and [Mg/Fe] > 0.0 at higher metallicities.
We selected stars from
Borkova & Marsakov (2005AZh....82..453B 2005AZh....82..453B, Cat. J/AZh/82/453; dwarfs),
Thevenin (1998yCat.3193....0T 1998yCat.3193....0T, Cat. III/193; giants),
Alves-Brito et al. (2010A&A...513A..35A 2010A&A...513A..35A, Cat. J/A+A/513/A35;
thin and thick disc stars),
Edvardsson et al. (1993A&AS..102..603E 1993A&AS..102..603E, Cat. J/A+AS/102/603;
thin and thick disc stars),
Bensby et al. (2005A&A...433..185B 2005A&A...433..185B, Cat. J/A+A/433/185;
thin and thick disc stars),
Stephens & Boesgaard (2002AJ....123.1647S 2002AJ....123.1647S; halo metal-poor dwarfs/subgiants),
Mishenina et al. (2004A&A...418..551M 2004A&A...418..551M, Cat. J/A+A/418/551; FGK dwarfs),
and Nissen & Schuster (2010A&A...511L..10N 2010A&A...511L..10N, Cat. J/A+A/511/L10; halo dwarf).
The estimates of the Mg abundances are not on the same scale in all references,
but they were only used for the purpose of sample selection.
--------------------------------------------------------------------------------
See also:
J/ApJS/152/251 : Indo-US library of coude feed stellar spectra (Valdes+, 2004)
III/218 : ELODIE archive (Prugniel+, 2001)
III/232 : STELIB: A library of stellar spectra at R∼2000
(Le Borgne+, 2003)
J/ApJS/94/687 : Old stellar populations. V. (Worthey+, 1994)
J/MNRAS/371/703 : MILES library of empirical spectra (Sanchez-Blazquez+, 2006)
J/MNRAS/374/664 : Stellar atmospheric parameters in MILES library
(Cenarro+, 2007)
J/AZh/82/453 : Catalogue of spectroscopic abundances in stars
(Borkova+, 2005)
B/pastel : The PASTEL catalogue (Soubiran+, 2016-)
III/193 : Chemical Abundances in Late-Type Stars (Thevenin, 1998)
J/A+A/513/A35 : Abundances of Galactic red giants (Alves-Brito+, 2010)
J/A+AS/102/603 : Chemical evolution of the galactic disk II.
(Edvardsson+, 1993)
J/A+A/433/185 : Detailed abundance analysis of 102 F and G dwarfs
(Bensby+, 2005)
J/A+A/418/551 : Galactic disk stars abundances and velocities
(Mishenina+, 2004)
J/A+A/511/L10 : Abundances and space velocities of 94 stars (Nissen+, 2010)
Byte-by-byte Description of file: table6.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 13 A13 --- Name Star name (Name)
14- 21 F8.2 K Teff Effective temperature (Teff)
23- 30 F8.2 K e_Teff ?=-1000 Mean error of Teff
(σFTeff) (G1)
32- 35 F4.2 [cm/s2] logg Logarithm of the stellar surface gravity
(logg)
37- 44 F8.2 [cm/s2] e_logg ?=-1000 Mean error of logg
(σFlogg) (G1)
46- 50 F5.2 [-] [Fe/H] Logarithm of the iron abundance ratio
([Fe/H])
52- 59 F8.2 [-] e_[Fe/H] ?=-1000 Mean error of [Fe/H]
(σF[Fe/H]) (G1)
61- 65 F5.2 [-] [alpha/Fe] Logarithm of the alpha-element/Fe abundance
ratio ([α/Fe]) (G2)
67- 74 F8.2 [-] e_[alpha/Fe] ?=-1000 Mean error of [alpha/Fe]
(σF[alpha/Fe]) (G1)
76- 80 F5.2 [km/s] logvt Logarithm of the stellar microturbulence
velocity (logΞt)
82- 89 F8.2 [km/s] e_logvt ?=-1000 Mean error of vt
(σFlogΞt) (G1)
91- 92 A2 --- Flag Flag for the quality of the spectral
fitting (Flag) (G3)
94- 96 I3 --- MILES MILES identifier number (MILES) (G4)
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table8.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 8 A8 --- Cluster Cluster name (1)
10- 20 A11 --- Name Star name (Name)
22- 28 F7.2 K Teff Effective temperature (Teff)
30- 37 F8.2 K e_Teff ?=-1000 Mean error of Teff
(σFTeff) (G1)
39- 42 F4.2 [cm/s2] logg Logarithm of the stellar surface gravity
(logg)
44- 51 F8.2 [cm/s2] e_logg ?=-1000 Mean error of logg
(σFlogg) (G1)
53- 57 F5.2 [-] [Fe/H] Logarithm of the iron abundance ratio
([Fe/H])
59- 66 F8.2 [-] e_[Fe/H] ?=-1000 Mean error of [Fe/H]
(σF[Fe/H]) (G1)
68- 72 F5.2 [-] [alpha/Fe] Logarithm of the alpha-element/Fe abundance
ratio ([α/Fe]) (G2)
74- 81 F8.2 [-] e_[alpha/Fe] ?=-1000 Mean error of [alpha/Fe]
(σF[alpha/Fe]) (G1)
83- 87 F5.2 [km/s] logvt Logarithm of the stellar microturbulence
velocity (logΞt)
89- 93 F5.2 [km/s] e_logvt ?=-1000 Mean error of vt
(σFlogΞt) (G1)
95- 96 A2 --- Flag Flag for the quality of the spectral
fitting (Flag) (G3)
98-100 I3 --- MILES MILES identifier number (MILES) (G4)
-------------------------------------------------------------------------------
Note (1): Cluster identification taken from Cenarro et al.
(2007MNRAS.374..664C 2007MNRAS.374..664C, Cat. J/MNRAS/374/664 tablea2).
-------------------------------------------------------------------------------
Byte-by-byte Description of file: table10.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 9 A9 --- Name Star name (Name)
11- 17 F7.2 K Teff Effective temperature (Teff)
19- 23 F5.2 K e_Teff ?=-1000 Mean error of Teff
(σFTeff) (G1)
25- 28 F4.2 [cm/s2] logg Logarithm of the stellar surface gravity
(logg)
30- 33 F4.2 [cm/s2] e_logg ?=-1000 Mean error of logg
(σFlogg) (G1)
35- 39 F5.2 [-] [Fe/H] Logarithm of the iron abundance ratio
([Fe/H])
41- 44 F4.2 [-] e_[Fe/H] ?=-1000 Mean error of [Fe/H]
(σF[Fe/H]) (G1)
46- 50 F5.2 [-] [alpha/Fe] Logarithm of the alpha-element/Fe
abundance ratio ([α/Fe]) (G2)
52- 55 F4.2 [-] e_[alpha/Fe] ?=-1000 Mean error of [alpha/Fe]
(σF[alpha/Fe]) (G1)
57- 61 F5.2 [km/s] logvt Logarithm of the stellar microturbulence
velocity (logΞt)
63- 66 F4.2 [km/s] e_logvt ?=-1000 Mean error of vt
(σFlogΞt) (G1)
68- 69 A2 --- Flag Flag for the quality of the spectral
fitting (Flag) (G3)
--------------------------------------------------------------------------------
Global notes:
Note (G1): The synthetic spectra were grouped in five different grids according
to their Teff. The range of free parameters in the FERRE library
(Allende Prieto et al. 2006ApJ...636..804A 2006ApJ...636..804A) used for each group is
listed in Table 3 section 3. To do so, each individual spectrum,
either observational or theoretical, was scaled by its average flux
(measured in the whole spectral range) before comparing them.
Hereafter, flux error spectra for the MILES stars, which were
obtained by propagating the errors during data reduction as
described in Section 2.1, were considered in the fitting process.
The optimization was carried out with version 4.7.2 of FERRE
(Allende Prieto et al. 2006ApJ...636..804A 2006ApJ...636..804A).
Note (G2): The ratio of alpha process elements to iron, also known as the
alpha-enhancement, is written as the logarithm of the alpha process
elements O, Ne, Mg, Si, S, Ca, and Ti.
Note (G3): Solutions coming from the multiple ferre libraries were compared,
and that providing the minimum x2 was adopted, discarding solutions
with parameters or abundances near or on the limits of the associated
spectral library. A minimum distance to the grid edge of 10 per cent
of the step size was enforced, to avoid issues related to
inaccuracies in the interpolations near the limits.
Following the procedure explained in the section 3.2 quality checks,
We construct a flagging scheme according to the quality of the
spectral fits as follows:
0 = Bad fit,
1 = Good fit,
2 = Good fit although close to a model grid edge,
3 = Good fit despite model edge proximity to all relevant model
libraries,
4 = Bad fit and model edge proximity,
5 = Bad fit and edge proximity for all relevant model libraries.
The same scheme applies if a second analysis is performed, which is
the case for all the stars with an edge proximity parameter estimate.
In this case, the flag digit is duplicated (e.g. 22 stands for a good
fit plus an edge proximity in the second analysis).
Note (G4): Taken from MILES
(Cenarro et al. 2007MNRAS.374..664C 2007MNRAS.374..664C, Cat. J/MNRAS/374/664)
(Sanchez-Blazquez et al. 2006MNRAS.371..703S 2006MNRAS.371..703S, Cat. J/MNRAS/371/703).
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Luc Trabelsi [CDS] 06-Jun-2024