J/AJ/144/168 Spectroscopy of Scl 1019417 and UMi 20103 (Kirby+, 2012)
Detailed abundances of two very metal-poor stars in dwarf galaxies.
Kirby E.N., Cohen J.G.
<Astron. J., 144, 168 (2012)>
=2012AJ....144..168K 2012AJ....144..168K
ADC_Keywords: Equivalent widths ; Abundances ; Galaxies, nearby ;
Stars, metal-deficient ; Spectroscopy
Keywords: galaxies: abundances - galaxies: dwarf - galaxies: evolution -
Local Group
Abstract:
The most metal-poor stars in dwarf spheroidal galaxies (dSphs) can
show the nucleosynthetic patterns of one or a few supernovae (SNe).
These SNe could have zero metallicity, making metal-poor dSph stars
the closest surviving links to Population III stars. Metal-poor dSph
stars also help to reveal the formation mechanism of the Milky Way
(MW) halo. We present the detailed abundances from Keck/HIRES
spectroscopy for two very metal-poor stars in two MW dSphs. One star,
in the Sculptor dSph, has [FeI/H]=-2.40. The other star, in the Ursa
Minor dSph, has [FeI/H]=-3.16. Both stars fall in the previously
discovered low-metallicity, high-[α/Fe] plateau. Most abundance
ratios of very metal-poor stars in these two dSphs are largely
consistent with very metal-poor halo stars. However, the abundances of
Na and some r-process elements lie at the lower end of the envelope
defined by inner halo stars of similar metallicity. We propose that
the metallicity dependence of SN yields is the cause. The earliest SNe
in low-mass dSphs have less gas to pollute than the earliest SNe in
massive halo progenitors. As a result, dSph stars at -3<[Fe/H]←2
sample SNe with [Fe/H]{Lt}-3, whereas halo stars in the same
metallicity range sample SNe with [Fe/H]~-3. Consequently,
enhancements in [Na/Fe] and [r/Fe] were deferred to higher metallicity
in dSphs than in the progenitors of the inner halo.
Description:
Target stars (Scl 1019417 and UMi 20103) were selected from the
catalog of medium-resolution spectroscopic abundance measurements of
Kirby et al. (2010, cat. J/ApJS/191/352). We observed Scl 1019417 (in
the Sculptor dwarf spheroidal galaxy) and UMi 20103 (in the Ursa Minor
dwarf spheroidal galaxy) with HIRES (High Resolution Echelle
Spectrometer) on Keck I telescope at Keck Observatory. We used the red
cross-disperser with a spectral range of 3927-8362Å. The slit width
was 1.15", and the slit length was 7". Scl 1019417 was observed on
2010 Nov 27 (R=29300) and UMi 20103 on 2010 Apr 2, 2011 Jun 6, and
2012 Feb 2 (R=34500). We included in our study two metal-poor
abundance standards previously observed with HIRES by J. Cohen: HD
115444 and HD 122563. Table 1 gives observational details for all of
our targets.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 124 6 HIRES observations and data quality
table2.dat 95 353 Line list with equivalent widths
table4.dat 56 140 Abundances
--------------------------------------------------------------------------------
See also:
II/271 : TASS Mark IV patches photometric catalog, version 2 (Droege+, 2007)
VII/233 : The 2MASS Extended sources (IPAC/UMass, 2003-2006)
II/237 : Stellar Photometry in Johnson's 11-color system (Ducati, 2002)
III/211 : Keck/HIRES Sky Line Atlas (Osterbrock+ 1997)
J/ApJ/708/560 : Spectroscopy of UMa II and Coma Ber (Frebel+, 2010)
J/ApJS/191/352 : Abundances in stars of MW dwarf satellites (Kirby+, 2010)
J/ApJ/719/931 : Chemical evolution of the UMi dSph (Cohen+, 2010)
J/ApJ/705/328 : Abundance measurements in Sculptor dSph (Kirby+, 2009)
J/ApJ/701/1053 : Abundances of 8 stars in the Draco dSph (Cohen+, 2009)
J/AJ/132/137 : Abundances of extremely metal-poor carbon stars (Cohen+, 2006)
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1 A1 --- f_Name [PS] P=program star, S=abundance standard
3- 13 A11 --- Name Star name (Scl=Sculptor, UMi=Ursa Minor)
15- 16 I2 h RAh Hour of Right Ascension (J2000)
18- 19 I2 min RAm Minute of Right Ascension (J2000))
21- 22 I2 s RAs Second of Right Ascension (J2000)
24 A1 --- f_RAs [a] Correction flag in RAm (1)
26 A1 --- DE- Sign of the Declination (J2000)
27- 28 I2 deg DEd Degree of Declination (J2000)
30- 31 I2 arcmin DEm Arcminute of Declination (J2000)
33- 34 I2 arcsec DEs Arcsecond of Declination (J2000)
36- 40 F5.2 mag Vmag [6.19/18.41]? V band magnitude (2)
42- 45 F4.2 mag V-I ? The V-I color index (2)
47- 50 F4.2 mag V-J ? The V-J color index (2)
52- 55 F4.2 mag V-K ? The V-K color index (2)
57- 67 A11 "Y:M:D" Date UT date of observation
69 I1 --- N1 Number of exposures
71- 73 I3 min T1 Exposure time
75 I1 --- N2 ? 2nd number of exposures
77- 79 I3 min T2 ? 2nd exposure time
81- 83 I3 min Exp ? Total exposure time
85- 87 F3.1 arcsec S Seeing
89- 92 I4 0.1nm lam0 [3185/3927] Spectral range: lower limit (Å)
94- 97 I4 0.1nm lam1 [5993/8364] Spectral range: upper limit (Å)
99-103 I5 --- Res [29300/37300] Resolving power (3)
105-107 I3 --- S/N [91/566]? Signal-to-noise ratio per FWHM
resolution element at 5750Å
109-115 F7.2 km/s HRV [-244.48/98.87] Heliocentric radial velocity
117-120 F4.2 km/s e_HRV Uncertainty in HRV
122-124 A3 --- EW Name of the column in table2 giving the
equivalent widths for this star (added at CDS)
--------------------------------------------------------------------------------
Note (1): a = In table 1 of the paper, the minute of Right Ascension (01')
for the star Scl 1019417 was most likely a mistake. This value was
corrected at CDS into 00'.
Note (2): The V magnitude for HD 115444 was taken from the TASS Mark IV
survey (cat. II/271), and the J and K magnitudes were taken from 2MASS
(cat. VII/233). All of the photometry for HD 122563 was taken from
Ducati (2002, cat. II/237).
Note (3): Defined as the Full Width at Half Maximum (FWHM) of unbroadened
spectral features divided by wavelength. This number depends on both
the spectrograph configuration and stellar macroturbulence.
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table2.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 2 A2 --- El Element identification
4- 5 A2 --- Ion Ionization state (I=neutral; II=ionized)
7- 13 F7.2 0.1nm lambda [3538.52/7797.59] Wavelength; in Å
15- 19 F5.3 eV EP Excitation potential
21- 27 F7.3 [-] loggf Logarithm of oscillator strength
29 A1 --- l_EW1 Limit flag on EW1
30- 34 F5.1 0.1pm EW1 ? HD 115444 equivalent width; in mÅ (1)
36- 39 F4.1 0.1pm e_EW1 ? Noise error in EW1; in mÅ (2)
41- 44 F4.1 0.1pm s_EW1 ? DAOSPEC error in EW1; in mÅ (3)
46 A1 --- l_EW2 Limit flag on EW2
47- 51 F5.1 0.1pm EW2 ? HD 122563 equivalent width; in mÅ (1)
53- 56 F4.1 0.1pm e_EW2 ? Noise error in EW2; in mÅ (2)
58- 61 F4.1 0.1pm s_EW2 ? DAOSPEC error in EW2; in mÅ (3)
63 A1 --- l_EW3 Limit flag on EW3
64- 68 F5.1 0.1pm EW3 ? Scl 1019417 equivalent width; in mÅ (1)
70- 73 F4.1 0.1pm e_EW3 ? Noise error in EW3; in mÅ (2)
75- 78 F4.1 0.1pm s_EW3 ? DAOSPEC error in EW3; in mÅ (3)
80 A1 --- l_EW4 Limit flag on EW4
81- 85 F5.1 0.1pm EW4 ? UMi 20103 equivalent width; in mÅ (1)
87- 90 F4.1 0.1pm e_EW4 ? Noise error in EW4; in mÅ (2)
92- 95 F4.1 0.1pm s_EW4 ? DAOSPEC error in EW4; in mÅ (3)
--------------------------------------------------------------------------------
Note (1): Measured from the original spectrum.
Note (2): This random error is the standard deviation among the equivalent
widths measurements from noise-added spectra (see Section 3.2 for details)
Note (3): DAOSPEC (Stetson & Pancino, 2008PASP..120.1332S 2008PASP..120.1332S) is the technique
we used to measure equivalent widths in each spectrum. The DAOSPEC
error estimate includes uncertainty on the equivalent width not only
due to spectral noise but also due to systematic errors caused by
blended lines and imprecise continuum placement
(see Section 3.1. and 3.2 for more details).
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table4.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 11 A11 --- Name Star name
13- 18 A6 --- El Element identification (1)
20- 21 A2 --- Ion Ionization state (I=neutral; II=ionized)
23- 24 I2 --- N ? Number of lines
26- 28 A3 --- f_N [syn] spectral synthesis for C (1)
30 A1 --- l_eps Limit flag on Eps
32- 36 F5.2 [-] eps ? Abundance ε of the species (2)
38 A1 --- l_[X/Fe] Limit flag on [X/Fe]
40- 44 F5.2 [Sun] [X/Fe] ? [X/Fe] abundance relative to the solar
abundance (3)
46 A1 --- f_[X/Fe] [a] [X/Fe] is relative to H ([Fe/H])
48- 51 F4.2 [Sun] e_[X/Fe] ? Uncertainty in ε due to spectral noise
(δnoiseε) (4)
53- 56 F4.2 [Sun] s_[X/Fe] ? Systematic error in ε due to
uncertainty in atomic parameters for each line
(δsysε) (4)
--------------------------------------------------------------------------------
Note (1): we measured the abundance of neutral carbon from spectral synthesis
of the CH molecular G band (see Section 4.4 for more details)
Note (2): ε(X)=12+log(n(X)/n(H)), where n(X) is the photospheric
number density of element X. See section 4.3 for more details.
Note (3):
[X/Fe]=(ε(X)-ε(Fe))-(εSun(X)-εSun(Fe)).
Solar abundances have been taken from Anders & Grevesse (1989GeCoA..53..197A 1989GeCoA..53..197A),
except εSun(Fe)=7.52.
Note (4): See Section 4.3. and Equation (2) for more details.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Greg Schwarz [AAS], Sylvain Guehenneux [CDS] 20-Dec-2013