ArticlePDF Available

α-process Elements in the Galaxy: A Possible GAIA Contribution


α-process Elements in the Galaxy: A Possible GAIA Contribution

Abstract and Figures

The sensitivity of stellar spectra to α/Fe abundance changes is investigated with the aim to be detected photometrically and employed for the scientific goals of the GAIA mission. A grid of plane parallel, line blanketed, flux constant, LTE model atmospheres with different [α/Fe] ratios was calculated. As a first step, the modeled stellar energy fluxes for solar-type stars and giants were computed and intercompared. The spectral sensitivity to α/Fe abundance changes is noticeable and has to be taken into account when selecting photometric filters for GAIA. The Ca II H and K lines and Mg I b triplet are the most sensitive direct indicators ofα/Fe abundance changes.
Content may be subject to copyright.
Institute of Theoretical Physics and Astronomy, Goštauto 12, Vilnius 2600, Lithuania
Department of Astronomy and Space Physics, Uppsala Astronomical Observatory, Box 515,
SE-751 20, Uppsala, Sweden
Abstract. The sensitivity of stellar spectra to α/Fe abundance changes is investigated with the aim
to be detected photometrically and employed for the scientific goals of the GAIA mission. A grid
of plane parallel, line blanketed, flux constant, LTE model atmospheres with different [α/Fe] ratios
was calculated. As a first step, the modelled stellar energy fluxes for solar-type stars and giants were
computed and intercompared. The spectral sensitivity to α/Fe abundance changes is noticeable and
has to be taken into account when selecting photometric filters for GAIA. The Ca
II HandKlines
and Mg
I b triplet are the most sensitive direct indicators of α/Fe abundance changes.
1. Introduction
The evolution of a galaxy is closely related with a gradual chemical enrichment.
The enrichment and spatial distribution of the chemical elements depend on vari-
ous galactic and stellar processes. In particular, the star formation (SF) history,
time delay between SF and enrichment of the interstellar medium (ISM), metal
dependency of the nucleosynthesis, galactic gas flows, and mixing processes in
the ISM are important. Since the individual elements are produced at various sites
and on different time scales, the observed abundances are very useful in describing
galactic evolution. The main archeological tracers of the chemical evolution are
the elements produced on a short time scale (10
years) by massive stars ending as
core-collapse supernovae (SN II) and on a longer time scale (10
years) by Type Ia
(SN Ia) supernova events. SNe II contribute to the enrichment of the interstellar
medium mainly with elements produced by the capture of α-particles (α-elements)
and from the r-process, and SNe Ia predominantly produce elements belonging
to the Fe peak. Consequently, one of the basic tools to constrain the evolution
of a galaxy is the analysis of relations between the ratios of [α-element/Fe] and
Fe abundances [Fe/H] for stars born at different times and in different parts of
a galaxy. For instance, theoretical evolutionary model of the Milky Way galaxy
recently proposed by Chiappini et al. (2001) predicts a slight decrease with distance
in the average [α/Fe] ratios in stars born in the Galactocentric distance range of
4–10 kpc and an increase with distance of this ratio in the range of 10–18 kpc.
Astrophysics and Space Science 280: 143–150, 2002.
© 2002 Kluwer Academic Publishers. Printed in the Netherlands.
A rst glance at the variation of α-elements shows that most of the metal-
poor stars in the Galaxy appear to have been formed with enhanced abundances
of oxygen and other α-elements (i.e., Ne, Mg, Si, S, Ar, Ca and Ti). For stars
with [Fe/H] ≤−1, the mean value of [α/Fe] lies between +0.3and+0.4, with
no discernible dependence on metallicity (cf. Pagel and Tautvaišien
e, 1995; Sam-
land, 1998). A more precise analysis shows that there is a significant population
of eld stars with [α/Fe] 0.0 (see Figure 4 by Nissen and Shuster, 1997). Sur-
veys by Shuster et al. (1993) and Carney et al. (1996) report evidence for about
0.1 0.2 dex variations in [α/Fe] at fixed [Fe/H]. Carney et al. (1997) have found
that the high-velocity subgiant with the apogalacticon over 20 kpc BD +80
has < [α/Fe] >=−0.29 ± 0.02 despite to its low metallicity, [Fe/H] =−1.86. At
the same time there are metal deficient giants with [α/Fe] of +0.7 dex as reported by
Giridhar et al. (2001). It is important to map the abundance pattern of α-elements
in the Galaxy and understand the origin of variations.
A central element of the GAIA mission is the determination of the star forma-
tion histories, as described by the temporal evolution of the star formation rate, and
the cumulative numbers of stars formed in the bulge, inner disk, Solar neighbour-
hood, outer disk and halo of our Galaxy (Perryman et al., 2001). Such information,
together with the kinematic information from GAIA, and complementary chemical
abundance information, again primarily from GAIA, may give us the full evolu-
tionary history of the Galaxy. Information on α/Fe abundance ratios in stars is very
important for their age determination (cf. VandenBerg et al., 2000; Salasnich et al.,
2000). In our study, a first attempt is made to investigate the sensitivity of stellar
spectra to α/Fe abundance variations and their detectability by GAIA photometry.
2. Model Atmospheres with Enhanced α/Fe Ratios
A grid of plane parallel, line blanketed, flux constant, LTE model atmospheres with
enhanced [α/Fe] ratios was calculated with the updated version of the MARCS
code (Gustafsson et al., 1975) using continuous opacities from Asplund et al.
(1997) and including UV line blanketing as described by Edvardsson et al. (1993).
The grid contains model atmospheres with effective temperatures 4500 T
6500 K, 1.5 log g 4.5, 2 ≤[Fe/H]≤0and0.0 [α/Fe] 0.4.
3. Sensitivity of Stellar Spectra to α/Fe Changes
In order to see what spectral regions are most sensitive to α/Fe changes, as a first
step, we computed stellar surface uxes with a wavelength sampling of R 20 000
for solar type stars and giants and investigated their ratios. For instance, Figure 1a
shows the smoothed ratio of the flux distributions with T
= 5500 K, log g = 4.0,
[Fe/H] =−0.4 and two different [α/Fe] ratios: 0.3 and 0.0 dex. The approximate
Mg I
Flux ratio
Flux ratio
Flux ratio
3000 4000 5000 6000 7000 8000 9000
Flux ratio
Figure 1. Sensitivities of the modelled surface energy flux ratios to the variations in fundamental
parameters and chemical abundances. The standard model has T
= 5500 K, log g = 4.0,
[Fe/H] =−0.4and[α/Fe] = 0.0. See Section 3 for more explanations.
Mg I
Flux ratio
Flux ratio
Flux ratio
Ca I
3000 4000 5000 6000 7000 8000 9000
Flux ratio
Figure 2. Sensitivities of the modelled surface energy flux ratios to the variations in fundamental
parameters and chemical abundances. The standard model has T
= 4500 K, log g = 3.0,
[Fe/H] =−0.4and[α/Fe]= 0.0. Notice that the vertical scaling differs from that in Figure 1.
Figure 3. The ratio of modelled surface energy uxes with [α/Fe]=+0.3forT
= 5500 K,
log g = 4.0and[Fe/H]=−0.4, plotted together with the indication of filter positions of the three
photometric systems proposed for GAIA. See Section 3 for more explanations.
difference of 0.3 dex was found between thin disk and thick disk stars in the solar
vicinity (cf. Fuhrmann, 1998; Tautvaišien
e et al., 2001). The major wavelength
features directly sensitive to α/Fe in dwarf stars are Ca
II H and K lines and the IR
triplet, OH bands around 3100 Å and also Mg I b triplet region. The latter region is
also affected by MgH molecular bands in cool stars. These features are indicated
below the curve in panel a. There is also an indirect effect of the α-element abund-
ances which is seen on the carbon molecular bands: the higher oxygen abundance
binds more free carbon into CO, which weakens other carbon molecular bands.
The strongest examples are indicated above the curve in panel a. It is obvious that
these secondary effects are quite dramatic and should be taken into account while
using carbon features for photometry.
In Figure 1b, the effect of decreasing the surface gravity by 1.0 dex is shown.
The effects are qualitatively similar to those of the α-element abundance increase in
several wavelength regions. A clear difference starts only bluewards of 3800 Å. The
strong gravity effect on the Ca
II lines makes their use for α-element determinations
quite dependent on a precise log g determination. It is interesting to notice the op-
posite gravity effects on Ca
II and Mg I features. The former are radiation damped
(thus not pressure-sensitive) and strengthened by the decreasing H
opacity and increasing degree of ionization, while the latter are pressure-broadened
and thus weakened both by the weaker gas pressure and higher degree of ionization.
The sensitivity of the spectrum to changes in metallicity and effective temperature
are shown in Figures 1c and 1d, respectively. The prominent feature of CN near
3850 Å is dependent on very many parameters, including the nitrogen abundance,
which is known to vary during stellar evolution.
Figure 2 shows sensitivities of the modelled surface flux ratios to the variations
in fundamental parameters and chemical abundances for a giant star with the stand-
ard model of T
= 4500 K, log g = 3.0, [Fe/H] =−0.4and[α/Fe] = 0.0. As it
is seen from the vertical scales, in giants the effects of [α/Fe] abundance changes
and other parameter variations are considerably larger than for the dwarfs.
In Figure 3 we display the three medium-band photometric systems proposed
for GAIA: 2A by Munari (1998), 3G by Høg et al. (2000), and 1F by Grenon et
al. (1999, the wide filters F 33, F 57 and F 67 are not displayed) along with the
modelled ratio of surface energy uxes with [α/Fe]=+0.3forT
= 5500 K,
log g = 4.0and[Fe/H]=−0.4. Figures 1–3 show that the lter centered at 3450 Å
in the 3G system is very useful in determining surface gravities, the lter centered
at 3750 Å may evaluate the CN feature, while the filter centered at 4050 Å – the H
and K lines. However, the lter at 5150 Å includes both C
,MgHandMgI b lines
which may cause a confusion. Here the lter of the 2A system centered exactly
on the Mg
I b triplet seems to be better. Filters of the 1F system are quite broad
and may cause difficulties in accounting for α-element variations. One filter in the
GAIA system could be set on the Ca
II IR triplet as well.
The first qualitative investigation of the sensitivity of stellar spectra to α/Fe
abundance variations, presented in this study, indicates that the spectral sensitivity
to α/Fe abundance changes are noticeable. It has both direct and indirect influ-
ence to stellar spectra. A possibility to employ the Ca
II H and K lines and Mg I b
triplet might be considered for the photometric determination of α/Fe abundance
ratios with GAIA photometry. The examples for the dwarf and giant stars with
[α/Fe] = 0.3, displayed in Figures 1 and 2, show that filters of about 80–100 Å
width centered on these features could be used. In the interval of the spectrum
3905–4005 Å with the Ca
II lines, the intensity of the spectrum drops down by
0.08 mag in the dwarf and by 0.11 mag in the giant. In the interval 5160–5240 Å
with the Mg
I b triplet, the intensity of the spectrum drops by 0.02 mag and 0.05 mag,
respectively. Assuming the mission-end photometric GAIA accuracy and three
slots of filters (ESA, 2000), from the Ca
II lines, the accuracy of [α/Fe] 0.1
might be preserved for the giants (with parameters under consideration) down to
about 17.0 mag and for the dwarfs down to about 16.3 mag. From the Mg
I b triplet,
the same accuracy of ±0.1 dex might be preserved for the giants down to about
17.3 mag and for the dwarfs down to about 15.6 mag. In case the accuracy of
±0.2 dex is also acceptable, the stars of 17.9 and 16.7 mag could be investigated
using the Ca
II lines and the stars of 18.2 and 16.5 mag using the Mg I b triplet,
The work presented in this paper marks the beginning of a large work to be
done in preparations for the photometric investigation of the α-elements in the
Galaxy. Under the assumption of known effective temperature, metallicity and sur-
face gravity, the carbon, nitrogen and α-element abundances might be determined
by means of photometric indices as well.
4. Conclusions
The spectral ux sensitivity to α/Fe abundance changes is noticeable and has to
be taken into account in GAIA photometry. The Ca
II H and K lines and Mg I b
triplet are most sensitive direct indicators of α/Fe abundance changes and might be
used for the photometric determination of α-element abundances. The photomet-
ric systems proposed for GAIA have to be carefully tested for accounting of the
α-element abundance determination.
Photometric classification of stars should provide as many physical parameters
as possible. Depending on the accuracy with which the fundamental parameters
are known, we should seek to determine abundances not only of α-elements but of
carbon and nitrogen as well.
We thank Vytautas Straižys, Michel Grenon and Vladas Vansevi
cius for help-
ful discussion. G.T. acknowledges support from the Nordic Research Academy
(REF NB00-N030) and NATO Linkage grant CRG.LG 972172. B.E. acknowledges
support by the Swedish Natural Sciences Research Council (NFR).
Asplund, M., Gustafsson, B., Kiselman, D. and Eriksson, K.: 1997, Astron. Astrophys. 318, 521.
Carney, B.W., Laird, J.B., Lathman, D.W. and Aguilar, L.A.: 1996, Astron. J. 112, 668.
Carney, B.W., Wright, J.S., Sneden, C., Laird, J.B., Aguilar, L.A. and Lathman, D.W.: 1997, Astron.
J. 114, 363.
Chiappini, C., Matteuci, F. and Romano, D.: 2001, Astrophys. J. 554, 1044.
Edvardsson, B., Andersen, J., Gustafsson, B., Lambert, D.L., Nissen, P.E. and Tomkin, J.: 1993,
Astron. Astrophys. 275, 101.
ESA: 2000, GAIA: Composition, formation and evolution of the galaxy, Technical Report,ESA-
Fuhrmann, K.: 1997, Astron. Astrophys. 338, 161.
Giridhar, S., Lambert, D.L., Gonzalez, G. and Pandey, G.: 2001, Publ. Astron. Soc. Pacific 113, 519.
Grenon, M., Jordi, C., Figueras, F. and Torra, J.: 1999, Tech. Rep. to the GAIA Photom. Working
Group, MG-PWG-002.
Gustafsson, B., Bell, R.A., Eriksson, K. and Nordlund, Å.: 1975, Astron. Astrophys. 42, 407.
Høg, E., Straižys, V. and Vansevi
cius, V.: 2000, Tech. Rep. from Copenhagen University Observatory,
Munari, U.: 1998, Baltic Astronomy 8, 123.
Nissen, P.E. and Shuster, W.J.: 1997, Astron. Astrophys. 326, 751.
Pagel, B.E.J. and Tautvaišien
e, G.: 1995, Mon. Not. R. Astron. Soc. 276, 505.
Perryman, M.A.C., de Boer, K.S., Gilmore, G. et al.: 2001, Astron. Astrophys. 369, 339.
Salasnich, B., Girardi, L., Weiss, A. and Chiosi, C.: 2000, Astron. Astrophys. 361, 1023.
Samland, M.: 1998, Astrophys. J. 496, 155.
Shuster, W.J., Parrao, L. and Contreras Marti
nez, M.E.: 1993, Astron. Astrophys. Suppl. Ser. 97, 951.
e, G., Edvardsson, B., Tuominen, I. and Ilyin, I.: 2001, Astron. Astrophys., in press, (astro-
VandenBerg, D.A., Swenson, F.J. and Alexander, D.R.: 2000, Astrophys. J. 532, 430.

Supplementary resource (1)

Full-text available
Stars provide most of the light we detect in the universe and enrich our knowledge of chemical evolution of galaxies. In order to have the real view we should know to which extent stars reflect the material from which they were formed and to which they expose effects of their internal evolution. Crucial chemical elements in this sense are carbon and nitrogen, an alteration of which in evolved stars is already a recognized fact. Abundances of oxygen and α-process elements also differ in galactic populations of stars to about a factor of three. In this contribution, on the basis of new stellar model atmospheres, we discuss the sensitivity of stellar spectra to C, N, O and α-process element abundances in order to take this effect into account in photometric observations at the Gaia orbiting observatory. The synthetic spectra of these evolved stars show an enhancement of the NH band at 3300-3500 Å by about 10% and several weakened CH and C2 features. Such spectral changes can be observed photometrically and should be taken into account in the photometric classification of giants. Carbon features in stellar atmospheres show a particularly complex behaviour being very dependent on mixing processes in stars, and on the abundance of oxygen which can also be altered by different reasons. NH bands are more independent indicators of mixing processes in stars. Abundances of α-process elements can be evaluated photometrically by using the direct indicators - Ca II H and K lines and Mg I b triplet.
The determination of the stellar parameters is of crucial importance for many fields of astrophysics, from the comprehension of the structure of the stars and their evolution, up to the determination of distances and the cosmological scale. This contribution reviews the main observational techniques and methods of calibration for the determination of the mass, radius, temperature, chemical composition and luminosity of stars.
Full-text available
The sensitivity of stellar spectra to C, N, O and α -process element abundances is discussed with the aim of taking this effect into account when selecting a photometric system for the Gaia orbiting observatory. On the basis of a spectrometric, photometric and theoretical study of spectra of evolved first-ascent giants and clump stars in the open cluster NGC 7789 it is demonstrated that evolutionary alterations of carbon and nitrogen abundances can cause noticeable spectral changes and, if not taken into account, may yield misleading photometric [Fe/H] determinations. Carbon features in stellar atmospheres show a particularly complex behavior being dependent on mixing processes in stars, on the stellar surface gravity and on the abundance of oxygen which can also be altered by different reasons. NH bands could serve for the evaluation of mixing processes in stars and the interpretation of carbon dominated spectral regions. Abundances of α -process elements can be evaluated photometrically by using the direct indicators -- Ca II H and K lines and Mg I b triplet.
If unsupported by appropriate diagnostic data, Gaia would yield immense numbers of positions and velocities of objects whose astrophysical nature would be unknown. The core science case requires measurement of luminosity, effective temperature, mass, age and chemical composition for the stellar populations in the Galaxy and its nearest neighbours. These quantities can be derived from the spectral energy distribution of the stars, i.e. through high-quality multi-band photometry. The current proposals for broad and medium band photometric systems, their capabilities for the astrophysical parametrization of the stars and hence the expected precision of the stellar populations characterization are reviewed.
Full-text available
We present a new collection of synthetic spectra finalized at the derivation of information about the enhancement of α-process elements in the atmospheres of F-GK stars. The spectra are computed at high resolution from fully consistent models and Opacity Distribution Functions (ODFs) for [α/Fe]= +0.0 and +0.4. Comparisons with observed spectra at R ≈ 2000 from the STELIB library are illustrated. Synthetic photometry in the Gaia photometric system, once the filter characteristics will be definitively assessed, could be computed and analyzed for disentangling α-enhanced stars from solar scaled chemical abundance ones.
Recent observational studies of the age-metallicity and [α/Fe] - [Fe/H] relations for stars in the solar neighbourhood are reviewed. It is still debated if there exists a well defined trend of [Fe/H] with age or if [Fe/H] has a substantial scatter at a given age. Systematic differences in [α/Fe] between stellar populations have recently been discovered, but a detailed mapping of how [α/Fe] varies as a function of stellar position and kinematics remains to be carried out. Gaia has the potential to give an important contribution to the solution of these problems, and hence to contribute very significantly to the study of Galactic chemical evolution and galaxy formation in general. The requirement to the accuracy of the determination of parameters and abundances of late-type stars is, however, quite high, i.e. σ(Teff ) 100 K, σ(MV) 0.2 mag, and σ[Fe/H] σ[α/Fe] 0.1 dex.
The European Gaia astrometry mission is due for launch in 2011. Gaia will rely on the proven principles of the ESA Hipparcos mission to create an all-sky survey of about one billion stars throughout our Galaxy and beyond, by observing all objects down to 20 mag. Through its massive measurement of stellar distances, motions and multicolour photometry, it will provide fundamental data necessary for unravelling the structure, formation and evolution of the Galaxy. This paper presents the design and performance of the broad- and medium-band set of photometric filters adopted as the baseline for Gaia. The 19 selected passbands (extending from the UV to the far-red), the criteria and the methodology on which this choice has been based are discussed in detail. We analyse the photometric capabilities for characterizing the luminosity, temperature, gravity and chemical composition of stars. We also discuss the automatic determination of these physical parameters for the large number of observations involved, for objects located throughout the entire Hertzsprung-Russell diagram. Finally, the capability of the photometric system (PS) to deal with the main Gaia science case is outlined.
Full-text available
We model the evolution of abundances of oxygen, iron, α and r-process elements in the Galactic disc using the instantaneous and delayed production approximations which enable the problem to be handled analytically. In addition to using the new greatly improved data published by Edvardsson et al. and the oxygen abundance distribution function of G dwarfs in the solar neighbourhood as improved by Sommer-Larsen, we take into account new survey data on metal-deficient stars which show that the thick-disc population extends to arbitrarily low metallicities, meaning that a significant proportion of stars previously classified as belonging to the halo, on grounds of low metallicity, actually belong to the disc and should be modelled as such, without bringing in any prior phase associated with the halo. Comparison with theoretical yields from Type II and Type Ia supernovae shows good agreement, as has previously been found by Tsujimoto et al. using a different approach.
Full-text available
We present the chemical compositions of four K giants CS 22877-1, CS 22166-16, CS22169-35 and BS 16085 - 0050 that have [Fe/H] in the range -2.4 to -3.1. Metal-poor stars with [Fe/H] < -2.5 are known to exhibit considerable star - to - star variations of many elements. This quartet confirms this conclusion. CS 22877-1 and CS 22166-16 are carbon-rich. There is significant spread for [$\alpha$/Fe] within our sample where [$\alpha$/Fe] is computed from the mean of the [Mg/Fe], and [Ca/Fe] ratios. BS 16085 - 0050 is remarkably $\alpha$ enriched with a mean [$\alpha$/Fe] of $+$0.7 but CS 22169-35 is $\alpha$-poor. The aluminium abundance also shows a significant variation over the sample. A parallel and unsuccessful search among high-velocity late-type stars for metal-poor stars is described. Comment: 14 pages (text), 4 tables, 3 figures, Accepted for publication in PASP
Full-text available
In this paper we adopt a chemical evolution model, which is an improved version of the Chiappini, Matteucci and Gratton (1997) model, assuming two main accretion episodes for the formation of the Galaxy. The present model takes into account in more detail than previously the halo density distribution and explores the effects of a threshold density in the star formation process, during both the halo and disk phases. In the comparison between model predictions and available data, we have focused our attention on abundance gradients as well as gas, stellar and star formation rate distributions along the disk. We suggest that the mechanism for the formation of the halo leaves detectable imprints on the chemical properties of the outer regions of the disk, whereas the evolution of the halo and the inner disk are almost completely disentangled. This is due to the fact that the halo and disk densities are comparable at large Galactocentric distances and therefore the gas lost from the halo can substantially contribute to building up the outer disk. We also show that the existence of a threshold density for the star formation rate, both in the halo and disk phase, is necessary to reproduce the majority of observational data in the solar vicinity and in the whole disk. Moreover, we predict that the abundance gradients along the Galactic disk must have increased with time and that the average [alpha/Fe] ratio in stars (halo plus disk) slightly decrease going from 4 to 10 Kpcs from the Galactic center. We also show that the same ratios increase substantially towards the outermost disk regions and the expected scatter in the stellar ages decreases, because the outermost regions are dominated by halo stars. Comment: 41 pages (including the figures), To be published in ApJ
Full-text available
The GAIA astrometric mission has recently been approved as one of the next two `cornerstones' of ESA's science programme, with a launch date target of not later than mid-2012. GAIA will provide positional and radial velocity measurements with the accuracies needed to produce a stereoscopic and kinematic census of about one billion stars throughout our Galaxy (and into the Local Group), amounting to about 1 per cent of the Galactic stellar population. GAIA's main scientific goal is to clarify the origin and history of our Galaxy, from a quantitative census of the stellar populations. It will advance questions such as when the stars in our Galaxy formed, when and how it was assembled, and its distribution of dark matter. The survey aims for completeness to V=20 mag, with accuracies of about 10 microarcsec at 15 mag. Combined with astrophysical information for each star, provided by on-board multi-colour photometry and (limited) spectroscopy, these data will have the precision necessary to quantify the early formation, and subsequent dynamical, chemical and star formation evolution of our Galaxy. Additional products include detection and orbital classification of tens of thousands of extra-Solar planetary systems, and a comprehensive survey of some 10^5-10^6 minor bodies in our Solar System, through galaxies in the nearby Universe, to some 500,000 distant quasars. It will provide a number of stringent new tests of general relativity and cosmology. The complete satellite system was evaluated as part of a detailed technology study, including a detailed payload design, corresponding accuracy assesments, and results from a prototype data reduction development.
We report [Fe/H] results for a sample of five high-velocity metal-poor halo population stars and five comparably metal-poor stars thought to have kinematics typical of an older, hotter disk population. We also derive abundances of lithium, oxygen, magnesium, silicon, calcium, titanium, and barium for many of the stars. Four of the candidate disk population stars are found to be subgiants, and a re-evaluation of their kinematics show that they have kinematics typical of the halo population. One star, G190-15, is confirmed to be a dwarf with disk-like kinematics: U = -10, V = -80, W = -90 \kms. It is indistinguishable in [X/Fe] from other metal-poor stars with higher velocities, however. The most interesting star is the high-velocity subgiant BD+80° 245, which is found to have <[``alpha "/Fe]> = -0.29+/-0.02 despite its low metallicity, [Fe/H] = -1.86. It is also extremely deficient in barium, [Ba/Fe] ~ -1.8, and has a large apogalacticon distance, over 20 kpc. It has experienced a very different chemical history than have other metal-poor stars found in the solar neighborhood.
Based on our expanded sample of metallicities and kinematics for a large sample of stars selected from the Lowell Proper Motion Catalog, we study several questions relating to the halo stellar population(s) in our Galaxy. For [m/H]≤-1.4, there does not seem to be any variation with [m/H] in the mean values of the V velocity (i.e., angular momentum related to that in the disk) or the Galactic orbital eccentricities. Further, in spite of the strong kinematical biases in our sample, stars with very low metallicities are found that have small V velocities (high orbital angular momenta) and low orbital eccentricities. These results contradict the model that the metal-poor stars are a single population that is only the relic of the earliest stages of the Galaxy's collapse. There are signs that some of the metal-poor stars in the solar neighborhood are due to accretion events and, perhaps, also to the earliest stages of the formation of the Galactic disk. Regarding accretion, we confirm Majewski's [ApJS, 78, 87 (1992)] finding of a retrograde rotation among stars that reach S kpc or more from the plane. These stars do not show any radial metallicity gradient, and may be younger on average than dynamically hot, metal-poor stars closer to the plane. These latter stars show net prograde rotation and a radial metallicity gradient, suggestive of a dissipative process in the earliest stages of disk formation. The correlation between metallicity and perigalacticon found by Ryan & Norris [AJ, 101, 1835 (1991a)] disappears when care is taken to exclude the stars that may have been accreted by our Galaxy. The field star results complement those for globular clusters found by other workers, notably Zinn (1993), who argued for two populations of metal-poor clusters, one apparently in retrograde rotation with no radial metallicity gradient and slightly younger ages, and the other with prograde rotation, a weak radial metallicity gradient, and slightly older ages. The field stars and globular clusters do differ slightly, however. Their metallicity distributions differ, with the field stars showing a larger fraction of the most metal-poor stars. This could be caused by accretion of Draco dwarf galaxy-like objects, with very low metallicities and no globular clusters. We see in our data, particularly in the V vs >Rapo< plane, possible signs of large-scale kinematic substructure suggestive of specific accretion events. We also see signs for the Preston et al. [AJ, 108, 538 (1994)] low-metallicity, intermediate kinematics, and younger age stellar population. However, the strength of the signal in our data suggests that a fairly large fraction of its stars may be old. On the other hand, the "away" versus "toward" mystery of Croswell et al. [Al, 93, 1445 (1987)] has disappeared: the numbers of stars approaching and receding from the plane agree with expectations. Finally, we point out that the model of Norris [ApJ, 431, 645 (1994)] for a proto-disk population that is hotter dynamically than the accreted halo components does not agree with our expanded data sample. We suggest that the proto-disk component was dynamically cooler when the mean metallicity was very low.
Formation and Evolution of the Galaxy, Consept and Technology Study Report
  • Esa Gaia Composition
ESA GAIA: Composition, Formation and Evolution of the Galaxy, Consept and Technology Study Report. ESA-SCI(2000)4
  • B W Carney
  • J S Wright
  • C Sneden
  • J B Laird
  • L A Aguilar
  • D W Lathman
Carney B.W., J.S. Wright, C. Sneden, J.B. Laird, L.A. Aguilar and D.W. Lathman. Astron. J., 114:363, 1997.
  • C Chiappini
  • F Matteuci
  • D Romano
Chiappini, C., F. Matteuci and D. Romano. Astrophys. J., 554:1044, 2001.
  • M Samland
Samland M. Astrophys. J., 496:155, 1998.