Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 359, 1085-1106 (2000)

Previous Section Next Section Title Page Table of Contents

1. Introduction

Oxygen is the most abundant metal and therefore of primary interest in cosmochemical studies. Abundance determinations at extragalactical distances have been restricted to studies of HII regions (see e.g. Skillman 1998) until recently. The generation of large telescopes presently coming into operation and their dedicated spectrographs now provide the tools for quantitative spectroscopy of luminous stars at distances as far as the Virgo and Fornax clusters (Kudritzki 1998, Kudritzki et al. 1999). Primary targets for ground based observations of extragalactic stars are A- & B-type supergiants as they appear brightest in the visual.

Being young stars, these supergiants reflect the present composition of the surrounding interstellar medium, offering an independent method for deriving abundances in addition to HII region studies. At the same time they are evolved objects with CNO-cycled gas mixed into the atmosphere. Furthermore, they are the progenitors of type II supernovae, the main contributors to the enrichment of the ISM with oxygen. Abundance analyses of supergiants are therefore of importance for the verification of evolutionary scenarios of massive stars (see Venn (1995a) for results on galactic A-type supergiants) as well as for the chemical evolution of galaxies.

For oxygen, reliable abundance determinations for supergiants are scarce. Previous analyses of late B- and A-type stars (abbreviated as BA-type stars) are mainly based on either LTE calculations or on rather simple non-LTE model atoms. Pioneering work on non-LTE effects for neutral oxygen was performed by Baschek et al. (1977) and the contributions so far culminated in a comprehensive model atom published by Takeda (1992). In the meantime, improved line-blanketed model atmospheres have become available (Kurucz 1991) and the accuracy of atomic data has been enhanced enormously due to the efforts of e.g. the Opacity Project (OP; see Seaton et al. (1994) for a general overview). A critical reinvestigation seems appropriate as abundances derived from different spectral lines are inconsistent. The point of interest will be the weak lines in the visible and the strong features in the near-infrared, especially the [FORMULA] 7771-5 triplet which is a powerful luminosity criterion and most prominent in supergiants (Faraggiana et al. 1988).

At lower temperatures, in the F and G stars, these triplet lines also have a long history as controversial abundance indicators when compared to other oxygen features. Starting with the work of Johnson et al. (1974) non-LTE effects have been discussed by a number of authors, see e.g. Kiselman (1993), Cavallo et al. (1997) or Reetz (1998) for recent results. Despite many promising approaches to the topic a comprehensive theoretical interpretation of the observations is still lacking.

Kudritzki (1992) first investigated the effects of a transsonic velocity field (i.e. the stellar wind) on the formation of weak ("pseudophotospheric") metal lines. Later, Lamers & Achmad (1994) concluded that the apparent high microturbulence velocities derived for A and F supergiants are due to mass loss. The OI spectrum in A-type supergiants (with both weak and extremely strong features) offers an excellent opportunity for verifying the predictions as these stars are slow rotators as compared to those of earlier spectral type where large rotation velocities are likely to mask the mass loss effects.

In this paper we present the basics for an accurate abundance determination of neutral oxygen in BA-type stars with special emphasis on supergiants. A comprehensive model atom for OI is presented in the next section together with a critical examination of the expected uncertainties. The results from our LTE and non-LTE line formation computations are described in Sect. 3 while in the following section a test sample of high S/N and high resolution spectra of three stars is analysed for oxygen abundances. In Sect. 5 we discuss the impact of micro- and macroscopic velocity fields on the line formation. Finally, a short summary is given.

The application of the model to the determination of oxygen abundances in extragalactic supergiants (see Przybilla et al. (1999) for first results) and the implications for galactic abundance gradients will be the subject of further investigation when a sufficiently large sample of observations becomes available.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: July 13, 2000