ForumSpringerAstron. Astrophys.
ForumWhats NewSearchOrders

Astron. Astrophys. 326, 647-654 (1997)

Previous Section Next Section Title Page Table of Contents

1. Introduction

X-ray selection has been shown to be a powerful way of selecting pre-main sequence (PMS) late-type stars, already using Einstein observations of star-forming regions (SFR), as for example shown by Walter et al. (1994) in the Sco-Cen SFR, where they identified several previously unknown PMS stars by studying the newly detected X-ray sources. This technique has recently been extensively applied to the stellar counterparts of soft X-ray sources detected in the RASS ( Alcala et al. 1995, Alcala et al. 1996, Alcala et al. 1997, Wichmann et al. 1996, Wichmann et al. 1997, Krautter et al. 1997, hereafter collectively referred to as the "RASS-WTTS papers") yielding a large number of candidate PMS stars, which are generally referred to as "Weak-line T-Tauri Stars" (WTTS), as they do not show any of the extreme spectral characteristics (strong emission lines, and large amount of "veiling") typical of classical T-Tauri stars (CTTS). The abundance of PMS stars identified in the RASS, even far away from obvious sites of star formation, has raised several questions about the mechanisms of low-mass star formation, about the recent history of star formation in the solar neighborhood, and about the mechanisms responsible for the spatial diffusion of newborn stars from their sites of formation.

These questions have for example been addressed in detail by Feigelson (1996), who has discussed various possible models for the diffusion of young stars from their place of birth, to match the WTTS population identified around the Chameleon star forming region investigated by Alcala et al. (1995). One persistent difficulty observed by Feigelson (1996) is that the putative age of the RASS WTTS population is consistently too young. One proposed way of accounting for such a large number of very young stars far away from their plausible birthplace is to assume the presence of a large number of spatially sparse, small sites of star formation which have, by the time these stars have been observed, dissipated away. The concentration of low-mass star formation in these small sites would challenge much of our understanding of low-mass star formation. As Feigelson (1996) remarks, these explanations must be considered as tentative, as the possibility that a fraction of the RASS high-lithium stars are already on the main sequence cannot be discarded.

Brice"no et al. (1997) reach, for the RASS stars, the same conclusion which was reached earlier for the Einstein Extended Medium Sensitivity Survey (EMSS) by Micela et al. (1993), namely that the majority of low-mass stars detected in X-ray flux-limited surveys at the flux levels typical of the RASS and of the EMSS are likely to be young main sequence stars, rather than PMS stars. To support this statement, Brice"no et al. (1997) use a numerical approach, similar to the one presented by Favata et al. (1992), and applied by Micela et al. (1993) and Sciortino et al. (1995) to the data from the EMSS, similarly concluding that the majority of the low-mass stars detected in the RASS are likely to be young main-sequence stars.

Very similar conclusions are reached by Guillot et al. (1996) who have developed a similar (although fully independent) model of galactic coronal X-ray source counts. Their work includes a more detailed modeling of the scale-height evolution of the young stellar populations, and thus succeeds in better predicting the low-latitude population. This model has been shown to match well the observations of low-latitude RASS fields ( Motch et al. 1997), again showing the predominance of young main-sequence coronal sources.

The large numbers of WTTS identified in the RASS-WTTS papers is in contrast with the small number of young (from a few times [FORMULA] up to [FORMULA] yr of age) main-sequence stars in the same samples. For example, the low number of non-PMS coronal sources detected by Alcala et al. (1995) contrasts sharply with the much larger number of such sources detected in the RASS by Motch et al. (1997). The contrast is even more striking if one considers that the latter sample has a shallower limiting flux. An analysis of the RASS-WTTS papers shows that the attribution of WTTS "nature" has been done on the basis of low-resolution spectra, and that no lithium abundance determination has been used to assess the (eventual) PMS status of individual sources. Either the eventual presence of a "strong" Li I doublet (the watershed feature used to discriminate between "WTTS" and "other active stars") is quoted as determining the "WTTS nature" of a source (without any quantitative definition of "strong", as in Alcala et al. 1995) or a single mass-independent, minimum equivalent width of the Li I doublet is taken as discriminating between WTTS and other active stars (as in Wichmann et al. 1996). In all the above works, the equivalent width of the Li I doublet is measured using spectra with resolution ranging from [FORMULA] Å to [FORMULA] Å.

Experience shows that the detection, and, a fortiori, the measurement of weak spectral features in low-resolution spectra is an uncertain operation, and that the measurement of spectral features whose equivalent width is a small fraction of the spectral resolution is likely to be fraught with large systematic as well as statistical errors. To assess the reliability of a WTTS identification process based only on low-resolution spectra, we have analyzed the low- ( [FORMULA] Å) and high-resolution ( [FORMULA] Å) spectra of a number of active stars spanning a wide range of spectral types and lithium abundances. We have also studied the influence of the usage of a single watershed value for the equivalent width of the Li I feature as determining whether a source is a WTTS or not, noting the systematic biases in the resulting samples of X-ray selected WTTS.

The present paper is structured as follow: Sect.  2briefly presents the sample of stars whose spectra have been studied here and describes the data reduction, Sect.  3discusses the derivation of Li I doublet equivalent widths and compares in detail the results obtained from low- and high-resolution spectra, Sect.  4discusses the effects of using a simple threshold in equivalent width of the Li I doublet in the identification of PMS sources, and Sect.  5discusses the apparent lack of main-sequence stars in the samples from the RASS-WTTS papers. Finally, Sect.  6discusses the implications of the findings of the present paper.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: October 15, 1997