Astron. Astrophys. 351, 477-486 (1999)

1. Introduction

Blue compact galaxies (BCGs) were first observed spectroscopically by Sargent & Searle (1970), who clearly established that the properties of these galaxies implied high star formation rates at low metallicities (Doublier et al. 1997). BCGs have been thought to represent a different and extreme environment for star formation compared to the Milky Way and many other nearby galaxies. They are very important for understanding the star formation process and galactic evolution (Kinney et al. 1993, Martin 1998). BCGs are characterized by their compact morphology and very blue UBV colors (Sage et al. 1992, Hunter & Thronson 1995). Their optical spectra show strong narrow emission lines superposed on a nearly featureless continuum, similar to the spectrum of HII regions (Izotov et al. 1997, Östlin et al. 1999). Radio observations at 21-cm have shown BCGs contain large amounts of neutral hydrogen. The mean value of for a sample of 122 BCGs is 0.16 (Krüger et al. 1995, Salzer & Norton 1998). Systematic spectroscopic studies of BCGs have shown that about one-third of BCGs have broad W-R bumps, mainly at that are characteristic of late WN stars (Conti 1991, Izotov et al. 1997).

Ever since their discovery, the question has arisen whether BCGs are truly young systems where star formation is occurring for the first time, or whether they are old galaxies with current starbursts superposed on an old underlying stellar population (Garnett et al. 1997, Lipovetsky et al. 1999). Because the star formation rate in BCGs is very high, the metallicity could have reached the observed value even within a time (Fanelli et al. 1988). Hence, one interpretation for the low metallicity, high gas content and high star formation rate is that BCGs are young objects, and they are being seen at the epoch of the formation of the first generation stars (). The other interpretation is that they are old objects in which star formation occurs in short bursts with long quiescent phases in between (Krüger et al. 1995, Gondhalekar et al. 1998, Östlin et al. 1999).

The low metal abundance together with the high star formation rates and large gas masses makes BCGs most suitable to determine the element abundance (Thuan et al. 1995, 1996), the primordial helium abundance (Izotov et al. 1994) and to study the variations of one chemical element relative to another (van Zee et al. 1998). It also provides a wealth of diagnostics for the study of intrinsic physical conditions (Izotov & Thuan 1999). The results of these papers are based on direct measurements of the emission line intensities, but according to Vaceli et al. (1997), the observed emission line intensities are affected by the underlying stellar absorption. Since the stellar absorption can affect substantially some fundamental emission lines used for the derivation of reddening and other physical and chemical properties, one of the first and most critical steps in the analysis of BCG spectroscopic properties is to quantify and remove the contribution of the stellar population.

If we can resolve the stellar population of a BCG, we can know its age and star formation regime. We can then subtract the stellar absorption line from the emission-line spectrum. With the launch of HST and 10-m class telescope, we are now witnessing a new era that allows us to analyze in detail nearby objects, such as Galactic HII regions or 30 Dor in the LMC and resolve old red giants in a few distant galaxies (Grebel 1999). Such studies allow us to resolve and study individual stars in massive star clusters. However, when one studies objects at larger distances, individual stars (except for some giants) are unresolved and hence we are limited to studying their global properties (Mas-Hesse & Kunth 1999). In this paper we have selected 10 BCGs and determined their stellar population by applying a population synthesis method based on star cluster integrated spectra. In a subsequent paper, we will apply an evolution population synthesis method to these galaxies and the results of the two papers will be considered together.

The outline of the paper is as follows. In Sect. 2 we describe the observations and data reduction. In Sect. 3 we present measurements of equivalent widths and continuum analysis for the BCG spectra. In Sect. 4 we carry out the population synthesis and give the results of computation. In Sect. 5 we subtract the stellar population synthesis spectra from the observed ones and study the resulting emission line spectra. The results are presented in Sect. 6. In Sect. 7, we summarize our conclusions. Throughout this paper, we use a Hubble constant of .

© European Southern Observatory (ESO) 1999

Online publication: November 3, 1999
helpdesk.link@springer.de