A Statistical Comparison of Line Strength Variations in Coma and Cluster Galaxies at z0.3

Introduction

Early type galaxies have long been known to exhibit strong systematic behaviour in many of their photometric properties. This behaviour manifests itself in several ways. For example, the colour-absolute magnitude relation (Visvanathan & Sandage 1977) and the fundamental plane relations (Faber & Jackson 1976; Dressler et al. 1987; Djorgovski & Davis 1987). The tight correlations among many different photometric properties of early type galaxies and yet the relatively small number of parameters required to describe their behaviour suggest a similar origin and evolutionary history for early type galaxies. Observations of distant clusters show that the colour-absolute magnitude relation is well formed at z0.5 (Ellis et al. 1997) placing even tighter constraints on the origin of the homogeneity of the photometric properties of early type galaxies.

On the other hand, it is not clear that broadband photometry can provide any unambiguous indicators of the star formation and chemical enrichment history in a galaxy without being able to fix the age or the metallicity via other means. Because of the degeneracy of age and metalllicity of broadband colours in old stellar populations (Worthey 1994), it is only under the assumption that all early type galaxies are old and stopped forming stars a long time ago that the colour-absolute magnitude relation becomes a simple sequence in metallicity.

When we turn to spectroscopy, the picture changes. We see in studies such as O'Connell (1976), Rose (1985a), Rose (1985b), Rose (1994), González (1993), and Jones (1996) that for a narrow range of photometric properties, in both single stellar populations and galaxies, there are large variations in spectral line strengths. The line strength variations relate directly to differences in the underlying stellar populations giving cause to question the single origin interpretation for early type galaxies derived from the broadband photometric relations. Hence, spectroscopy provides a way to break through the ambiguity of broadband colours and place tighter constraints on the formation and evolutionary processes of early type galaxies.

However, integrated spectroscopy of galaxies at the present epoch has an inherent ambiguity of its own because integrated galaxy light is just that, integrated, so the properties derived from such a study will always be luminosity weighted averages of the different constituent stellar populations and even multi-wavelength studies of galaxies have yet to uniquely identify all of the contributors to the integrated light. For this reason, we need to observe, directly, intermediate stages of galaxy evolution to act as signposts to direct us from the period of formation to the present day. Not only will these direct observations tell us how galaxies are changing in time, but they will act as stakes in the ground to constrain galaxy formation models, so that one day models will have to reproduce quasar absorption systems, morphological number counts of galaxies, the galaxy luminosity function, and present day observations of galaxy populations, as well as the integrated spectroscopic properties of the intermediate stages of galaxy evolution.

As a first step towards this goal, we have undertaken a spectroscopic comparison between galaxies in a nearby rich galaxy cluster and three similarly rich clusters at z0.3. In this paper we present a statistical analysis of the spectral variations in the distant cluster galaxies relative to the nearby galaxies, concentrating on what type of information can be derived from the sample with a full discussion of the implications for the galaxy stellar populations to await a future article. Section 2 describes the sources of the data, Section 3 presents the analysis, and Section 4 is the discussion of the results.

© Copyright Astronomical Society of Australia 1997