Astron. Astrophys. 355, 900-914 (2000)

Astron. Astrophys. 355, 900-914 (2000)

1. Introduction

The development of a new generation of multi-object spectrographs, exemplified by the `Two degree Field', or 2dF, multi-fibre spectrograph on the Anglo-Australian Telescope (AAT), has opened up whole new areas of astronomical survey science. One particular area, which we discuss in this paper, is the opportunity to make a truly complete spectroscopic survey of a given area on the sky, down to well determined, faint limits, irrespective of image morphology or any other preselection of target type.

The Fornax Spectroscopic Survey , or FSS, seeks to exploit the huge multiplexing advantage of 2dF by surveying a region of 12 square degrees centred on the Fornax Cluster of galaxies. It will encompass both cluster galaxies, of a wide range of types and magnitudes, and background and foreground galaxies (over a similarly wide range of morphologies), as well as Galactic stars, QSOs and any unusual or rare objects.

Although many surveys of nearby clusters have been made over the past 20 years or more, these are all limited in several crucial aspects. Spectroscopic surveys exist, but typically only of the few dozen brightest cluster galaxies (and any background interlopers in the top few magnitudes of the cluster luminosity function). Photometric surveys, of course, go much deeper, but such studies must be of a statistical nature (e.g. subtracting off the expected background numbers; Smith et al. 1997), or rely on subjective judgements of likely cluster membership based on morphology, surface brightness or colour (e.g. Ferguson 1989). Of particular concern is the surface brightness; low surface brightness galaxies (LSBGs) seen towards a cluster are conventionally assumed to be members, while apparently faint, but high surface brightness galaxies (HSBGs) are presumed to be luminous objects in the background (e.g. Sandage et al. 1985). The failure of either assumption, i.e. the existence of large background LSBGs (such as the serendipitously discovered Malin 1; Bothun et al. 1987) or of a population of high surface brightness (compact) dwarfs in the cluster (Drinkwater & Gregg 1998), can have a dramatic effect on our perception of the galaxy population as a whole. Furthermore, it is possible that a population of extremely compact galaxies (either in the cluster or beyond) could masquerade as stars and hence be missed altogether from galaxy samples. Examples have previously been found in, for example, QSO surveys, but again these are serendipitous discoveries and hard to quantify (see Drinkwater et al. 1999a = Paper II, and references therein).

Few previous attempts at all-object surveys have been made. The one most similar to ours is probably that of Morton and Tritton in the early 1980s. They obtained around 600 objective prism spectra and 100 slit spectra for objects in a 0.31 square degree region of background sky (i.e. no prominent cluster) over the course of a 5 year period (Morton et al. 1985). More recently Colless et al. (1993) obtained spectra of about 100 objects in a small area of sky and small magnitude range in order to investigate the completeness of faint galaxy redshift surveys.

Our overall survey will therefore represent a huge increase in the volume of data and in addition will give a uniquely complete picture of a cluster of galaxies. It is worth noting that the huge galaxy surveys planned, with 2dF (Folkes et al. 1999; Colless 1999) or the Sloan Digital Survey (Gunn 1995; Loveday & Pier 1998) will not address such problems, since their galaxy samples will be pre-selected from photometric surveys and will only include objects classified as galaxies and not of too low surface brightness, thus removing both ends of any potentially wide range of galaxy parameters.

In the present paper we discuss the design and aims of our all-object Fornax Spectroscopic Survey (FSS) and present initial results on the velocity distributions. Sect. 2 gives a technical definition of the survey, describing the relevant features of the 2dF spectrograph, the selection of our target catalogue and the calibration of this input catalogue. In Sect. 3 we discuss the scientific aims of the survey and summarise the types and numbers of objects we expect to observe. In Sect. 4 we discuss the spectroscopic observations and observational strategy. We describe the technique we have developed to identify and classify objects automatically from the 2dF spectra and give some examples from our initial observations. Sect. 5 gives the initial redshift results and Sect. 6 summarises the survey work to date.

Online publication: March 21, 2000
helpdesk.link@springer.de