Introduction

Despite steady progress in the understanding of the radio emission from clusters of galaxies (e.g. Mills and Hoskins 1977; Owen et al. 1982; Fanti et al. 1983; Reynolds 1986; Andernach et al. 1988; Unewisse 1993; Slee et al. 1994; Ledlow and Owen 1995a, 1995b, 1996), it is still not clear what effect, if any, the cluster environment has on the likelihood of a galaxy becoming a strong radio emitter. In recent years one of the major observing programs on the Molonglo Observatory Synthesis Telescope (MOST; see Robertson, 1991, and references therein, for a description) has been a survey of 261 southern clusters from the catalogue of Abell, Corwin and Olowin (1989; ACO), with the aim of extending our current knowledge of the radio properties of clusters. The survey samples two overlapping subsets of the ACO catalogue. The first consists of all clusters with a distance class D 4, equivalent to redshifts The second consists of all clusters with a Molonglo Reference Catalogue source (MRC; Large et al. 1981) lying within a projected separation of 670 kpc from the centre ( km s Mpc), as listed by Robertson and Roach (1990).

One advantage of the Molonglo Cluster Survey over previous radio surveys arises from the wide cosec() field of view of the MOST, which covers an area out to the Abell radius (ACO) when centred on clusters at redshift . The sensitivity of the MOST (the weakest catalogued sources being 5-10 mJy) is about the same as the most recent previous surveys mentioned above. In addition, the observing frequency of 843 MHz is suitable for detection of steep spectrum sources which are most often found in clusters, and are less readily detected at higher frequencies. Another advantage is the excellent ability of the MOST to image the extended, diffuse structures (such as the rare halo and relic sources) peculiar to cluster fields.

In deep radio surveys there are far fewer sources optically identified with cluster members than there are faint, random background sources. For this reason, in some earlier surveys the likely number of cluster-related sources at a given radius from the cluster centre was estimated from the statistical excess over the background. Ideally, we would like to select only those sources associated with the clusters, in order to gain insight into the physics that governs the radio properties of cluster galaxies. Selection of cluster sources first requires reliable optical identifications. This paper presents an identification method developed and tested on a sample of 27 clusters from the MOST survey. Its major advantage is the use of relatively new tools, such as the COSMOS/UKST Southern Sky Object Catalogue and the Digitized Sky Survey, resulting in more efficient and reliable optical identifications than was possible in the past.

Section 2 describes the procedure for cataloguing the radio sources in the MOST images. Section 3 presents an automated method for making optical identifications on the basis of positional coincidence and for determining their reliability. Section 4 discusses radio-optical overlays as the only viable method for identifying extended, well resolved sources which make up a small but important minority of our source lists. In section 5 we present the results of the identification procedure, and compare them with earlier results from Unewisse (1993). Section 6 outlines a first attempt to separate the cluster and non-cluster radio galaxies, and section 7 states our conclusions.

Throughout this paper a Hubble constant of = 75 km s Mpc is assumed, giving an Abell radius of  Mpc.

© Copyright Astronomical Society of Australia 1997