A Catalogue of MOST Clusters

   
The Abell-Corwin-Olowin Cluster Catalogue

One of the first systematic studies of clusters of galaxies was carried out by [Abell1958] who compiled a catalogue of 2712 rich clusters. Abell defined a cluster as an overdensity of galaxies within a fixed radius of a given centre, now called the Abell Radius, R_A, and given by

$$R_A = \frac{1.7}{z} \quad \mbox{arcmin}$$ (2.1)

where z is the redshift of the cluster. The clusters in the Abell catalogue are predominantly located in the northern hemisphere (with declination, $\delta \ge -27^\circ$) and were observed as part of the Palomar Sky Survey. Abell selected a subset of the clusters for statistical comparison, based on the criteria of richness, compactness, distance and galactic latitude.

[Abell et al.1989] later extended the catalogue to the southern sky. They also made the existing catalogue more complete, and removed some systematic errors in the magnitude scale. The resulting all-sky catalogue (the ACO catalogue) contains 4073 clusters. The ACO clusters are categorized in several different ways. The richness class, R, is a measure of the surface density of galaxies projected onto the sky within the bounds of the cluster. R ranges from R=0 (30-50 galaxies) to R=5 (>300 galaxies) in the ACO catalogue. The distance class, D, of a cluster is a slightly ambiguous grouping based on redshift. There is some overlap between groups, for example D=3 has redshift range 0.026 < z < 0.051 and D=4 has redshift range 0.041 < z < 0.072. The clusters studied in this project are a distance limited ($D \le 4$) subset of the ACO catalogue.

Criteria for Sample Selection

The cluster sample for this study is based on observations of $\sim 130$ clusters, carried out with the Molonglo Observatory Synthesis Telescope (MOST) between 1983 and 1993. A catalogue of the reduced data and images, in B1950 co-ordinates, was compiled by [Ye1993].

Before any analysis could be carried out on the clusters, the sample was limited to those clusters satisfying certain conditions. This ensured we could make valid statistical comparisons with the remaining sample and that any possible bias in the results was minimized. The following constraints were applied, resulting in the sample that was used for the rest of this project:

• Only declinations south of $-30^\circ$ were kept, to ensure that the detection of diffuse sources was not limited by the sensitivity or beamshape of MOST.
• Clusters with distance class 1 were removed so that the range of redshifts in the sample was reduced. This was to make comparisons of source densities between clusters valid.
• Several clusters were removed from the sample because the quality of the original radio image was very poor, making any source counts or diffuse source detections unreliable.
• Several clusters were removed because a very bright source was present, reducing the sensitivity to faint emission in the field and making the detection of a diffuse source impossible.

The sky distribution of the final sample of 92 clusters is shown in Figure 2.1.

  

Figure: An equiangular plot showing the distribution of sample clusters on the southern sky. The co-ordinate system is represented by radial lines (Right Ascension) and circles (Declination). The centre of the diagram is the South Pole and the band marked by dotted lines shows $\pm 10^\circ$ in galactic latitude.

The Cluster Data

The catalogue compiled by [Ye1993] includes some information about each cluster, mainly drawn directly from the ACO catalogue. We have collated these data in Tables A.1 and B.1. Since then more information has become available, for example; more accurate redshifts, X-ray data/images and optical data/images. Table C.1 gives the information obtained by our search of the literature and electronic databases. We sent batch queries to NED (NASA Extragalactic Database), ARNIE (ARchive Network InterfacE), STScI (Space Telescope Science Institute) and other large WWW databases. The remaining data were found in the literature using the ADS (Astrophysical Data System) Abstract Query service.

Radio Images

Ye's catalogue contained MOST images of the clusters in B1950 co-ordinates. The FITS files associated with these images were viewed in kview [Gooch1996], using a flux density range of -6 mJy $\le S_\nu \le$ 30 mJy so that everything from the bright sources to the background noise was visible. Each image was overlaid with a circle of radius $\frac{1}{3}R_A$. Other information, for example the optical images, was most readily available in J2000 coordinates so we regridded the radio images to J2000 coordinates.

For several clusters we also retrieved images from the NVSS (NRAO VLA Sky Survey) [Condon et al.1998]. The NVSS is a 1.4 GHz radio survey carried out by the National Radio Astronomy Observatory (NRAO) with the Very Large Array (VLA). The survey covers the sky north of $\delta = -40^\circ$ so we had overlap in the range $-30^\circ < \delta < -40^\circ$.

Optical Images

It was important to get optical images of the regions covered by the clusters so that optical identifications could be made for sources within each cluster. This is one way to determine whether a source is a cluster member or a foreground or background source.

An image centred on each cluster, was retrieved from the DSS (Digitized Sky Survey) [Lasker et al.1992]. The DSS is a project run by the Catalogs and Surveys Branch of the Space Telescope Science Institute which has digitized the photographic plates from the Palomar and UK Schmidt telescopes. The individual plates which each cover 6.5 $\times$ 6.5 degrees on the sky have been combined and are available on CDROM.

X-Ray Data and Images

The starting point in the search for X-ray data was the XBACs (X-ray Brightest Abell Clusters) catalogue [Ebeling et al.1996] which provided X-ray luminosities for 26 sources in our sample. Pointed ROSAT images for 6 of the clusters were obtained from LEDAS (LEicester Database and Archive Service.) We smoothed the images by convolving them with a 60'' Gaussian so that the image has a continuous rather than discrete appearance, making it easier to determine the overall distribution of the X-ray emission.

Recently we obtained X-ray data from the REFLEX (ROSAT-ESO Flux Limited X-ray) cluster survey [Böhringer1999]. The REFLEX survey is a complete flux-limited sample of about 450 clusters which covers the southern hemisphere ( $\delta < 2.5^\circ$). This had X-ray data for 11 of our clusters, in addition to the XBACs data. Since REFLEX is complete, any clusters in our sample without a listed X-ray luminosity are weak X-ray sources, with $f_X < 3\times 10^{-12} \mbox{ erg s}^{-1} \mbox{cm}^{-2}$.

Redshifts

The redshifts quoted in Ye's catalogue are all from the original ACO catalogue. This listing is incomplete and in some cases inaccurate due to the method[Abell et al.1989] used to calculate the redshifts. They estimated the redshifts from a set of empirically derived relationships, for example

$$\log(cz) = 0.108[m_{10}(S) - 11.5] + 3.26 \qquad\mbox{for} \quad m_{10}(S) \le 12.5$$ (2.2)

where m₁₀(S) is the magnitude of the tenth brightest cluster member. Comparisons of measured and estimated redshifts for ACO clusters show significant discrepancies [Unewisse1993], which is why the broader distance classes mentioned in Section 2.1 were developed.

Our search of the literature found measured redshift values for 43 clusters. The redshift distribution of the sample is shown in Figure 4.1.

A Web Based Catalogue

An important consideration when assembling a large database is data accessibility. Traditionally, numerical data have been presented in paper-based catalogues and there is still a tendency to publish papers containing large tables of data which are not always available electronically. There are many advantages of storing data and images electronically and most organizations now release large catalogues on CDROM or make them available on the web. Electronic data can be edited, reformatted and distributed much more efficiently than hard copies as well as having the advantage of being searchable.

Our objective was to create a comprehensive collection of data for our cluster sample, based on the following criteria:

• Easy to maintain, modify and extend as the project developed
• Automatically up to date
• Accessible to others in the department and the research community
• Possible to view subsets of the data - such as particular clusters or properties
• Possible to view and obtain images associated with the clusters
• Reduce paper usage

To best incorporate these features we developed a web catalogue which is located at
http://www.astrop.physics.usyd.edu.au/RELICS/.

The main catalogue consists of a central file containing all the cluster data. A set of checkboxes on the web page allows the user to customize the table of data, and the relevant HTML files are generated by AWK scripts. Hence, only the central data file ever needs to be updated which reduces the possibility of errors and inconsistencies.

The catalogue can be searched to find data and images for a particular cluster. Again AWK scripts are used to read the data file and generate a page with all the available data, a greyscale image and other image links. The images are available in different formats so that they can be viewed on the screen quickly (JPEG) or downloaded for high quality printing (compressed postscript).

All references for data such as redshifts are shown as links to the relevant ADS online journal article, making it a straightforward task to look for background or associated information. The complete table of data is also available as a plain text file, making it easy to copy without having to change formats.

Electronic catalogues are an essential part of research in astronomy and astrophysics. There is a huge amount of data freely available online but accessing it would be impossible without flexible databases. Although the amount of data in this project is relatively small, the techniques used to create the catalogue apply to much larger datasets. The resulting catalogue is easy to access and use, portable across computer platforms (eg: Windows, UNIX) and easy to maintain.