A radio supernova search with the MOST

Archival data from the Molonglo Observatory Synthesis Telescope (MOST) now span more than a decade (1982-93), and represent a valuable resource for studies of time-varying radio emission. The telescope's operating frequency of 843 MHz is well suited to observing RSNe, and the fixed format and resolution make it easy to compare archival data with new observations. To investigate the feasibility of a large-scale MOST search for radio supernovae, we carried out a pilot study in late 1992/93.

Our targets were nearby (heliocentric radial velocity v km/s), spiral (type Sa or later) galaxies from the Revised Shapley-Ames Catalog (RSA; Sandage and Tammann 1981) which had already been observed with MOST at least twice during the period 1982-92. The 12 galaxies which met these criteria were re-observed with MOST during 1992-93. We excluded NGC 1313, the parent galaxy of SN1978K, on the possibly dubious grounds that it was known to contain the kind of object we were looking for. Six of the 12 galaxies have had one or more recent optical supernovae.

To test for the presence of time-varying radio emission in MOST maps, we would ideally like to perform a complete subtraction of the two maps. In our pilot study, however, the pairs of maps available for comparison often had slightly different field centres which made proper subtraction difficult with the existing reduction software. As an alternative, one of us (DCW) made a list of the positions and flux densities of all sources in each pair of MOST maps being studied (to date, we have only compared the 1992/93 map with the most recent archive map). We then compared the two lists and noted any significant changes.

We estimate that we can detect the following:

• A `new' source (i.e. one which is present in the later map but not the earlier one) at a level of 10mJy or above.
• Changes in source brightness of 20% or more for a source seen in two maps (for sources of 10mJy and above).
• The disappearance (due to fading) of an `old' source which is present in an earlier map at a level of 10mJy or above, but has fallen below the detection limit in the later map.

While these numbers are probably conservative, they give us a benchmark for estimating the radio supernova rate. A further restriction is that a new source close to a strong galaxy core may be hard to recognize. In principle, therefore, sensitivity to new sources may be lower within about 1 arcmin of the nucleus.

© Copyright Astronomical Society of Australia 1997