Tiaan Bezuidenhout

Unlike traditional single-dish radio telescopes, radio interferometers work by adding together the signals of many separate antennas. By adding delays to the signals before they are added together, we can effectively focus its view to a narrow (but sensitive) "beam" on the sky. Additionally, we can form multiple beams beams at once, each focused on a different point on the sky, and in that way create a kind of fly's-eye view across a relatively wide portion of the sky. This has already proven effective in searches for transients, i.e. sources that are impermanent or come in and out of view, aiding in the discovery of the mysterious Fast Radio Bursts (FRBs). These are extremely energetic, millisecond long flashes from outside the universe whose origins are uncertain.

One problem for transient surveys is that of localisation, which is usually done by creating an image of the field of a transient candidate. You need to store the raw data from all of the hundreds of antennas in order to create such an image, which is often prohibited by cost. However, if you detect a new transient, but can't pinpoint its position on the sky, you may not be able to re-observe it to gather more data. Here I present an alternative for localisation without having to make an image. All we need to know is how sensitive each beam is at different parts of the sky and how bright the source was in that beam. This allows us to create a map of the likelihood that the source is located at any given point on the sky.

This method, called Tied-Array Beam Localisation (TABLo) has been tested with MeerKAT detections of pulsars whose positions are known exactly, and we've shown that it's able to localise a source to within a few arcseconds (i.e. about 1/3600 the size of the moon). This is much more precise than has been possible previously without imaging.