PUMA: The Positional Update and Matching Algorithm

Abstract We present new software to cross-match low-frequency radio catalogues: the Positional Update and Matching Algorithm. The Positional Update and Matching Algorithm combines a positional Bayesian probabilistic approach with spectral matching criteria, allowing for confusing sources in the matching process. We go on to create a radio sky model using Positional Update and Matching Algorithm based on the Murchison Widefield Array Commissioning Survey, and are able to automatically cross-match ~ 98.5% of sources. Using the characteristics of this sky model, we create simple simulated mock catalogues on which to test the Positional Update and Matching Algorithm, and find that Positional Update and Matching Algorithm can reliably find the correct spectral indices of sources, along with being able to recover ionospheric offsets. Finally, we use this sky model to calibrate and remove foreground sources from simulated interferometric data, generated using OSKAR (the Oxford University visibility generator). We demonstrate that there is a substantial improvement in foreground source removal when using higher frequency and higher resolution source positions, even when correcting positions by an average of 0.3 arcmin given a synthesised beam-width of ~ 2.3 arcmin.