Bayesian Source Discrimination in Radio Interferometry

TL;DR

This paper introduces BaSC, a Bayesian method that directly uses interferometric visibility data for source discrimination, outperforming traditional image-based techniques like SExtractor, especially in complex scenarios.

Contribution

The paper presents BaSC, a novel Bayesian code that processes visibility data directly, improving source discrimination accuracy over existing image-based methods.

Findings

BaSC outperforms SExtractor in discriminating nearby sources.

The Bayesian resolving formula is validated with simulated data.

Potential for resolving sources below the FWHM of the restoring beam.

Abstract

Methods currently in use for locating and characterising sources in radio interferometry maps are designed for processing images, and require interferometric maps to be preprocessed so as to resemble conventional images. We demonstrate a Bayesian code - BaSC - that takes into account the interferometric visibility data despite working with more computationally manageable image domain data products. This method is better able to discriminate nearby sources than the commonly used SExtractor, and has potential even in more complicated cases. We also demonstrate the correctness of the Bayesian resolving formula for simulated data, and its implications for source discrimination at distances below the full width half maximum of the restoring beam.