Source finding in the era of the SKA (precursors): Aegean 2.0

TL;DR

This paper discusses the challenges of source finding in large, high-resolution sky images from SKA precursors and presents enhancements to the Aegean software to address these issues, including handling spatial correlations and variable backgrounds.

Contribution

It introduces new methods in the Aegean source finding package to improve performance on large, complex datasets with spatially varying properties.

Findings

Enhanced Aegean handles larger datasets efficiently.

Addresses source finding with spatially correlated data.

Introduces priorized fitting for improved accuracy.

Abstract

In the era of the SKA precursors, telescopes are producing deeper, larger images of the sky on increasingly small time-scales. The greater size and volume of images place an increased demand on the software that we use to create catalogues, and so our source finding algorithms need to evolve accordingly. In this paper we discuss some of the logistical and technical challenges that result from the increased size and volume of images that are to be analysed, and demonstrate how the Aegean source finding package has evolved to address these challenges. In particular we address the issues of source finding on spatially correlated data, and on images in which the background, noise, and point spread function, vary across the sky. We also introduce the concept of forced or priorized fitting.