Thesis Supervisors: Paola Andreani, Carlos De Breuck and Fabrizia Guglielmetti

Abstract

This project leverages advanced machine learning algorithms to explore the ALMA archive in search of cold gas and dust within the circumgalactic medium (CGM). The CGM serves as an intermediate reservoir of matter between galaxy discs and the intergalactic medium (IGM), playing a central role in key aspects of galaxy formation and evolution, including the missing baryon problem and the regulation and quenching of star formation (the so-called feedback mechanism).

The focus is on galaxies at redshift z∼2 (the so-called Cosmic Noon), a period when massive galaxies are thought to undergo intense feedback processes and be sustained by cold streams. These dynamics are expected to foster the formation of a cold CGM through the baryon cycle, offering a unique window into galaxy evolution.

The project introduces an innovative data reduction approach, employing specialised tools such as RESOLVE, which are optimised for detecting low signal-to-noise signals in interferometric data. This initiative encompasses the development, refinement, and application of these tools to ALMA data, with the potential for broader application to datasets from other facilities in the future.