Radio and infrared surveys for active galactic nuclei behind the Magellanic Clouds

Pennock, Clara Marie

Abstract

I present an analysis of a new 120 deg2 radio continuum image of the Large Magellanic Cloud (LMC) at 888 MHz with a bandwidth of 288 MHz and beam size of 13.′′9×12.′′1, from the Australian Square Kilometre Array Pathfinder (ASKAP). I constructed a catalogue of 54,612 sources reaching down to <0.2 mJy and explore the sources by cross-matching with surveys at other wavelengths. I find sources are predominantly extragalactic, display synchrotron emission associated with AGN, and star-forming galaxies become more prominent below 3 mJy compared to AGN.
I employ machine learning to separate the stellar from the extragalactic in the Magellanic Clouds. The t-SNE algorithm is used with multi-wavelength data from Gaia EDR3, VISTA survey of the Magellanic Clouds (VMC), AllWISE and ASKAP to cluster similar radio sources together. This separates AGN, galaxies, blazars and stellar sources. The probabilistic random forest classifier is trained on known sources with data from optical to mid-IR. This yielded accuracies of 0.93 ± 0.01 (SMC) and 0.91 ± 0.01 (LMC) when tested on known sources. I classify the 31,169,627 sources in the VMC SMC field to find that classes distribute across colour-colour plots and the SMC field as expected, except for in the highest density regions where there is an over-density of AGN due to blending and photometry mismatches.
Following the discovery of SAGE0536AGN (z ∼ 0.14), with the strongest 10-μm silicate emission ever observed for an AGN, I discovered SAGE0534AGN (z ∼ 1.01), a similar AGN but with less extreme silicate emission. Both originally mistaken as evolved stars in the Magellanic Clouds. Lack of star-formation implies we are seeing the central engine of the AGN without contribution from the host galaxy. They could be a key link in galaxy evolution. I searched for more of these sources using the SMC t- SNE clusters to find they are grouped with AGN (0.13 < z < 1.23) separated from the rest, suggesting a rare class. Their host galaxies appear to be either in or transitioning into the green valley, where AGN properties, such as the torus width, X-ray luminosity, radio loudness/spectral index and Eddington ratio, appear to be tracing the transition.

Citation

Pennock, C. M. (2023). Radio and infrared surveys for active galactic nuclei behind the Magellanic Clouds. (Thesis). Keele University

• Thesis Type: Thesis
• Deposit Date: Jul 11, 2023
• Publicly Available Date: Jul 11, 2023
• Award Date: 2023-06