Towards a complete census and characterisation of the distant quasar population: a panchromatic view

Abstract

A supermassive black hole (SMBH) is found at the centre of most massive galaxies. These SMBHs primarily grow during a phase of efficient accretion, where they are observed as active galactic nuclei (AGN), or in the most luminous cases, as quasars. The early growth of SMBHs is believed to predominantly occur in a dust-obscured phase: identifying this obscured phase is observationally challenging, limiting our understanding of SMBH and, by connection, host galaxy growth. In this thesis, I aim to extend our current knowledge on obscured quasars by performing the largest multi-wavelength characterization of an obscuration-unbiased quasar sample to date, and leading the design of the first large-scale optical spectroscopic survey targeting the obscured infrared (IR) quasar population.

I first utilise deep multi-wavelength data in the COSMOS field to identify a complete, obscuration-unbiased sample of 578 IR quasars () at through detailed UV-to-far IR spectral energy distribution (SED) fitting. To further investigate the properties of the sample, I complement our SED constraints with X-ray and radio observations from and the Karl G. Jansky Very Large Array, respectively. Overall, I find that IR quasars are undetected in the X-ray band due to heavily obscured or even Compton thick () column densities, showing the effectiveness of the mid-IR (MIR) band in identifying obscured AGNs. I show that of IR quasars are obscured () and identify significant differences in the average properties between the obscured and unobscured quasars. Notably, obscured quasars exhibit star-formation rates (SFR) higher and produce stronger AGN-driven radio emission, with a radio-loudness parameter higher than unobscured quasars, despite similar stellar masses. These findings argue against the simple AGN orientation model since the radio and star-formation properties of IR quasars are largely independent of the torus orientation.

Motivated by the large SFR found in obscured quasars, I then utilise ALMA archival data to explore the contribution of host galaxy obscuration to the total quasar obscuration. I analyse the 24 IR quasars with ALMA photometry and submillimetre host galaxy sizes and a control sample of submillimetre galaxies (SMG) with similar SFRs but not hosting quasar activity. I find that: (1) at , the SMGs and IR quasars have remarkably compact submillimeter sizes () and, therefore, can have a host galaxy column density high enough to heavily obscure the quasar, even reaching Compton-thick levels in extreme cases, and (2) given the large host galaxy column densities estimated for the systems, host galaxy obscuration can significantly contribute to an increase in the obscured quasar fraction. Consequently, I demonstrate that a substantial part of the quasar obscuration can come from the host galaxy when the quasars are embedded in compact starbursts.


Finally, to extend the previous IR quasar studies to much larger datasets, I present the design and predictions of the 4MOST IR AGN survey, the first large-scale optical spectroscopic survey characterizing obscured MIR AGN. The 4MOST IR AGN survey will observe obscured IR AGN over an area of down to a magnitude limit of . The aim of the survey is to complement the 4MOST X-ray survey that will follow up of -selected AGN. I identify the 4MOST IR AGN sample by applying the 90 AGN reliability colour selection to the unWISE catalogue, selecting the most obscured sources with . This MIR-optical colour-selection method predominantly identifies obscured AGN, as validated by using X-ray spectral constraints and UV-to-far IR SED template modelling in four well-observed fields. I find that the AGN reliability of our obscured IR AGN selection approach ranges 80-87, where 60 of the identified AGN are likely to be IR quasars, and around 70 of them are obscured by . I finally predict that around 80 of the sample will be undetected by due to extreme absorption.


Overall, my thesis has demonstrated a clear link between AGN obscuration and star formation in galaxies. I show that there is a close connection between starburst galaxies and obscured quasars, and that the relationship between obscured and unobscured quasars is more complicated than that predicted by the standard torus orientated model. Instead, my results are more consistent with obscured and unobscured quasars being predominantly distinct phases, potentially linked in an evolutionary sequence where a significant fraction of obscured quasars represents an early growth phase, closely connected to a period of enhanced star formation. However, the limited size of my current IR quasar sample prevents us from proving an evolutionary relationship between different quasar populations, or exploring alternative scenarios. The 4MOST IR AGN survey will overcome the current statistical limitations by extending my current IR quasar research to times larger datasets, and has the potential to unveil the relationship between different quasar populations, and to ultimately reveal the role of quasars in the co-evolution of SMBHs and galaxies.