̽��ϵ��

Topic

Galaxy interactions and active galactic nuclei: insights from cosmological simulations

Department of Physics and Astronomy

Date & location

• Tuesday, April 14, 2026
• 8:30 A.M.
• Clearihue Building, Room B007

Examining Committee

Supervisory Committee

• Dr. Sara Ellison, Department of Physics and Astronomy, ̽��ϵ�� (Supervisor)
• Dr. David Patton, Department of Physics and Astronomy, UVic (Co-Supervisor)
• Dr. Arif Babul, Department of Physics and Astronomy, UVic (Member)
• Dr. Scott McIndoe, Department of Chemistry, UVic (Outside Member)

External Examiner

• Prof. Peter Johansson, Department of Physics, University of Helsinki

Chair of Oral Examination

• Dr. Kathy Gaul, School of Exercise Science, Physical and Health Education, UVic

Abstract

Idealized numerical simulations of galaxy mergers have long predicted that in a merger event, the angular momentum of gas is efficiently drained, leading to gaseous inflows that increase the central gas density. These inflows are expected to fuel both bursts of star formation and increased accretion onto the central supermassive black hole (SMBH), which should increase the presence of visual signatures of active galactic nuclei (AGN). Numerous observational studies find that AGN are present in excess in merging systems, compared with non-merger controls. However, the observational literature studying the contribution of galaxy mergers to the AGN population is more complicated, with some studies finding the majority of highly luminous AGN are in mergers while others find that AGN hosts are no more likely to be mergers than similar non-AGN control galaxies. While idealized simulations have provided the foundation for a predicted merger-AGN connection, they are not as well suited to investigate populations of galaxy mergers. Therefore, in order to determine the theoretically expected contribution of mergers to the AGN population, one needs to study AGN triggering within a cosmologically representative merger sample. In my dissertation, I present an analysis of galaxy mergers drawn from the IllustrisTNG cosmological simulations. Mergers identified in large box cosmological simulations preserve the ‘natural’ diversity in galaxy characteristics (such as gas fractions, star formation rates, environments) and merger characteristics (such as mass ratios, orbital parameters), and are therefore an ideal dataset to investigate the frequency of luminous AGN in mergers and their contribution to the simulated AGN population. While the strongest inflows are predicted to occur in major post-mergers, my research investigates AGN triggering in the regimes of galaxy interactions that are predicted to be weaker (than major post-mergers) but are more common in the galaxy population: the pair phase and the minor (and mini) merger regime. In my research, I find that if one only considers major post-mergers, their contribution to the overall AGN population is subdominant. However, after accounting for AGN in pairs and minor/mini mergers, I find that interactions are associated with the majority (up to 94% depending on minimum mass ratio considered) of the most luminous AGN population in the simulations. Notably, moderate luminosity AGN are still overwhelmingly associated with secular galaxies that have not had any recent interactions. To address the apparent tension between my findings and numerous observational studies which do not find a majority of AGN are in mergers, I demonstrate that the visual identification of interacting galaxies is challenging in mini and minor mergers. Concerning major mergers, I initially demonstrate that, in TNG, mergers rarely host luminous AGN at any point during the pre and post-merger phase. However, cosmological simulations only provide sparse temporal sampling of the SMBH accretion rate histories in mergers. In order to further investigate AGN in merging systems, I develop and present the Merger zOOm-in Simulation Ensemble (MOOSE), a suite of zoom-in cosmological simulations of galaxy mergers. The suite allows me to investigate AGN stochasticity with higher temporal sensitivity than cosmological simulations (an improvement from ∼160 Myr between simulation snapshots to 25 Myr). Using MOOSE, I am able to better sample the SMBH accretion rate history of mergers, and I determine that a large fraction (40%) of MOOSE major mergers will host a luminous AGN (with an Eddington ratio exceeding 10%) at some point along the merger sequence. I demonstrate that the fraction of mergers hosting an AGN is highly sensitive to the temporal snapshot spacing, and that if I down-sample MOOSE to 150+ Myr spacing (commonly used in cosmological simulations), the fraction of mergers hosting AGN drops by half. Finally, I demonstrate that these events are not localized to near coalescence, and can occur at both wide pair separations and in the late post-merger stage. Overall, my results suggest that wide pairs, late stage post-mergers, minor, and mini mergers may all contribute to a ‘hidden’ population of merger-enhanced AGN which do not show identifiable features of a recent interaction. Furthermore, my results support a prediction that mergers are the dominant contributor to the population of the most luminous AGN. Finally, my results caution that while cosmological simulations are an excellent tool to study populations of AGN or mergers, the coarse snapshot sampling will preferentially lose luminous AGN events along the merger sequence, resulting in missing AGN events when studying any single merger event over time.