Scott Wilkinson

Topic

Mergers, black holes, and the end of star formation: a simulation-boosted multi-wavelength approach to understanding galaxy quenching

Department of Physics and Astronomy

Date & location

• Thursday, April 16, 2026
• 8:30 A.M.
• Clearihue Building, Room B007

Examining Committee

Supervisory Committee

• Dr. Sara Ellison, Department of Physics and Astronomy, University of Victoria (Co-Supervisor)
• Dr. Toby Brown, Department of Physics and Astronomy, UVic (Co-Supervisor)
• Dr. Colin Bradley, Department of Mechanical Engineering, UVic (Outside Member)

External Examiner

• Prof. Dr. Dominika Wylezalek, Fakultät für Physik und Astronomie, Universität Heidelberg

Chair of Oral Examination

• Dr. Michael McGuire, Department of Electrical and Computer Engineering, UVic

Abstract

Over cosmic time, galaxies are transitioning from star-forming to quiescent in a process known as quenching. Star formation will necessarily halt if there is no molecular gas available to fuel star formation, or if molecular gas is present but unable to form stars. Although there are many proposed galaxy-scale events that can lead to the end of star formation in galaxies, this work focuses primarily on the role of galaxy mergers and the energetic feedback processes from active galactic nuclei (AGN) and their affect on the molecular gas in galaxies.

Surveys of galaxies observed in both optical (for observing the stellar component) and millimetre wavelengths (for observing the molecular gas component) allow for the exchange between gas and stars to be studied. In the existing surveys of galaxies observed in optical and millimetre, there is a trade-off between the resolution of the observations and the sample size of the survey. In this work, I create a dataset that strikes a balance between large but unresolved surveys and the higher resolution surveys with smaller samples. In a significant technical undertaking, I combine archival products from the Sloan Digital Sky Survey (SDSS) and the Atacama Large Millimetre Array (ALMA) to create a “semi-resolved” survey of 277 galaxies, meaning the inner and outer regions of each galaxy are independently resolved. I call this dataset the SDSS-ALMA Legacy Value Archival Gas Exploration (SALVAGE). I use SALVAGE to better understand global scaling relations, such as the star forming main sequence (SFMS). On the SFMS, I find that the global star formation rate (SFR) of a galaxy is driven largely by the star formation efficiency of the inner region. Meanwhile, above and below the SFMS, I find that the global , of a galaxy is driven more by the availability of gas in its inner region, even more so than its global gas reservoir. My results distinguish the central few kiloparsecs as the most consequential region for galaxy evolution at low redshift, likely due to the role of both mergers and AGN in driving gas inwards and either ejecting or consuming molecular gas from the inner regions.

Although theory suggests that AGN feedback can eject gas from the central region of a galaxy, previous studies of the global molecular gas reservoirs of AGN hosts have concluded that they are not systematically depleted. I use SALVAGE to select 70 AGN and test if they have depleted molecular gas in the central ∼1-2 kpc. Using several multiwavelength definitions of AGN and three metrics of gas depletion in comparison with a matched non-AGN control sample, I conclude that AGN do not have depleted gas in the central ∼2 kpc. The lack of depletion could be a result of the resolution of the data or a temporal effect where AGN-driven gas depletion is only seen after the AGN is turned off. Indeed, many galaxies in SALVAGE are found to have centrally depleted molecular gas, but they are found in both the AGN and non-AGN samples.

To explore the causes of depleted central molecular gas in galaxies, I select a sample of 28 central gas deficient galaxies (CGDs) from SALVAGE. I demonstrate that CGDs are examples of galaxies that are quenching inside-out, but that they are less likely to be AGN than matched controls. I use a cosmological simulation called IllustrisTNG50 to select an analogous sample of 70 CGDs, which allows me to “rewind the clock” of the simulation to see what caused the central gas to be removed. Tracing these galaxies back 0 to 8 Gyr in the past to the moment they became a CGD, I find that (78±5%) are associated with the onset of kinetic mode AGN feedback, which is known to expel gas from the central regions. Galaxies with centrally depleted gas reservoirs are therefore likely caused by AGN feedback in the distant past, but I identify several tests that will be needed to confirm this result.

Galaxy mergers can induce gas inflows leading to a starburst and ultimately halt star formation. Reliably identifying galaxy mergers in optical imaging can depend on the resolution and depth of the imaging as well as the method used to identify the mergers. No previous work has conducted a comprehensive assessment of the combined effect of all three of these variables on our ability to identify recent galaxy mergers. I select 424 known galaxy mergers (and 424 non-merger controls) from a cosmological simulation called IllustrisTNG100 and create synthetic optical images. I degrade each galaxy image to 36 different image qualities, in terms of their depth and resolution, spanning the range of present and forthcoming ground-based optical imaging surveys. I apply a suite of commonly-used merger identification methods and assess the completeness of recovered mergers. Even in perfect quality images, I find that most methods recover only 37-55% of mergers and I show that the completeness depends on image quality, galaxy properties, and the viewing angle of the observer. The values of completeness and false positive rate reported in this work can be used to account for the number of missed mergers in any ground-based imaging survey.

Post-starburst galaxies (PSBs) are defined as having had a recent burst of star formation that has subsequently halted. PSBs are therefore an excellent probe of rapid quenching mechanisms and have long been associated with galaxy mergers. Recent works using resolved spectroscopy have expanded the classification of PSBs into central PSBs (cPSBs), where the quenching is centrally concentrated, and ring PSBs (rPSBs) where the quenching is occurring in a ring. Previous works that use only optical imaging have led to conflicting results regarding the local mechanism causing central quenching versus outer quenching in a ring and whether one can evolve into the other. I use a new survey called the Kiloparsec Investigation of Local Objects’ Gas And Star-formation (KILOGAS) to identify 12,096 PSB regions and study their molecular gas content. By comparing the molecular gas in PSB regions to typical star forming regions, I find that the local process that inhibits star formation depends on the galactocentric radius. Furthermore, by comparing the molecular gas profiles of cPSBs and rPSBs, I find that rPSBs with ongoing central star formation are consistent with evolving into centrally quenched PSB systems.

This body of work demonstrates (1) the importance of combined optical and millimetre data for understanding galaxy evolution, (2) the latent ability of archival data for assembling large and diverse datasets, (3) the relevance of the central few kiloparsecs for the evolution of galaxies as a whole, and (4) the utility of cosmological simulations for interpreting observational results.