Files and links (4)
CC BY V4.0, Open Access
Dissertation
SEARCHING FOR ACTIVE GALACTIC NUCLEI IN LOW-MASS GALAXIES
Washington State University
Doctor of Philosophy (PhD), Washington State University
05/2025
DOI:
https://doi.org/10.7273/000007424
Abstract
It is now understood that supermassive black holes (SMBHs, log MBH /M⊙ ≳ 5) occupy the centers of all massive galaxies (log M∗/M⊙ ≳ 10). Despite their ubiquitous nature, much about these extreme systems remains a mystery. The origin of these SMBHs is still unknown, though astronomers have narrowed their formation pathways to a few potential seeding mechanisms. These models predict observable differences in the resulting population of BHs, particularly in low-mass galaxies. Additionally, constraining these models involves finding and studying the still elusive population of intermediate mass BHs (2 ≲ log MBH /M⊙ ≲ 6).
Observations of these low-mass systems remain scarce, however; the sphere of influence of a 105 M⊙ BH would only be large enough to be resolvable within the Local Group. Astronomers have then learned to detect indirect signatures of BH activity, namely through interactions with their environment. Among these are active galactic nuclei, complex structures of gas and other material accreting onto a central massive BH. Because of the extreme astrophysical processes at play, these objects radiate a tremendous amount of energy throughout the entire electromagnetic spectrum. Not only do their signals contain a myriad of unique AGN signatures, but their light has been shown to vary stochastically across time. This photometric variability can be modeled mathematically, providing a prolific tool for finding AGNs that is less likely to overlook low-mass and low-metallicity AGNs.
With the advent of large time-domain surveys, recent works have been able to search for AGN activity in tens of thousands of galaxies, identifying hundreds of AGNs in dwarf galaxies. In anticipation of the next generation of surveys, this work seeks to understand the resulting population of low-mass AGN candidates and examine alternative methods to find them. We assess the results of various AGN selection methods on a small group of variability-selected AGNs, finding X-ray emission is less than half of the sample. We present the results a large scale variability search, finding over one thousand new AGN candidates in dwarf galaxies. Despite this large number of new candidates whose masses are consistent with previous relations, emission line diagnostics and broad Hα emission only detect BH activity in a small fraction of our resulting sample. Finally, we explore a variety of machine learning algorithms with a particular emphasis on the low-mass regime. Gradient Boosted Trees and a Neural Network are found to be the most capable of finding AGNs in the low-mass range without sacrificing generalizability for massive galaxies.
Metrics
1 File views/ downloads
8 Record Views
Details
- Title
- SEARCHING FOR ACTIVE GALACTIC NUCLEI IN LOW-MASS GALAXIES
- Creators
- Alexander Scott Messick
- Contributors
- Vivienne Baldassare (Chair)Matthew Duez (Committee Member)Guy Worthey (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Department of Physics and Astronomy
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Publisher
- Washington State University
- Number of pages
- 182
- Identifiers
- 99901221151801842
- Language
- English
- Resource Type
- Dissertation