Measuring the Gas-phase Metallicity of AGN-host galaxies with Bayesian Methods

Abstract

Chemical abundances in the interstellar medium record the history of galaxy assembly, and thus investigating abundances in galaxies hosting active galactic nuclei (AGN) is crucial for understanding the role of AGN regulation in the evolution of their host galaxies. In this thesis, we use Bayesian methods to measure gas-phase metallicity in both local AGN and non-AGN hosts with a consistent method that makes it possible to compare both their mean metallicities and their metallicity fluctuation fields.

We find that at fixed stellar mass, AGN-host galaxies generally have higher metallicity and smaller metallicity correlation length than non-AGN galaxies. These differences are primarily because AGN hosts have systematically lower star formation rates (SFR) at fixed stellar mass. Our results show that in the local universe, AGN only indirectly influence both the overall mean metallicity and the metallicity fluctuation fields by suppressing star formation activity, and are consistent with a simple model whereby the main effect of an AGN is to reduce accretion of fresh gas onto galaxies, thereby both suppressing the SFR and inhibiting the dilution of metal concentrations by accreting metal-poor gas. In addition, we find that while both SFR and stellar mass are positively correlated with metallicity correlation length, the former is more fundamental, implying that fluctuations in the metallicity distribution within galaxies are driven more by short-term responses to physical processes such as star formation that can change much faster than a Hubble time.

We prepare for future observations of AGN hosts at intermediate redshift (z ~ 1-3), where AGN activity is more prevalent and stronger, by evaluating a range of strategies to derive metallicites from the data that will come from the next generation of integral field spectroscopy instruments. The strategies we use to measure metallicity in AGN hosts in the Local Universe will be challenging to apply to these data because they require access to emission lines over a wide wavelength range that will be expensive or impossible to obtain. We show that there are reduced line suites accessible over a narrower wavelength range that will make reasonably good metallicity measurements possible on these instrument over certain intermediate redshift windows.

Finally, we propose future works, including improving our Bayesian methods for deriving metalilcities by incorporating additional sources of ionizing radiation, investigating spatially resolved scaling relations, situating our results for the relationship between SFR, metallicity, and AGN activity in the context of galaxy evolution models, and designing strategies for making metallicity fluctuation maps for future IFS observations. Show more