Pith. sign in

REVIEW

The Galaxy Activity, Torus, and Outflow Survey (GATOS). Black hole mass estimation using machine learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2411.18200 v1 pith:GM6HVKJP submitted 2024-11-27 astro-ph.GA

The Galaxy Activity, Torus, and Outflow Survey (GATOS). Black hole mass estimation using machine learning

classification astro-ph.GA
keywords massesgalaxiesgatoslearningmachineblackestimatehole
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
abstract

The detailed feeding and feedback mechanisms of Active Galactic Nuclei (AGN) are not yet well known. For low-luminosity and obscured AGN, as well as late-type galaxies, determining the central black hole (BH) masses is challenging. Our goal with the GATOS sample is to study circum-nuclear regions and better estimate BH masses with more precision than scaling relations offer. Using ALMA's high spatial resolution, we resolve CO(3-2) emissions within ~100 pc around the supermassive black hole (SMBH) in seven GATOS galaxies to estimate their BH masses when sufficient gas is present. We study seven bright ($L_{AGN}(14-150\mathrm{keV}) \geq 10^{42}\mathrm{erg/s}$), nearby (<28 Mpc) galaxies from the GATOS core sample. For comparison, we searched the literature for previous BH mass estimates and made additional calculations using the \mbh~ - $\sigma$ relation and the fundamental plane of BH activity. We developed a supervised machine learning method to estimate BH masses from position-velocity diagrams or first-moment maps using ALMA CO(3-2) observations. Numerical simulations with a wide range of parameters created the training, validation, and test sets. Seven galaxies provided enough gas for BH mass estimations: NGC4388, NGC5506, NGC5643, NGC6300, NGC7314, NGC7465, and NGC~7582. Our BH masses, ranging from 6.39 to 7.18 log$(M_{BH}/M_\odot)$, align with previous estimates. Additionally, our machine learning method provides robust error estimations with confidence intervals and offers greater potential than scaling relations. This work is a first step toward an automated \mbh estimation method using machine learning.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.