Abstract

Abstract We use bulge-type classifications of 809 representative Sloan Digital Sky Survey galaxies by Gadotti to classify a large sample of galaxies into real bulges (classical or elliptical) and pseudobulges using Random Forest. We use structural and stellar population predictors that can easily be measured without image decomposition. Multiple parameters—such as the central-mass density with 1 kpc, concentration index, Sérsic index, and velocity dispersion—result in accurate bulge classifications when combined together. We classify ∼44,500 face-on galaxies above stellar mass of 1010 M ⊙ and redshift 0.02 < z < 0.07 into real bulges or pseudobulges with 93% ± 2% accuracy. We show that ∼75%–90% of active galactic nuclei (AGNs) identified by the optical line ratio diagnostic are hosted by real bulges. The pseudobulge fraction significantly decreases with AGN signature as the line ratios change from indicating pure star formation (∼54% ± 4%), to composite of star formation and AGN (∼18% ± 3%), and to AGN-dominated galaxies (∼5% ± 3%). Using the dust-corrected [O iii] luminosity as an AGN accretion indicator, and the stellar mass and radius as proxies for a black hole mass, we find that AGNs in real bulges have lower Eddington ratios than AGNs in pseudobulges. Real bulges have a wide range of AGN and star formation activities, although most of them are weak AGNs. For both bulge-types, their Eddington ratios are correlated with specific star formation rates (SSFR). Real bulges have a lower specific accretion rate but higher AGN fraction than pseudobulges do at similar SSFRs.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call