Physics of Coevolution of Galaxies and Supermassive Black Holes

TL;DR

This paper introduces a physically based model for the coevolution of galaxies and supermassive black holes, emphasizing the role of star formation feedback over AGN feedback in galaxy evolution.

Contribution

The model proposes that SMBH growth occurs mainly post-starburst via recycled gas, challenging the idea that SMBHs grow significantly during starburst phases.

Findings

SMBH growth is fueled by recycled gas in the post-starburst phase.

Star formation feedback is as strong as AGN feedback in quenching star formation.

Predicted SMBH-bulge relations align with observations.

Abstract

A new model for coevolution of galaxies and supermassive black holes (SMBH) is presented that is physically based. The evolutionary track starts with an event that triggers a significant starburst in the central region of a galaxy. In this model, the main SMBH growth takes place in post-starburst phase fueled by recycled gas from inner bulge stars in a self-regulated fashion on a time scale that is substantially longer than 100Myrs and at a diminishing Eddington ratio with time. We argue that the SMBH cannot gorge itself during the starburst phase, despite the abundant supply of cold gas, because star formation is a preferred mode of gas consumption in such an environment than accretion to the central SMBH. We also show that feedback from star formation is at least as strong as that from AGN and thus, if star formation is in need of being quenched, AGN feedback generally does not play the primary role. The predicted relation between SMBH mass and bulge mass/velocity dispersion is consistent with observations. A clear prediction is that early-type galaxy hosts of high Eddingtion rate AGNs are expected to be light-blue to green in optical color, gradually evolving to the red sequences with decreasing AGN luminosity. A suite of falsifiable predictions and implications with respect to relationships between various types of galaxies and AGN, and others, are made. For those where comparisons to extant observations are possible, the model appears to be in good standing.