top of page

BLACK HOLE-GALAXY CO-EVOLUTION

The finding of tight correlations between the mass of supermassive black holes and their host galaxy properties (e.g. bulge mass, stellar velocity dispersion; Magorrian et al. 1998, Kormendy & Ho 2013) strongly suggest that the growth of galaxies is linked to that of their central black holes. This black hole-galaxy co-evolution can be explained if a fraction of the AGN energy output couples with the galactic medium and regulates star formation (and thus the growth of the host galaxy).

Screen Shot 2019-01-16 at 11.21.09 AM.pn

The Mbh-𝛔 correlation (where Mbh is black hole mass and 𝛔 the bulge stellar velocity disperion) found for massive galaxies suggests that supermassive black holes and their host galaxies grow in tandem. Credit: Kormendy & Ho (2013).

Challenging the black hole-galaxy scaling relations

The high-mass end. The synchronized supermassive black hole-galaxy evolution paradigm is challenged by the finding in Mezcua et al. (2018a) that those supermassive black holes at the high-mass end of the scaling relations and residing in brightest cluster galaxies grow more rapidly than their host galaxies, which implies that the local supermassive black hole-galaxy scaling relations do not hold for these extreme objects. The finding is in agreement with studies of the ratio between a supermassive black hole's growth rate and the growth rate of stars in its host galaxy, which is found to be much higher for brightest cluster galaxies than for less massive galaxies (Yang et al. 2018). These results are of significant importance for understanding whether black holes and galaxies co-evolve, and were the highlight of a NASA press release.

A tour of black hole growth in Chandra Deep Field South. Credit: NASA/Chandra X-ray Observatory.

At the low-mass end of the scaling relations, we found that the Mbh-𝛔 correlation flattens and forms a plume (Mezcua 2017; Martín-Navarro & Mezcua 2018, ApJ Letters), as expected from seed black hole formation models in which intermediate-mass black holes in local dwarf galaxies are the fossils of seed black holes formed by direct collapse (e.g. Volonteri & Natarajan 2009; van Wassenhove et al. 2010). The break in the Mbh-𝛔 is found to occur at a stellar mass remarkably similar to that of the galaxy luminosity function and to mark a transition from the AGN-dominated feedback regime of high-mass galaxies to a supernovae-driven regime in low-mass systems (Martín-Navarro & Mezcua 2018, ApJ Letters), in agreement with numerical simulations. See:

Screen Shot 2019-01-16 at 11.13.51 AM.pn

The Mbh-𝛔 correlation combining the sample of low-mass galaxies of Martín-Navarro & Mezcua (2018; black diamonds) with that of van den Bosch (2016) for higher mass galaxies (light circles). The green line indicates the best-fitting trend for 𝛔~70 km/s galaxies, while the red line is the relation expected for more massive objects (~200 km/s). A characteristic 𝛔trans~100 km/s marks the transition between low-mass (supernovae regulated) and high-mass (AGN regulated) scaling relations, and corresponds to a stellar mass of 3 x 10^10 solar masses as predicted by numerical simulations. Credit: Martín-Navarro & Mezcua (2018), ApJ Letters.

More recently, we have found a sample of seven dwarf galaxies out to redshift z=0.9 whose AGN are powered by black holes of more than 10^7 solar masses, this is, by supermassive black holes rather than intermediate-mass black holes!! These black holes are way more massive than expected from black hole-galaxy scaling relations, which means that they tend to grow faster than their host galaxies, contradicting models of synchronized growth. See:

The Mbh-Mstellar correlation showing the offset of the seven AGN dwarf galaxies at redshift z=0.4-0.9. We show for comparison the low-z AGN of Reines & Volonteri (2015) (RV2015) and the high-z AGN of Suh et al. (2020) (Suh+2020), whose masses have been computed using the same procedure and parameters as in our sample. The solid line shows the local + high-z Mbh-Mstellar correlation  found for the combination of the RV2015 and Suh+2020 samples with a 1sigma scatter of 0.5 dex.

Credit: Mezcua et al.  (2023), ApJ Letters.

bottom of page