P-A Colloquium: How AGN Feedback Might Work

Speaker:  Megan Donahue, Michigan State University

Title:  How AGN Feedback Might Work

View Speaker Schedule

Refreshments at 3:30 pm in 1400 BPS.

Date: Thu, 19 Jan 2017, 4:10 pm - 5:10 pm  

Type: Colloquium

Location: 1415 BPS Bldg.

One of the biggest challenges to understanding how galaxies work is why some galaxies are forming stars today and some galaxies stopped forming stars a long time ago. The answer to this question seems to be that the central supermassive black hole, one found in the center of nearly every galaxy above a certain mass limit, seems to control the formation of stars in its host galaxy by regulating the thermodynamic state of the hot, diffuse, gaseous halo surrounding the galaxy. While the supermassive black hole is accreting gas, it can also inject a large amount of energy into its surroundings; it is during this phase the black hole reveals itself to observers as an Active Galactic Nucleus, or AGN. The gaseous halo, the “circumgalactic medium”, or CGM, is most easily studied in the most massive galaxies in the universe, the central galaxies of clusters of galaxies. The CGM is where most of the baryons of a galaxy actually resides. It is the most likely source of fuel for making new stars and for feeding the supermassive black hole. This solution, known as “AGN feedback”, however, has not been easy to model or to fully understand. I will describe recent progress towards understanding how AGN feedback might work, guided by the striking relationship between the processes (such as star formation, AGN activity, and co-existence of multiphase gas) in cluster central galaxies and massive ellipticals, and the thermodynamic state of their circumgalactic gas. These observations have led us to the formulation of the precipitation model, in which a multiphase gas state emerges when the gas cooling time is approximately 10 times the free-fall time at that position in the galaxy. We predict that when simulations also reproduce this emergent phenomenon (e.g. with feedback that includes kinetic energy), many of the scaling relations we see for galaxies (e.g. mass & metallicity, mass & efficiency of converting baryons into stars, supermassive black hole mass and velocity dispersion, age of stellar populations/star formation history and mass) will naturally and robustly be reproduced, even for galaxies where we cannot yet detect their CGM. This model also implies that AGN feedback may be relevant for low-mass galaxies. I will discuss how future observations and simulations could test these ideas.