Research Interests

My primary research interests are observational study of quasars, supermassive black holes, and the evolutionary relationship between black holes and their host galaxies.


Quasars are nature's most extreme phenomena. They belong to a class of objects called active galactic nuclei (AGN), which, as the name suggests, are galaxies with unusually active central regions. AGN come in a variety of luminosities and other properties, with quasars on the extreme end. A single, luminous quasar can outshine a thousand Milky Way-like galaxies, and yet all of this light emanates from a region not much larger than our solar system. The central "engine" of an AGN is believed to be a supermassive black hole — a black hole that is hundreds of thousands to tens of billions of times the mass of the Sun — actively feeding on gaseous and stellar material within its immediate vicinity. This material reaches relativistic speeds as it careens down to the black hole, becomes super-heated, and shines brightly enough to be observed from billions of light-years away.

Quasar accretion disk and jet [Credit: Unknown]

The majority of my research is conducted using data from the Sloan Digital Sky Survey. The SDSS is a public archive containing vast amounts of data, including spectra for over 120,000 quasars.

Supermassive black holes and galaxies

One of the significant developments of 20th century astrophysics was the discovery that supermassive black holes lurk in the centers of almost all galaxies. These enormous black holes are believed to be relics of ancient quasars long since deprived of their fuel. By the turn of this century, it was found that the masses of these black holes correlate strongly with properties of their host galaxies, such as bulge luminosity (and therefore bulge mass) and the stellar velocity dispersion. The latter is the statistical distribution of the random velocities of stars orbiting in the galaxy bulge. Since this depends on the mass and radius of the bulge, the velocity dispersion is a measure of the bulge potential well. The relationship between the mass of the supermassive black hole and the velocity dispersion of its host galaxy is known as the "M-sigma" relationship (M for the mass of the black hole, and the Greek letter sigma for the velocity dispersion). The ubiquity of supermassive black holes in the local universe as well as in distant quasars, coupled with the strong correlations, suggest some kind of evolutionary link between the growth of galaxies and the growth of supermassive black holes. The nature of the link is puzzling, because supermassive black holes are typically only about 1/1000th the mass of their host galaxy bulges (see here for a recently discovered exception to this), which means they exert effectively no direct gravitational influence on the stars in the galaxies.

The way the black hole - galaxy correlations change with time is key to understanding the black hole - galaxy link. Any real change with time is referred to as evolution. The big question is whether black holes grow substantially before, after, or coevally with their host galaxies. In order to address this question, we compare the black hole - galaxy correlations at several different epochs in cosmic history. The further back in time you want to investigate something, the further away you have to look. This presents a challenge to those of us who wish to look very far back in time, because it's exceedingly difficult to obtain detailed observations of such distant galaxies. This is where quasars come in. Given the immense brightness of quasars, they are observable over a vast range of distances and therefore cosmic history.

The bulk of my work for the last decade has involved using quasars to investigate the M-sigma relationship over billions of years of cosmic history. Instead of trying to measure the black hole mass and galaxy properties directly, these quantities can be inferred using measured properties from quasar spectra.

Frustratingly, the results of various black hole - galaxy evolution studies have painted a confused picture, with many groups disagreeing on the magnitude of the evolution, the direction of the evolution, and even whether evolution has taken place at all. One of the important discoveries in my work has been the role of observational biases in such studies, which can produce a false signal of evolution. These biases can be significant, and addressing them is a crucial part of ongoing work in this area.

Quasar with dusty torus and jets
A supermassive black hole obscured by a dusty torus; polar jets are shown emanating from near the black hole's event horizon [Credit: Unknown]

Simplified M-galaxy relationship
Graphical representation of the black hole mass - galaxy bulge mass correlation [Credit: K. Cordes & S. Brown (STScI)]

The latest word on the M-sigma relationship
[Credit: Gültekin et al. (2009)]

Other topics of interest

Big galaxies: The most massive black holes in the universe weigh in at 5 billion times the mass of the Sun or more. Quasar demographics indicate a space density of 200 such black holes per cubic gigaparsec of space. The M-sigma correlation suggests a similar space density of commensurately large galaxies to host the eventual relics of these quasars. However, large-scale searches have failed to detect such galaxies. As part of the effort to look for these giant host galaxies, I obtained high-quality spectra of six of the largest-velocity dispersion galaxies in the SDSS using the Hobby-Eberly Telescope. The goal was to determine whether higher-quality data would reveal larger dispersions than those indicated by the SDSS data. They did not. However, given that these galaxies have the highest-known dispersions, they are still of great interest in terms of the M-sigma relationship. One implication of the failed search is that the majority of the most massive black holes in the universe reside in comparatively modest galaxies — I am therefore very interested in what sort of black holes lurk at the centers of these galaxies. I am currently working with a collaborator to obtain Chandra and EVLA observations of the nuclei of these galaxies in order to determine the masses of their central black holes.

Recoiling black holes: During major mergers of galaxies, the supermassive black holes residing in each of the galaxies orbit each other for a while and eventually coalesce into a single black hole. Numerical relativity simulations predict that, under certain circumstances, these coalescing black holes will be imparted with a net velocity — a "kick" — and in extreme cases even flung out of the merged galaxies. Such kicked black holes are predicted to have distinct observational signatures in the spectra of AGN. Our group conducted a search for these kicked black holes in the spectra of SDSS quasars, but found no significant evidence for their existence. This was unexpected given the frequency with which major mergers occurred during the quasar epoch. This negative result was featured in a press release (see below). [Project led by Erin Bonning]

Accretion disk temperatures: The accretion disk fueling a quasar is the source of most of the ultraviolet and optical emission in a quasar spectrum. Theoretical models of accretion disks indicate that hotter accretion disks should be bluer in color than cooler disks. However, in a study of accretion disk temperatures versus continuum colors in SDSS quasar spectra, our group found deviation from this expected trend, even to the extent that some quasars exhibited the opposite trend — hotter disks tending to be redder in color. The highly-deviant objects were found to be accreting material at a high rate, suggesting that accretion rate may be a factor in the temperature - color relationship for quasar accretion disks. [Project led by Erin Bonning]

Iron abundances in quasars: AGN spectra show a wide range of Fe II abundances. Recent work carried out with collaborators indicates that this range of abundances can be explained by a model in which the gas-phase Fe II is depleted by solidifying into grains beyond the dust sublimation radius in the broad line region of AGN. [Project led by Greg Shields]

In the press

The negative result for recoiling black holes was sufficiently interesting that McDonald Observatory issued a press release. We presented our result at the 2007 summer meeting of the American Astronomical Society, along with other groups presenting on the same topic. The press covered the event, and it was picked up by Nature, New Scientist, Astronomy magazine, and MSNBC.

In the summer of 2009, I was interviewed by Discovery News (affiliated with the Discovery Channel) about the implications of the unexpectedly large mass for the supermassive black hole in the giant galaxy M87.