Accretion Disks in AGNs

I briefly review the theoretical models of radiatively efficient, geometrically thin and optically thick accretion disk spectra that currently exist for AGN. I then discuss three recent observational developments that have real potential to teach us about the physics of these flows. Finally, I present results on the most recent, thermodynamically consistent simulations of magnetorotational turbulence and discuss what these simulations are suggesting about the vertical structure of accretion disks. 1. Current Standard Accretion Disk Models of AGN Accretion disks around supermassive black holes continue to be the generally accepted paradigm for the central engine of AGN. I focus in this review on “standard disks”: those that are geometrically thin and optically thick. These models are, by construction, radiatively efficient. For radiatively inefficient disks, see the papers by Mineshige and by Yuan in these proceedings. Currently all theoretical models of the photon emission of accretion disks (radiatively efficient or otherwise) are based on the alpha prescription for the anomalous stress introduced by Shakura & Sunyaev (1973). A large number of such models for standard disks have been constructed over the years, e.g. Malkan (1983); Sun & Malkan (1989); Laor & Netzer (1989); Laor, Netzer & Piran (1990), to name just a few. The most sophisticated of these models have been developed in a series of papers by Hubeny & Hubeny (1997, 1998) and Hubeny et al. (2000, 2001), and are available to the community through the IDL package AGNSPEC, written by Ivan Hubeny. The models are constructed by first dividing the accretion disk into a set of discrete annuli, and then computing a one-dimensional vertical interior and atmosphere structure based on certain assumptions in each annulus. Using the locally calculated spectrum, photon geodesics from the disk to infinity are computed to then produce a predicted model spectrum for an observer at some viewing angle. Some aspects of the physics of accretion disks are treated very well by the AGNSPEC models. They fully account for the relativistic radial structure of the disk (Novikov & Thorne 1973; Riffert & Herold 1995) and relativistic Doppler shifts, gravitational redshifts and light bending in a Kerr spacetime (Agol 1997). The models include a detailed non-LTE treatment of all abundant elements and ions in the disk, and include continuum opacities due to bound-free and free-free transitions of these ions, as well as Comptonization. (No lines are included in the models, however.) There are many ad hoc assumptions that continue to plague these models, however. First there are assumptions that relate to the radial structure of the