5. accretion disk atmospheres

physical applications for astro-tomography

The applications of Astro-tomography to accretion problems are widespread. To date, however, Astro-tomography has been employed mostly to map the spatial and velocity structure of accretion flows. Although with the aid of Astro-tomography the modes of stellar accretion are well-understood, the detailed physics is still missing. What are the physical conditions in accretion disk atmospheres? How is energy transported in disks? What mechanism provokes the onset of disk outbursts? By unlocking the spatial information in integrated accretion disk spectra, tomography has provided us with the opportunity to address these questions from an observational perspective. This is the topic of my current research.

We are attempting to map the physical properties of accretion disks using sophisticated atmosphere codes, ground- and space-based observations and indirect imaging methods. We are concerned with the two fundamental regimes: 1. Irradiated disks whose structures are governed by the energy from hot central objects absorbed by the disk atmospheres, and 2. Quiescent disks whose structures are governed by the energy released by viscous stresses between shearing layers within the atmospheres. The experiments are to measure atmospheric conditions (size, shape, temperature, density, microturbulence) from spatially-resolved spectra, and our incentives are to determine the effects of irradiation on accretion disk structure and yield direct observational constraints relevant to the fundamental long-term problems of accretion disk research -- the mechanisms behind viscous dissipation, angular momentum transport and outbursts.

A central irradiating source heats an accretion disk atmosphere and structural instabilities rapidly result in irradiation-driven warping and precession (Wijers & Pringle 1998). Warps have been found in both AGN and planetary disks but the most classic example is the eclipsing X-ray binary star Her X-1. The precessing disk in this object provides an occulting body for the central X-ray source on a quasi-period of 20 orbital cycles and a shadow over the X-ray heated face of the companion star which varies on the same timescale. Disk models recreate these observations easily, but the problem of fitting three-dimensional disks to one-dimensional X-ray and optical light curves is poorly constrained. In order to put useful constraints on the disk shape we require a more robust diagnostic than standard photometry. We have developed ``Echo tomography'' to solve this problem.

This technique employs the X-ray pulses from the central object as it spins on a 1.2-s period. Some regions of the disk are shielded from the central source because of its warped shape. The X-rays reprocess off exposed regions of the disk atmosphere but there is a time delay due to the light-travel time between the X-ray source and the reprocessing site. The range of time delays between X-ray and UV/optical pulsations is a transfer function which defines a set of iso-delay curves over the surface of the disk. A sample of transfer functions measured over a full precession cycle allow a spatial map of the irradiated regions of the disk to be constructed, as Fig. 6 illustrates.

The UV/optical pulsation amplitude as a function of wavelength yields a temperature map of the disk. Furthermore gas reprocessing times provide a time-independent component to the time delay transfer functions which yields the gas density of the disk. Consequently echo tomography provides maps of both spatial structure and the physical properties of the disk atmosphere.

The physical process of viscous dissipation is poorly understood. Accretion results from the transfer of angular momentum outwards by an unknown viscous action between shearing material. Progress in accretion theory has been made largely by packing viscosity physics inside a dimensionless parameter. More sophisticated approaches which regard viscosity to be the result of magnetic stresses, hydrodynamic turbulence and/or tidal action have been encouraging but theorists have enormous freedom to maneuver in an area which has been poorly constrained by observation.

From an observational perspective, spectra contain potentially the information to determine fundamental atmospheric properties of quiescent atmospheres. However a broad range of physical conditions occur within the atmosphere of any accretion disk where the cool outer disk is a very different environment to the hot inner region. Consequently the characteristics of emergent disk spectra are strong functions of spatial location. Astrophysical disks cannot be resolved in general because of their size and distance, and therefore our problem is that classical observations provide only a global picture of accretion disk behaviour.

Our solution to this problem combines sophisticated atmosphere models with indirect imaging techniques and applies these to spectroscopic observations of quiescent disks to determine the local physical properties of their atmospheres, and resolve their spatial dependence.

This research combines three diskiplines:

i) Ground- and space-based observation: Quiescent accretion disks are hosted by white dwarf accretors in cataclysmic variables and black hole accretors in X-ray transients. We will acquire spectroscopic data of accretion disks with high time- and velocity-resolution over several energy bands, employing HST to detect the inner disk regions in the UV, and ground-based optical and IR instruments to detect the outer disk.

ii) Indirect accretion disk imaging: The accretion disk targets are too small to resolve directly, therefore we intend to use maximum entropy tomography techniques to spatially resolve disk spectra. We will construct maps of disks in velocity coordinates from orbital line profile variations and transform these to spatial coordinates by assuming a velocity field. Disk eclipses by the companion star will provide spectrally-resolved spatial maps more directly. The advantage of the resolution provided by these techniques is that we can diskriminate between contributions from the disk, stellar components and localized shocked material forming spots or waves. This fulfills one of the original scientific goals of disk tomography.

iii) Disk atmosphere models: The diskipline of disk atmosphere building has now reached a sophistication with which the physical state of disks can be realistically modelled. Our models can include the effects of external irradiation, vertical gravity gradients, finite optical depths, energy dissipation, turbulence and supersonic velocity gradients. Our current code has been used with notable success in probing the structure of absorption curtains in accretion disks and to investigate optically thick, steady-state atmospheres. A successful fit to synthetic data is provided in Fig. 7.