Accretion Disk Formation Volume Data
		 ------------------------------------
	
	The volume data here is from a simulation of "accretion disk"
formation carried out on the Fujitsu VP2200 at the Australian National
University Supercomputer Facility by Gustav Meglicki
(gustav@arp.anu.edu.au).

	The data is raw byte, of dimension 104x104x104. There are
files for pressure, temperature, density, and velocity components
(vx,vy and vz).

	I've include a few pictures I generated using an experimental
volume renderer/raytracer we've developed for our MIMD parallel
Fujitsu AP1000. The first is of the density, the second of the
density(red->green colormap) and the temperature (red->blue colormap),
illustrating the high temperature jet perpendicular to the plane of
the accretion disk.

	I've appended Gustav's technical description.

Enjoy,
Drew

// Drew Whitehouse,                E-mail: Drew.Whitehouse@anu.edu.au
// Visualization Group,             or     drw900@anusf.anu.edu.au  
// Australian National University, Fax   : +61 (0)6 247 3425 
// Supercomputer Facility.         Phone : +61 (0)6 249 5985
// GPO Box 4, Canberra ACT Australia 2601.

	We have two masses equal to the mass of the Sun separated by
	two solar radii and rotating around the common centre of mass.
	Gaseous matter leaks from one of the "stars" and spirals onto
	a compact companion. There is a constant mass injection rate
	and the initial velocity of gas which leaks out is equal to
	"thermal velocity". The model accounts for pressure and 
	viscous forces, but not for radiative cooling. Viscosity model
	is based on the Gingold-Lattanzio-Monaghan expression which
	yields shear and bulk visosity of comparable magnitude in
	continuum limit. The values of viscosity parameters used in
	the model correspond to those used in simulations of shock
	tube experiments, but result in excessive heating of the
	system which slows down the formation of the disk.

	The simulation was carried out using Smoothed Particle Hydrodynamics
	with controlled merging and splitting of particles. The data files 
	were produced by interpolating parameters from 240,000 particles 
	onto a 1,000,000 points Eulerian grid and scaling them down to
	produce rectangular bricks of bytes.