I am currently offering several PhD/honours projects in computational astrophysics. Please
contact me if you are interested.
Propagation of warps in accretion discs
 |
Many astrophysical accretion disc surrounding stars and compact objects are known to be warped. Warping
itself can be driven by a variety of mechanisms including radiation from the central star and precession induced by a spinning black
hole. However the theory of warp propagation in discs is very new and is only starting to be verified by full numerical
calculations. Early calculations show good agreement with theory in some respects whilst disagreement in others. The goal of this
project is to understand the fundamental physics of warp propagation by performing high-resolution numerical simulations of warps around stars
and black holes and comparing them with analytic warp propagation theory in order to understand the differences and/or verify the
analytic theory.
|
Application of Smoothed Particle Magnetohydrodynamics to the magneto-rotational instability
 |
Accretion of material onto stars and compact objects such as neutron stars and black holes is one of the most powerful sources of
energy in the universe. Nevertheless, how gas manages to accrete at all remains a problem, since accretion involves the removal of
all of the excess angular momentum. Observations reveal nature's way -- through the formation of accretion discs. However though
such discs are increasingly well observed, the means by which they accrete has only recently become clear - by the action of an
instability related to the presence of magnetic fields in the disc, causing the onset of turbulence which results in the necessary
energy loss. Whilst a good many simulations have now been performed showing the action of this so-called "Magneto-Rotational
Instability" in discs, most of these calculations have been limited to discs which are quite symmetric. Using a recently developed method for solving the
equations of Magnetohydrodynamics in the Smoothed Particle Hydrodynamics numerical method, the idea of this project is to be able to
simulate highly asymmetric discs (i.e., 3D discs with precession and warps) self-consistently, without relying on parameterised
prescriptions for disc accretion.
|
The origin of turbulence in molecular clouds
 |
Molecular clouds in the Galaxy are observed to be highly turbulent, with supersonic motions, which has strong
implications for how stars form from such clouds. However the origin of such energetic motions is as yet unclear, though there are
many good candidates in the form of supernovae, HII regions, protostellar jets and outflows as well as driving by Galactic spiral
arms. The aim of this project is to perform simulations of supersonic, magnetised turbulence with various types of physical driving
mechanisms (as distinct from adding energy artificially in fourier space which is the usual "cheats way"), comparing the results
with observations to see whether or not signatures of the actual driving mechanism can be found. Ultimately it is hoped that the origin of turbulence
in molecular clouds can be clearly pinned down.
|
Honours projects
Preliminaries to any of the above PhD projects can be adopted as honours projects, as well as those listed below:
Shearing box simulations of the magneto-rotational instability
Global calculations of accretion discs in 3D (as described above) require tremendous numerical resolution, however the basic workings of discs via the magnetorotational
instability can be understood by performing simulations in a tiny "shearing box" cut out from the disc midplane. The aim of this
project is to verify a new method for performing simulations of the magneto-rotational instability in discs by demonstrating that
the method can simulate the action of the MRI in the shearing box approximation.
Hamiltonian methods for timestepping in SPH
In recent years it has been realised that integrators for solving ordinary differential equations can be derived from
first-principles approaches using Hamilton's equations. These techniques in practise show significant advantages in accuracy and
stability, because they respect certain symmetries of the underlying physical system, such as the invariance of the equations to the
direction of time and the preservation of area in phase space.
The Smoothed Particle Hydrodynamics numerical method is a unique method for solving the equations of fluid dynamics, primarily developed
right here at Monash by Joe Monaghan and others. Since SPH itself is Hamiltonian and can be derived from a variational principle,
many of the powerful Hamiltonian methods for ODEs can be directly applied and should lead to substantial improvements in accuracy. This project therefore involves
implementing, comparing and assessing the suitability of several of these techniques for use with SPH.