| Abstract This project investigates the dynamics of astrophysical plasmas in a range of environments using numerical simulations. The plasma is described by the magnetohydrodynamic equations, a set of equations that couple together the ordinary hydrodynamic equations and Maxwell's equations. We have access to several computer codes, some which are newly developed and are parallelised using MPI. The problem that has received the most attention from us is the issue of the angular momentum transport in accretion discs. It is widely acknowledged that it is necessary to have a turbulent mechanism for driving the accretion since the molecular viscosity is too small by several orders of magnitude. A long-standing problem in accretion disc theory has been to find the cause of the turbulence. The most natural candidate is the differential rotation of the accretion disc, which can only be understood within a magnetohydrodynamic framework, since the Keplerian rotation in a geometrically thin accretion disc is hydrodynamically stable but magnetohydrodynamically unstable. Several groups, including ours, have demonstrated in numerical simulations that the magnetohydrodynamical instability leads to turbulence in the accretion disc. There is now working going on in refining the simulations such that they allow for a more accurate modelling of the accretion disc and in particular its vertical structure. Together with A. Brandenburg, Copenhagen, and N. E. Haugen, Trondheim, we are developing a new code for simulating the turbulence in discs with a strong radiation pressure. These simulations will be able to address the long-standing problem of the stability of discs that are dominated by radiation pressure. A related problem that we are working on is to understand the interaction between an accretion disc and a magnetized star, such as the neutron star in an X-ray pulsar or a T Tauri-star, a young solar-like star that has not yet lost its protoplanetary disc. We are here presented with a problem, in which we must at the same time resolve small-scale turbulence, and capture the global structure of the flow and the magnetic field. Even with the development of a new computer code that can handle a complex geometry this will demand the use of the most powerful computers available. Together with K. Galsgaard, St. Andrews and Copenhagen, we are also studying the role of magnetohydrodynamic processes in the solar corona and the solar wind. Both the heating of the solar corona and the acceleration of the solar wind cannot be understood without taking into account the dynamical processes in the solar magnetic field. In contrast to most of astronomy there is a direct practical interest in these studies, since the interaction of the solar wind with the Earth has been known to lead to the disruption of satellite communications and electrical power supplies. So far we have re-written an old code of Galsgaard's using MPI, and are now ready to use this for production runs, in which we will study the acceleration of the fast solar wind by Alfv'en waves and the dynamics of oscillating coronal loops. |
