Thermal and magnetic structure of the solar corona: 3D magneto-hydrodynamic modeling

Pencil Code (3D MHD): compressible hydrodynamic flow simulations with magnetic fields

The corona is the million degrees hot outer atmosphere of the Sun. How the plasma in the corona can be heated to temperatures more than 100 times higher than on the surface is one of the most interesting questions in stellar astrophysics. In recent years numerical experiments became feasible that model the corona of the Sun in a self-consistent way and allow to synthesize emission from the corona. This way one can directly compare images and spectra derived from the model with those in real observations.

For these numerical experiments we solve the time-dependent equations of magnetohydrodynamics in three dimensions. The magnetic field is driven by surface motions and by this currents are induced in the upper atmosphere that heat the coronal plasma through Ohmic dissipation. To be able to synthesize the coronal emission one has to fully account for the energy balance including not only the Ohmic heating but in particular also for the radiative losses and heat conduction. We solve these equations using the Pencil code. Members of our group are active developers of this code. With this we run numerical models on various super-computers, because the computational effort is significant (up to 15 million CPU hours per simulation run).

Using these numerical models we can understand many aspects of the corona above active regions (i.e. a pair or group of sunspots). For example we now understand why and how the hot loops in the corona form and evolve. Besides the many successes of the 3D MHD models open questions exist, one of the most interesting is to understand the large-scale thermal structure of an active region corona.

The main goal of this thesis project would be to run large-scale models for an active region corona to understand why the corona of an active region typically consists of loops hosting 5 to 10 million K hot plasma, while the larger-scale loops are heated to only 1 million K. This is a fundamental question to be answered and a new 3D MHD numerical model should provide the key to this.

To run these models successfully some modifications and tests will have to be applied to the Pencil code. This is mainly to change the numerical treatment of heat conduction and Lorentz force to speed-up the code. Tests showed that this can provide the required speed-up to run the planned simulations.

More information on the Solar and stellar coronae group and the group's work can be found at the Solar and stellar coronae group homepage. For questions please contact Hardi Peter.