Fabian Klein

Fabian Klein  ( Germany )


fklein @ ari.uni-heidelberg.de

Analysis, acceleration and parallelisation of astrophysical codes with application for projects

This is a computational and numerical project, related to analysis, implementation and optimization of astrophysical simulation codes and their use in current projects. Hydrodynamic codes to model gaseous disks in the galactic center and their interactions with stellar systems are in the focus of this work.

We consider a central star cluster dynamically interacting with the super massive black hole(SMBH) and assume the presence of an accretion disk (AD), which may have existed during the past billion years in the center of our Galaxy. The AD has an important role in these interactions as the dissipative force acting on stars in the disk results in an increased mass ow to the SMBH and creates asymmetries in the phase space distribution due to its rotation, and it traps stars which form a central stellar disk, as observed currently. The basic modeling idea for the disk is a vertically extended Keplerian disk, a modified disk from allowing [Shakura and Sunyaev, 1973], including a fully self-consistent treatment of stellar dynamics including the dissipative force originating from star- gas ram pressure eects. [Just et al. 2012, Kennedy et al. 2016]. This project's first and main goal is to do a full hydrodynamical simulation of the disk in order to investigate its lifetime and stability as well as any bi-directional feedback of star-disk and disk-SMBH interactions. We are using currently the PLUTO Code  [Mignone et al., 2007] in order to carry out the numerical simulations including massively parallel runs as offered by PLUTO. This example will also be used to gauge PLUTO's abilities and run benchmarks and profiling in order to determine further uses of it. A comparison to other hydrodynamic codes is also planned. The final goal of the project is to combine N-body simulations and the hydrodynamical simulation to include the star-gas interactions and feedbacks.


Supervisor:   Rainer Spurzem  ( ARI )

 
loading content