DJp.2.02 — Anisotropic thermal conduction with magnetic fields in galaxy clusters

Date & Time

Aug 10th at 6:00 PM until 6:00 PM




Rating ( votes)

Author(s): Alexander Arth1, Klaus Dolag1, Alexander Beck1, Margarita Petkova1, Harald Lesch1

Institution(s): 1. LMU Munich

Magnetic fields play an important role for the propagation and diffusion of charged particles, which are responsible for thermal conduction. In this poster, we present an implementation of thermal conduction including the anisotropic effects of magnetic fields for smoothed particle hydrodynamics (SPH). The anisotropic thermal conduction is mainly proceeding parallel to magnetic fields and suppressed perpendicular to the fields. We derive the SPH formalism for the anisotropic heat transport and solve the corresponding equation with an implicit conjugate gradient scheme. We discuss several issues of unphysical heat transport in the cases of extreme ansiotropies or unmagnetized regions and present possible numerical workarounds. We implement our algorithm into the cosmological simulation code GADGET and study its behaviour in several test cases. In general, we reproduce the analytical solutions of our idealised test problems, and obtain good results in cosmological simulations of galaxy cluster formations. Within galaxy clusters, the anisotropic conduction produces a net heat transport similar to an isotropic Spitzer conduction model with low efficiency. In contrast to isotropic conduction our new formalism allows small-scale structure in the temperature distribution to remain stable, because of their decoupling caused by magnetic field lines. Compared to observations, strong isotropic conduction leads to an oversmoothed temperature distribution within clusters, while the results obtained with anisotropic thermal conduction reproduce the observed temperature fluctuations well. A proper treatment of heat transport is crucial especially in the outskirts of clusters and also in high density regions. It's connection to the local dynamical state of the cluster also might contribute to the observed bimodal distribution of cool core and non cool core clusters. Our new scheme significantly advances the modelling of thermal conduction in numerical simulations and overall gives better results compared to observations.