paper

Simulations of magnetic fields in isolated disk galaxies

arXiv:1212.1452 · doi:10.1093/mnras/stt428

Abstract

Magnetic fields are known to be dynamically important in the interstellar medium of our own Galaxy, and they are ubiquitously observed in diffuse gas in the halos of galaxies and galaxy clusters. Yet, magnetic fields have typically been neglected in studies of the formation of galaxies, leaving their global influence on galaxy formation largely unclear. We extend our MHD implementation in the moving-mesh code Arepo to cosmological problems which include radiative cooling and the formation of stars. In particular, we replace our previously employed divergence cleaning approach with a Powell 8-wave scheme, which turns out to be significantly more stable, even in very dynamic environments. We verify the improved accuracy through simulations of the MRI in accretion disks, that reproduce its correct linear growth rate. Using this new MHD code, we simulate the formation of isolated disk galaxies similar to the Milky Way using idealized initial conditions with and without magnetic fields. We find that the magnetic field is quickly amplified in the initial starburst and the differential rotation of the forming disk until it eventually saturates when it becomes comparable to the thermal pressure. The additional pressure component leads to a lower star formation rate at late times compared to simulations without magnetic fields, and induces changes in the spiral arm structures of the gas disk. In addition, we observe highly magnetized fountain-like outflows from the disk. These results are robust with numerical resolution and are largely independent of the initial magnetic seed field assumed in the initial conditions, as the amplification process is rapid and self-regulated. Our findings suggest an important influence of magnetic fields on galaxy formation and evolution, cautioning against their neglect in theoretical models of structure formation.

19 pages, 15 figures, submitted to MNRAS, comments welcome

References in corpus (13)

Cited by in corpus (236)