The Magnetised Interstellar Medium: Simulating Realistic Star Forming Environments with Stellar Feedback and Magnetic Fields

Abstract

In this thesis I present my investigations of star formation on a unique simulation scale, bridging the gap in theory between galaxy scale and cloud scale star formation studies. Rather than simulating an isolated galaxy or molecular cloud, these models are extracted from a pre-evolved galaxy simulation. As such these models are in a unique position to study the formation and evolution of populations of giant molecular clouds (GMCs) within a realistic galactic environment. In the first study I performed hydrodynamic (HD) and magnetohydrodynamic (MHD) simulations of sub galactic regions including photoionising and supernova feedback. I aimed to improve the initial conditions of the region extraction models by including an initial population of stars. Supernovae from the initial population are important in trig- gering star formation and driving gas motions on larger scales above 100 pc, whilst the ionising feedback contribution from the initial population has less impact, since many members of the initial population have cleared out gas around them in the prior model. In terms of overall star formation rates though, the initial population has a relatively small effect, and the feedback does not for example suppress subsequent star formation. Next I investigated star formation from subparsec to kpc scales with MHD models of a cloud structure and a section of galactic spiral arm. I aimed to understand how magnetic fields affect star formation, cloud formation and how feedback couples with magnetic fields on scales of clouds and clumps. I found that magnetic fields overall suppress star formation by ∼10% with a weak field (5 µG), and ∼50% with a stronger field (50 µG). Cluster masses are reduced by about 40% with a strong field but show little change with a weak field. Identified clouds tend to be aligned parallel to the field with a weak field, and become perpendicularly aligned with a stronger field, whereas on clump scales the alignment is more random. The magnetic field - density relation of clouds and clumps in the models agree with the Zeeman measurements in the weaker field models, whilst the strongest field models show a relation which is too flat compared to the observations.