A bayesian approach to the study of dark matter in disk galaxies
2015-10-01T15:59:29Z (GMT) by
Studies of the rotation of disk galaxies have long been used to infer the presence and distribution of dark matter within them. Here I present a new Markov Chain Monte Carlo (MCMC) method to explore the extensive and complex parameter space created by the possible combinations of dark and luminous matter in these galaxies. I present exhaustive testing of this method to ensure it can retrieve dark matter halo parameters from artificial data, and apply it to real galaxies from The HI Nearby Galaxy Survey (THINGS) and other sources. The results of these studies can shed some light on how disk galaxies form and evolve. Chapters 1 and 2 provide background for the physics and statistical methods respectively. Chapter 3 shows the testing of the MCMC method on artificial data, and applies it to DDO 154 to find a more robust constraint on the inner log slope than previous methods. Chapter 4 applies this method to a broad range of galaxies taken from the THINGS survey, constrains their physical properties, and presents a simple model of feedback to compare with. Chapter 5 applies the method to M33, mapping a degeneracy between the log slope of the dark matter halo and the mass-to-light ratio, that excludes the combination of a cored halo and a light stellar disk. Chapter 6 extends the MCMC method to an earlier stage of analysis by marginalising over the parameter space of possible disk models for simulated galaxies. Chapter 7 presents conclusions and discusses future work that can lead on from this thesis.