Constraining a spin 1 dark matter model through CMB observations

Abstract

During the last century, overwhelming evidence has been presented to infer the existence of additional mass in the Universe, not accounted for by the observable matter content. This unknown form of matter has come to be known as Dark Matter (DM), and the leading hypothesis is that it mainly consists of new, yet-to-be discovered particles. There are different methods of probing DM, one of which is to study how annihilating DM particles would have injected energy in the early Universe and altered the appearance of the Cosmic Microwave Background (CMB) power spectrum, an approach previously used to set constraints on generic DM models. In this thesis, I investigate the prospects for using this approach to constrain certain DM model parameters. Considering a specific spin 1 DM model, the impact of its annihilation on the CMB power spectrum was investigated by implementing the model Lagrangian in GAMBIT, from which the annihilation cross section was evaluated, and used to determine the energy injected in the early Universe and how the power spectrum would have been affected. Comparing with CMB data, the likelihood for the specific set of parameter values was evaluated and then used to further sample the parameter space, allowing statistical exclusion limits to be set. Combining this with additional constraints, set from the observed DM relic density, it was possible to exclude large regions in the parameter space for the investigated model and set stringent limits on the corresponding DM mass, m 2 ( 102, 103) GeV, for the specific cases considered. It was thus found that the CMB power spectrum can be used to constrain model parameters, and used in combination with relic density constraints to exclude large regions in the parameter space. Not only can these results be used in future research when considering the same model, but this thesis provides a practical example of how to implement and use GAMBIT to set constraints on model parameters from observational data, which can also be incorporated for new models in future research.