In this thesis, I present my work related to the characterization of diffuse Galactic foregrounds for observing the polarization of the Cosmic Microwave Background (CMB) radiation, and the impact of these foregrounds on the measurement of cosmological parameters. One of the most important future challenges for cosmology is the potential detection of polarization B-modes of the CMB. Inflation is a theory that explains the extremely early Universe, and solves several problems that were present in classical cosmology. It describes the anisotropies observed in the current Universe as primordial quantum fluctuations stretched by rapid exponential expansion. A key prediction of inflation is the production of a background of primordial gravitational waves, which could be detected through the associated large-scale B-mode signal in the CMB polarization. The amplitude of the B-mode signal, which depends on the energy scale of inflation, is parametrized by the tensor-to-scalar ratio r. Diffuse emission from within our Galaxy, and other extra-Galactic sources, collectively referred to as CMB foregrounds, obscure a fraction of the cosmological signal from the CMB radiation. This is a huge problem, because they have to be cleaned using data analysis methods, called component separation. A significant challenge for the potential detection of the primordial B-mode signal is that it can be extremely small, to the extent that it can be dominated even by the residual foreground contamination after component separation. In this work, we characterize these foreground residuals and assess their impact on the cosmological parameters. We create a method to simulate observations of the microwave sky, including diffuse Galactic foregrounds, CMB realizations and instrumental noise. These simulations are used to propagate errors on the characterization of foregrounds through the analysis procedures employed in the observations of the CMB, including component separation, angular power spectra calculation and cosmological parameter estimation. We estimate the bias and the Ïƒ error for the tensor-to-scalar ratio, to quantify the impact of the foreground residuals in the cosmological signal. We also propose a novel method to model these residuals when determining cosmological parameters, in order to avoid a bias on the r parameter. We performed forecasts and optimization analyses for two proposed CMB polarization experiments: the Simon Observatory, a funded ground-based telescope that will observe the polarization of the CMB from the Atacama desert in Chile, and CORE, a proposed next-generation CMB satellite experiment. All of our work shows that the issue of foreground residuals must be considered very carefully in future studies. Foreground spectral parameters must be modelled very accurately, with errors < 0.5%, if we wish to measure a value r âˆ¼ 10^âˆ’3. These foreground residuals can easily be mistaken as primordial cosmological signals, so our work motivates further research into developing new data analysis techniques.
|Date of Award
|31 Dec 2018
- The University of Manchester
|Anna Bonaldi (Supervisor) & Michael Brown (Supervisor)
- Cosmic Microwave Background