Searching For Evidence Of New Physics In The Cosmic Microwave Background And Large Scale Distribution Of Galaxies PDF Download

In this dissertation I describe a set of three projects all related to the statistics of the Cosmic Microwave Background (CMB). In the first of these, I analyze the relationship between two CMB statistics, called S[subscript 1/2] C2, and also re-interpret the significance of S[subscript 1/2]. The S[subscript 1/2] statistic quantifies an unexpected feature which can be seen in the angular two-point correlation function C([theta]) of the observed CMB temperature field: C([theta]) is very close to zero on angular scales larger than about cos−1(1/2) = 60°. The C2 statistic is the variance of the five quadrupole moments of the CMB temperature field. Using CMB simulations based on the [lambda]CDM cosmological model in which the simulated C2 values are near the observed one, the p-value of S[subscript 1/2], or the fraction of simulations that have an S[subscript 1/2] value below the observed S[subscript 1/2] value, increases from 0.007 to 0.08. The p-value of C2 is 0.039. I show that these low p-value s are not independent of each other. Furthermore, corrections for a "look-elsewhere effect'' in the interpretation of the observed value of S[subscript 1/2] are large. This complication in interpretation arises due to the a posteriori nature of the S[subscript 1/2] statistic: the statistic was created after the observation of C([theta]) that motivated measuring it. The S[subscript 1/2] project is strongly data-driven, involving analysis of CMB maps from the Planck mission. The rest of the dissertation is aimed at understanding the value of future cosmological data. For these projects I calculate the expected uncertainties in [lambda] CDM + m[subscript nu] + w[subscript DE] cosmological parameter constraints, using a simple model for a linear galaxy bias, through a combination of future CMB lensing observations with future large scale structure (LSS) observations. The combination of CMB lensing information with that of a 3-dimensional map of galaxy number densities divided into redshift bins helps to break the degeneracy between the amplitude of the matter power spectrum P(k) and the galaxy bias, thus enabling certain parameters of interest to be better constrained in this combination of observables than with CMB lensing or galaxy clustering alone. In one case, the focus is constraints on the sum of the mass of neutrinos [sigma]m[subscript nu], without relying on cosmological optical depth information, and in the second case, is constraints on a time-varying Dark Energy equation of state (EoS) parameter w[subscript DE](a). In both of these cases, the main future data sets that I include in the forecasts are the CMB-S4 lensing map, the CMB-S4 primary CMB maps, and a 3-dimensional LSST galaxy map, divided into tomographic redshift bins, projected into 2-dimensional galaxy maps. Cosmic shear information is not included in these forecasts in order to show that CMB lensing + LSST galaxy clustering can provide a probe of neutrino mass and EoS that is complementary to methods which include cosmic shear. In the case of neutrino mass, I also include a forecast for the Dark Energy Spectroscopic Instrument (DESI) Baryon Acoustic Oscillation (BAO). I show that the combination of LSST galaxy clustering, CMB-S4 lensing, and DESI BAO should be able to achieve constraints on the neutrino mass sum of 24 meV, an independent measurement that is competitive with or slightly better than the optical-depth-limited constraint of 29 meV possible with CMB-S4 + DESI BAO, or 32 meV with LSST galaxy 2-point correlations + galaxy-shear correlations Planck. For the case of the EoS parameter, adding the CMB-S4 lensing map to the redshift-binned LSST galaxy clustering plus S4 primary CMB power spectra increases the Dark Energy Task Force figure of merit (FoM) by a factor of between 2 and 4. The FoM from these data is comparable to that from LSST cosmic shear + Planck in the absence of a CMB lensing map. For the EoS parameter forecast, I also point out how changes to the distance-redshift relation are more important for detection of a departure from w[subscript DE] -1$ than changes to the shape or amplitude of the matter power spectrum.