Gravitational waves as a probe for cosmology and non-standard physics

Abstract

The direct detection of Gravitational Waves (GWs) in 2015 revealed an entirely new way of studying the Universe and fundamental physics. This tremendous achievement, almost 100 years since the initial study of GWs by Einstein, allows us to see into regions invisible to our standard Electromagnetic (EM) observations, from the primordial universe to the interior of neutron stars.

In this thesis, we exploit this new probe to investigate a number of physical characteristics of the Universe and to test our current understanding of Einstein’s General Theory of Relativity. To accomplish this, we use powerful cosmological simulations, which are able to replicate the observed large-scale structure down to complex, non-linear scales.

More specifically: In Chapter 1, we review the current paradigm of cosmology, the LCDM model and its limitations. We motivate some of the fundamental assumptions and problems that we aim to tackle in the next chapters, and provide a brief introduction to GWs and their propagation in spacetime. In Chapter 2, we describe our numerical simulations and their outputs for the values of gravitational potentials.

In Chapter 3, we exploit GWs Standard Sirens, i.e. GWs observations where we are able to locate their counterpart sources in the EM spectrum, in order to constrain the level of inhomogeneities in our Universe. In parallel, we test possible degeneracies of the former with theories that modify general relativity.

In Chapter 4, we study the presence of inhomogeneities on the propagation of GWs. We derive a modified propagation equation due to the presence of matter anisotropies. We solve the latter numerically, exploiting realistic environments from numerical simulations. For this, we develop a novel ray-tracing approach. As a result, we are able to derive potential effects on the stochastic GW background and examine lensing observables and their cosmological applications.

Finally, in Chapter 5, we investigate the power of galaxies’ clustering on a statistical method of determination of the Hubble-Lemaˆıtre parameter based on GWs. Since the majority of GWs observations will not have an EM counterpart, we can combine in a statistical way, the galaxies that reside within the GWs sky localisation region, in order to determine H0. We estimate the capabilities of the method, in realistic environments with clustered galaxies, from numerical simulations. At the same time, we evaluate the importance of clustering, when galaxies are missing from survey catalogues, due to observational limitations.