Studying Unmodeled Physics from Gravitational Wave Data

Abstract

This thesis explores the detection and analysis of unmodeled physics in Gravitational Wave (GW) data. To this end, we develop the SCoRe framework, which uses the Correlated Residual Power Spectrum (CRPS) between pairs of detectors to identify deviations from our Standard Model (SM) of GW. This model includes General Relativity (GR) as the theory describing gravity, binary Black Holes (BHs) and Neutron Star (NS) merging as the sources of GWs, our model of the noise in the detectors, and the template waveform models used for data analysis. The thesis starts with a theoretical overview of GW physics, including an overview of GR, and how it describes the way GWs are generated and how they propagate and interact with matter. We then discuss the practical aspects of GW detection: the modelling of the noise in the detectors and the data analysis techniques used to extract and interpret GW signals. Next, we describe the SCoRe framework in Chapter 2, which is designed to distinguish between noise and deviations from the SM, while also shedding light on the underlying physics of the deviation. We detail its three main components: cross-correlating residual power between detectors, projecting onto physically motivated or agnostic bases, and combining information from multiple events by assuming a dependence of the unmodeled physics on the source parameters. To illustrate the method, we apply the SCoRe framework to toy models in Chapter 3. We demonstrate how the method can recover unmodeled signals without prior assumptions about their form, how to choose the timescale of cross-correlation, and how the method can be used to perform a null test of the SM. In Chapter 4, we then forecast the precision with which the SCoRe method can recover a deviation from the SM from a population of Binary Black Hole (BBH) mergers observed by a network of third-generation GW detectors. As the method leverages the dependence of the deviation from the SM on the source parameters, we investigate the effect the distribution of these parameters has on the method. For a model where the deviation scales with the chirp mass as a decaying power law, we show that the precision of the constraints on the deviation decreases as the power law becomes steeper. This has implications for constraining higher-dimensional operators in Effective Field Theories (EFTs) of gravity: higher dimensional operators correspond to steeper power laws and are, therefore, harder to constrain with the method. Finally, in Chapter 5, we illustrate another approach to testing the SM of GWs, where the GW signal in an alternative theory of gravity is numerically computed. We give an overview of the mathematical challenges and describe a method, the “fixing the equations” method, which aims to reduce pathologies in evolving EFTs of gravity by controlling energy flow to high frequencies.