Gravitational waveform modelling for inference of spin precession in binary black hole mergers

Thomas, Lucy Maja ORCID: 0000-0003-3271-6436 (2023). Gravitational waveform modelling for inference of spin precession in binary black hole mergers. University of Birmingham. Ph.D.

Preview

Thomas2023PhD.pdf
Text
Available under License All rights reserved.
Download (42MB) | Preview

Abstract

Detections of gravitational waves from the mergers of compact binaries have revolutionised our understanding of the universe. As well as confirming a key prediction of Einstein’s Theory of General Relativity, they have allowed unique opportunities to probe strong gravity, observe previously dark populations of binary black holes, and constrain binary formation channels.

Spin precession occurs due to the coupling of the spins of individual black holes with their orbital motion, resulting in complex and rich dynamics that encode valuable astrophysical information. To date no confident detection of a precessing binary merger has been made, in part due to systematics in the waveform models used to infer the existence of precession. However precession has been shown to exist at the level of the population, making individual detections a tantalising possibility.

This thesis addresses two key questions in waveform modelling of precession. Firstly, we develop a strategy to reduce the high dimensionality of precessing binaries, which may be crucial for ensuring accuracy of precessing waveform models through calibration to numerical relativity. Secondly, we utilise the power of artificial neural networks to build a waveform model that faithfully mimics a highly accurate precessing multipolar waveform, with a fraction of the evaluation cost. Since the detection of precession is predicated on waveform models which accurately contain the relevant physics and can be practically used for inference, striking this delicate balance between efficiency and accuracy is vital.

Finally, we consider the impact of low frequency sensitivity upon detections of precession in a black hole binary population. Next-generation ground-based detectors are expected to improve upon the low frequency sensitivity of current instruments, providing access to more inspiral content in binary mergers. We show that this will allow us to converge upon the true underlying population spin distribution faster, which could lead to more accurate constraints upon the formation pathways of binary black holes.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):