Trick or Treat?
: Juggling Noise Artefacts in Gravitational-Wave Data

  • Simone Mozzon

Student thesis: Doctoral Thesis


This is an exciting time for gravitational-wave astronomy. Since the first observation of a gravitational-wave signal from a binary black hole merger, the rate of detections has increased by over an order of magnitude. However, detecting gravitational waves and estimating the properties of their sources still present huge challenges. The output of gravitational-wave interferometers,
such as LIGO and Virgo, is dominated by noise which can be strongly modulated by disturbances from the environment. Consequently, interferometric data present non-Gaussian transients and also longer duration artefacts. Both can jeopardise the accuracy of astrophysical analyses, which generally assume that the noise is Gaussian and stationary.
In this thesis I investigate how noise transients affect searches for gravitational-wave signals from compact binary coalescences, proposing new methods to mitigate their effects. I present a new approach to include information from LIGO auxiliary sensors in the PyCBC search for compact binary coalescences. This helps the search to deal with periods of poor data quality, increasing the number of detectable gravitational-wave events by up to 20%. In addition, I included
corrections to account for slow variations of the detector noise, also called non-stationarity, further increasing the sensitive volume of the PyCBC search by 5%.
Finally, I examine how non-stationarity impacts the estimation of gravitational-wave source parameters. Focusing on parameters relevant to cosmological analysis with gravitational waves, I demonstrate that variations in the noise can bias the estimation of the luminosity distance by up to 6.8%. This work demonstrates that assessing the stationarity of gravitational-wave data
is crucial to obtain accurate estimates of the signals’ properties.
Date of Award14 Feb 2023
Original languageEnglish
Awarding Institution
  • University of Portsmouth
SupervisorLaura Nuttall (Supervisor), Ian Harry (Supervisor) & Kazuya Koyama (Supervisor)

Cite this