New Zealand Statistical Association 2024 Conference

Amin Boumerdassi

The University of Auckland, Laboratoire des 2 Infinis (Toulouse)

Investigating the impact of gravitational wave glitches on the parameter estimation of extreme mass ratio inspirals

This is joint work with Ollie Burke, Matt Edwards, Avi Vajpeyi, Ruiting Mao

Extreme mass ratio inspirals (EMRI) are a type of astrophysical event in which a compact and massive object such as two black holes of highly non-equal masses merge into each other. These result in the emission of gravitational waves (GW) – the ripples in spacetime famously observed in 2015 by the LIGO collaboration. Many types of GW events can be parameterised by known astrophysical models, which permits estimation with Bayesian inference and MCMC. EMRIs are no exception to this, however parameter estimation of EMRIs is faced with numerous difficulties. These include things such as the 14-dimensional parameterisation of EMRIs, years-long observation times and highly multimodal likelihoods. One possible challenge for EMRI parameter estimation is the presence of GW glitches – a type of frequent, short-duration, high-amplitude noise event present in GW detectors. They are not true GW events, rather coming from local and often unexplained disturbances to the detector. For many types of GW sources, glitches are known to induce biases in the posterior distribution – however to date, no study has been conducted on the impact of glitches on EMRI parameter estimation. This project aims to investigate whether glitches result in biased EMRI posteriors, and to what degree this may be the case.