Assessing the imprint of eccentricity in GW signatures using two independent waveform models

TL;DR

This study analyzes gravitational wave signals from 162 sources using two waveform models to assess orbital eccentricity, finding limited evidence for eccentricity in some black hole mergers.

Contribution

It provides a comprehensive comparison of eccentricity estimates across two independent waveform models on a large dataset of GW events.

Findings

Consistent conclusions disfavor eccentricity hypothesis overall.

Potential eccentricity detected in three specific GW events.

Eccentricity evidence varies with waveform models and analysis settings.

Abstract

The gravitational wave signal from merging compact binaries encodes information about their orbital and intrinsic properties. Over the last few years, state-of-the-art waveform models have begun to incorporate the effects of orbital eccentricity into their estimated signal. Over a similar period, many groups have applied these waveforms to characterize whether the imprint of eccentricity is present and, if so, measure this time-evolving property (at a suitably-defined reference point). In this work, we present a comprehensive analysis of 162 confident sources identified in the O3 and O4a observing runs of the International Gravitational Wave Network (LIGO-Virgo-KAGRA). Using the RIFT parameter inference engine, we employ two independently implemented waveform models (SEOBNRv5EHM and TEOBResumS-Dali) which account for orbital eccentricity and the effects of aligned compact object spins. Using these two waveforms, we find consistent conclusions that disfavor the eccentric hypothesis. Unlike previous work, among binary black hole candidates, we find potential evidence for eccentricity in three events: GW200129, GW231001, and GW231123. For the latter two events, the evidence for eccentricity is ambiguous, with different degrees of support from different waveforms. Consistent with previous work, we find conclusions obtained about GW200129 can be sensitive to analysis settings, as expected, given the nonstationary noise present.