Parameter Estimation with Targeted Eccentric Numerical-Relativity Simulations for GW200208_22 and GW190620

TL;DR

This study analyzes gravitational wave events for eccentricity using numerical relativity simulations and Bayesian methods, finding evidence supporting eccentric orbits in GW200208_22 and GW190620.

Contribution

The paper introduces a new approach combining targeted numerical relativity simulations with Bayesian analysis to estimate eccentricity in gravitational wave events.

Findings

Eccentricity estimates for GW200208_22 around 0.2 at 20Hz.

Eccentricity estimates for GW190620 around 0.19 at 10Hz.

Targeted simulations improve likelihood matching for waveform analysis.

Abstract

We have analyzed LVK gravitational wave events that show some evidence of eccentricity from TEOBResumS modeling parameter estimations and have confronted them independently with full numerical generated waveforms from our bank of nearly two thousand simulations of binary black holes. We have used RIFT for Bayesian parameter estimation and found that GW200208_22 KDE estimates favor eccentricities $e_{20} = 0.198_{-0.180}^{+0.119}$ upon entering the LVK band at $\sim20$Hz within a $90\%$ confidence interval. Within this event analysis we employed 42 new targeted full numerical relativity simulations and we have thus found a top improved likelihood $\ln\mathcal{L}$ matching waveform, compared to model-based analysis, with an estimated eccentricity at 20Hz, $e_{20}=0.200$, thus reinforcing the eccentric hypothesis of the binary. We have also used our full bank of numerical waveforms on GW190620 finding that the KDE estimate favors eccentricities at 10 Hz in $e_{10}=0.190_{-0.186}^{+0.046}$. New specifically targeted simulations will be required to narrow these eccentricity ranges.