Measuring the eccentricity of GW170817 and GW190425

TL;DR

This paper measures the eccentricity of two neutron star mergers, GW170817 and GW190425, finding low eccentricities consistent with formation in the field, but highlights the limitations of current methods due to prior dependence and waveform degeneracies.

Contribution

It provides the first direct gravitational-wave measurement of the eccentricity of GW170817 and GW190425, emphasizing the importance of full parameter estimation with accurate waveform models.

Findings

Eccentricity at 10 Hz is constrained to e ≤ 0.024 for GW170817 and e ≤ 0.048 for GW190425.

Results are consistent with binaries formed in the field, having nearly circular orbits.

Significant prior dependence indicates limited information in the current signals.

Abstract

Two binary neutron star mergers, GW170817 and GW190425, have been detected by Advanced LIGO and Virgo. These signals were detected by matched-filter searches that assume the star's orbit has circularized by the time their gravitational-wave emission is observable. This suggests that their eccentricity is low, but full parameter estimation of their eccentricity has not yet been performed. We use gravitational-wave observations to measure the eccentricity of GW170817 and GW190425. We find that the eccentricity at a gravitational-wave frequency of 10 Hz is $e \leq 0.024$ and $e \leq 0.048$ for GW170817 and GW190425, respectively (90% confidence). This is consistent with the binaries being formed in the field, as such systems are expected to have circularized to $e \leq 10^{-4}$ by the time they reach the LIGO-Virgo band. Our constraint is a factor of two smaller that an estimate based on GW170817 being detected by searches that neglect eccentricity. However, we caution that we find significant prior dependence in our limits, suggesting that there is limited information in the signals. We note that other techniques used to constrain binary neutron star eccentricity without full parameter estimation may miss degeneracies in the waveform, and that for future signals it will be important to perform full parameter estimation with accurate waveform templates.