Constraints on nonlinear tides due to $p$-$g$ mode coupling from the neutron-star merger GW170817

TL;DR

This paper investigates the potential impact of nonlinear p-g mode coupling tides on gravitational-wave signals from neutron star mergers, using GW170817 data to constrain the effect and find it is strongly disfavored in measurable regions.

Contribution

It provides the first Bayesian analysis constraining nonlinear tide effects from p-g mode coupling in GW170817, highlighting degeneracies and disfavoring detectable nonlinear tide effects.

Findings

GW170817 data is consistent with p-g mode coupling due to parameter degeneracy.

Regions with measurable nonlinear tide effects are strongly disfavored by Bayes factors.

Parameter degeneracy explains the apparent consistency with nonlinear tide models.

Abstract

It has been suggested by Weinberg et al. (2013) that an instability due to the nonlinear coupling of a neutron star's tide to its $p$- and $g$-modes could affect the gravitational-wave phase evolution of a neutron-star binary. Weinberg (2016) suggests that this instability can turn on as the gravitational-waves pass through the sensitive band of ground-based detectors, although the size of the effect is not known. The discovery of the binary neutron star merger GW170817 provides an opportunity to look for evidence of nonlinear tides from $p$-$g$ mode coupling. We compute Bayesian evidences that compare waveform models that include the $p$-$g$ mode coupling to models that do not. Using the waveform model and priors of Essick et al. (2016) we find that the observation of GW170817 is consistent with $p$-$g$ mode coupling, in agreement with Abbott, B. P., et al. (2019a). We investigate the properties of the model and find that this consistency is due to degeneracy in a large region of the parameter space between the the model that includes nonlinear tides from $p$-$g$ mode coupling and the standard post-Newtonian model that does not. We investigate the consistency of the GW170817 signal with regions of the parameter space where the effect of nonlinear tides is not degenerate with the standard model. Regions of the nonlinear tide parameter space that have a fitting factor of less than 99$\%$ (98.5$\%$) are disfavored by a Bayes factor of 15 (25). We conclude that the consistency of the GW170817 signal with the model of Essick et al. (2016) is due to parameter degeneracy and that regions where nonlinear tides produce a measurable effect are strongly disfavored.