Detecting Tidal Resonances in Binary Neutron Stars

Abstract: As a binary neutron star inspirals due to the emission of gravitational waves, the rising tidal frequency resonantly excites vibrational modes. These oscillations are seismological probes of the rich stellar interior, yet it remains to be established whether gravitational-wave interferometers can measure them. Here, we present the first fully Bayesian study of the capability of the Einstein Telescope to detect tidal resonances. We simulate one year of observations and analyse the 200 loudest signals. We find that the Einstein Telescope can identify resonant modes and is sensitive to gravitational-wave phase shifts as small as $ΔΦ\approx 0.03$ for favourable events. We further show that neglecting resonances can bias the inferred tidal deformabilities. These results establish tidal resonances as a measurable route for asteroseismology with future detectors.

• The paper presents a Bayesian detection framework using Monte Carlo simulations on 200 BNS events to identify transient tidal resonances.
• It quantifies phase shifts in gravitational waveforms (ΔΦ ≈ 0.03–0.1) and assesses their impact on tidal deformability measurements.
• The study highlights the necessity of incorporating resonance modeling to avoid systematic bias in inferring neutron star matter parameters.

Bayesian Detection of Tidal Resonances in Binary Neutron Star Inspirals with the Einstein Telescope

Introduction

The prospect of directly probing the dense-matter equation of state (EoS) via neutron star (NS) gravitational-wave (GW) signals is a central objective for third-generation GW observatories such as the Einstein Telescope (ET). The study "Detecting Tidal Resonances in Binary Neutron Stars" (2606.06376) delivers a detailed Bayesian assessment of whether tidal resonances—transient excitation of NS oscillation modes—can be detected through GW analysis with the ET, and quantifies their effects on measurement of the system's intrinsic parameters. The novelty lies in the exhaustive simulation of one year of ET observations and statistical inference over the 200 highest signal-to-noise ratio (SNR) binary neutron star (BNS) coalescences.

Resonant Mode Excitation in Binary Neutron Stars

During the late inspiral of a BNS system, increasing GW frequencies bring the system through the resonance condition $|m|\Omega \simeq \omega_\alpha$ for specific non-radial stellar oscillation modes. The resulting mode excitation extracts orbital energy—generating a phase shift $\Delta\Phi_\alpha$ in the GW waveform that depends on stellar structure, the resonance mode’s mass-multipole overlap, and the orbital configuration (Equation 2 in the paper). For typical masses and radii, theoretically motivated phase shifts for g-modes and inertial modes are $\Delta\Phi \sim 0.01$–$0.1$. This regime is inaccessible to current second-generation detectors but promises observability with ET-class sensitivity.

Simulation Framework and Bayesian Analysis

Monte Carlo simulations of 200 BNS systems were performed, injecting mergers with a physical parameter distribution reflecting expected astrophysical rates, equations of state, and spin populations. Both resonance and non-resonance (control) populations were considered. Signals were injected into synthesized ET-D noise, and Bayesian inference conducted using the Bilby package and nested-sampling (dynesty) to estimate Bayesian evidences for two hypotheses: presence or absence of tidal resonance signatures.

Event detectability is determined by the logarithmic Bayes factor $x = \log B$, comparing resonance and no-resonance models. A false-alarm probability $P_{\mathrm{FA}}$ is defined using the background distribution from non-resonant signals (see below for the figure).

Figure 1: Distributions of Bayes factors $B$ for resonance (orange) and non-resonance (blue) populations, with a detection threshold set to $\ln B \approx 1.73$ corresponding to $5\sigma$ significance.

A five-sigma detection threshold was placed at $x_{\textrm{th}} \approx 1.73$, above which resonance detection is claimed. The detection efficiency is then directly quantified from the foreground population.

Statistical Sensitivity to Resonances

Analysis reveals that for the top 200 SNR events in one year of ET observations—$\Delta\Phi_\alpha$0–$\Delta\Phi_\alpha$1—tidal resonances will be confidently detectable in $\Delta\Phi_\alpha$232% of loud events. The limiting factor for detectability is the amplitude of the phase shift, not the resonant frequency. In detail, the efficiency function shows $\Delta\Phi_\alpha$3 at $\Delta\Phi_\alpha$4, rising rapidly with phase shift.

Figure 2: Log Bayes factor as a function of maximal injected phase shift $\Delta\Phi_\alpha$5, with the SNR of each event indicated by color. The threshold for detection is marked, and the minimal phase shift resolvable by ET is estimated at $\Delta\Phi_\alpha$6 for the loudest events.

Empirically, for favorable sources ($\Delta\Phi_\alpha$7), the minimal detectable resonance is $\Delta\Phi_\alpha$8. This is a substantial improvement over previous second-generation detector projections, and implies that dynamical tides due to physical g-modes and inertial modes are within the reach of terrestrial GW observation.

Bias in Tidal Parameter Inference

Neglect of resonant mode effects in waveform modeling induces systematic bias in inference of other NS matter parameters, notably the mass-weighted tidal deformability $\Delta\Phi_\alpha$9. The analysis demonstrates that for resonances with $\Delta\Phi \sim 0.01$0, omission of the phase jump leads to a measurable overestimation of $\Delta\Phi \sim 0.01$1.

Figure 3: Posterior distributions for the inferred mass-weighted tidal deformability $\Delta\Phi \sim 0.01$2 with (orange) and without (blue) resonance modeling. For a signal with $\Delta\Phi \sim 0.01$3, omission leads to significant bias.

The impact is most pronounced for high-SNR and large-resonance events; for smaller shifts or low-SNR signals, the bias diminishes.

Implications for Neutron Star Structure and Dense Matter Theory

Mode-resolved phase shifts measurable by ET span the theoretically expected range for NS g-modes and inertial modes. Recent microscopic EoS and oscillation-mode calculations estimate typical quadrupole moment overlaps and resonance frequencies compatible with the $\Delta\Phi \sim 0.01$4 values detectable by ET. Interface modes (core-crust) and discontinuity g-modes, associated with sharp EoS transitions, induce smaller shifts, potentially accessible only to detectors surpassing ET sensitivity or in joint ET–Cosmic Explorer observing campaigns.

The detectability of these resonances opens a new avenue for GW asteroseismology, providing a direct probe of composition gradients, stratification, and exotic matter phases through dense matter sensitivity of the relevant mode overlaps and frequencies.

Conclusions

This study establishes that fully Bayesian analyses with the sensitivity and bandwidth of the Einstein Telescope can robustly detect tidal resonances in BNS inspirals over realistic astrophysical populations. Phase shifts as small as $\Delta\Phi \sim 0.01$5 are observable in optimal cases, with substantial fractions of loud events yielding confident detections. These results elevate tidal resonance detection from theoretical possibility to observational reality in the third-generation era.

Beyond practical implications for GW parameter estimation—mandating resonance modeling to avoid tidal deformability bias—the results motivate more detailed nuclear-theory–informed modeling of mode spectra, with direct consequences for EoS inference in future multi-messenger campaigns.

Predicted future developments include implementation of hierarchical Bayesian models linking resonance parameters directly to dense-matter microphysics, joint analysis across detector networks, and expansion to include multiple mode excitations and non-adiabatic effects in waveform families suitable for imminent ET/CE observations.