Short Gravitational-Wave Transients as Probes of Cosmic Domain Walls

Source: arXiv:2606.06478 · Published 2026-06-04 · By Tore Boybeyi, Doga Veske, David Maibach

TL;DR

This paper addresses the anomaly of very short gravitational-wave transients GW190521 and GW231123, whose inferred parameters (high total masses and spins) challenge standard binary black hole (BBH) formation theories. The authors propose and test a novel alternative hypothesis that these signals arise from cosmic domain walls—macroscopic topological dark matter (TDM) defects formed by scalar fields with discrete vacua. They develop and fit physically motivated domain wall signal templates to LIGO Hanford and Livingston strain data, comparing the domain wall hypothesis to the conventional BBH interpretation using Bayesian model selection.

Although the BBH model remains favored for both events with Bayes factors of log10(B_BBH/TDM) = 12.2 and 11.3, these values are anomalously weak compared to simulated BBH injections, suggesting morphological degeneracies. Importantly, a first joint multi-event fit constraining shared dark matter parameters across both GW190521 and GW231123 finds consistent scalar field mass and photon coupling parameters, supporting the possibility of a common underlying domain wall origin. Moreover, injections of domain wall signals recovered under the BBH hypothesis yield artificially large spin parameters, revealing a morphological degeneracy that could mask genuine domain wall signals with standard BBH templates. This work pioneers a multi-event parameter consistency test for topological dark matter with gravitational-wave detectors, offering a new discriminant in future analyses as short transient catalogs grow.

Key findings

• Bayesian model selection favors binary black hole over topological dark matter hypothesis with log10 Bayes factors 12.2 (GW231123) and 11.3 (GW190521), but these are weaker than expected for true BBH signals.
• Joint multi-event fit constrains shared dark matter scalar mass m_phi ≈ 8 x 10^-13 eV and photon sector parameter m_0 ≈ 6 x 10^-12 eV consistent across GW190521 and GW231123 within 0.07 dex (0.4 sigma).
• Injected domain wall signals are systematically recovered as BBH signals with artificially large spin magnitudes, indicating strong morphological degeneracy between hypotheses.
• Signal duration scale from domain wall thickness and velocity (v_DW ~ 0.026c) matches LIGO sensitivity band and inter-site delay constraints for coherent detection.
• The domain wall template’s intrinsic temporal symmetry limits its ability to reproduce asymmetric chirp morphologies characteristic of BBH signals, partially explaining lower likelihoods.
• Observed domain wall coupling g_DW posterior median ~10^-27 corresponds to extremely small fractional variation in fine-structure constant, far below current atomic clock detection sensitivity.
• Local domain wall density estimates (~1 per 100 AU) are compatible with subdominant dark matter contribution respecting Zel’dovich bound and existing laboratory constraints.
• Multi-event parameter consistency testing provides a new discriminant that can sharpen with more short-duration transient detections beyond traditional BBH triggers.

Methodology — deep read

The authors begin with a physical model of topological dark matter in the form of cosmic domain walls formed by a real scalar field with a discrete periodic potential. The scalar mass m_phi and the periodic potential parameters set the domain wall thickness that determines the transient signal duration. Interactions with the electromagnetic sector are modeled via couplings that induce transient variations in the fine-structure constant and an effective photon mass term, yielding changes in phase and optical path length in interferometers.

The domain wall crossing signal template is described analytically as a sinusoidal envelope modulated by a kink soliton profile traveling with velocity v_DW and direction ˆn, producing phase shifts in the detector arms measurable as strain. The template has 8 parameters including m_phi, the photon sector parameter m_0, coupling g_DW, wall velocity/direction, absolute time t_0, and an integer N_ratio controlling oscillation cycles.

The data used comes from publicly released strain data from LIGO Hanford (H1) and Livingston (L1) observatories during the first part of the fourth observing run. Around each event trigger (GW231123 and GW190521) 1.6-second segments of 10-200 Hz whitened strain data are analyzed. Bayesian inference is performed with the bilby software and dynesty nested sampler to calculate posterior distributions and model evidences for both the domain wall template and the BBH waveform model (NRSur7dq4).

Model comparison is quantified by Bayes factors between BBH and domain wall hypotheses. To calibrate significance, injections of maximum-a-posteriori BBH waveforms in nearby noise segments are analyzed to form a background distribution. A joint analysis constraining a single common scalar field across both events is conducted, fitting domain wall parameters m_phi, m_0, and g_DW as shared while allowing event-specific wall velocities and directions.

Reconstruction of injected domain wall signals under the BBH hypothesis reveals morphologies that mimic highly spinning BBHs, demonstrating a degeneracy that complicates discrimination with standard pipelines optimized for BBH triggers. The study notes selection bias due to BBH template searches only targeting near-coincident (≤10 ms) signals, whereas domain walls may produce longer inter-site delays (~seconds).

The analysis pipeline is fully described with data conditioning steps (whitening, bandpass), priors (physically motivated), likelihood formulation assuming Gaussian noise, and detailed domain wall detector response modeling including size, dispersive, and center-of-mass acceleration effects. Posterior corner plots and robustness tests are reported in supplemental materials. The code release status is not specifically stated, but data used is public LIGO open data.

Technical innovations

• Introduction of a physically motivated time-domain domain wall crossing template in gravitational-wave strain data, parameterized by scalar mass, photon mass coupling, and wall velocity.
• First joint Bayesian fit constraining shared topological dark matter parameters across multiple short gravitational-wave transients to test a common scalar field origin.
• Demonstration that domain wall signals can be morphologically degenerate with very short, high-spin binary black hole waveforms, leading to systematic misclassification under standard BBH templates.
• Proposal and application of multi-event parameter consistency tests as a new discriminant in dark matter searches using high-precision interferometric data.

Datasets

• LIGO Hanford and Livingston strain data — publicly available — from first part of fourth observing run (O4 segments around GW190521 and GW231123 triggers)

Baselines vs proposed

• BBH baseline injection for GW231123: median log10 Bayes factor BBBH/TDM = 23.1 vs observed log10 Bayes factor = 12.2
• BBH baseline injection for GW190521: median log10 Bayes factor BBBH/TDM = 40.9 vs observed log10 Bayes factor = 11.3
• SNR for GW231123: SNR_BBH = 20.8 vs SNR_TDM = 20.1 (3% difference)
• SNR for GW190521: SNR_BBH = 15.2 vs SNR_TDM = 14.5 (about 5% difference)

Figures from the paper

Figures are reproduced from the source paper for academic discussion. Original copyright: the paper authors. See arXiv:2606.06478.

Fig 2: Posteriors on the shared dark matter parameters and

Fig 2 (page 5).

Fig 3 (page 5).

Fig 4 (page 5).

Fig 5 (page 5).

Fig 6 (page 5).

Fig 1: BBH posterior for GW231123 using the NRSur7dq4

Fig 5: shows the resulting Bayes-factor distributions.

Limitations

• Bayes factors favor BBH but are anomalously low compared to injection baselines, indicating current methods may insufficiently discriminate domain wall signals.
• Domain wall template enforces temporal symmetry limiting ability to model the asymmetric chirp structure of BBH signals, biasing fits against TDM hypothesis.
• Analysis limited to two events; statistical power and parameter consistency tests require a larger catalog of short-duration transients to improve discriminative power.
• Selection bias exists because coincident triggers are formed within light-travel time windows (~10 ms), while domain walls may produce delays on order seconds.
• Coupling g_DW remains largely prior dominated due to extremely weak expected signal amplitudes, limiting precision of amplitude constraints.
• No dedicated unmodeled or domain wall-specific search pipeline was implemented; results rely on comparison to BBH template-based analyses.
• Scalar field coupling only modeled through photon sector interactions; inclusion of fermion couplings or multi-channel effects is left for future work.

Open questions / follow-ons

• Can dedicated unmodeled or domain wall-specific search pipelines improve sensitivity and discrimination of short transient topological dark matter signatures?
• How will increasing the catalog of short-duration gravitational-wave events impact the statistical power of multi-event parameter consistency tests for domain wall dark matter?
• What is the impact of including higher-dimensional or fermion-sector couplings of scalar fields on the domain wall signal morphology and detection prospects?
• How can future detectors or multi-network timing enable discrimination of domain wall induced longer inter-site delays characteristic of sub-luminal wall velocities?

Why it matters for bot defense

For bot-defense or CAPTCHA practitioners, this work exemplifies a rigorous approach to discriminating morphologically similar but physically distinct transient signals by leveraging multi-event parameter consistency and joint Bayesian model comparisons. While the domain is gravitational-wave physics, the principle of building and validating physically motivated alternative signal templates alongside standard models—and testing for systematic misclassification or degenerate morphologies—is broadly applicable to anomaly detection and bot-defense mechanisms.

The identification of parameter degeneracies that may mask genuine nonstandard signals highlights the importance of interpretability and multi-event corroboration in complex classification tasks. Similarly, the use of multi-source spatial and temporal correlations to enhance detection fidelity could inspire analogous strategies in multi-channel bot or CAPTCHA challenge response analyses, where coordinated temporal patterns or parameter correlations across sessions may signal genuine or spoof behaviors.

Cite

@article{arxiv2606_06478,
  title={ Short Gravitational-Wave Transients as Probes of Cosmic Domain Walls },
  author={ Tore Boybeyi and Doga Veske and David Maibach },
  journal={arXiv preprint arXiv:2606.06478},
  year={ 2026 },
  url={https://arxiv.org/abs/2606.06478}
}

Read the full paper

• arXiv:2606.06478 (abstract)
• arXiv:2606.06478 (PDF)