Gravitational-Wave Propagation and Polarization Tests

Gravitational waves test the propagation and polarization content of the low-energy gravitational field over long baselines. Modified speed, damping, dispersion, birefringence, extra polarizations, and stochastic backgrounds correspond to different operators and require different detector-network observables. Agreement with general relativity constrains those models; it does not identify a microscopic completion.

Required background. Quantum-Gravity Observables and Test Taxonomy fixes the inference hierarchy. Graviton and Matter Nonlocal Form Factors supplies the EFT origin of modified propagation. Helpful background. Bubble Sources and Cosmological Gravitational-Wave Propagation separates source spectra from transport, and Tensor Modes and Primordial Gravitons supplies the cosmological tensor baseline.

From a wave equation to detector strain

For circular polarization $A=R,L$, a broad parameterization on an expanding background is

$$h_A''+(2+\nu_A)\mathcal{H}h_A' +\left[c_{T,A}^2k^2+a^2\mu^2\right]h_A=0.$$

The friction correction $\nu_A$ changes amplitude distance, $c_{T,A}$ changes arrival time and phase, and $\mu$ gives massive dispersion. A helicity dependence produces amplitude or velocity birefringence. Extra scalar or vector fields add polarization tensors rather than merely modifying the two tensor modes.

For a source with sky direction $\hat n$, detector $I$ records

$$d_I(f)=\sum_P F_I^P(\hat n,\psi) h_P(f;\vartheta,\alpha)+n_I(f),$$

where $P$ labels polarizations, $\vartheta$ source parameters, and $\alpha$ propagation coefficients. A network is essential: a single interferometer can trade polarization content against inclination, distance, and antenna response.

First application: dispersive waveform propagation

Take

$$E^2=p^2+m_g^2+A_\alpha p^\alpha .$$

Stationary-phase propagation changes the frequency-domain signal to

$$\widetilde h(f) =\widetilde h_{\mathrm{GR}}(f;\vartheta) \exp\!\left[i\,\delta\Psi(f;A_\alpha,m_g,z)\right].$$

For a massive graviton, the leading phase scales as $f^{-1}$ times a distance factor; a general power-law dispersion produces a corresponding frequency power, with logarithmic special cases. The first application is to propagate a compact-binary waveform through this kernel and evaluate the network likelihood

$$\log\mathcal{L} =-\frac12\sum_I \big(d_I-h_I\mid d_I-h_I\big)_I ,$$

jointly sampling masses, spins, eccentricity, calibration splines, waveform uncertainty, and the propagation coefficient. The noise-weighted inner product must use the analyzed data segment’s estimated spectrum.

The multimessenger neutron-star event GW170817 and its gamma-ray counterpart strongly constrained the difference between gravitational and electromagnetic propagation speeds at the observed frequencies Abbott et al. 2017. This does not force $c_T=1$ at every frequency or epoch; it constrains models after accounting for emission-time uncertainty and line-of-sight effects.

Polarization and stochastic-background logic

Tensor, vector, and scalar polarizations produce distinct antenna patterns. Null streams—linear combinations that cancel tensor responses for a known sky position—can test additional components, but their interpretation depends on network geometry and calibration. A non-GR polarization search constrains the tested signal family; it does not prove that unmodeled power is a new field.

For an isotropic stochastic background, cross-correlation between detectors has expectation

$$\langle \widetilde s_I^*(f)\widetilde s_J(f')\rangle \propto \delta(f-f')\sum_P \gamma_{IJ}^P(f)\Omega_P(f),$$

with overlap functions $\gamma_{IJ}^P$. Separating polarizations or spectral components requires sufficiently different baselines. Correlated magnetic or environmental noise is an adversarial alternative, not a small afterthought.

At the 10 August 2026 cutoff, the LVK O4b tests combined 77 new high-significance events with 91 previously tested events and reported consistency with general relativity across the tested channels LVK 2026 official summary. These are increasingly stringent null constraints on parameterized deviations, not evidence that ultraviolet quantum gravity has been observed.

Adversarial control: waveform error as dispersion

Inject a general-relativistic signal generated with a waveform family or eccentricity treatment absent from the recovery model. Add frequency-dependent calibration error within its measured uncertainty. If the analysis returns nonzero $A_\alpha$, the “dispersion” parameter is absorbing source or instrument error. Repeat with multiple waveform families, calibration realizations, detector subsets, and time slides.

A propagation interpretation strengthens if the coefficient is common across sources, scales with distance as predicted, survives waveform changes, and is consistent between detector networks. Even then the result is an anomaly in a low-energy operator until conventional propagation, environmental correlations, and alternative fields are excluded.

Common pitfalls

Treating every phase correction as propagation. Source-generation modifications can have the same frequency power. Redshift and distance scaling, multiple source classes, and a joint generation–propagation model are required.

Calling a tensor-mode detection a graviton detection. Classical gravitational waves are coherent excitations well described without resolving individual quanta. Their detection confirms dynamical tensor gravity, not microscopic graviton counting.

The chapter overview contains the structure diagram and validity and failure diagram. They are embedded there once so that their shared chapter-level context is not repeated on every article.

For the chapter-wide comparison of assumptions, counterevidence, falsifiers, and claim ceilings, see the claim-domain table.

References

• Abbott, B. P., et al. “Gravitational Waves and Gamma-Rays from a Binary Neutron Star Merger: GW170817 and GRB 170817A.” Astrophysical Journal Letters 848, L13 (2017). DOI.
• LIGO Scientific, Virgo, and KAGRA Collaborations. “Testing General Relativity with the Latest and Loudest Compact Binary Merger Observations.” Official O4b science summary (2026). Collaboration page.
• Mirshekari, S., N. Yunes, and C. M. Will. “Constraining Generic Lorentz Violation and the Speed of the Graviton with Gravitational Waves.” Physical Review D 85, 024041 (2012). DOI.