Orbital Decay and Radiative Damping

Scrollable formula. Use Left and Right arrow keys to reveal hidden content.

Scrollable table. Use Left and Right arrow keys to reveal additional columns.

Scrollable figure. Use Left and Right arrow keys to inspect hidden content.

Orbital period decay is often presented as a direct observation of gravitational radiation. The timing fit actually returns an apparent $\dot P_b$ in the chosen barycentric time coordinate. Proper motion, differential Galactic acceleration, changing masses, and other line-of-sight effects also change the observed period. Only after those contributions and their covariance are removed may the intrinsic decay be compared with a radiative prediction. The empirical examples include results through 2026-08-19.

Required background. Probability and Statistical Inference supplies the likelihood, nuisance marginalization, and correlated uncertainty propagation needed for the corrected decay.

Helpful background. Weak-Field Gravity supplies the propagation and Galactic weak-field terms; Relativistic Two-Body Dynamics supplies the orbit; and Gravitational Radiation and Waveforms supplies the quadrupole-flux prediction.

The measured and intrinsic period derivatives

For an orbit whose line-of-sight velocity changes slowly, the observed derivative decomposes schematically as

\begin{aligned} \dot P_b^{\rm obs}&=\dot P_b^{\rm int}+\dot P_b^{\rm Shk}+\dot P_b^{\rm Gal},\\ \dot P_b^{\rm int}&=\dot P_b^{\rm rad}+\dot P_b^{\dot m}+\dot P_b^{\rm tide}+\cdots . \end{aligned}

The Shklovskii term is an apparent radial acceleration caused by transverse motion,

\frac{\dot P_b^{\rm Shk}}{P_b}=\mu^2D,

in units with $c=1$ , where $\mu$ is total proper motion and $D$ is distance. The Galactic term is $\dot P_b^{\rm Gal}/P_b=\Delta a_\parallel$ , the difference between the line-of-sight Galactic accelerations of the binary and the Solar-System barycenter. Restoring SI units divides both accelerations by $c$ . Spin-down mass loss, tides, mass transfer, or a changing gravitational coupling can supply additional intrinsic terms; whether they are treated as corrections or alternative signals depends on the declared hypothesis.

Here “intrinsic” includes every physical change of the binary’s orbital period. The radiation-only term $\dot P_b^{\rm rad}$ is obtained only after intrinsic mass loss, tides, or other declared processes are modeled as well. This prevents counting mass loss both inside $\dot P_b^{\rm int}$ and as an extra correction to it.

The positive Shklovskii sign follows directly from geometry. A source at distance $D$ with constant transverse velocity $v_\perp$ has distance $\sqrt{D^2+v_\perp^2t^2}\simeq D+v_\perp^2t^2/(2D)$ near the reference epoch. Its apparent radial acceleration is $v_\perp^2/D=\mu^2D$ . Thus, in SI units,

\frac{\dot P_b^{\rm Shk}}{P_b} =2.43\times10^{-21}\ {\rm s^{-1}} \left(\frac{\mu}{1\ {\rm mas\,yr^{-1}}}\right)^2 \left(\frac{D}{1\ {\rm kpc}}\right),

where mas denotes milliarcseconds and kpc denotes kiloparsecs. An illustrative one-day binary at $D=1$ kpc with $\mu=10$ mas per year has $\dot P_b^{\rm Shk}\simeq2.1\times10^{-14}$ . This is an apparent increase in period even though the source has no physical acceleration in this straight-line example.

General relativity predicts the leading quadrupolar decay

\dot P_b^{\rm GR}=-\frac{192\pi}{5}(G\mathcal M n)^{5/3} \frac{1+\tfrac{73}{24}e^2+\tfrac{37}{96}e^4}{(1-e^2)^{7/2}},

where $n=2\pi/P_b$ and $\mathcal M$ is the chirp mass. The masses used here must be inferred from PK observables not already circularly constrained by the same decay measurement. Higher-order radiative contributions enter when their size approaches the experimental uncertainty.

Freire and Wex set out the extrinsic-period correction in their 2024 review, §3.5.1 and Eq. (35). The Double Pulsar analysis applies the Shklovskii, Galactic, and spin-down mass-loss terms before forming its damping comparison, and it propagates distance and mass information through the prediction Kramer et al. 2021, §VI.B.1–5 and Eqs. (39)–(48). Thus the headline agreement is a property of a corrected timing and radiation model, not of the raw fitted derivative.

A likelihood for the damping ratio

Let the measured vector contain $\dot P_b^{\rm obs}$ , parallax or another distance indicator, proper-motion components, sky position, Keplerian and PK parameters, and any mass-loss proxy. Let $\boldsymbol\eta$ contain the Galactic-potential, ephemeris, timing-noise, and distance nuisance parameters. A transparent comparison uses

R(\boldsymbol\eta)= \frac{\dot P_b^{\rm obs}-\dot P_b^{\rm Shk}(\boldsymbol\eta)-\dot P_b^{\rm Gal}(\boldsymbol\eta)-\dot P_b^{\dot m}(\boldsymbol\eta)} {\dot P_b^{\rm GR}(m_p,m_c,e,P_b)}

inside a normalized joint likelihood for the original measurements. Under the GR null, $R=1$ after all corrections. Quoting an uncertainty obtained by adding independent error bars is inadequate when parallax, proper motion, masses, and timing parameters share covariance. Priors on distance and Galactic-potential parameters must be explicit, as must whether an interval is Bayesian credible or frequentist confidence.

The dataset identity includes the primary TOA release or the precisely described observing span, telescope/backend and profile calibration, clock correction, Solar-System ephemeris, dispersion and red-noise models, and the astrometric release supplying distance or proper motion. Epoch and frequency cuts, interference excision, conjunction handling, and profile-quality thresholds define the selection. A single well-timed system is not a complete sample of compact binaries; it tests damping in that system’s masses, eccentricity, environment, and timing regime.

Reproduction and adversarial checks

A bounded rerun recipe to reproduce the comparison begins with the published fitted quantities and covariance, not with a rounded ratio. Recompute each correction with explicit units, recover the eccentricity function in the GR flux, draw nuisance parameters jointly, and then form $R$ . Repeat with an independent distance estimate and at least two defensible Galactic-potential models. A synthetic injection with $R=1$ should be recovered without bias; a deliberately displaced acceleration should appear in the correction channel rather than being absorbed into radiation damping.

Important systematics include parallax zero point, proper-motion covariance, vertical and radial Galactic acceleration, Solar position and velocity, dispersion variation, red timing noise, clock and ephemeris errors, unmodeled mass loss, and PK mass covariance. A future distance revision, timing baseline, ephemeris, or Galactic-potential determination can move the inferred intrinsic derivative and therefore requires a fresh comparison. Agreement with quadrupole damping in one binary does not exclude dipole radiation that vanishes for equal charges or alternatives screened in that environment.

Exercise

A binary has $P_b>0$ , distance $D>0$ , and nonzero proper motion $\mu$ . Is its intrinsic decay more negative or less negative than the uncorrected observed value after removing only the Shklovskii term?

Solution

The Shklovskii contribution obeys $\dot P_b^{\rm Shk}=P_b\mu^2D>0$ , so it makes the observed derivative more positive. Removing it gives $\dot P_b^{\rm int}=\dot P_b^{\rm obs}-\dot P_b^{\rm Shk}$ when all other terms are temporarily neglected. Therefore the corrected intrinsic derivative is more negative than the raw value. This sign check is useful because accidentally adding the correction would weaken or even reverse an apparent damping signal.

References

Freire, P. C. C., and Wex, N. (2024), “Gravity Experiments with Radio Pulsars,” Living Reviews in Relativity 27, article 5. DOI.
Kramer, M., et al. (2021), “Strong-Field Gravity Tests with the Double Pulsar,” Physical Review X 11, 041050. DOI.