Compact Binaries and Gravitational Waves

This path is for graduate students and researchers who need to connect compact-binary dynamics to gravitational-wave observables across several approximation regimes. It begins by identifying invariant scales, moves through post-Newtonian, post-Minkowskian, effective-field-theory, self-force, effective-one-body, tidal, and radiation-reaction descriptions, and then crosses numerical relativity, waveform construction, and detector inference.

The goal is not to nominate one universally best method. It is to decide which small parameter and observable control a problem, how descriptions overlap, and where calibration, numerical error, detector noise, priors, and selection effects enter the conclusion.

Helpful background. Einstein equations supplies the geometric dynamics, Bianchi consistency, and stress–energy coupling used across approximation schemes. Ordinary differential equations supplies multiscale evolution, stability, singular-point, and boundary-value reasoning. Probability and statistical inference supplies likelihoods, priors, posterior checks, uncertainty propagation, and selection effects.

A binary is a chain of regimes and observables

A coordinate separation or effective potential can change under a gauge or canonical transformation, while invariant relations such as binding energy versus orbital frequency, scattering angle, redshift variables, radiated flux, and detector strain support cross-method comparison. The useful question is therefore not “Which formalism is true?” but “Which controlled expansion or calibrated representation predicts the declared observable in this region of parameter space?”

Every phase keeps source dynamics, radiation, numerical completion, waveform assembly, and detector inference as connected but distinct layers.

What you will be able to do

After completing the route and its assessment work, you should be able to:

1. Identify the scale hierarchy and validity domain of a compact-binary approximation.
2. Compare post-Newtonian, self-force, effective-one-body, finite-size, and numerical descriptions without treating them as interchangeable.
3. Connect source dynamics to waveform construction and detector inference.
4. Assess parameter uncertainty, model systematics, and selection effects in a cross-regime capstone.

Check your starting point

1. What distinguishes a coordinate orbital radius from an invariant frequency relation?
2. Can you identify the fast orbital timescale and slow radiation-reaction timescale in an adiabatic inspiral?
3. Why is a posterior distribution not the likelihood, and why does a detected sample not directly reproduce the source population?

Answers and review

1. A coordinate radius depends on representation; a relation between suitably defined observables can be compared across gauges or methods. Review the relativistic dynamics context through the Graduate General Relativity Self-Study Path.
2. Orbital motion varies on $T_{\rm orb}\sim\Omega^{-1}$, while dissipative parameters drift on a longer time $T_{\rm rr}$ when losses per orbit are small. Review multiscale ODE reasoning at Differential Equations for Relativity.
3. The posterior is proportional to likelihood times prior, with normalization by the evidence. Detection probability depends on source properties, so population inference must include a selection normalization. Review this at Probability and Inference for Gravity.

Route from orbital scales to detector inference

Phase 1 — Identify regimes and analytic descriptions

1. Scales, Invariants, and Regimes — define masses, mass ratio, velocities, compactness, timescales, bound and scattering sectors, and comparison observables.
2. Post-Newtonian Dynamics — organize weak-field, slow-motion corrections with gauge and truncation visible.
3. Post-Minkowskian Dynamics and Scattering — expand in gravitational coupling at arbitrary velocity and use asymptotic scattering data.
4. Worldline Effective Field Theory — separate compact-object scales, operators, matching coefficients, and regularization.
5. Gravitational Self-Force and Regularization — control the small-mass-ratio expansion and distinguish singular and regular fields.
6. Effective-One-Body Dynamics — assemble invariant analytic input and declared calibration into a resummed model.

Phase 2 — Finite size and secular evolution

7. Tides and Finite-Size Effects — connect object response coefficients to orbital operators, resonance, dissipation, and validity.
8. Radiation Reaction and Multiscale Evolution — compare local forces, balance laws, adiabatic and post-adiabatic evolution, and secular error.

Phase 3 — Strong-field completion and waveforms

9. Spacetime Evolution — connect formulation, gauge, boundaries, constraints, and horizon diagnostics to a controlled strong-field evolution.
10. Compact-Binary Waveforms — map source dynamics into strain modes through inspiral, merger, and remnant regimes.
11. Approximants, Hybrids, and Surrogates — track calibration domains, alignment, reduced representations, and waveform error.

Phase 4 — Inference and observed systems

12. Compact-Binary Parameter Estimation — separate source and detector frames, likelihood, priors, degeneracies, calibration, and waveform uncertainty.
13. Binary Black-Hole Observations — interpret event and population claims with selection and model dependence visible.

A representative reasoning pattern: estimate before choosing a method

For a leading-order quasi-circular binary, restore $c$ temporarily and define

$$x=\left(\frac{\pi GM f}{c^3}\right)^{2/3}, \qquad \frac{v}{c}\simeq x^{1/2},$$

where $M$ and the gravitational-wave frequency $f$ are in the detector frame. For $M=60M_\odot$ and $f=100\,\mathrm{Hz}$, use

$$\frac{GM_\odot}{c^3}\simeq4.93\,\mu\mathrm{s}.$$

Then

$$\frac{\pi GM f}{c^3} \simeq \pi(60)(4.93\times10^{-6}\,\mathrm{s})(100\,\mathrm{s}^{-1}) \simeq0.093,$$

so $x\simeq0.21$ and $v/c\simeq0.45$. The parameter is not uniformly tiny. This quick estimate warns that a low-order post-Newtonian truncation alone may be inadequate near this frequency and motivates resummation, numerical input, or an error comparison.

The estimate is a regime diagnostic, not an error bar. Spin, eccentricity, mass ratio, higher-order coefficients, waveform harmonic, and the distinction between source- and detector-frame mass all matter before drawing a quantitative accuracy conclusion.

Check your understanding

Checkpoint 1. Compare without identifying

How would you test a post-Newtonian and a self-force result in an overlap regime without comparing gauge-dependent orbital radii term by term?

Conceptual hint

Choose an observable both methods predict.

Setup hint

Expand in the symmetric mass ratio or small mass ratio at fixed invariant frequency.

Next-step hint

State truncation orders and translate conventions first.

Solution

Select an invariant relation such as binding energy $E(\Omega)$, periastron advance, redshift, or scattering angle. Translate mass, spin, frequency, and normalization conventions, then expand the PN expression in the small-mass-ratio limit or expand the self-force result in the weak field. Compare coefficients only in the shared domain and propagate both truncation errors. Agreement tests the overlap; disagreement must first be separated from gauge, variable, and subtraction differences.

Checkpoint 2. From waveform error to population bias

Why can a small mismatch between two waveform models fail to guarantee unbiased population inference?

Conceptual hint

Event-level and population-level questions use different accumulations of information.

Setup hint

Mismatch depends on detector noise and alignment.

Next-step hint

Bias can correlate across many events and selection can be model dependent.

Solution

A mismatch is a noise-weighted waveform-distance diagnostic under specified detector sensitivity, parameter alignment, and signal. It is not automatically a bound on parameter bias. A coherent modeling offset can shift individual likelihoods in the same direction, accumulate across events, and alter detection efficiency or the selection function. Population inference therefore requires injection or predictive tests that propagate waveform, calibration, noise, prior, and selection uncertainties through the hierarchical likelihood.

Transfer task — Move from inspiral to scattering

Take a method you used for a quasi-circular inspiral and apply it to an unbound encounter. State which variables and checks transfer and which parts of the formulation must be replaced.

Solution outline

Mass ratios, spins, asymptotic energy and angular momentum, dimensional analysis, flux conservation, and a weak-field overlap can transfer. Orbital frequency as a global label, adiabatic circular balance, and a phase accumulated over many quasi-circular cycles generally do not. Replace them by scattering angle, impulse, closest-approach regime, radiated four-momentum, and capture versus escape conditions. A PM organization may be more natural than a velocity expansion. If dissipation changes an otherwise unbound orbit into capture, compare incoming and outgoing invariant data rather than interpreting a coordinate trajectory alone.

Capstone — Cross-regime binary claim

Choose a binary with declared masses, spins, orbital sector, and frequency or impact-parameter range. Build a traceable prediction from dynamics to inference that:

• identifies at least two approximation parameters and their domains;
• compares two analytic descriptions through a shared invariant;
• states how finite-size or absorption effects enter, even if negligible;
• uses or describes numerical-relativity completion with constraint and flux checks;
• builds a waveform with an explicit error budget;
• propagates waveform, detector, prior, and selection uncertainty to a bounded source or population conclusion.

Criterion	What a complete solution shows
Regime control	Small parameters, domains, overlaps, and breakdown criteria are quantitative.
Cross-method comparison	A shared invariant and translated conventions support the comparison.
Dynamics-to-waveform chain	Numerical or analytic inputs, hybridization, and error propagation are explicit.
Inference	Likelihood, prior, nuisance uncertainty, and selection support no stronger claim than warranted.

Completion and onward routes

Completion requires both checkpoints, the scattering transfer task, prediction-and-check records for the strong-field and waveform calculations, and a capstone satisfying all four criteria.

Continue to Black-Hole Perturbation Theory for response coefficients and small-mass-ratio foundations, Computational Relativity for production evolutions and waveform extraction, or Relativistic Matter and Compact Objects for neutron-star structure and material response.