Numerical Initial Data and Equilibria

Numerical initial data translate a desired black hole, binary, or star into fields on one hypersurface that satisfy Einstein’s constraints and carry controlled physical content. The distinction between those two aims organizes this chapter. Elliptic convergence verifies the constraints; global charges, quasi-equilibrium diagnostics, surface regularity, junk radiation, and short evolutions test whether the solved slice represents the intended system. The chapter treats algorithms and evidence, not the continuum derivation or astrophysical interpretation of the data.

Helpful background. Formulations of General Relativity supplies constraint decompositions, Ordinary Differential Equations supplies nonlinear solution concepts, Hypersurfaces and Foliations supplies slice geometry, and Relativistic Matter and Compact Objects supplies black-hole and stellar equilibrium observables.

Enter this chapter

The Hamiltonian and momentum constraints are underdetermined until free data, topology, boundary conditions, and a decomposition are chosen. Consequently, two solutions can drive their residuals to zero yet differ in gravitational-wave content or global charges. Cook 2000, §§2–4 develops the major constructions and makes this freedom explicit.

Every calculation in the chapter records the conformal and hypersurface conventions, free fields, inner and outer boundaries, solver and precision, refinement sequence, independent constraint operator, charge surfaces, and rejection criteria. A solver’s iteration norm is never substituted for the continuum constraint. At least one check must change representation: an analytic isolated solution, independent discretization, global identity, or short evolution.

Route through the six pages

Elliptic Constraint Solvers begins with nonlinear solution, independent residuals, and the Schwarzschild conformal-factor benchmark. Conformal Decompositions of Initial Data then compares transverse–traceless and thin-sandwich choices while separating numerical error from genuine free-data differences.

Puncture and Excision Black-Hole Data contrasts two representations of black-hole interiors through exterior invariants and junk radiation. Quasi-Equilibrium Binary Initial Data adds orbital angular velocity and radial approach, closing an iterative loop with short evolutions and eccentricity fits.

Relativistic Stellar Equilibria uses TOV models, rotating-star virial identities, surfaces, and oscillation spectra as benchmarks. Mass, Momentum, Spin, and Constraint Checks finally combines local residuals with surface sequences, horizon measures, and relations whose stationarity assumptions are declared.

A hierarchy of evidence

The first layer is algebraic: nonlinear and linear solvers reach tolerances below discretization error. The second is local: Hamiltonian and momentum constraints decrease under resolution when re-evaluated independently. The third is global: ADM, Komar where applicable, and horizon quantities stabilize across surfaces and outer boundaries. The fourth is dynamical: a nominal equilibrium stays stationary or exhibits only convergent, understood gauge and truncation transients.

These layers are complementary. A stable ADM mass can coexist with a localized constraint defect; a tiny residual can coexist with the wrong asymptotic normalization; and a stationary coordinate profile can hide drift in invariant charges. For stars, Paschalidis and Stergioulas 2017, §2.7 and §4 shows how equilibrium construction, virial tests, and dynamical properties intersect.

The rerun record should preserve raw fields, domain maps, constraint norms, charge values on every surface, horizon data, solver histories, and environment metadata. The executable interfaces on these pages specify conforming implementations rather than completed runs; only recorded outputs and independent checks count as computational evidence.

Boundaries of the chapter

Constraint satisfaction does not establish astrophysical uniqueness, exact helical symmetry, low eccentricity, low junk radiation, or long-term evolution stability. Conformal flatness, maximal slicing, puncture topology, horizon boundary conditions, rotation laws, and EOS choices are modeling assumptions whose regime must be stated. The chapter equips the reader to quantify what a finite initial-data calculation supports and where that support ends.

Review the chapter

Diagnostic. A binary solve has eighth-order constraint convergence and stable horizon masses. Its ADM energy changes by one percent when the outer boundary doubles, while the measured eccentricity is $10^{-3}$. Can the eccentricity be quoted to two significant digits?

Solution

The evidence is insufficient. A one-percent ADM-energy change cannot be converted directly into an eccentricity error: the observables have different sensitivities, and the charge extraction itself may be responsible. Repeat the orbital fit at both boundary radii, separate extraction error from changes in the solved data, and cross that study with resolution. Two significant digits require a stable eccentricity estimate and an uncertainty small enough to support them. Local constraint convergence and horizon stability alone do not supply that uncertainty.

Exercises

For time-symmetric Schwarzschild puncture data, $\psi=1+m/(2r)$ with $m=GM$. Find the minimum of the areal radius $R=\psi^2r$, and use its area to recover the physical horizon mass. Explain what happens as $r\to0$.

Solution

Expanding gives $R=r+m+m^2/(4r)$, so $dR/dr=1-m^2/(4r^2)$ vanishes at $r=m/2$. There $R=2m$, $A=4\pi R^2=16\pi m^2$, and $\sqrt{A/(16\pi G^2)}=m/G=M$. As $r\to0$, $R\to\infty$: this initial slice approaches a second asymptotically flat end. Its puncture is not the spacetime curvature singularity. This simultaneously checks coordinate interpretation, horizon normalization, and the distinction between a physical mass and a geometric length.

References

• Cook, G. B. (2000). “Initial data for numerical relativity.” Living Reviews in Relativity 3, 5. DOI.
• Paschalidis, V., and Stergioulas, N. (2017). “Rotating stars in relativity.” Living Reviews in Relativity 20, 7. arXiv.