Initial-Boundary Value Problems in AdS and Finite Domains

Einstein evolution on a domain with timelike boundary requires a joint choice of initial geometry, hyperbolic reduction, gauge, incoming physical data, incoming constraint control, and corner compatibility. A finite artificial worldtube and the conformal boundary of an asymptotically AdS spacetime are geometrically different, but both lie outside the domain of dependence of the initial slice alone after causal signals from the wall can arrive. Boundary data are therefore part of the continuum problem, not a numerical afterthought.

Helpful background. The Einstein Equation supplies constraints and evolution; Hypersurface Normals and Tangential Projectors supplies initial, timelike, and corner geometry; The Gibbons–Hawking–York Boundary Term supplies the variational boundary problem; PDEs and Green Functions for Gravitation supplies characteristic and energy methods; Formulations of General Relativity orients the volume; and Causality and Global Structure supplies domains of dependence and timelike infinity.

Geometry and analytic structure

Let a spacelike slice $S_t$ meet a timelike boundary $\mathcal T$ in a two-surface $C_t$. With outward unit spacelike normal $s^a$, the boundary metric

$$h_{ab}=g_{ab}-s_as_b$$

and boundary second fundamental form

$$\mathcal K_{ab}=h_a{}^ch_b{}^d\nabla_cs_d$$

obey Gauss–Codazzi constraints. They cannot both be prescribed arbitrarily. After choosing a hyperbolic reduction, the boundary normal symbol divides the evolution variables into incoming, outgoing, and tangential characteristic fields. Only the incoming fields accept independent boundary data, and their allowed relations must close an energy estimate.

Friedrich and Nagy turn a geometric vacuum problem with smooth timelike boundary into a maximally dissipative symmetric-hyperbolic system and prove a local existence result under smooth compatibility conditions in Friedrich and Nagy 1999, §§3–8 and Theorem 8.1, pp. 624–652. Their precise data and uniqueness qualifications belong to that frame formulation; other reductions use different characteristic variables.

A coherent route through the chapter

The six leaves form a dependency chain:

1. Timelike Boundary Geometry and Data decomposes intrinsic, extrinsic, cut, and characteristic data and checks the boundary constraints.
2. Boundary Data and Constraint Preservation in Einstein Evolution separates initial, gauge, physical, constraint, and corner layers in a complete reduced IBVP.
3. Constraint-Preserving Boundary Conditions derives the subsidiary incoming-mode condition and the zero-constraint energy implication.
4. Radiative, Reflective, and Dissipative Boundaries separates continuum stability from physical reflection and finite-radius error.
5. Asymptotically AdS Boundary Data gives the dimension-specific Fefferman–Graham coefficients and radial constraints for four bulk dimensions.
6. Corner Compatibility and Well-Posedness constructs the differentiated compatibility hierarchy and relates its depth to regularity.

Read the first three before choosing a radiative condition: a boundary formula has no analytic meaning until the incoming variables and subsidiary system are known. Read the final corner page after any boundary choice, because changing the boundary operator changes its compatibility jet.

Four tests for every boundary prescription

Test	Required evidence	A failure indicates
Geometric admissibility	Gauss–Codazzi and initial constraints	Candidate data do not embed in an Einstein solution
Main-system stability	Boundary symbol and energy estimate	No continuous dependence or controlled trace
Constraint preservation	Complete incoming subsidiary inventory	Boundary-driven violation of Einstein equivalence
Physical fidelity	Reflection/flux calculation in physical fields	Stable but inaccurate exterior modeling

Corner compatibility is a fifth test. It asks whether the initial and boundary traces and their differentiated equations describe one solution at $C_0$. Kreiss and collaborators provide a structured harmonic example with strong boundary estimates in Kreiss et al. 2007, Theorems 1–3, pp. 3–10, PDF.

Finite worldtubes versus AdS infinity

A finite artificial worldtube removes an exterior region that contains independent Cauchy data and backscatter. A local Sommerfeld condition can only approximate that exterior; moving the worldtube tests the resulting modeling error. An asymptotically AdS conformal boundary is selected by the geometry, but it is timelike and likewise requires a boundary condition. Common stationary fixed-boundary-metric data are reflective and create a closed cavity. A time-dependent prescribed metric can drive the bulk; Dirichlet data alone do not imply conserved energy.

Near four-dimensional asymptotically locally AdS infinity,

$$G=\frac{\ell^2}{z^2} \left[dz^2+ (g_{(0)}+z^2g_{(2)}+z^3g_{(3)}+\cdots)_{ij}dx^idx^j \right].$$

The leading conformal metric fixes $g_{(2)}$, while the radial equations make $g_{(3)}$ transverse and traceless. These formal asymptotic data do not replace the hyperbolic boundary condition or its corner compatibility.

Scope and limitations

The chapter states continuum local architecture. It does not supply a production discretization, an exact absorbing condition at finite radius, complete low-regularity theorems, or a holographic interpretation of AdS coefficients. A discrete implementation needs its own stability and convergence analysis. Long-time existence, global geometric uniqueness, and nonlinear AdS stability lie beyond the local boundary results developed here.

Review the chapter

Suppose a boundary condition makes the main reduced Einstein system strongly well posed and sets two physical incoming fields to zero. Name three further checks required before concluding that it defines a smooth vacuum Einstein evolution.

Solution

First, derive the subsidiary constraint system and ensure that every incoming constraint field has homogeneous boundary data compatible with zero constraints; main-system stability alone does not prove $C=0$. Second, supply admissible incoming gauge data so the boundary prescription is complete rather than underdetermined. Third, match the initial and boundary data at the corner through every order required by the claimed regularity. One should also check that the initial Hamiltonian and momentum constraints hold. If the boundary is intended to model an isolated exterior, a reflection calculation is needed even after these mathematical checks pass.

References

• Friedrich, Helmut, and Gabriel Nagy. “The Initial Boundary Value Problem for Einstein’s Vacuum Field Equation.” Communications in Mathematical Physics 201 (1999): 619–655. DOI.
• Kreiss, Heinz-Otto, Oscar Reula, Olivier Sarbach, and Jeffrey Winicour. “Well-Posed Initial-Boundary Value Problem for the Harmonic Einstein Equations Using Energy Estimates.” Classical and Quantum Gravity 24, no. 23 (2007): 5973–5984. DOI; Open PDF.