Constraint Equations, Initial Data, and Gluing

The Einstein constraints define which intrinsic metric and extrinsic curvature can arise on a spacelike slice. This chapter treats that constraint surface as a geometric-elliptic problem: it identifies deformation obstructions, constructs solutions by conformal methods, distinguishes spatial from null asymptotics, and imposes gluing or trapped-boundary conditions without violating the field equations.

Helpful background. The Einstein Equation supplies the normal projections, Causality and Global Structure supplies developments and asymptotic boundaries, PDEs and Green Functions for Gravitation reviews elliptic equations, Analysis for Mathematical Relativity reviews weighted spaces, and Hypersurface Normals and Tangential Projectors fixes the geometry of $(h,K)$.

Admissible geometry on a slice

With future unit normal $n^\mu$, this chapter uses

$$K_{ij}=-h_i{}^\mu h_j{}^\nu\nabla_\mu n_\nu.$$

The vacuum constraints are

$$R(h)+K^2-K_{ij}K^{ij}=2\Lambda, \qquad D^j(K_{ij}-Kh_{ij})=0,$$

with the corresponding matter sources when present. These four equations are underdetermined but not freely solvable. Spatial diffeomorphisms create gauge directions; the adjoint linearized operator detects Killing initial data; topology, Yamabe type, asymptotics, and boundaries constrain existence.

Three recurring distinctions prevent overstatement:

• approximate splicing is not exact constraint gluing;
• a conformal ansatz is not a global one-to-one parametrization of all data;
• an inner excision surface is not automatically a MOTS, apparent horizon, or event horizon.

The operator, conformal, and deformation viewpoints are complementary rather than competing. The common framework—from normal projections through conformal elliptic equations and gluing—is developed in Bartnik and Isenberg 2004, §2, pp. 3–7 and §§4.1–4.6, pp. 14–28. Every construction below must finally be checked by substituting the reconstructed $(h,K)$ into both constraints with the same sign convention.

The six constructions in order

1. The Einstein Constraint Map identifies the linearized operator, its adjoint, and KID obstructions.
2. Solvability of the Conformal Constraint Equations gives the CMC classification and the limits of non-CMC results.
3. Conformal Thin-Sandwich Data encodes an intended trace-free metric velocity through shift and lapse data.
4. Asymptotically Flat and Hyperboloidal Initial Data separates spatial decay and ADM charges from conformal null asymptotics.
5. Gluing and Localized Deformation explains exact corrections, transition regions, and finite-dimensional balancing.
6. Inner Boundaries and Trapped-Surface Data derives the oriented null-expansion boundary condition.

Together they answer both halves of the construction problem: which local degrees of freedom may be specified, and which global or boundary compatibility conditions prevent them from extending to exact initial data.

Construction and deformation routes

For an operator-first route, begin with The Einstein Constraint Map. It explains why KIDs form finite-dimensional cokernel obstructions. Then read Gluing and Localized Deformation to see how nondegeneracy or charge balancing permits exact corrections.

For a construction route, read Solvability of the Conformal Constraint Equations followed by Conformal Thin-Sandwich Data. The first separates CMC classification from non-CMC gaps; the second translates intended metric velocity into lapse–shift-aware elliptic data.

For asymptotic or black-hole data, read Asymptotically Flat and Hyperboloidal Initial Data and Inner Boundaries and Trapped-Surface Data. These pages distinguish ADM from Bondi-type structures and derive the oriented null-expansion condition on an inner boundary.

The chapter constructs continuum data. Numerical elliptic discretization, solver convergence, and binary-data production belong to Computational Relativity; global black-hole interpretation belongs to the horizon and global-structure treatments.

Before choosing a method, specify what must remain fixed. Prescribing a conformal class and mean curvature suggests the conformal method. Requiring exact agreement with given data outside two small regions instead calls for a localized gluing theorem and its local no-KID hypothesis. Smooth vacuum data satisfying that hypothesis can acquire a neck while remaining exactly unchanged outside those regions Chruściel, Isenberg, and Pollack 2005, Theorem 1.1. The result allows non-CMC data, but the hypothesis excludes highly symmetric local geometries such as flat Minkowski data. Thus “small modification” describes the desired size of a change; it does not identify which elliptic solvability theorem applies.

Review the chapter

Suppose one wants to alter vacuum data only inside an annulus while leaving both sides exactly unchanged. Why is small constraint error not by itself sufficient?

Solution

The compactly supported correction must solve a linearized constraint equation whose cokernel is $\ker D\Phi^*$. Any local KID gives an orthogonality condition on the residual. Smallness permits nonlinear iteration only after those finite-dimensional obstructions are removed, usually by a no-KID hypothesis or by balancing model parameters and charges.

Once admissible data are available, Local Well-Posedness and Maximal Development proves that they generate a locally unique spacetime and then constructs the maximal globally hyperbolic development.

References

• Bartnik, Robert, and Jim Isenberg. “The Constraint Equations.” In The Einstein Equations and the Large Scale Behavior of Gravitational Fields, 1–38. Birkhäuser, 2004. arXiv:gr-qc/0405092.
• Chruściel, Piotr T., James Isenberg, and Daniel Pollack. “Initial Data Engineering.” Communications in Mathematical Physics 257 (2005): 29–42. DOI. Open PDF.

Further reading

• Carlotto, Alessandro. “The General Relativistic Constraint Equations.” Living Reviews in Relativity 24 (2021): 2. DOI.