Black-Hole Geometry and Mechanics

Black holes are not defined by a large curvature, a dark appearance, or a particular coordinate radius. They are causal and geometric structures whose meaning depends on the question being asked. This volume builds a controlled path from the global black-hole region to stationary and quasi-local horizons, then develops the charges, variational identities, classification results, extremal limits, and extraction processes that constitute classical black-hole mechanics. The central discipline is to state the hypotheses under which two horizon notions coincide—and to keep them distinct when those hypotheses fail.

Helpful background. Exact Solutions of Einstein’s Equations supplies the named solutions used as recurring checks; Causality and Global Structure supplies conformal infinity and global causal boundaries; The Einstein Equation supplies the field equation and energy conditions used in horizon mechanics.

The geometric question

In an asymptotically flat spacetime with a suitable future null infinity $\mathscr I^+$, the global definition begins with the points from which a future-directed causal signal can escape. The black-hole region and its future event horizon are

$$\mathcal B=\mathcal M\setminus J^-(\mathscr I^+), \qquad \mathcal H^+=\partial J^-(\mathscr I^+).$$

This definition is exact, but it is intentionally global. Determining whether a point belongs to $J^-(\mathscr I^+)$ requires the full future development of the spacetime. Consequently, an event horizon cannot in general be found from measurements in an arbitrarily small neighborhood, and its location can react to matter that has not yet arrived. The opening chapter, What Is a Black Hole?, develops this statement without confusing it with curvature singularities, trapped surfaces, or the surface of a material body.

The same formula cannot simply be copied into every setting. Anti-de Sitter spacetime has a timelike conformal boundary; causal reachability is defined relative to a chosen boundary component, while field evolution additionally requires boundary conditions. De Sitter observers encounter cosmological horizons rather than an asymptotically flat $\mathscr I^+$; numerical evolutions usually need a surface that can be located on current slices. The question “where is the black hole?” is therefore incomplete until asymptotics, available data, slicing, equilibrium, and intended observable are specified. Booth’s comparison in Booth 2005, §§2–5 is a useful guide to the resulting hierarchy.

Three layers of horizon structure

The volume uses three layers, none of which automatically replaces the other two.

Global horizons are causal boundaries. Their definition is invariant and does not require stationarity, but it depends on the complete spacetime and on the chosen asymptotic region. Event, Killing, Apparent, Trapping, Isolated, and Dynamical Horizons gives a side-by-side taxonomy and identifies the assumptions behind familiar coincidences.

Stationary horizons exploit symmetry. If a Killing field $\chi^a$ becomes normal to a null hypersurface, that hypersurface is a Killing horizon. For an asymptotically flat rotating black hole one normally fixes $\chi^a=t^a+\Omega_H\phi^a$ using the unit time translation at infinity and the $2\pi$-periodic axial field. Its non-affinity defines surface gravity,

$$\chi^b\nabla_b\chi^a=\kappa\chi^a \quad\text{on the horizon}.$$

Because $\kappa$ rescales with $\chi^a$, normalization is physical data, not a decorative convention. Equilibrium Black Holes and Horizon Anatomy extracts invariant horizon data from Schwarzschild, Reissner–Nordström, Kerr, and Kerr–Newman, while Killing Horizons and Bifurcate Geometry develops generators, bifurcation surfaces, adapted coordinates, and the degenerate/nondegenerate distinction.

Quasi-local horizons use the geometry of closed spacelike sections, most often their two future null expansions. They can be located without knowing the entire future and can describe a black hole that is locally in equilibrium even while radiation remains elsewhere. Their advantages come with choices: foliation, null-normal scaling, outer/inner conditions, and signature must all be recorded. Quasi-Local and Isolated-Horizon Mechanics develops this layer in detail. The modern status and remaining transition questions are reviewed by Ashtekar and Krishnan in Ashtekar and Krishnan 2025, §§2–5.

The mechanics spine

Once the horizon, generator, asymptotics, and field theory are fixed, geometry becomes mechanics. Conserved Charges, Smarr Relations, and First Laws compares ADM, Komar, horizon, and covariant-phase-space charges and explains why their agreement requires hypotheses. In four-dimensional asymptotically flat Einstein–Maxwell theory, with the standard normalization and variations among stationary solutions, the first law takes the form

$$\delta M =\frac{\kappa}{8\pi G}\,\delta A_H +\Omega_H\,\delta J +\Phi_H\,\delta Q.$$

Each term has the dimensions of energy, and the formula is invariant under an electromagnetic gauge shift only after the potential is defined relative to the chosen boundary. It is a variational identity on a specified solution space, not an unrestricted finite-difference formula. The related Smarr identity follows from scaling in this particular theory and dimension; it is not a universal equation for arbitrary couplings or asymptotics.

The Laws of Black-Hole Mechanics states the zeroth, first, second, and third laws with their separate hypotheses. The area theorem, for example, combines the Raychaudhuri equation with an energy or null convergence condition and global predictability assumptions. It does not say that every marginal surface on every slice grows. Bardeen, Carter, and Hawking give the classical architecture in Bardeen, Carter, and Hawking 1973, Open PDF, pp. 161–170; Hawking radiation, generalized entropy, and evaporation belong to the later semiclassical volumes.

Reader routes through the remaining chapters

After the definitions and mechanics spine, choose a route matched to the problem.

• Stationary Classification, Rigidity, and No-Hair Results explains what four-dimensional uniqueness theorems actually assume and what their physical conclusions do—and do not—establish. Complete theorem proofs remain in Mathematical Relativity. The theorem architecture and its hypothesis boundaries are reviewed in Chruściel, Costa, and Heusler 2012, §§3.1–3.4.
• Extremal Horizons and Near-Horizon Geometry separates coincident roots, vanishing surface gravity, degenerate horizon data, scaling limits, and perturbative instability questions. Kunduri and Lucietti’s Kunduri and Lucietti 2013, §§2–6 supply the classification framework. For compact oriented two-dimensional Einstein–Maxwell extremal-horizon data, intrinsic rigidity now supplies the axial symmetry rather than assuming it; together with the static sector, this completes the compact four-dimensional Einstein–Maxwell–$\Lambda$ near-horizon classification Colling, Katona, and Lucietti 2025, Theorems 1.1–1.2 and §§1–2. This closure does not extend automatically to higher-dimensional or noncompact sections.
• Ergoregions, Penrose Processes, and Classical Energy Extraction follows negative Killing energy through particle splitting, irreducible mass, and the horizon-flux condition for superradiance. Amplification spectra and astrophysical engines are handed onward.
• Black Holes in de Sitter Spacetimes treats black-hole and cosmological horizons, unequal surface gravities, coalescence limits, and the difficulty of defining a single global energy.
• Black Holes in Anti-de Sitter Spacetimes makes the timelike boundary, frame at infinity, topology, and fixed versus variable cosmological constant explicit, without importing holographic interpretation.
• Higher-Dimensional Black Holes and Black Objects shows how rotation, topology, compact directions, and nonuniqueness change beyond four dimensions. Emparan and Reall survey the principal families in Emparan and Reall 2008, §§3–9.

These routes are complementary. A numerical relativist may move from taxonomy directly to quasi-local mechanics; a mathematical relativist may proceed from Killing horizons to rigidity and extremal classification; a reader studying rotating holes may follow Kerr anatomy through extraction and then compare Kerr–AdS and Myers–Perry.

Boundaries that protect the conclusions

Several tempting identifications are wrong without extra assumptions. An event horizon need not be a Killing horizon. An apparent horizon is slice dependent and need not coincide with the event horizon. A trapped surface is not a curvature singularity, and a singularity theorem concludes causal geodesic incompleteness rather than the existence of a singular point. Vanishing surface gravity is not, in every theory and setting, equivalent to saturation of a Kerr–Newman parameter bound. “No hair” is not one theorem independent of dimension, matter, asymptotics, topology, and regularity. Thermodynamic language does not by itself turn a classical area identity into a microscopic entropy derivation.

The volume repeatedly tests conclusions against Schwarzschild, Kerr–Newman, Vaidya-type evolution, non-asymptotically-flat boundaries, extremal limits, and higher-dimensional black objects. A statement that survives only one benchmark is presented as family-specific rather than universal.

Check your route

A numerical evolution supplies one spacelike slice containing a closed surface $S$ with outgoing null expansion $\theta_{(\ell)}=0$ and ingoing expansion $\theta_{(n)}\lt0$. What can be concluded immediately, and what cannot?

Solution

The data identify $S$ as a marginally outer trapped surface relative to the chosen future null normals. With additional outer and stability conditions it may be a leaf of an apparent, trapping, or dynamical horizon. The single slice does not determine whether $S$ lies on the event horizon, because that requires the complete future development and the relevant asymptotic boundary. It also does not establish stationarity, the existence of a Killing field, or a curvature singularity. The shortest route is therefore the horizon-taxonomy chapter followed by quasi-local mechanics.

References

• Ashtekar, Abhay, and Badri Krishnan. “Quasi-Local Black Hole Horizons: Recent Advances.” Living Reviews in Relativity 28 (2025): article 8. DOI.
• Bardeen, James M., Brandon Carter, and Stephen W. Hawking. “The Four Laws of Black Hole Mechanics.” Communications in Mathematical Physics 31 (1973): 161–170. DOI.
• Booth, Ivan. “Black-Hole Boundaries.” Canadian Journal of Physics 83 (2005): 1073–1099. DOI.
• Chruściel, Piotr T., João Lopes Costa, and Markus Heusler. “Stationary Black Holes: Uniqueness and Beyond.” Living Reviews in Relativity 15 (2012): 7. DOI.
• Colling, Alex, David Katona, and James Lucietti. “Rigidity of the Extremal Kerr–Newman Horizon.” Letters in Mathematical Physics 115 (2025): 19. DOI. Open PDF.
• Emparan, Roberto, and Harvey S. Reall. “Black Holes in Higher Dimensions.” Living Reviews in Relativity 11 (2008): 6. DOI.
• Kunduri, Hari K., and James Lucietti. “Classification of Near-Horizon Geometries of Extremal Black Holes.” Living Reviews in Relativity 16 (2013): 8. DOI.