Stationary and Axisymmetric Spacetimes

Stationary axisymmetry is the shared geometric language of rotating equilibrium spacetimes. Before any Kerr formula is introduced, two commuting Killing fields determine an orbit metric, circularity decides whether a $2+2$ reduction is legitimate, and norm, twist, and electromagnetic potentials compress the field equations. Multipoles and orbit-space boundaries then turn asymptotic and degeneracy data into invariant statements.

Helpful background. The Einstein Equation supplies vacuum and electrovac dynamics; The Riemann Curvature Tensor supplies Ricci and regularity identities; Killing Fields and Isometry Groups supplies the symmetry action; Formulations of General Relativity supplies reduced systems; Causality and Global Structure supplies global qualifications.

The reduction problem

The chapter assumes a four-dimensional spacetime with commuting stationary and axial Killing fields $(\xi,\eta)$. The stationary generator is normalized at a specified asymptotically flat end, and the axial generator has $2\pi$-periodic orbits. Their Gram matrix

$$\lambda_{AB}=g(K_A,K_B), \qquad \rho^2=-\det\lambda,$$

is the recurring invariant. It separates data intrinsic to the group orbits from data on the two-dimensional quotient.

This reduction is conditional. Two Killing fields do not automatically make their orthogonal planes integrable; Einstein’s equation and the matter model must enforce circularity. Even after circularity, local Weyl coordinates need not extend globally without topology and regularity assumptions. The review Chruściel, Costa, and Heusler 2012, §§3.2.3, 4.4, and 8.3 makes these dependencies explicit.

Routes through the chapter

The geometry-first route begins with stationary and axial Killing geometry. It distinguishes stationarity, staticity, and a regular rotation axis without relying on component form. Continue to the proof-critical page on circularity and orthogonal transitivity, which states the dimension, regular-axis, connectedness, Ricci, and symmetry hypotheses needed to eliminate orbit-meridional cross terms.

The field-equation route then uses Weyl–Papapetrou form and orbit space to derive $\rho=\sqrt{-\det\lambda}$ as a harmonic coordinate. Twist and Ernst potentials combine the stationary norm and twist into a complex scalar obeying the Ernst equation; the electrovac extension adds a complex electromagnetic potential. Ernst introduced the vacuum formulation in Part I, pp. 1175–1178 and the electrovac extension in Part II, pp. 1415–1417.

The interpretation route starts with relativistic multipole moments, where conformal completion and an asymptotic gauge prevent arbitrary expansion coefficients from masquerading as moments. Finish with ergoregions, Killing horizons, and axis structure, which identifies the degenerating Killing combination on each orbit-space boundary interval.

Distinctions to keep fixed

Several terms that sound similar answer different questions:

• Stationary versus static: $\mathcal L_\xi g=0$ expresses time-translation symmetry; $\xi\wedge d\xi^\flat=0$ additionally expresses hypersurface orthogonality.
• Axisymmetric versus circular: an axial $U(1)$ action supplies $\eta$; circularity says the planes orthogonal to $(\xi,\eta)$ form surfaces.
• Stationary limit versus Killing horizon: $g(\xi,\xi)=0$ locates a stationary-limit surface; a horizon is generated by a null normal combination $\xi+\Omega_H\eta$.
• Axis versus generic $\rho=0$ boundary: on an axis the periodic generator vanishes; on a horizon interval a nonzero Killing combination becomes null.
• Expansion coefficient versus multipole: invariant moments require a controlled asymptotic frame or the Geroch–Hansen construction.

This common stationary structure underlies the study of Kerr domains, uniqueness, rotating stars, and black-hole mechanics. Those subjects use the reduction developed here but require additional hypotheses and tools.

A recurring calculation

On every circular stationary-axisymmetric metric, compute the Killing block before interpreting a coordinate surface. Its determinant gives $\rho$; its normalized timelike eigen-direction identifies stationary observers where they exist; its kernel at $\rho=0$ distinguishes axes from candidate horizons. Then check twist, regularity, and the field equations. This sequence is robust under meridional coordinate changes.

For static vacuum, the scheme collapses to a useful calibration: $\omega=0$, $f=e^{2U}$, $U$ solves the axisymmetric Laplace equation, and the remaining conformal factor follows by quadrature. Stationary vacuum restores nonlinearity through the twist potential, while electrovac introduces electric and magnetic potentials. The hierarchy explains why solution-generating methods act naturally on reduced potentials rather than directly on four-dimensional metric components.

Review the chapter

Conceptual check. Why is $g_{t\phi}=0$ not an invariant definition of staticity?

Solution

Metric cross terms depend on the chart and on which Killing combination is called $\partial_t$. Staticity asks whether a physically selected timelike Killing field is hypersurface orthogonal, equivalently whether its twist vanishes. Rotating coordinates can give Minkowski spacetime a nonzero $g_{t\phi}$ even though an inertial hypersurface-orthogonal Killing field still exists.

Proof check. In vacuum, what two logically distinct steps prove circularity?

Solution

First, the Killing identity and $R_{\mu\nu}=0$ show that the two circularity scalars are constant. Second, a regular axis sets those constants to zero because $\eta=0$ there. Frobenius then gives the orthogonal surfaces. Omitting the axis step leaves potentially nonzero constants.

Transfer task. On a segment with $\det\lambda=0$, how do you distinguish an axis from a horizon candidate?

Solution

Find the kernel of $\lambda_{AB}$. If it selects the periodic axial field and that field vanishes, the segment is an axis and must pass elementary flatness. If it selects a nonzero null combination $\xi+\Omega\eta$, test whether that combination is normal to a smooth null hypersurface and whether a regular extension exists. Global causal data are still needed before assigning an event horizon.

References

• Chruściel, Piotr T., João Lopes Costa, and Markus Heusler. “Stationary Black Holes: Uniqueness and Beyond.” Living Reviews in Relativity 15 (2012): 7. DOI: 10.12942/lrr-2012-7.
• Ernst, Frederick J. “New Formulation of the Axially Symmetric Gravitational Field Problem. I.” Physical Review 167, no. 5 (1968): 1175–1178. DOI: 10.1103/PhysRev.167.1175.
• Ernst, Frederick J. “New Formulation of the Axially Symmetric Gravitational Field Problem. II.” Physical Review 168, no. 5 (1968): 1415–1417. DOI: 10.1103/PhysRev.168.1415.