Angular Harmonics, Separability, and Hidden Structure

Black-hole perturbation equations become tractable when their angular dependence is organized by the geometry and spin weight of the field. This chapter builds that machinery in three layers: spherical bases for tensorial data, spheroidal bases for rotating backgrounds, and hidden-symmetry operators that explain when separation is exact. It then shows how a named deformation breaks the operator structure and generates calculable mode couplings. The central lesson is diagnostic: a convenient harmonic expansion always exists under suitable regularity, but separated radial equations exist only when the bulk operator and its boundary domain preserve the same angular sectors.

Helpful background. Black-Hole Geometry and Mechanics supplies Schwarzschild and Kerr geometry; PDEs and Green Functions for Gravitation supplies eigenfunction expansions and boundary domains; Formulations of General Relativity locates perturbation variables within the field equations; Frames, Coframes, and Anholonomy supplies dyads, tetrads, and their local transformation laws.

The six-part route through angular structure

Read the pages in order when building a separation calculation from scratch. The first two fix the round-sphere basis and phases; the third introduces rotation; the next two explain the hidden geometry and its lift to field operators; the last tests what survives under deformation.

Page	Object built	Decisive check
Scalar, Vector, and Tensor Spherical Harmonics	Parity-resolved bases for scalars, covectors, and symmetric tensors	Low-$\ell$ exceptional sectors are treated separately
Spin-Weighted Spherical Harmonics	Dyad-covariant spin-$s$ basis and eth ladders	Ladder signs, conjugation, and unit-sphere norm agree
Spin-Weighted Spheroidal Harmonics	Massless Kerr angular eigenproblem at $c=a\omega$	Matrix and continued-fraction eigenvalues converge to the same branch
Killing Tensors and Carter Separability	Quadratic integral and separable Kerr coordinates	The $a\to0$ limit recovers total angular momentum
Symmetry Operators for Field Equations	Commuting Carter operator and higher-spin extensions	Both the local commutator and boundary domain are preserved
Mode Coupling and Broken Separability	Sparse projected systems for a specified deformation	Selection rules follow from the operator; truncation error is checked at fixed source, frequency, domain, and norm

The route is cumulative but not circular. Harmonics can be defined without a black-hole field equation. Hidden symmetry explains why a particular equation becomes diagonal in a particular basis. Numerical separation constants then implement that structure rather than supply evidence for it by themselves.

From geometry to radial equations

On the round sphere, scalar harmonics diagonalize the Laplacian. Covariant differentiation and the unit-sphere volume form generate even- and odd-parity vector and tensor harmonics, with separate monopole and dipole analysis where the generic formulas vanish. A complex dyad packages tensor components into spin-weighted scalars. Under a dyad rotation $m^A\mapsto e^{i\chi}m^A$, a spin-$s$ component transforms as $\eta\mapsto e^{is\chi}\eta$; the basis must transform with the same phase.

Kerr rotation changes the angular operator. For massless fields described by the Teukolsky equation, the dimensionless deformation is $c=a\omega$, and the appropriate eigenfunctions are spin-weighted spheroidal harmonics. They mix round-sphere multipoles at fixed $s$ and $m$. This mixing does not mean Kerr is nonseparable: it means the spherical basis is not the eigenbasis of the Kerr angular operator. The spheroidal basis resums precisely that angular mixing and leaves one radial equation for each separated label.

The deeper reason is a hierarchy of commuting structures. Stationarity and axisymmetry give the eigenvalues $\omega$ and $m$. The principal tensor generates a Killing tensor, whose quadratic contraction is the Carter constant for geodesics. For the scalar field it lifts to a second-order differential operator commuting with the Kerr wave operator. Common eigenfunctions of these operators yield the radial–angular product ansatz. Frolov, Krtouš, and Kubizňák develop this chain from principal tensors to field separation in 2017, §§ 2.2–2.3, pp. 16–22 and §§ 6.2–6.6, pp. 117–131.

Convention bridge and independent checks

For the massless Teukolsky problem, three separation constants commonly describe the same angular eigenproblem:

$$E_{s\ell m}=A_{s\ell m}+s(s+1), \qquad \lambda_{s\ell m}=A_{s\ell m}+c^2-2mc.$$

In this chapter $A_{s\ell m}(0)=\ell(\ell+1)-s(s+1)$. Recording only a numerical value without the $A/E/\lambda$ dictionary, time phase, and sign of $c$ makes cross-source comparison ambiguous. Berti, Cardoso, and Casals give the angular equation, series recurrence, and corrected convention relations in 2006, §§ II–IV, pp. 3–12.

For a scalar of mass $m_{\rm s}$, the angular equation instead depends on $\gamma^2=a^2(\omega^2-m_{\rm s}^2)$. Its separation constants obey

$$A=A_{0\ell m}(\gamma),\qquad \lambda=A+a^2\omega^2-2am\omega.$$

The radial shift still contains $a^2\omega^2$, not $\gamma^2$. Killing Tensors and Carter Separability derives this distinction; only the massless limit permits replacing $\gamma$ by $c=a\omega$ throughout that dictionary.

A compact set of checks catches most mistakes before the radial problem is solved:

• At $s=0$, spin-weighted harmonics reduce to the Condon–Shortley $Y_{\ell m}$.
• Applying $ð$ then $\barð$ gives the declared negative angular eigenvalue, with both ladder signs visible.
• At $c=0$, the spheroidal eigenvalue and eigenvector reduce to the spherical branch; for real $c$, the angular matrix is Hermitian.
• At $a=0$, the Carter convention $\mathcal K$ approaches $L^2$, while the alternative $Q$ approaches $L^2-L_z^2$.
• A proposed symmetry operator must commute locally and must map the chosen boundary domain to itself.

These are independent checks because they probe phases, normalizations, limits, and domains rather than repeating the same algebra in a second notation.

Diagnosing a failure of separation

Begin with the full equation and write the operator expansion explicitly, for example $\mathcal L_\varepsilon=\mathcal L_0+\varepsilon\delta\mathcal L+\cdots$. Project each term onto the normalized angular basis. The resulting Wigner or Gaunt coefficients state which $(\ell,m)$ channels couple. For an axisymmetric quadrupole $P_2(\cos\theta)$, $m$ and parity are preserved and $\Delta\ell=0,\pm2$. A generic statement that “nonsphericity mixes modes” is not enough to choose a truncation or check an implementation.

Next determine what broke. If only the basis is non-diagonal, as with using spherical rather than spheroidal harmonics on Kerr, a new eigenbasis can restore exact separation. If the bulk commutator fails, one may still obtain a controlled coupled-mode system or a perturbative near-identity reduction. If the boundary surface depends on angle, even a separable bulk operator can have a coupled domain. If the coupling is a nonlinear source, the background linear operator may remain separable while its driven modes communicate.

Finally hold the source, frequency, domain, boundary data, and error norm fixed while refining the angular cutoff and radial resolution. Small deformation alone is not sufficient near a degeneracy or a strongly nonnormal response, where poor conditioning can enhance mixing. Compare with an independently converged direct two-dimensional calculation when possible; if both methods share a radial discretization, converge it separately. Agreement tests the chosen equation and boundary problem, not omitted orders in the background model or a different evolution problem.

Review the chapter

Suppose a massless scalar calculation on Kerr is expanded in ordinary spherical harmonics and the angular matrix has nonzero entries for many neighboring $\ell$ values at fixed $m$. A colleague concludes that the Kerr scalar wave equation is not separable. Identify the missing step and give two checks that distinguish basis mixing from genuinely broken separability.

Solution

The spherical basis diagonalizes the $c=0$ angular operator, not the Kerr operator at $c=a\omega\ne0$. The missing step is to solve the spin-weighted spheroidal eigenproblem and transform to its eigenbasis. In that basis each regular angular eigenfunction has one separation constant and the scalar Kerr equation reduces to a radial ordinary differential equation.

First, compute the truncated spheroidal matrix and verify that its eigenvector tail decreases and its eigenvalue converges as $j_{\max}$ increases. This shows that the many spherical coefficients constitute one resolved spheroidal eigenfunction. Second, construct the Carter symmetry operator and verify $[\Box,\mathcal C]=0$ on the bulk domain. Its common eigenvalue with $i\partial_t$ and $-i\partial_\phi$ supplies the invariant operator reason for separation. If instead a specified deformation produces a nonzero commutator, or the boundary condition is not preserved by $\mathcal C$, the coupling is genuine for that operator-domain pair.

General representation theory remains mathematical preparation rather than a topic developed here. The classification and exact geometry of principal tensors remain in Exact Solutions. This chapter uses those results to build and test black-hole perturbation mode expansions without claiming global mode completeness or stability.

References

• Berti, Emanuele, Vitor Cardoso, and Marc Casals. “Eigenvalues and Eigenfunctions of Spin-Weighted Spheroidal Harmonics in Four and Higher Dimensions.” Physical Review D 73 (2006): 024013; correction 109902. arXiv, correction DOI.
• Frolov, Valeri P., Pavel Krtouš, and David Kubizňák. “Black Holes, Hidden Symmetries, and Complete Integrability.” Living Reviews in Relativity 20 (2017): 6. arXiv.
• Goldberg, Joshua N., A. J. Macfarlane, Ezra T. Newman, Fritz Rohrlich, and E. C. G. Sudarshan. “Spin-s Spherical Harmonics and ð.” Journal of Mathematical Physics 8 (1967): 2155–2161. DOI.