Nariai and Bertotti-Robinson Geometries

Scrollable formula. Use Left and Right arrow keys to reveal hidden content.

Scrollable table. Use Left and Right arrow keys to reveal additional columns.

Scrollable figure. Use Left and Right arrow keys to inspect hidden content.

Nariai and Bertotti–Robinson are homogeneous because each is a direct product of two homogeneous constant-curvature factors. They are not four-dimensionally constant-curvature: planes tangent to a factor and planes mixing the factors have different sectional curvatures. Nariai is an Einstein vacuum product with positive $\Lambda$ ; Bertotti–Robinson balances opposite factor curvatures with a uniform nonnull Maxwell field.

Required background. The Riemann Curvature Tensor supplies sectional and product curvature; The Einstein Equation supplies the vacuum and Einstein–Maxwell checks.

Helpful background. Killing Fields and Isometry Groups explains homogeneity without maximal isotropy.

Nariai: dS₂ × S²

Let both factors have radius $L>0$ . A static product chart is

ds^2=L^2\left[-(1-x^2)d\tau^2 +\frac{dx^2}{1-x^2}+d\Omega_2^2\right], \qquad -1<x<1.

The first bracket is a static patch of two-dimensional de Sitter space; the sphere has constant radius $L$ . On each block,

R^{(2)}_{AB}=L^{-2}g^{(2)}_{AB}.

Therefore the four-dimensional Ricci tensor satisfies

R_{\mu\nu}=L^{-2}g_{\mu\nu}, \qquad R=4L^{-2},

and the vacuum Einstein equation holds with $\Lambda=L^{-2}$ . The surfaces $x=\pm1$ are static-patch Killing horizons of the dS $_2$ factor, not curvature singularities.

The areal radius never varies: every rotational orbit has area $4\pi L^2$ . Consequently Nariai cannot be written on an open set in a gauge that uses areal radius as a quotient coordinate. This is why it appears as an exceptional Birkhoff branch. It still possesses an additional local Killing field—indeed, the product has the full factor isometry group—but it is not part of the varying-radius Kottler coordinate family.

Nariai is the exceptional constant-areal-radius branch of spherical vacuum Einstein gravity with $\Lambda>0$ . It also arises as a controlled near-coincident-horizon limit of Schwarzschild–de Sitter, but one must rescale coordinates while the roots coalesce; merely substituting a repeated root into the original static chart produces a degenerate expression. Schleich and Witt derive the branch directly and display it in horizon-regular coordinates Schleich and Witt 2010, Theorem 1 and Eqs. (10), (15)–(16), printed pp. 3–5, Open PDF.

In that controlled limit, proper separation encoded by rescaled coordinates remains finite even while the roots approach one another in the original areal coordinate. The limiting product is not globally identical to an extremal Kottler extension; it is the nondegenerate geometry obtained after zooming into the shrinking inter-horizon region. Limit, family member, and maximal extension are three distinct claims.

Bertotti–Robinson: AdS₂ × S²

A convenient equal-radius form is

ds^2=L^2\left[-(1+x^2)d\tau^2 +\frac{dx^2}{1+x^2}+d\Omega_2^2\right], \qquad x\in\mathbb R,\quad \tau\in\mathbb R.

The first factor is the universal cover of AdS $_2$ and the second is $S^2$ . Their Ricci tensors have opposite signs:

R_{ab}=-L^{-2}g_{ab}, \qquad R_{ij}=+L^{-2}g_{ij}, \qquad R=0.

Use Gaussian Maxwell normalization

T_{\mu\nu}=\frac1{4\pi} \left(F_{\mu\alpha}F_\nu{}^\alpha -\frac14g_{\mu\nu}F_{\alpha\beta}F^{\alpha\beta}\right).

In an oriented orthonormal coframe, take

F=E\,e^0\wedge e^1+B\,e^2\wedge e^3, \qquad G(E^2+B^2)=L^{-2}.

Maxwell’s equations hold because the factor volume forms are closed and covariantly constant. Its stress tensor is $-(E^2+B^2)g_{ab}/(8\pi)$ on AdS $_2$ and the opposite sign on $S^2$ . Hence $R_{\mu\nu}=8\pi G T_{\mu\nu}$ with $\Lambda=0$ . Electric–magnetic duality changes $E$ and $B$ while preserving the metric through $E^2+B^2$ .

The magnetic term requires at least two gauge-potential patches on the sphere when magnetic charge is nonzero, even though the two-form $F$ is globally defined. The need for these patches reflects the gauge field’s global structure; it does not indicate a metric singularity. The electric description uses the AdS $_2$ volume form; a duality rotation moves continuously between electric and magnetic support without changing the Ricci tensor.

Bertotti–Robinson is the near-horizon geometry of extremal Reissner–Nordström after a scaling limit. Bičák identifies the Robinson–Bertotti universe as the asymptotic throat geometry of the extremal solution Bičák 2000, §3, printed pp. 39–40, Open PDF.

Homogeneous but not maximally symmetric

Both geometries have constant curvature components in adapted frames and transitive factor isometries. Yet their mixed Riemann components vanish. In Nariai, a plane tangent to either factor has curvature $+L^{-2}$ , while a mixed plane has zero curvature. In Bertotti–Robinson, factor planes have $-L^{-2}$ and $+L^{-2}$ , while mixed planes again have zero curvature. No four-dimensional isotropy can rotate all these planes into one another.

The Weyl decomposition sharpens that comparison. For a direct product of two constant-curvature two-spaces with curvatures $k_1$ and $k_2$ , every Weyl block is proportional to $k_1+k_2$ . Nariai has $k_1=k_2=L^{-2}$ , so its Weyl tensor is nonzero and it is Petrov D. Bertotti–Robinson instead has $k_1=-L^{-2}$ and $k_2=+L^{-2}$ , so the sum cancels: its Weyl tensor vanishes and it is Petrov O. Podolský and Ortaggio exhibit the conformally flat electrovacuum explicitly and place Bertotti–Robinson under type O while placing Nariai under type D in Podolský and Ortaggio 2003, §3.1, printed p. 5, Eq. (25), and Table 2, printed p. 7, Open PDF.

Conformal flatness does not make Bertotti–Robinson maximally symmetric. Its Ricci tensor has opposite eigenvalues on the two factors, and its factor and mixed sectional curvatures still differ. Nariai and Bertotti–Robinson therefore illustrate two separate warnings: constant scalar invariants do not imply maximal symmetry, and neither does a vanishing Weyl tensor unless the Ricci curvature also has the constant-curvature form.

The comparison also separates matter from homogeneity. Nariai’s equal positive Ricci eigenvalues can be supplied by one positive cosmological constant. Bertotti–Robinson’s opposite eigenvalues cannot: its Maxwell field is the exact source supporting the curvature split. Neither homogeneity nor occurrence as a near-horizon limit establishes dynamical stability or astrophysical realization.

Their controlled limits also preserve different data. The Nariai scaling retains the geometry between nearly coincident Schwarzschild–de Sitter roots and fixes a positive sphere radius through $\Lambda$ . The Bertotti–Robinson scaling retains the infinitely deep extremal Reissner–Nordström throat and fixes the sphere through charge. In both cases the limiting spacetime gains symmetry and is not obtained by setting the original radial coordinate equal to its horizon value. Checking the source equation after the scaling is therefore as important as checking the limiting metric.

Check your understanding

Why can no cosmological constant by itself support equal-radius AdS $_2\times S^2$ ?

Solution

A pure cosmological-constant vacuum requires one eigenvalue $R_{\mu\nu}=\Lambda g_{\mu\nu}$ on every tangent direction. In AdS $_2\times S^2$ , the Ricci eigenvalue is $-L^{-2}$ on the first factor and $+L^{-2}$ on the second. No single $\Lambda$ equals both. The traceless Maxwell stress tensor supplies exactly this sign split, with total scalar curvature zero.

References

Bičák, Jiří. “Selected Solutions of Einstein’s Field Equations: Their Role in General Relativity and Astrophysics.” Einstein’s Field Equations and Their Physical Implications, Lecture Notes in Physics 540 (2000), 1–126. arXiv:gr-qc/0004016, §3, printed pp. 31–41.
Podolský, Jiří, and Marcello Ortaggio. “Explicit Kundt Type II and N Solutions as Gravitational Waves in Various Type D and O Universes.” Classical and Quantum Gravity 20, no. 9 (2003): 1685–1701. doi:10.1088/0264-9381/20/9/307; arXiv:gr-qc/0212073, Open PDF.
Schleich, Kristin, and Donald M. Witt. “A Simple Proof of Birkhoff’s Theorem for Cosmological Constant.” Journal of Mathematical Physics 51 (2010), 112502. doi:10.1063/1.3503447.