Hyperboloidal Framework in Relativistic Systems

• Hyperboloidal Framework is a geometric formulation that uses hyperboloidal slicing, conformal compactification, and scri-fixing to create a compact domain bounded by null surfaces.
• It replaces artificial outer-boundary conditions with regularity requirements at future null infinity and the event horizon, enabling accurate treatment of quasinormal modes.
• The framework underpins numerical relativity and perturbation theory by improving long-term stability, facilitating efficient waveform extraction, and ensuring smooth asymptotic behavior.

The hyperboloidal framework is a geometric and analytic formulation of hyperbolic problems in which spacelike hypersurfaces are chosen to approach future null infinity $\mathcal I^+$ and, in black-hole spacetimes, to intersect the future event horizon $\mathcal H^+$. Combined with conformal compactification and scri-fixing, it replaces formulations tied to spatial infinity $i^0$ and, for maximally extended black holes, the bifurcation sphere $\mathcal B$, by a compact domain bounded by the physically relevant null surfaces. In this setting, outgoing and ingoing conditions are encoded as regularity at $\mathcal I^+$ and $\mathcal H^+$, rather than by artificial outer-boundary prescriptions, and time-harmonic objects such as quasinormal modes become geometrically regularized (Macedo et al., 2024, Zenginoglu et al., 2010).

1. Geometric foundations

At the conceptual core are hyperboloidal surfaces: smooth spacelike hypersurfaces that asymptote to future null infinity and, in black-hole exteriors, intersect the future event horizon. In asymptotically flat settings this replaces standard $t=\mathrm{const}$ slices, which accumulate at $i^0$ and, in extended black-hole spacetimes, intersect $\mathcal B$. The resulting pathologies are not merely coordinate inconveniences. In the frequency domain they manifest as the familiar divergence of time-harmonic perturbations near $i^0$ and $\mathcal H^+$0; in the time domain they force artificial timelike boundaries and finite-radius wave extraction. Hyperboloidal slicing removes these features by aligning the foliation with outgoing and ingoing characteristics near the physical boundaries (Macedo et al., 2024).

The second structural ingredient is conformal compactification. In Penrose’s formulation one passes from the physical metric $\mathcal H^+$1 to a conformal metric

$\mathcal H^+$2

so that null infinity becomes a finite boundary in the unphysical spacetime. Scri-fixing then chooses coordinates in which $\mathcal H^+$3 sits at a fixed coordinate location. In practical implementations this yields a finite computational domain whose outer boundary is null, with no incoming physical characteristic at $\mathcal H^+$4 (Macedo et al., 2024, Demirboğa et al., 25 May 2026).

A common misconception is that the framework is just radial compactification. The literature instead treats it as a coupled construction: hyperboloidal foliation, conformal compactification, and a gauge choice adapted to the causal structure. In numerical relativity this is closely tied to conformal gauges such as the preferred conformal gauge, where $\mathcal H^+$5 at scri, and to hyperboloidal gauge drivers that maintain scri-fixing during evolution (Vañó-Viñuales et al., 2017).

2. Coordinate constructions and gauge structure

The basic time transformation is a height-function shift. In the simplest Schwarzschild presentation one writes

$\mathcal H^+$6

with $\mathcal H^+$7 chosen so that $\mathcal H^+$8 approaches retarded time at large radius and ingoing Eddington–Finkelstein time near the horizon. Using the tortoise coordinate $\mathcal H^+$9, the standard asymptotics are

$i^0$0

so that the hyperboloid-rescaled frequency-domain field becomes bounded at both ends (Macedo et al., 2024).

A simple global compactification is

$i^0$1

which sends $i^0$2 to $i^0$3. In more elaborate settings the compactification is applied only in an outer layer, leaving an interior Cauchy region untouched. This “hyperboloidal layer” strategy is particularly attractive when existing interior solvers are to be retained (Macedo et al., 2024, Bernuzzi et al., 7 Aug 2025).

For static spherically symmetric geometries the gauge freedom separates naturally into a height function and a radial compactification function. A widely used parametrization is

$i^0$4

with $i^0$5 and boost function $i^0$6, where $i^0$7. In the hyperboloidal minimal gauge one sets $i^0$8, equivalently $i^0$9, and constructs $\mathcal B$0 from the tortoise coordinate by sign changes of precisely identified singular or regular pieces. The paper distinguishes two strategies: “in-out,” in which the height function follows from the tortoise coordinate by changing the sign of the terms singular at future null infinity, and “out-in,” in which the sign change occurs in the tortoise coordinate’s regular terms (Macedo, 2023).

In Kerr, the generic hyperboloidal map is built on ingoing Kerr coordinates,

$\mathcal B$1

and the minimal gauge is refined into two branches. The radial-function-fixing minimal gauge produces a smooth extremal limit to extremal Kerr, whereas the Cauchy-horizon-fixing minimal gauge produces a limit to the near-horizon extremal Kerr geometry. This bifurcation is one of the distinctive structural results of the Kerr hyperboloidal literature (Macedo, 2019).

3. Hyperboloidal wave equations, quasinormal modes, and operator theory

For Schwarzschild perturbations, the Regge–Wheeler and Zerilli master equations have the standard form

$\mathcal B$2

With the time-harmonic ansatz $\mathcal B$3, quasinormal-mode conditions are purely ingoing at $\mathcal B$4 and purely outgoing at $\mathcal B$5,

$\mathcal B$6

After the hyperboloidal transformation,

$\mathcal B$7

the rescaled field $\mathcal B$8 is bounded at both boundaries if $\mathcal B$9 has the asymptotics above. In compactified coordinates, the boundary conditions are replaced by regularity at $\mathcal I^+$0 and $\mathcal I^+$1 (Macedo et al., 2024).

The resulting frequency-domain problem is a singular ODE on a compact interval,

$\mathcal I^+$2

with

$\mathcal I^+$3

so $\mathcal I^+$4. This degeneracy is not a defect. It is the mechanism by which the boundary equations reduce to first-order regularity relations selecting the physical solution at each end (Macedo et al., 2024).

The operator-theoretic interpretation then becomes natural. In one dimension, quasinormal modes are poles of the analytically continued Green’s function and zeros of the Wronskian of outgoing solutions. In the hyperboloidal formulation they also arise as eigenvalues or resonances of the generator of time translations on a suitable hyperboloidal Hilbert space, or equivalently as eigenvalues of a non-selfadjoint operator pencil

$\mathcal I^+$5

This is the setting in which recent work emphasizes resolvent analysis, pseudospectra, non-modal amplification, and spectral instability (Macedo et al., 2024).

For Kerr, the separated Teukolsky ansatz takes the form

$\mathcal I^+$6

and the hyperboloidal transform regularizes both the horizon and null infinity after appropriate spin-dependent field rescalings. In the frequency domain the minimal gauge provides the spacetime counterpart of Leaver’s formalism: the regularization factors familiar from continued-fraction methods emerge directly from the hyperboloidal slicing, and the non-extremal and extremal regularization factors correspond to distinct minimal-gauge limits (Macedo, 2019).

4. Numerical realizations and representative implementations

The framework has developed into a family of concrete computational strategies spanning fixed-background wave propagation, black-hole perturbation theory, self-force calculations, free hyperboloidal evolution in 3+1 formulations, exact conformal matching across multiple compactified patches, and waveform propagation from worldtubes to null infinity (Zenginoglu et al., 2010, Leather, 2024, Demirboğa et al., 25 May 2026, Álvares et al., 22 May 2025, Vaswani et al., 24 Jun 2026, Bernuzzi et al., 7 Aug 2025).

Paper	Setting	Characteristic feature
(Zenginoglu et al., 2010)	Scalar fields on Minkowski and Schwarzschild	3D spectral evolution with scri-fixing
(Leather, 2024)	Lorenz-gauge GSF in Schwarzschild	Hyperboloidal compactification plus Chebyshev spectral domains
(Demirboğa et al., 25 May 2026)	Global scalar scattering in Minkowski	Exact conformal matching of $\mathcal I^+$7, $\mathcal I^+$8, and $\mathcal I^+$9
(Álvares et al., 22 May 2025)	Einstein–Maxwell–Klein–Gordon	Free hyperboloidal GBSSN/Z4 evolution to $\mathcal H^+$0
(Vaswani et al., 24 Jun 2026)	Teukolsky equation with point particle	Comoving compactified hyperboloidal coordinates
(Bernuzzi et al., 7 Aug 2025)	3+1 NR waveform extraction	Hyperboloidal RWZ propagation from a timelike worldtube

In practice, several numerical motifs recur. One is spectral discretization on compact intervals. For frequency-domain RW/Zerilli and Lorenz-gauge self-force systems, Chebyshev–Gauss–Lobatto collocation on subdomains separated by the particle worldline turns singular ODEs into matrix polynomials in $\mathcal H^+$1, with jump conditions and boundary regularity enforced algebraically rather than by external Sommerfeld data (Leather, 2024). Another is hyperboloidal layering, where compactification is activated only beyond a finite radius; this keeps the interior numerics close to standard Cauchy formulations while still moving $\mathcal H^+$2 onto the grid (Zenginoglu et al., 2010, Bernuzzi et al., 7 Aug 2025).

A more global construction appears in Minkowski scattering, where a past hyperboloidal domain attached to $\mathcal H^+$3, a central Penrose domain around $\mathcal H^+$4, and a future hyperboloidal domain attached to $\mathcal H^+$5 are matched exactly along identical conformal hypersurfaces. In that framework the matching surfaces agree pointwise in compactified radial coordinate and conformal factor, so no interpolation between scri-fixing gauges is required (Demirboğa et al., 25 May 2026).

Particle-sourced perturbation theory has generated a separate hyperboloidal line of development. In the 1+1 Teukolsky framework with comoving compactified coordinates, the particle is held at fixed $\mathcal H^+$6, which simplifies the jump conditions. The hyperboloidal compactification is reported to evade the nonphysical growing modes that appear in some noncompactified characteristic schemes, and long-term stable $\mathcal H^+$7 evolutions are demonstrated (Vaswani et al., 24 Jun 2026).

At the level of full evolution systems, hyperboloidal techniques have been integrated with conformal Z4 and generalized BSSN. In spherical symmetry, free hyperboloidal evolution of the Einstein–Maxwell–Klein–Gordon system uses a time-independent conformal factor, scri-fixing gauge drivers for lapse and shift, and an adapted Lorenz gauge for the electromagnetic scalar potential. The framework reaches $\mathcal H^+$8 directly and extracts scalar and electric signals there (Álvares et al., 22 May 2025).

5. Initial data, smooth scri, and logarithmic obstructions

The hyperboloidal framework is not only a slicing prescription for evolution equations; it also has a distinct initial-data theory. A vacuum initial data triple $\mathcal H^+$9 is hyperboloidal if $t=\mathrm{const}$0 extends to future null infinity under conformal compactification, the rescaled metric extends smoothly to the boundary, the mean curvature has a positive limit there, and the rescaled trace-free part of $t=\mathrm{const}$1 extends smoothly as well. In the parabolic–hyperbolic formulation of the constraints, the unknowns $t=\mathrm{const}$2 satisfy a coupled parabolic–hyperbolic system along a radial foliation, with $t=\mathrm{const}$3 ensuring the parabolic character of the lapse equation (Csukás et al., 14 Mar 2025).

A central issue is the appearance of polyhomogeneous expansions with logarithmic terms at scri. Building on Andersson–Chruściel and extending a result of Beyer–Ritchie, recent work shows that the existence of well-defined Bondi mass and Bondi angular momentum forces the disappearance of low-order logarithmic coefficients and places the data in the smooth subclass. In particular, under the stated assumptions one obtains

$t=\mathrm{const}$4

together with the algebraic conditions

$t=\mathrm{const}$5

With additional boundary relations,

$t=\mathrm{const}$6

the constrained fields extend smoothly to the conformal boundary (Csukás et al., 14 Mar 2025).

This makes precise a point that is sometimes blurred in numerical discussions: regular hyperboloidal evolution presupposes, or else must dynamically preserve, nontrivial asymptotic structure. Smoothness at null infinity is not automatic. It depends on the interplay among the slicing, the conformal completion, the choice of free data, and the physical charges defined at infinity (Csukás et al., 14 Mar 2025).

6. Advantages, limitations, and active directions

The main advantages are consistent across the literature. Hyperboloidal formulations eliminate artificial timelike outer boundaries, place radiation extraction at the physically correct location $t=\mathrm{const}$7, regularize frequency-domain boundary behavior at $t=\mathrm{const}$8 and $t=\mathrm{const}$9, and often improve long-time numerical stability by aligning the computational domain with the causal structure of the open system (Macedo et al., 2024, Zenginoglu et al., 2010). In perturbation theory, they unify horizon and null-infinity treatments on a single compact interval. In numerical relativity, they offer a route to Bondi-level waveform information without extrapolation, as illustrated by perturbative hyperboloidal extraction from 3+1 worldtubes (Bernuzzi et al., 7 Aug 2025).

The limitations are equally clear. Nonlinear source terms need not remain well behaved after conformal rescaling. In global scalar scattering on Minkowski space, quintic and septic semilinear sources exhibit approximately fourth-order convergence, but the cubic case degrades to first order because the conformally rescaled source remains non-vanishing at $i^0$0; the paper identifies this as a boundary-regularity issue near compactified boundaries and $i^0$1 (Demirboğa et al., 25 May 2026). In free hyperboloidal EMKG evolution, the adapted Lorenz gauge is validated only in spherical symmetry, and extension to 3D is explicitly said to require careful hyperbolicity analysis (Álvares et al., 22 May 2025).

There are also structural trade-offs. Hyperboloidal layers preserve interior solvers but can be less efficient than pure hyperboloidal foliations; perturbative extraction to $i^0$2 is inexpensive and often close to CCE quality, but by construction it does not capture nonlinear propagation effects outside the worldtube (Zenginoglu et al., 2010, Bernuzzi et al., 7 Aug 2025). In Kerr, the minimal gauge exposes two distinct extremal limits, which is mathematically illuminating but also signals that gauge choice changes the global conformal picture in a substantive way (Macedo, 2019).

Active directions include optimal hyperboloidal gauges and layer designs, full Einstein evolutions on hyperboloidal slices, second-order and quadratic quasinormal-mode problems, pseudospectral analysis in environmental or modified-gravity backgrounds, treatment of spatial infinity through Friedrich’s cylinder, and extensions to non-asymptotically flat or non-vacuum systems (Macedo et al., 2024, Csukás et al., 14 Mar 2025). A plausible implication is that the framework is evolving from a boundary-treatment technology into a broader geometric language for open relativistic systems, linking asymptotic analysis, non-selfadjoint spectral theory, and long-time numerical evolution on the same conformally compactified domain.