The Stretched Horizon Limit

We consider four-dimensional general relativity with a positive cosmological constant, $Λ$, in the presence of a boundary, $Γ$, of finite spatial size. The boundary is located near a cosmological event horizon, and is subject to boundary conditions that fix the conformal class of the induced metric, and, $K$, the trace of the extrinsic curvature along $Γ$. The proximity of $Γ$ to the horizon is controlled by the dimensionless parameter ${K}{Λ^{-\frac{1}{2}}}$. We provide an exhaustive analysis of linearised gravitational perturbations for the setup. This is performed both for a $Γ$ encasing a portion of the static patch that ends just before the cosmological horizon (pole patch), as well as a $Γ$ containing only the region near the cosmological horizon (cosmic patch). In the pole patch, we uncover a layered hierarchy of modes: ordinary normal modes, a novel type of boundary gapless mode, and boundary soft modes of frequency $ω\approx \pm 2πi T_{\text{dS}}$, with $T_{\text{dS}}$ the horizon temperature. Minkowskian behaviour is recovered only for angular momenta $l \gtrsim {K}{Λ^{-\frac{1}{2}}}$ which can be made parametrically large, thus attenuating previously found growing modes. In the cosmic patch, we uncover sound and shear fluid-dynamical modes that we interpret in terms of a conformal fluid with shear viscosity over entropy density ratio $\tfracη{s} = \tfrac{1}{4π}$ and vanishing bulk viscosity $ζ=0$. The fluid dynamical sector is shown to admit a non-linear treatment. We describe a scaling regime in which the stretched horizon gravitational dynamics is dictated by a universal Rindler geometry, independent to the details of the infilling horizon. We briefly discuss quantitative features that distinguish cosmological and black hole horizons away from the Rindler regime.

💡 Research Summary

The paper investigates four‑dimensional general relativity with a positive cosmological constant Λ in the presence of a finite‑size timelike boundary Γ that is placed arbitrarily close to the de Sitter (dS) cosmological horizon. The authors fix two pieces of boundary data: (i) the conformal class of the induced metric on Γ (chosen to be the round S²×ℝ metric) and (ii) the trace of the extrinsic curvature K, which is taken to be constant over the boundary. The dimensionless combination K Λ⁻¹ᐟ² controls how close Γ is to the true horizon. Two distinct configurations are studied: the “pole patch”, where Γ encloses a region that ends just before the horizon (ψ→0 in static coordinates), and the “cosmic patch”, where Γ surrounds only the near‑horizon region (ψ→π/2).

The authors develop a systematic linear‑perturbation framework. Perturbations are decomposed into scalar and vector spherical harmonics labelled by (ℓ,m) and a complex frequency ω measured with respect to the inertial clock at the centre of the static patch. The presence of the timelike boundary introduces a new sector of boundary modes that would be pure gauge in the absence of Γ but become physical because the boundary conditions break full diffeomorphism invariance.

Pole patch (K Λ⁻¹ᐟ² → –∞).
In this limit the spectrum splits into three families:

1. Normal modes – a discrete tower of bulk excitations in both scalar and vector sectors. Their frequencies are spaced by Δω ≈ π log|K Λ⁻¹ᐟ²| (eq. 4.23), which shrinks to zero as the boundary approaches the horizon, effectively yielding a quasi‑continuous spectrum. Scattering off the boundary reproduces the phase shift (4.40) that matches the quasinormal mode structure of the empty dS static patch up to O((K Λ⁻¹ᐟ²)⁻²) corrections. These modes leave the boundary Weyl factor ω(t,Ω) untouched and are labelled by (ℓ,m,n).

2. Gapless boundary modes – a novel set of real‑frequency scalar excitations with dispersion ω ≈ ±√{ℓ(ℓ+1)} √{2/|K Λ⁻¹ᐟ²|} (eq. 4.14). They are “gapless” because the frequency depends continuously on ℓ, and they are localized near Γ: as K Λ⁻¹ᐟ² → –∞ their radial profiles become sharply peaked at the boundary (Fig. 7). Their dynamics is parametrically slower than that of normal modes, suggesting a Goldstone‑like origin associated with the breaking of diffeomorphisms at the timelike surface.

3. Soft modes – modes with purely imaginary frequency ω ≈ ±2πi T_dS, where T_dS = 1/(2πℓ) is the de Sitter temperature. These are reminiscent of the “membrane‑parity” soft hair discussed in black‑hole contexts. Their presence attenuates the previously identified exponentially growing modes in the pole patch, rendering the stretched horizon dynamically more stable.

A crucial observation is that for angular momenta ℓ ≳ |K Λ⁻¹ᐟ²| the spectrum becomes effectively Minkowskian; only in this regime does the usual flat‑space intuition apply, thereby explaining why earlier analyses that ignored the large‑ℓ sector over‑estimated instabilities.

Cosmic patch (K Λ⁻¹ᐟ² → +∞).
When the boundary encloses the horizon, the low‑frequency sector reorganizes into fluid‑dynamical excitations. By mapping the linearised Einstein equations onto effective fluid variables (velocity, pressure, temperature), the authors identify:

• Sound (acoustic) modes with ω ≈ ±c_s k, c_s = 1/√3, corresponding to longitudinal pressure waves on the stretched horizon.
• Shear (diffusive) modes with ω ≈ –i (η/(ε+P)) k², where η/s = 1/(4π) and bulk viscosity ζ = 0. This matches the universal shear viscosity bound known from AdS/CFT and membrane‑paradigm studies.

The fluid is conformal, as reflected by the tracelessness of the boundary stress tensor (eq. 2.3). Section 6 demonstrates that the Einstein equations admit a fully non‑linear realization of the Navier‑Stokes equations on Γ, confirming that the stretched horizon behaves as a genuine dynamical membrane rather than a mere linear probe.

Rindler universal regime.
Both patches share a common asymptotic regime when |K Λ⁻¹ᐟ²| → ∞. In this limit the geometry near Γ reduces to a universal Rindler wedge: ds² ≈ –ρ² dτ² + dρ² + dx_⊥². All perturbative modes become governed by the same Rindler dynamics, and the distinction between black‑hole and cosmological horizons collapses to differences in global thermodynamic quantities (temperature, entropy) rather than in local dynamics. The authors term this the “stretched horizon limit” and argue that it provides a clean setting where the membrane paradigm can be applied uniformly to any Killing horizon.

Conclusions and outlook.
The work establishes that (i) fixing the conformal class and extrinsic curvature on a timelike surface yields a well‑posed variational problem and a rich spectrum of boundary‑induced modes; (ii) in the pole patch, gapless and soft modes soften previously reported instabilities, while large‑ℓ modes restore flat‑space behaviour; (iii) in the cosmic patch, the stretched horizon is precisely described by a conformal fluid with η/s = 1/4π and vanishing bulk viscosity, and this description survives non‑linear extensions; (iv) the universal Rindler limit unifies the dynamics of black‑hole and cosmological horizons.

Future directions suggested include: extending the analysis to time‑dependent boundaries, higher‑dimensional de Sitter or anti‑de Sitter backgrounds, incorporating quantum corrections to probe the micro‑state structure of the dS horizon, and exploring holographic interpretations of the gapless boundary sector. The paper thus provides a comprehensive bridge between horizon thermodynamics, membrane fluid dynamics, and the underlying gravitational field theory in a cosmological setting.