The merger of a black hole with a cosmological horizon

Hangzhi Wang; Maxime Gadioux

REVIEW 3 major objections 4 minor 15 references

The merger of a black hole with a cosmological horizon

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper argues that the zero-cosmological-constant limit of a Schwarzschild-de Sitter black-hole/cosmological-horizon merger is the Emparan-Martínez infinite-mass-ratio Schwarzschild merger, and uses this to regularise the otherwise…

desk verdict Solid finite SdS merger geometry with a genuinely new caustic analysis; the Λ→0 reduction to the Emparan–Martínez merger is argued rather than proved, and the factor-2 discrepancy in the regularized area is a real loose end. read the letter →

arxiv 2412.04551 v1 pith:5PRHWZQP submitted 2024-12-05 gr-qc

Maxime Gadioux , Hangzhi Wang This is my paper · ORCID

classification gr-qc MSC 83C57 PACS 04.70.-s

keywords eventhorizonSchwarzschild-deSittercosmologicalblackholemergercausticsGerochlimitareaincreaseEmparan-Martínez

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper studies the merger of a black hole with the cosmological horizon in Schwarzschild-de Sitter spacetime, as seen by an accelerating observer at future null infinity. Its central claim is that taking the cosmological constant to zero, with the black-hole mass fixed, does not give an isolated Schwarzschild black hole but rather the Emparan-Martínez infinite-mass-ratio Schwarzschild merger. If correct, the finite Schwarzschild-de Sitter system provides a concrete regularisation of the divergent area growth that afflicts the infinite-mass-ratio model. The paper also shows that caustic generators contribute most of the area increase and that the merger obeys a universal scaling law for conical-singularity opening angles.

What carries the argument

The central mechanism is Geroch's limit-of-spacetimes formalism (promoting Λ to a scalar on a five-dimensional manifold and attaching orthonormal tetrads to a reference point) applied with reference point p at future null infinity and conformal coordinate x=1/r. This choice lets the region near I+ survive the limit, unlike earlier constructions. The event horizon is built by evolving null geodesics backwards from p, classifying generators into black-hole, cosmological-horizon, and caustic generators; the caustic line is where the two horizons first touch and where conical singularities appear. This machinery converts a parameter-to-infinity problem (infinite mass ratio in the Emparan-Martínez limit) into a finite one (Λ→0 of Schwarzschild-de Sitter).

What would settle it

One could construct explicitly a family of orthonormal tetrads along a curve P(Λ) tending to p and compute the resulting Geroch-limit metric; if the limit spacetime's event horizon is not the Emparan-Martínez null hypersurface (for instance, if the limit depends on the chosen reference curve), the central claim would fail. A more direct check would be to compare the induced metric on the limiting horizon cross-sections with the Emparan-Martínez solution.

Watch

Extended reading notes

Core claim

The discovery is that the Λ→0 limit of the Schwarzschild-de Sitter merger, defined through Geroch's limits-of-spacetimes construction with a reference point p on future null infinity and a conformal compactification x=1/r, is the Emparan-Martínez infinite-mass-ratio Schwarzschild merger. This identification is supported numerically: extrapolating the relative area increase of black-hole generators to Λ=0 gives 0.24174, matching Emparan-Martínez to four significant figures, and including the associated caustic generators gives 0.79287, within 0.1% of the earlier result. The identification regularises the otherwise divergent area growth of the large black hole: dividing by the geometric mean of the initial areas, the total regularised area increase tends to 1 in the Λ→0 limit.

Load-bearing premise

The identification of the Λ→0 limit with the Emparan-Martínez merger rests on the assumption that Geroch's construction, with the reference point on future null infinity and the conformal compactification x=1/r, actually converges to a well-defined limit spacetime and that this limit's horizon is the Emparan-Martínez one, even though the required orthonormal tetrads are not explicitly constructed.

Editorial extensions

If this is right

The Schwarzschild-de Sitter merger supplies a finite regularisation of the divergent area increase in the Emparan-Martínez infinite-mass-ratio model; the total regularised area increase is exactly 1 in the limit Λ→0.
Caustic generators dominate the area increase for all M√Λ in (0,1/3), including as Λ→0, so any physical account of area growth in such mergers must centre on caustics.
The relative area increase of black-hole generators in the Λ→0 limit matches Emparan-Martínez (0.24174 vs four significant figures), confirming the limit identification.
In the Nariai limit M√Λ→1/3 the contributions from black-hole and cosmological-horizon generators to the area increase become equal, as expected when the two horizons coincide.
The conical-singularity opening angle scales as (T⋆−T)^{1/2} near merger, corroborating the locally universal merger prediction.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the Geroch-limit identification is correct, similar finite regularisations could be sought for other extreme-mass-ratio mergers, e.g. Kerr-de Sitter limits, where quantities like radiated energy or momentum currently diverge.
The factor-of-two mismatch between the regularised area increase (1) and the finite-mass head-on formula (2) suggests that the counting of horizon degrees of freedom in the infinite-mass-ratio limit is ambiguous; a quasilocal definition of the large black hole's area may resolve it.
The non-monotonic duration of the merger as a function of M√Λ is foliation-dependent; a foliation-independent measure of merger duration would be needed before interpreting this feature physically.
The technique of promoting a parameter to a scalar field, compactifying, and taking a Geroch limit may offer a systematic way to regularise divergences in other parameterised families of spacetimes.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Here's my take. The paper gives a clean, finite model of a black hole merging with a cosmological horizon in Schwarzschild–de Sitter. That part is genuinely new and mostly well done. The horizon cross-sections, the caustic line, the dominance of caustic generators in area growth, and the consistency checks against A(−∞) and A(∞) all look solid. The analytic expansion in Eq. (28) is a nice check on the numerics, and the conical-singularity scaling agreement with Ref. [8] is a real point in its favour.

The soft spot is the Λ→0 limit. The authors claim, following Geroch, that the limit is the Emparan–Martínez infinite-mass-ratio Schwarzschild merger. That is the load-bearing step for the regularisation claim, and it is not proved. The reference point p is placed at u=x=0, which is on I+ in the conformal compactification, not a point of any physical SdS spacetime, and no family of orthonormal tetrads is supplied. The phrase "we seem to recover the precise construction" is an honest admission that the limit is being asserted rather than established. What supports it is the numerical agreement: δA = 0.24174 for black-hole generators and 0.79287 including the relevant caustic generators, both within 0.1% of Emparan–Martínez. That is good evidence, but it is not a convergence proof.

The unresolved factor-of-2 discrepancy between the regularised area increase from the SdS limit (ΔAreg → 1) and the finite-mass formula (ΔAreg = 2) is a direct symptom of the same problem. If the Λ→0 limit were exactly the EM merger, the same normalisation should give the same number. The authors acknowledge this and offer heuristics (half-area geometric mean, E = M/√2 at R = 4μ), but none of that resolves the inconsistency. This does not undermine the horizon-geometry results, but it does mean the regularisation claim should be treated as a conjecture, not a theorem.

Minor point: no code or data is released, though the error estimates and analytic checks give reasonable confidence in the numerics.

Who is this for? People working on exactly solvable horizon mergers, caustics, and the Emparan–Martínez program. They will get real value from the SdS model and the area decomposition. I would send it to peer review, but the referee should be asked to press on Section 3.2.1. If the Geroch limit cannot be made rigorous, the paper should be revised to present the Λ→0 reduction as a well-supported conjecture, with the factor-2 issue discussed as an open problem. As it stands, the central geometric content is solid and the limit claim is interesting but unproven.

Referee Report

3 major / 4 minor

Summary. The paper studies the event horizon of a Schwarzschild-de Sitter spacetime as seen by an observer at future null infinity, modelling the merger of the black hole horizon with the observer's cosmological horizon. The authors numerically construct the event horizon, identify caustic generators, compute the area increase and its decomposition among black-hole, cosmological-horizon, and caustic generators, and check the universal conical-singularity scaling near merger. The central conceptual claim is that the Λ→0 limit of this SdS merger reproduces the Emparan-Martínez infinite-mass-ratio Schwarzschild merger, which the authors use to propose a regularisation of the divergent area increase in that model. The paper also reports an unresolved factor-of-2 discrepancy between this regularised area increase and the corresponding finite-mass two-black-hole formula.

Significance. If the central claim were rigorously established, the paper would provide a useful finite-parameter regularisation of divergent quantities in extreme-mass-ratio mergers and a concrete testbed for the universal properties of horizon caustics. The numerical study is careful: the initial and final areas A(−∞) and A(∞) are exact, Eq. (28) is an analytic expansion of exact formulas, the caustic-opening-angle exponents agree with the universal prediction of Ref. [8], and the extrapolated relative area increases (0.24174 and 0.79287) agree with the Emparan-Martínez values to high precision. However, both pillars of the main claim—the Geroch-limit identification and the consistency of the area regularisation—are not fully established, which substantially limits the significance as the paper stands.

major comments (3)

[§3.2.1, Eqs. (31)–(33)] The claimed Λ→0 limit is not established as a Geroch limit of spacetimes. Geroch's construction requires a curve of points P(Λ) inside each physical spacetime and a family of orthonormal tetrads at those points with a well-defined limit. Here the reference point p is placed at u=x=0 on I+ in the conformally rescaled metric (32), i.e. at x=0, where the physical metric (31) is singular and which is not a point of any SdS spacetime. No family of orthonormal tetrads is constructed, and the text explicitly states that 'we seem to recover the precise construction' rather than proving it. Consequently, the identification of the limiting event horizon with the Emparan-Martínez horizon rests on the numerical agreement of the δA values rather than a convergence theorem for the horizons H_Λ=∂J^-(p). The manuscript should either supply a genuine Geroch-limit argument with points in the physical spacetimes and a convergent tetrad family, or explicitly label the EM identification as a conjecture supported by numerical evidence.
[§3.2.1, Eqs. (34)–(37)] The factor-of-2 discrepancy between the regularised area increase ∆Areg(∞)→1 obtained from the SdS merger and the value ∆Areg=2 from the finite-mass Schwarzschild formula (37) is acknowledged but not resolved. This discrepancy is load-bearing: if the Λ→0 limit were exactly the Emparan-Martínez merger, the same regularisation prescription (division by the geometric mean of the initial areas) should produce the same number in both limits. The suggested fixes—halving the cosmological-horizon area, or replacing M by E=M/√2 at R=4μ—are explicitly heuristic and do not resolve the inconsistency. Since the abstract credits the SdS construction with regularising the otherwise divergent area increase, the unresolved mismatch must be either resolved or clearly demoted from a definite result to an open problem.
[§2, paragraphs after Eq. (15) and footnote 3] The classification of generators into black-hole, cosmological-horizon, and caustic generators relies on the assumption that no generator intersections occur before the symmetric ±q pair meets at θ=π (or before the generator reaches H−). The text states only that 'We have not found any evidence' for such intersections. Since the past endpoint of a generator is defined as the first intersection with any other generator, an unobserved earlier intersection would change both the caustic structure and the area decomposition shown in Figures 4–7. The paper should either prove this no-earlier-intersection property or state it as an explicit assumption and assess how the extrapolated values 0.24174 and 0.79287 might change if it fails.

minor comments (4)

[§2, after Eq. (8)] The phrase 'duetothefreedominchoosinganewaffineparameter' is missing spaces; it should read 'due to the freedom in choosing a new affine parameter'.
[§2, Eq. (14)] The definition of T contains a logarithm that should be typeset with unambiguous absolute-value bars; the current expression can be misread as log((r_C−r)/(2M)) without the vertical bars.
[§3.1, Figure 3] The vertical axis of Figure 3 is unlabelled; please add an axis label such as '−u⋆/M' for clarity.
[§3.2.1, after Eq. (32)] The statement 'the metric trivially reduces to that of a conformally Schwarzschild spacetime' would be easier to verify if the conformal factor and the relation between x and the usual Schwarzschild compactification were displayed explicitly.

Circularity Check

1 steps flagged · score 4.0 of 10

One regularization result reduces by construction (the 'regularised area increase = 1' is fixed by the chosen divisor), but the main Λ→0 identification with the Emparan–Martínez merger is independently supported.

self definitional [Section 3.2.1, Eqs. (28), (34)–(35) and surrounding text]
"In order to regularise this quantity, we divide it by the factor 8√3πM/√Λ, which in the limit of small cosmological constant is the geometric mean of the initial areas of the black hole and the cosmological horizon. Thus, the regularised area increase is given by (34). We then have, from (28), ∆Areg(∞) = 1 − M√Λ/√3 + O(M²Λ). Thus, we see that in the merger of two Schwarzschild black holes in the infinite mass ratio limit, the regularised area increases by only one unit."

Equation (28) gives ∆A(∞)=8√3πM/√Λ+O(M²). The divisor introduced in Eq. (34), 8√3πM/√Λ, is the leading coefficient of that very expansion (equivalently the geometric mean 4πr_B r_C with r_B≈2M, r_C≈√(3/Λ)). Dividing a quantity by its own leading coefficient forces the limit to be 1; the 'only one unit' statement is therefore a normalization, not a derived result. The paper acknowledges the ambiguity through the factor-2 mismatch with Eq. (37), which shows the value depends on the chosen regularisation scale. The main Λ→0 identification with the Emparan–Martínez merger is not circular—it is supported by the extrapolated δA values—but this particular regularization claim reduces by construction.

full rationale

The paper's central numerical derivation is self-contained: horizon generators are obtained by solving the null geodesic equations, areas are computed from the metric, and comparisons with Emparan–Martínez values (0.24174, 0.79287) and the conical-opening-angle scaling are external checks, not fitted inputs. The Geroch-limit discussion in Section 3.2.1 is admittedly incomplete—'we seem to recover the precise construction'—but incompleteness is not circularity. Refs. [3] and [8] are by authors of this paper, but they are used for terminology/method and for a universal scaling law that is independently tested against the numerics; they are not load-bearing self-citations that force the conclusions. The one genuine circular step is the regularised area increase: the normalisation factor in Eq. (34) is exactly the leading coefficient of Eq. (28), so the limit ∆Areg(∞)=1 is fixed by the definition of the regularisation, not by new physics. Because this is a corollary rather than the main Λ→0 identification, and because the authors openly discuss the resulting factor-of-2 ambiguity, the overall circularity is partial but not structural.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters are fitted to derive the central results; M and Lambda are physical spacetime parameters. The axioms are standard horizon definitions plus the specific Geroch-limit assumptions. No new physical entities are introduced; caustic generators and crease sets are geometric constructs from prior literature.

assumptions (4)

domain assumption The event horizon of the merger is the boundary of the causal past of a point p at future null infinity, H+ = J^-(p).
Standard GR definition of an event horizon for an accelerating observer, adapted from Emparan-Martinez; stated in Section 2.
domain assumption Geroch's limits-of-spacetimes formalism provides a valid framework for taking Lambda to zero.
Section 3.2.1 invokes Ref. [9] but does not explicitly construct the required family of orthonormal tetrads attached to the reference point.
ad hoc to paper Generators with equal and opposite impact parameter intersect at theta = pi, and no other generator intersections occur before reaching H- or the caustic.
The authors state in Section 2 that they have not found evidence for other intersections, but no proof is given. If false, not all past endpoints would be caustic points.
ad hoc to paper The conformal compactification x = 1/r with reference point p at future null infinity preserves the horizon generator structure in the Lambda-to-zero limit.
This is the crux of the Geroch limit argument in Section 3.2.1, Eqs. (31)-(33), and is used to identify the limit with the Emparan-Martinez merger.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The merger of a black hole with a cosmological horizon." pith.science (2026). https://pith.science/paper/5PRHWZQP

@misc{pith2026241204551,
  author       = {Pith},
  title        = {Pith review of: The merger of a black hole with a cosmological horizon},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5PRHWZQP}},
  note         = {Machine review of arXiv:2412.04551}
}

read the original abstract

In recent years there have been many studies on exactly solvable black hole mergers, based on a model by Emparan and Martinez where the mass of one black hole is blown up to infinity. Here we replace the large black hole by a cosmological horizon, and study how it merges with a black hole in the Schwarzschild-de Sitter spacetime by considering an observer positioned at future null infinity. We describe the geometry of the horizon over time, including the role that caustics play in the merger process, and also examine the growth of the horizon area. We argue that in the limit of zero cosmological constant, the system reduces to the Emparan-Martinez Schwarzschild merger. This allows us to regularise the increase in the area during the merger, which otherwise diverges.

Figures

Figures reproduced from arXiv: 2412.04551 by the authors.

**Figure 2.** Evolution of the event horizon with Λ = 0.1, M = 0.5. At early times, there are two nearly-spherical disconnected horizon cross-sections. These develop increasingly-stretched and pointed tips which approach each other. The tips are caustic points. At T ≈ −1.953 the tips touch: this is the point of merger in this foliation. At later times, the now connected horizon becomes convex and smooths out into a spherical shap… view at source ↗

**Figure 3.** Duration of the merger, characterised by [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: Area increase ∆A(∞), in units of 1/Λ, as a function of M √ Λ. The area increase from black hole, cosmological horizon and caustic generators are plotted with blue discs, green squares and orange triangles, respectively. The red stars give the total area increase determ…

**Figure 5.** Figure 5: Area increase ∆A(T) for the merger with Λ = 0.1, M = 0.5. Contributions from black hole, cosmological horizon and caustic generators are shown in dash-dotted blue, dotted green, and dashed orange, respectively. The solid red curve gives the sum of these. For smaller M …

**Figure 6.** Figure 6: Relative area increase δA(∞) as a function of M √ Λ. The relative increase from black hole generators and cosmological horizon generators are plotted with blue discs and green squares, respectively. The orange triangles represent the black hole generators plus the caus…

**Figure 7.** Figure 7: Regularised area increase ∆Areg(∞) as a function of M √ Λ. The area increase from black hole, cosmological horizon and caustic generators are plotted with blue discs, green squares and orange triangles, respectively. The red stars give the total regularised area increa…

**Figure 8.** Figure 8: Opening angle of the conical singularities for the merger with [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

15 extracted references · 13 canonical work pages

[8]

Gadioux and H

M. Gadioux and H. S. Reall,Creases, corners, and caustics: Properties of nonsmooth structures on black hole horizons , Phys. Rev. D 108 (2023) 084021, [2303.15512]

arXiv 2023
[1]

Exact Event Horizon of a Black Hole Merger

R. Emparan and M. Martínez,Exact Event Horizon of a Black Hole Merger , Class. Quant. Grav. 33 (2016) 155003, [1603.00712]. 20

work page Pith review arXiv 2016
[2]

Emparan, M

R. Emparan, M. Martínez and M. Zilhao,Black hole fusion in the extreme mass ratio limit , Phys. Rev. D 97 (2018) 044004, [1708.08868]

arXiv 2018
[3]

Evolution of creases on the event horizon of a black hole merger

M. Gadioux, R. A. Hennigar and H. S. Reall,Evolution of creases on the event horizon of a black hole merger , Phys. Rev. D 110 (2024) 084029, [2407.07962]

work page Pith review arXiv 2024
[4]

Precursory collapse in Neutron Star-Black Hole mergers

R. Emparan and D. Marín,Precursory collapse in Neutron Star-Black Hole mergers , Phys. Rev. D 102 (2020) 024009, [2004.08143]

work page Pith review arXiv 2020
[5]

D. M. Pina, M. Orselli and D. Pica,Event horizon of a charged black hole binary merger , Phys. Rev. D 106 (2022) 084012, [2204.08841]

work page Pith review arXiv 2022
[6]

J. M. Dias, A. M. Frassino, V. D. Paccoia and J. V. Rocha,Black hole-wormhole collisions and the emergence of islands , Phys. Rev. D 107 (2023) 124056, [2304.06098]

work page Pith review arXiv 2023
[7]

J. M. Dias, A. M. Frassino, D. C. Lopes, V. D. Paccoia and J. V. Rocha,The impact of higher derivative corrections to General Relativity on black hole mergers , 2407.12947

work page Pith review arXiv

Show all 15 references

[9]

R. P. Geroch,Limits of spacetimes, Commun. Math. Phys. 13 (1969) 180–193

1969
[10]

Hackmann and C

E. Hackmann and C. Lammerzahl,Geodesic equation in Schwarzschild- (anti-) de Sitter space-times: Analytical solutions and applications , Phys. Rev. D 78 (2008) 024035, [1505.07973]

2008 arXiv
[11]

Bengtsson, S

I. Bengtsson, S. Holst and E. Jakobsson,Classics Illustrated: Limits of Spacetimes , Class. Quant. Grav. 31 (2014) 205008, [1406.4326]

2014 arXiv
[12]

Bugden and C

M. Bugden and C. Paganini,The Λ to zero limit of spacetimes and its physical interpretation , Class. Quant. Grav. 36 (2019) 045003, [1810.00436]

2019 arXiv
[13]

G. W. Gibbons and S. W. Hawking,Cosmological Event Horizons, Thermodynamics, and Particle Creation, Phys. Rev. D 15 (1977) 2738–2751

1977
[14]

R. C. Myers and A. Singh,Entanglement Entropy for Singular Surfaces , JHEP 09 (2012) 013, [1206.5225]

2012 arXiv
[15]

Jeffrey and D

A. Jeffrey and D. Zwillinger, eds.,Table of integrals, series and products / I.S. Gradshteyn and I.M. Ryzhik . Elsevier, Amsterdam ; Boston, 7th ed., 2007. 21

2007

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [8]

Gadioux and H

M. Gadioux and H. S. Reall,Creases, corners, and caustics: Properties of nonsmooth structures on black hole horizons , Phys. Rev. D 108 (2023) 084021, [2303.15512]

arXiv 2023

[2] [1]

Exact Event Horizon of a Black Hole Merger

R. Emparan and M. Martínez,Exact Event Horizon of a Black Hole Merger , Class. Quant. Grav. 33 (2016) 155003, [1603.00712]. 20

work page Pith review arXiv 2016

[3] [2]

Emparan, M

R. Emparan, M. Martínez and M. Zilhao,Black hole fusion in the extreme mass ratio limit , Phys. Rev. D 97 (2018) 044004, [1708.08868]

arXiv 2018

[4] [3]

Evolution of creases on the event horizon of a black hole merger

M. Gadioux, R. A. Hennigar and H. S. Reall,Evolution of creases on the event horizon of a black hole merger , Phys. Rev. D 110 (2024) 084029, [2407.07962]

work page Pith review arXiv 2024

[5] [4]

Precursory collapse in Neutron Star-Black Hole mergers

R. Emparan and D. Marín,Precursory collapse in Neutron Star-Black Hole mergers , Phys. Rev. D 102 (2020) 024009, [2004.08143]

work page Pith review arXiv 2020

[6] [5]

D. M. Pina, M. Orselli and D. Pica,Event horizon of a charged black hole binary merger , Phys. Rev. D 106 (2022) 084012, [2204.08841]

work page Pith review arXiv 2022

[7] [6]

J. M. Dias, A. M. Frassino, V. D. Paccoia and J. V. Rocha,Black hole-wormhole collisions and the emergence of islands , Phys. Rev. D 107 (2023) 124056, [2304.06098]

work page Pith review arXiv 2023

[8] [7]

J. M. Dias, A. M. Frassino, D. C. Lopes, V. D. Paccoia and J. V. Rocha,The impact of higher derivative corrections to General Relativity on black hole mergers , 2407.12947

work page Pith review arXiv