Concentration of Small Hawking Type Surfaces

Alexander Friedrich

REVIEW 3 major objections 4 minor 1 cited by

Concentration of Small Hawking Type Surfaces

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Small Hawking-energy surfaces concentrate only at stationary points of a specific curvature combination.

desk verdict A genuinely new extension of the Willmore concentration machinery to Hawking-type functionals, with explicit K-dependent formulas, but the classification of concentration points is conditional on an energy bound that the definition does not guarantee. read the letter →

arxiv 1909.02388 v2 pith:4EAWKWBQ submitted 2019-09-05 math.DG math-phmath.APmath.MP

Alexander Friedrich This is my paper · ORCID

classification math.DGmath-phmath.APmath.MP MSC 53C2149Q1058E12

keywords Hawkingenergyquasi-localWillmorefunctionalgeneralizedbubbletreearea-constrainedminimizerconcentrationpointsmallsurfaces

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

This paper sets out to show that the Hawking energy, a quasi-local notion of gravitational energy assigned to surfaces in a spacetime, can be studied on small surfaces in a spacelike slice as a generalized Willmore functional. The central claim is that area-constrained critical surfaces of Hawking type functionals share the compactness, regularity, and concentration behavior established for the Willmore functional: minimizers exist as haunted bubble trees, become embedded spheres at small area, and concentrate only at points characterized by vanishing of a vector built from moments of the lower-order term. For the Hawking energy itself, the characterization reads $\nabla_M(Sc_p + \frac{3}{5}\operatorname{tr}K_p^2 + \frac{1}{5}|K_p|^2)=0$, and the small-sphere expansion is $E[\Sigma] - \frac{1}{12}(|\Sigma|/4\pi)^{3/2}(Sc_p+\frac{3}{5}\operatorname{tr}K_p^2+\frac{1}{5}|K_p|^2)=O(|\Sigma|^2)$. If correct, this gives a geometric selection rule for where small Hawking-energy maximizers can sit, and it exposes a contrast with light-cone expansions, in which the energy density $\rho=(Sc+(\operatorname{tr}K)^2-|K|^2)/16\pi$ appears instead.

What carries the argument

The central object is the Hawking type functional $H[\Sigma]=W[\Sigma]+\int_\Sigma L(x,\nu)d\mu$, whose area-constrained Euler-Lagrange equation is the generalized Willmore equation $\Delta H+H|\mathring A|^2+HQ+\gamma(\mathring A,S)+2\lambda H+T=0$. The argument runs on three pieces of machinery. First, the compactness and regularity theory for haunted bubble trees, stratified surfaces made of spheres joined at points with some components allowed to be constant, which supplies the minimizers of Theorem 2.9. Second, the small-surface roundness estimates of Section 3.1, which bound the trace-free second fundamental form by $C|\Sigma|$ and force the mean curvature close to $2/R$, making concentrating surfaces nearly umbilical. Third, the approximation of a small critical surface by a round sphere with controlled center of mass, which converts the first variation into moment integrals $c_{(\alpha_1,\dots,\alpha_k)}(F,a)$ over the unit sphere; these integrals define the concentration vector $V_p$ and gradient vector $W_p$. For the Hawking functional, explicit sphere integrals telescope to $\frac{4}{5}\partial_\alpha(3\operatorname{tr}K^2+|K|^2)$, which is the mechanism behind the main gradient condition.

What would settle it

Exhibit, in a CB-bounded spacelike slice where $\nabla_M(Sc+\frac{3}{5}\operatorname{tr}K^2+\frac{1}{5}|K|^2)$ never vanishes, a sequence of area-constrained critical spheres of the Hawking functional concentrating at some point; the theorem would force the gradient to vanish there, so such a sequence would refute the characterization. A more direct check is to compute $H[\Sigma_r]$ for a candidate concentrating sequence: if the value exceeds $4\pi+\epsilon_0^2$, the roundness estimates on which the conclusion rests do not apply.

Watch

Extended reading notes

Core claim

The central discovery is that the Hawking type functional $H[\Sigma]=W[\Sigma]+\int_\Sigma L(x,\nu)\,d\mu$ is a generalized Willmore functional whenever $L$ is smooth and bounded, and this single fact drives the whole argument. From it, the paper proves an existence and regularity theorem: on a compact CB-bounded three-manifold, the infimum of $H$ among haunted, branched, immersed bubble trees of any prescribed area is attained; for sufficiently small area every minimizer is an embedded sphere contained in a normal coordinate neighborhood, with $|H[\Sigma_a]-4\pi|\le C a$. Following the small-surface Willmore program, the paper then characterizes concentration points. For a general Hawking type functional, a concentration point must satisfy the vanishing of a vector $V_p$ of moment integrals of $L$; when $V_p$ vanishes, the further condition $\nabla_M Sc_p - W_p=0$ holds for an explicitly computed vector $W_p$. For the functional corresponding to the Hawking energy, $L=-\frac14(\operatorname{tr}_\Sigma K)^2$ is even in the normal, so $V_p$ vanishes identically, and the computation collapses to $\nabla_M(Sc_p+\frac{3}{5}\operatorname{tr}K_p^2+\frac{1}{5}|K_p|^2)=0$.

Load-bearing premise

The classification of concentration points assumes every concentrating family obeys the small-energy bound $H[\Sigma_r]\le 4\pi+\epsilon_0^2$, but the proof establishes this bound only for the global area-constrained minimizers produced by the existence theorem, not for arbitrary area-constrained critical surfaces.

Editorial extensions

If this is right

Small area-constrained maximizers of the Hawking energy are embedded round spheres; no necks or bubble trees can form in the small-area limit.
Concentration points of Hawking-energy maximizers are stationary points of $Sc+\frac35\operatorname{tr}K^2+\frac15|K|^2$, not of the energy density $\rho$.
The expansion $E[\Sigma]=\frac{1}{12}(|\Sigma|/4\pi)^{3/2}(Sc_p+\frac35\operatorname{tr}K_p^2+\frac15|K_p|^2)+O(|\Sigma|^2)$ gives a concrete quasi-local estimate for the Hawking energy of small spheres.
Any Hawking type functional with $L$ even in the normal obeys the same concentration principle, with a computable vector $W_p$ replacing the scalar gradient.

Reading between the lines

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

Beyond the paper: if the small-energy bound is truly necessary, area-constrained critical surfaces with energy above $4\pi+\epsilon_0^2$ could concentrate at points outside the theorem's characterization; constructing such sequences would delimit the theorem's scope.
Beyond the paper: the discrepancy between the spacelike-slice expansion and the known light-cone expansion suggests that the Hawking energy evaluated on spacelike spheres is not a direct measure of local energy density; comparing both expansions on the same spacetime could reveal which quantity the functional actually tracks.
Beyond the paper: the moment-integral machinery should transfer to other quasi-local energies of the form $W+\int L$ with $L$ even, producing gradient conditions $\nabla(Sc-w)=0$ for computable functions $w$; such conditions would give selection rules for the small surfaces that extremize those energies.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

The main novelty is real: Friedrich treats Hawking-type functionals H_L = W + ∫ L(x,ν) as generalized Willmore functionals and derives explicit first-order concentration conditions and a small-sphere expansion with new K-dependent terms. The concentration vector and the expansion coeﬃcients (3/5 trK² + 1/5 |K|²) are not in Lamm–Metzger or Horowitz–Schmidt, and the paper reduces correctly to the earlier Willmore results when L = 0. The comparison with Horowitz–Schmidt is handled sensibly: the discrepancy comes from spacelike versus null surfaces, and the author says so plainly. Appendix B's sphere integrals are explicit and the algebra checks out. I also credit the paper for being open about what is imported from [5] and [11] rather than hiding those dependencies.

The soft spot is exactly what the stress-test note identiﬁes, and it is a genuine gap between the abstract and the theorems. Definition 3.4 deﬁnes a concentration point using arbitrary spherical area-constrained critical surfaces, with no bound on H[Σ_r]. Theorems 3.6(2) and 3.7 classify only those concentration points whose surfaces satisfy H ≤ 4π + ε₀². That bound is proved in Theorem 2.9 for the minimizers constructed there, but nothing in the deﬁnition forces a critical concentrating family to be minimizing. Without that bound, Proposition 3.2 and Corollary 3.3 (roundness, mean-curvature control) do not apply, and the De Lellis–Müller approximation that underlies the concentration-vector computation is unavailable. So the phrase “characterize the concentration points” overstates what is proven. The theorem as stated is conditional; the deﬁnition should either include the energy bound or the result should be phrased as a characterization of concentration points arising from minimizers.

Is this fatal? Not to the core mathematical content. The main new calculations are independent of that gap: Theorem 1.3 and Corollary 3.9 are expansion results with their own hypotheses, and they stand. The conditional theorem is still a valuable structural statement for the minimizing families that the paper constructs. What needs work is framing and precision: either weaken the abstract or strengthen Deﬁnition 3.4 so it guarantees the small-energy hypothesis. The deferred proof of Corollary 3.3 and the last estimate of Proposition 3.2 are also worth noting, but they are standard material from [11], and this is a companion-paper situation, not a hidden circularity.

Who gets value from this: geometric analysts working on Willmore-type functionals and quasi-local mass. It deserves a serious referee, and I would send it to peer review rather than desk-reject. With a modest revision tightening the concentration-point deﬁnition and the claims, it would be a solid contribution.

Recommendation: accept with revisions, after a careful referee check of the companion papers.

Referee Report

3 major / 4 minor

Summary. The paper introduces Hawking type functionals H[Σ] = W[Σ] + ∫_Σ L(x,ν)dµ on surfaces in a Riemannian three-manifold, motivated by the Hawking energy of surfaces in a spacelike slice. It claims that such functionals are generalized Willmore functionals, that area-constrained minimizers exist as haunted bubble trees and are embedded spheres for small area (Theorem 2.9), and that small spherical critical surfaces obey roundness and energy estimates (Section 3.1). The central analytic result is a concentration-point characterization: at a concentration point p with concentrating surfaces satisfying the small-energy bound H[Σ_r] ≤ 4π + ε0², the gradient of Sc_p + (3/5)tr K_p² + (1/5)|K_p|² vanishes for the Hawking-energy functional (Theorem 3.7). The paper also gives an expansion of the Hawking energy on small spheres (Corollary 3.9). The proofs rely on the companion paper [5] for existence and regularity and on [10, 11] for several estimates, and the main classification is conditional on the small-energy hypothesis.

Significance. If the results hold, they provide a systematic Willmore-type treatment of the Hawking energy in arbitrary spacelike slices without symmetry, with an explicit expansion that differs from the Horowitz–Schmidt light-cone expansion. The explicit sphere integrals in Appendix B and the vector computations in Theorem 3.7 are concrete and verifiable, and the paper makes a falsifiable prediction about where small area-constrained maximizers of the Hawking energy concentrate. The main limitations are the dependence on companion/previous works and the mismatch between the definition of concentration point and the energy hypothesis needed for the classification; these are repairable but currently leave the abstract's 'characterize the concentration points' stronger than the theorems prove.

major comments (3)

[Definition 3.4, Theorems 3.6(2) and 3.7] Definition 3.4 defines a concentration point without any bound on H[Σ_r], yet Theorem 3.6(2) and Theorem 3.7 classify only concentration points whose concentrating surfaces satisfy H[Σ_r] ≤ 4π + ε0², with ε0 from Proposition 3.2. That energy bound is established in Theorem 2.9 only for the area-constrained minimizers used in Theorem 3.6(1); nothing in Definition 3.4 forces an arbitrary concentrating family to consist of minimizers. Without H ≤ 4π + ε0², Proposition 3.2 and Corollary 3.3 (roundness, |H − 2/R|, |H^{-1}| bounds) are unavailable, so the De Lellis–Müller approximation in Theorem A.8 and the subsequent expansion of δ_f L cannot be justified. This is a load-bearing gap between the statement and the proof. I recommend either adding the small-energy bound to Definition 3.4, or restating the classification theorems explicitly for 'small-energy concentration points' and adjusting the abstract accordingly.
[Proposition 3.2, Theorem 3.6(2)] Proposition 3.2 requires H(Σ) ≤ 4π + ε² for some ε ∈ (0,ε0) together with |Σ| ≤ ε², but Theorem 3.6(2) assumes only H[Σ_r] ≤ 4π + ε0² for the fixed threshold ε0. As stated, the latter condition does not imply the former: a surface with H = 4π + ε0² cannot be fed into Proposition 3.2 with ε < ε0. The proof silently treats these hypotheses as interchangeable. This is easily fixed by restating Proposition 3.2 with ε ∈ (0,ε0] or by replacing ε0 in Theorems 3.6 and 3.7 with a smaller constant obtained from Proposition 3.2, but the current text needs a correction or an explicit remark.
[Section 2 and Section 3.1] Theorem 2.9 (and hence Theorem 1.1) is not self-contained: existence, regularity, and compactness of area-constrained minimizers are imported from the companion paper [5], while the crucial final estimate in Proposition 3.2 and the entire Corollary 3.3 are deferred to [11]. This is legitimate organization, but the main results inherit every unproved statement of those works. The manuscript should state this dependence clearly in the theorem statements or in a remark at the start of Section 3, rather than only in proof sketches.

minor comments (4)

[Page 2, Theorem 1.1] There is a typo: 'iniﬁmum' should be 'infimum'.
[Lemma 2.7 proof] The integration by parts written as ∫_Σ dVL(∇Σ f)dµ = ∫_Σ divΣ(f dVL) − f divΣ(dVL(x,ν)) dµ uses the notation dVL(x,ν), which is not defined; it appears to denote the Σ-trace of ∇^M dVL evaluated on (·,ν), but this should be clarified to make the derivation of (2.3) verifiable.
[Corollary 3.9] The statement of Corollary 3.9 assumes only L ∈ C¹, while the proof uses Taylor expansion of L in both variables and the estimates of Section 3.1, which require boundedness of dTML, HessV L, and ∇M dVL. Please align the hypotheses (e.g., assume L ∈ C² with the bounds listed at the beginning of Section 3).
[Abstract and Introduction] The abstract states that the paper 'characterizes the concentration points' without mentioning the small-energy hypothesis H[Σ_r] ≤ 4π + ε0² that is present in Theorems 1.2, 3.6(2), and 3.7; the abstract should include this condition to match the theorems.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: concentration classification follows from first-variation estimates and explicit sphere integrals; the companion-paper citation [5] supplies independent existence theory.

full rationale

The derivation chain is not circular. Theorem 1.2 / Theorem 3.6(2) / Theorem 3.7 are obtained by fixing an area-constrained critical sphere Σ_r, using the first-variation identity λδ_f A = δ_f (U+V+L), estimating each term in adapted normal coordinates, and pulling the L-variation back to a Euclidean sphere via the De Lellis–Müller rigidity estimates (Theorem A.8). The final condition ∇_M(Sc_p + 3/5 trK² + 1/5 |K|²) = 0 is a computed consequence of the explicit sphere integrals in Appendix B; it is not assumed, fitted, or encoded in the definition of the Hawking type functional. The small-surface expansion (Corollary 3.9) combines the Lamm–Metzger Willmore expansion with evaluations of c(L,p); again no fitted parameter is relabelled as a prediction. The only non-self-contained input is the author's companion paper [5], cited for compactness/regularity and existence of minimizers (Theorem 2.9). That is a same-author dependency, but [5] proves those results in a separate framework and is not used to assume the concentration or expansion conclusions; it is independent support rather than a circular reduction. The conditional energy bound H[Σ_r] ≤ 4π + ε0², which is assumed in the classification but not guaranteed by Definition 3.4 for arbitrary critical surfaces, is a gap or missing hypothesis in the statement, not a circular step: it does not make the conclusion equivalent to the input. The analysis is therefore self-contained apart from normal external citations, and no circularity is present.

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

No data were fitted; all constants are existential. The mathematical load is carried by bounded-geometry hypotheses and by importing compactness and regularity from the companion paper [5] and rigidity estimates from [10, 11].

assumptions (4)

domain assumption M has CB-bounded geometry: injectivity radius bounded below uniformly, and |Rm| + |∇Rm| bounded.
Used throughout: normal coordinates, comparison of Euclidean and curved quantities in Appendix A, and estimates in Section 3.
domain assumption L, dTM L, Hess_V L, and ∇^M dV L are smooth and bounded on TM, with bounded norm C_L.
Assumed so that Hawking type functionals satisfy the generalized Willmore conditions and so variation terms can be controlled in Section 3.1.
domain assumption The compactness, regularity, and minimization theorems for generalized Willmore functionals from [5, Theorems 4.5 and 5.6] hold for the class of haunted, branched bubble trees.
Theorem 2.9 is proved by direct reference to the author's companion paper; this paper does not reprove those theorems.
standard math De Lellis-Muller rigidity and the Lamm-Metzger surface estimates (Theorem A.8, Corollaries A.5 and A.10, Lemma A.11) are correct as quoted.
Used in Theorem 3.6 to replace Σ by a round sphere for the expansion of δL and to compute the concentration vector.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Concentration of Small Hawking Type Surfaces." pith.science (2026). https://pith.science/paper/4EAWKWBQ

@misc{pith2026190902388,
  author       = {Pith},
  title        = {Pith review of: Concentration of Small Hawking Type Surfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4EAWKWBQ}},
  note         = {Machine review of arXiv:1909.02388}
}

read the original abstract

We investigate the Hawking energy of small surfaces in space times without symmetry assumptions by introducing the notion of Hawking type functionals. In particular, we find that Hawking type functionals are generalized Willmore functionals which allows us to find area constrained, minimizing, immersed, haunted bubble trees. These bubble trees are smooth spheres provided their area is small enough. Following a similar analysis of the Willmore functional conducted by T. Lamm and J. Metzger we characterize the concentration points of area constrained, critical surfaces for Hawking type functionals and the Hawking energy. Moreover, we determine their expansion on small surfaces.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Minimizers of Generalized Willmore Functionals
math.DG 2019-09 conditional novelty 6.0 of 10

A generalized Willmore framework yields existence of area-constrained, and area-volume-constrained, minimizers among haunted bubble trees, with partial regularity for critical points.

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages · cited by 1 Pith paper

[5]

Minimizers of Generalized Willmore Functionals

Alexander Friedrich. Minimizers of Generalized Willmo re Functionals. arXiv e-prints, September 2019, 1909.02381

work page Pith review arXiv 2019
[11]

Minimizers of the Willmore functional with a small area constraint

Tobias Lamm and Jan Metzger. Minimizers of the Willmore functional with a small area constraint. Annales de l’Institut Henri Poincare. Annales: Analyse Non Lineaire/Nonlinear Analysis, 30(3):497– 518, 2013, arXiv:1201.1887v2

work page Pith review arXiv 2013
[1]

Bubble tree of a class of confo rmal mappings and applications to Willmore functional

Jingyi Chen and Yuxiang Li. Bubble tree of a class of confo rmal mappings and applications to Willmore functional. American Journal of Mathematics , 136, August 2014

work page 2014
[2]

Optimal rigidity es timates for nearly umbilical surfaces

Camillo De Lellis and Stefan Müller. Optimal rigidity es timates for nearly umbilical surfaces. Journal of Diﬀerential Geometry , 69(1):075–110, 2005

work page 2005
[3]

A C 0 estimate for nearly umbilical surfaces

Camillo De Lellis and Stefan Müller. A C 0 estimate for nearly umbilical surfaces. Calculus of Variations and Partial Diﬀerential Equations , 26(3):283–296, 2006

work page 2006
[4]

Normally Hyperbolic Invariant Manifolds - The Noncompact C ase

Jaap Eldering. Normally Hyperbolic Invariant Manifolds - The Noncompact C ase. Altlantis Press, 2013

work page 2013
[6]

Horowitz and B.G

G.T. Horowitz and B.G. Schmidt. Note on gravitational en ergy. Proceedings of the Royal Society of London A , 381:215–224, 1982

work page 1982
[7]

Geometric evolut ion equations for hypersurfaces

Gerhard Huisken and Alexander Polden. Geometric evolut ion equations for hypersurfaces. Lecture Notes in Mathematics , pages 45–84, 1999

work page 1999

Show all 17 references

[8]

W 2, 2-conformal immersions of closed Riemann surfaces into Rn

Ernst Kuwert and Yuxiang Li. W 2, 2-conformal immersions of closed Riemann surfaces into Rn . Communications in Analysis and Geometry , 20:313–340, 2012

2012
[9]

Removability of point singul arities of Willmore surfaces

Ernst Kuwert and Schätzle. Removability of point singul arities of Willmore surfaces. Annals of Math- ematics, 160:315–357, 2004

2004
[10]

Small surfaces of Willmore type in Riemannian manifolds

Tobias Lamm and Jan Metzger. Small surfaces of Willmore type in Riemannian manifolds. Intl. Math. Res. Notices, pages 3786–3813, 2010, arXiv:0909.0590v2

2010 arXiv
[12]

A New Conformal Invariant an d Its Applications to the Wilmore Con- jecture and the First Eigenvalue of Compact Surfaces

Peter Li and Sing-Tung Yau. A New Conformal Invariant an d Its Applications to the Wilmore Con- jecture and the First Eigenvalue of Compact Surfaces. Inventiones Mathematicae, 69:269–291, 1982

1982
[13]

Reﬁned position estimates for surfaces of Willmore type in Riemannian manifolds

Jan Metzger. Reﬁned position estimates for surfaces of Willmore type in Riemannian manifolds. arXiv e-prints, Aug 2019, 1908.11577

2019 arXiv
[14]

Some results about the existence of cri tical points for the Willmore functional

Andrea Mondino. Some results about the existence of cri tical points for the Willmore functional. Mathematische Zeitschrift , 266(3):583–622, Nov 2010

2010
[15]

Willmore spheres i n compact Riemannian manifolds

Andrea Mondino and Tristan Rivière. Willmore spheres i n compact Riemannian manifolds. Advances in Mathematics , 232(1):608 – 676, 2013

2013
[16]

Szabados

László B. Szabados. Quasi-Local Energy-Momentum and A ngular Momentum in GR: A Review Ar- ticle. Living Reviews in Relativity , 7(1):4, Mar 2004

2004
[17]

Hawking S. W. Gravitational Radiation in an Expanding U niverse. Journal of Mathematical Physics , 9(4):598–604, 1968. University of Potsdam, Institute of Ma thema tics, Karl-Lie bknecht-Str. 24-25, D-14476 Potsdam OT Golm

1968

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [5]

Minimizers of Generalized Willmore Functionals

Alexander Friedrich. Minimizers of Generalized Willmo re Functionals. arXiv e-prints, September 2019, 1909.02381

work page Pith review arXiv 2019

[2] [11]

Minimizers of the Willmore functional with a small area constraint

Tobias Lamm and Jan Metzger. Minimizers of the Willmore functional with a small area constraint. Annales de l’Institut Henri Poincare. Annales: Analyse Non Lineaire/Nonlinear Analysis, 30(3):497– 518, 2013, arXiv:1201.1887v2

work page Pith review arXiv 2013

[3] [1]

Bubble tree of a class of confo rmal mappings and applications to Willmore functional

Jingyi Chen and Yuxiang Li. Bubble tree of a class of confo rmal mappings and applications to Willmore functional. American Journal of Mathematics , 136, August 2014

work page 2014

[4] [2]

Optimal rigidity es timates for nearly umbilical surfaces

Camillo De Lellis and Stefan Müller. Optimal rigidity es timates for nearly umbilical surfaces. Journal of Diﬀerential Geometry , 69(1):075–110, 2005

work page 2005

[5] [3]

A C 0 estimate for nearly umbilical surfaces

Camillo De Lellis and Stefan Müller. A C 0 estimate for nearly umbilical surfaces. Calculus of Variations and Partial Diﬀerential Equations , 26(3):283–296, 2006

work page 2006

[6] [4]

Normally Hyperbolic Invariant Manifolds - The Noncompact C ase

Jaap Eldering. Normally Hyperbolic Invariant Manifolds - The Noncompact C ase. Altlantis Press, 2013

work page 2013

[7] [6]

Horowitz and B.G

G.T. Horowitz and B.G. Schmidt. Note on gravitational en ergy. Proceedings of the Royal Society of London A , 381:215–224, 1982

work page 1982

[8] [7]

Geometric evolut ion equations for hypersurfaces

Gerhard Huisken and Alexander Polden. Geometric evolut ion equations for hypersurfaces. Lecture Notes in Mathematics , pages 45–84, 1999

work page 1999