The vanishing rate of Weil-Petersson sectional curvatures

Scott A. Wolpert

REVIEW 2 major objections 4 minor 17 references

The vanishing rate of Weil-Petersson sectional curvatures

T0 review · 2 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 This paper proves that for surfaces with short geodesics, every Weil-Petersson sectional curvature is at most −C*σ7, where σ is the product of the small geodesic lengths.

desk verdict Wolpert quantifies the Weil-Petersson sectional curvature vanishing rate with a σ^7 bound and illustrates the expected σ^3 rate; the result is solid in outline, but Proposition 3's collar-crossing lemma needs proof. read the letter →

arxiv 1908.09859 v1 pith:SGTMEK26 submitted 2019-08-26 math.DG

Scott A. Wolpert This is my paper · ORCID

classification math.DG MSC 30F6053C2132G1531A10

keywords Weil-PeterssonmetricsectionalcurvatureTeichmüllerspacemodulihyperbolicsurfacesGreen'sfunctiongeodesic-lengthfunctionscollardecomposition

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 determine how quickly the Weil-Petersson metric on moduli space becomes flat as a hyperbolic surface develops short closed geodesics. Its main theorem states that there is a constant $C_*$ depending only on the topological type such that every sectional curvature is at most $-C_*\sigma^7$, where $\sigma$ is the product of the lengths of all short geodesics. In other words, even though the metric looks like a product near the boundary and has nearly flat 2-planes, its negative curvature cannot vanish faster than this seventh-power rate. The paper also gives examples showing that the actual vanishing rate is often faster than the general bound, and argues that the optimal exponent is probably 3.

What carries the argument

The load-bearing mechanism is the propagation decay of the Green's function $G(p,q)$ of $\Delta = -2(D-2)^{-1}$, written as a sum of $e^{-2d(p,\gamma q)}$ over the uniformization group. The twisting-number parameterization of simple arcs -- by integer winding around each collar annulus -- turns this sum into geometric series indexed by twists, and each full crossing of a collar of length $\ell_\alpha$ contributes a factor $\ell_\alpha^3$ to the lower bound, while a half crossing contributes $\ell_\alpha$. These estimates yield Corollary 4, $G(p,q) \ge C''\sigma^3$. The curvature step then uses the identity $R_{\alpha\bar\beta\gamma\bar\delta} = (\alpha\bar\beta, \gamma\bar\delta) + (\alpha\bar\delta, \gamma\bar\beta)$ together with Bochner's formula for sectional curvature and the Hölder inequalities that originally established negativity, to compare quartic pairings and extract the $\sigma^7$ bound.

What would settle it

One concrete check is to compute, on a sequence of hyperbolic surfaces degenerating along a fixed simple closed geodesic, the ratio of any sectional curvature to $\sigma^7$; if the ratio tends to $0$ for some 2-plane, the theorem is false. A more direct test of the intermediate claim is to search for a sequence with short geodesics in which a shortest simple arc enters and leaves the same collar through the same boundary: if such arcs exist for arbitrarily small collar lengths, Proposition 3's lower bound fails.

Watch

Extended reading notes

Core claim

The central claim is a definite lower bound on the size of Weil-Petersson sectional curvature in the degenerating direction: for surfaces with sufficiently small geodesics, the curvature of any 2-plane is at most $-C_*\sigma^7$, with $\sigma$ the product of the small geodesic lengths. The engine is a propagation-decay estimate for the Green's function of the operator $\Delta = -2(D-2)^{-1}$ built from the hyperbolic Laplacian $D$. Crossing a collar around a short geodesic of length $\ell_\alpha$ suppresses the Green's function by a factor proportional to $\ell_\alpha^3$, while crossing half a collar suppresses it by $\ell_\alpha$; summing over all simple arcs by their twisting numbers gives the uniform lower bound $G(p,q) \ge C''\sigma^3$. Combined with Hölder and mean-value estimates for harmonic Beltrami differentials, this forces the quartic pairings that build the curvature tensor to differ by at least a multiple of $\sigma^7$, giving the theorem.

Load-bearing premise

The proof assumes that once the tubes around the short closed geodesics are narrow enough, a shortest path between two points never leaves a tube and then re-enters it through the same opening; the paper does not say how narrow that is, and this control is what makes the $\sigma^3$ lower bound on the Green's function hold.

Editorial extensions

If this is right

If the theorem is correct, then every sequence of surfaces degenerating to a boundary stratum has sectional curvatures bounded away from zero at scale $\sigma^7$: the metric cannot become infinitely flat relative to the product of the short lengths.
The Green's function estimate gives a uniform rule for interactions across a degenerating collar: a full collar of length $\ell$ contributes a factor $\ell^3$ and a half-collar contributes $\ell$ in every curvature pairing.
For pinching along a standard homology basis, where a minimal arc crosses at most two half-collars, the relevant bound is the product of the two smallest lengths to the third power, not $\sigma^7$.
In the three model cases -- two thick regions joined by a collar, a collar adjacent to a thick region, and two collars adjacent to a thick region -- the vanishing rates are $\ell^3$, $\ell$, and $\ell_1\ell_2$, respectively, consistent with the paper's conjecture that the optimal universal exponent is 3.

Reading between the lines

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

A numerical test on explicit degenerating families could check whether the ratio of sectional curvature to $\sigma^7$ stays bounded below; if the observed exponent is consistently smaller than 7, the proof's symmetric product $\sigma^7$ is not the sharp mechanism.
The same twisting-number counting of propagation paths might predict vanishing rates for other Green's-function quantities on Teichmüller space, such as the Hessian of geodesic-length functions or the asymptotics of Ricci and scalar curvature.
Making the 'sufficiently small' collar threshold in Proposition 3 explicit would turn the existence constant $C_*$ into a computable quantity, which could sharpen quantitative convexity and geodesic arguments near the boundary.

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

Colleague,

Here is the short version. The paper proves that near the boundary of moduli space, every Weil-Petersson sectional curvature is bounded above by −C* σ^7, where σ is the product of the short geodesic lengths. That is a real step forward: prior work, including Wolpert's own, mostly said the curvature tends to zero without quantifying the rate. The three examples (adjacent thick regions, a collar adjacent to a thick region, two collars) give vanishing rates ℓ^3, ℓ, and ℓ1ℓ2, and the author is upfront that he expects the true optimal exponent to be 3 rather than 7. The proof of Theorem 5 is a chain of inequalities that is coherent on reading. The Green's function decay picture — crossing a collar costs ℓ^3, a half-collar costs ℓ — is the right organizing idea, and the examples are consistent with it.

The soft spot is exactly where the stress-test note puts it. Proposition 3 needs the fact that a shortest simple geodesic crosses each collar at most once. The argument given is a compactness heuristic: as core lengths go to zero, collars converge to pairs of cusp regions, and for cusp regions a simple geodesic cannot enter the sub-region of area one-half and come back out the same side. That last statement is asserted without proof, and the threshold for 'sufficiently small' is not quantified. This is load-bearing: Corollary 4's σ^3 lower bound for G(p,q) depends on it, and Theorem 5 inherits it. I don't think it is wrong — the heuristic is plausible and probably standard — but it is a genuine geometric lemma hiding inside 'It follows that.' A referee should insist on a proof.

Other reservations are minor. The constants are existential throughout, which is normal for this kind of estimate, though it means the theorem is qualitative in its constants. The reliance on the author's earlier papers [Wlp86] and [Wlp12] is appropriate; those are independent derivations, and self-citation here is not a problem.

Bottom line: this is a solid, useful paper for people working in Teichmüller theory and Weil-Petersson geometry. The main theorem is new and the argument holds up in outline. It deserves serious peer review. I would send it to a referee with a specific request to prove the collar-crossing lemma in Proposition 3. If that lemma gets pinned down, I would be happy to see it published.

Referee Report

2 major / 4 minor

Summary. The paper proves a quantitative bound on the Weil-Petersson sectional curvature of moduli space near the boundary. The main theorem (Theorem 5) states that for surfaces with short geodesics, every sectional curvature is at most -C*sigma^7, where sigma is the product of the small geodesic lengths. The proof combines a lower bound for the Green's function kernel (Proposition 3 and Corollary 4), obtained by summing over twisting families of simple arcs using Dehn coordinates, with a careful analysis of a chain of inequalities for the curvature tensor. The paper also analyzes three examples with vanishing rates of order l^3, l, and l_a l_b, and speculates that the optimal exponent in general is three.

Significance. If the proof is completed, this is a significant quantitative refinement of the negativity of the Weil-Petersson metric. It gives the first explicit polynomial rate, in terms of the product of short geodesic lengths, at which sectional curvatures tend to zero near the boundary of moduli space. The method, combining Dehn parametrization with exponential-distance sums, is potentially useful for other problems in Teichmuller theory. The examples provide concrete and plausible vanishing rates, and the paper correctly distinguishes the proved bound from the heuristic expectation of an optimal exponent.

major comments (2)

[§4, proof of Proposition 3] The assertion that, for sufficiently small core lengths, a simple geodesic does not enter and leave a collar by crossing the same boundary is not proved. The passage beginning 'First we adjust the size of the collars...' states that for cusp regions of unit area a simple geodesic cannot enter the cusp subregion of area one-half, and then concludes from compactness that the analogous property holds for collars with sufficiently small core length. Neither the cusp statement nor the transfer to collars is justified. The compact-open convergence of collars to cusp regions does not by itself control geodesic arcs whose lengths diverge as log(1/l) while the core length l tends to zero, and no quantitative threshold for 'sufficiently small' is given. Since this property is used to conclude that the shortest geodesic eta_pq crosses each collar at most once, and hence underpins the lower bound in Proposition 3 and Corollary 4, Theorem 5 depends on this step. Please replace this passage with a precise lemma and proof, or provide a reference to an existing result that covers it.
[§5, proof of Theorem 5, last paragraph] The conclusion that the difference of the second and last terms of (5) is bounded below by a positive multiple of sigma^3 rests on the comparison of the upper bound for (|f|,|f|) with the lower bound for (|mu_1|^2,|mu_2|^2). The upper bound is obtained from the L^1 bound (7) together with the mean value inequality and inequality (2), giving a pointwise bound of order rho^{-1} sigma^2 on the complement of the cusp regions. For this quantity to be O(sigma), as is later used to conclude (|f|,|f|) = O(sigma^3), the paper needs to state explicitly the convention for 'small' geodesic lengths and the resulting comparison between rho and sigma, namely that the small lengths are bounded above by a fixed constant and that the number of short curves is bounded by the topology. As written, the argument appears to assume an unstated comparability of the short lengths. Please make this convention and the O(rho^{-1} sigma^2) = O(sigma) step explicit.

minor comments (4)

[§2, equations (1) and (2)] The constant C' is used both in the mean value inequality (1) and in the cusp bound (2), but these are different constants. Renaming one of them would avoid confusion.
[§5, notation (f,h)] The definition 'we write (f,h) = ∫ f(p)∆ h(q)dA' appears to have a typo: the second variable should be evaluated at p, so that (f,h) = ∫ f(p)(∆h)(p)dA(p). As printed, the notation is ambiguous.
[Introduction, Theorem statement] The phrase 'for σ the product of small geodesic-lengths' is ambiguous. It should be stated that σ is the product of the lengths of all simple closed geodesics shorter than a fixed constant, and that the constant C* may depend on that threshold and on the topological type. If there are no short geodesics, the statement should be formulated separately or the product taken over an empty set.
[§6, expectation paragraph] The final paragraph presents the expectation that the optimal exponent is three. This is clearly labeled as an expectation, but it may be helpful to state more explicitly that it is not part of the proved theorem, since a reader could otherwise mistake it for a result.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: Theorem 5 is proved from Green's-function lower bounds and prior curvature formulas, not from the target bound itself.

full rationale

Walking the derivation chain: Theorem 5 reduces the sectional curvature estimate to inequality chain (5), then uses Corollary 4's lower bound G(p,q) ≥ C''σ^3 together with mean-value and cusp-norm bounds. Corollary 4 is derived from Proposition 3, which proves K(p,q) ≥ C''σ^3 by Dehn parametrization and explicit comparison arcs across collars; the bound is proved, not assumed. The curvature tensor expression is cited from the author's earlier [Wlp86], and the geodesic-length gradient expansions in the examples from [Wlp12], but these are prior published results with independent derivations; they do not assume the σ^7 theorem. No parameter is fitted to a subset of data and then renamed as a prediction, and no uniqueness theorem is imported from the authors' prior work to forbid alternatives. The genuinely delicate point is the unquantified compactness claim inside Proposition 3 that, for sufficiently small core lengths, a simple geodesic cannot enter and leave a collar through the same boundary; the paper asserts 'It follows' without a proof or threshold. This is an omitted-support/correctness risk, not circularity: the claim is not equivalent to the target result, and the target bound is not fed back into its own proof. Accordingly, there is no circular step.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The central claim rests on standard hyperbolic geometry and spectral theory, plus several prior results by the author: the Weil-Petersson curvature formula [Wlp86], gradient and Hessian comparisons [Wlp08], and geodesic-length gradient expansions [Wlp12]. These are treated as background axioms rather than re-derived. The cusp-norm inequality with constant 1/8 is an ad hoc technical bound asserted in the proof. No free parameters are fitted to data; all constants are existential and depend only on topological type.

assumptions (6)

domain assumption The Weil-Petersson Riemann tensor has the formula R_{αβ̄γδ̄} = (αβ̄,γδ̄) + (αδ̄,γβ̄) from [Wlp86, Theorem 4.2].
Used at the start of Section 5 to write the curvature numerator (4); it is imported from a prior paper by the same author.
standard math The Green's function G for -2(D-2)^{-1} is positive and has the uniformization group sum representation G(p,q) = Σ_{γ∈Γ} -2Q1(d(p,γq)) from [Fay77].
Used in Section 3 and Corollary 4 to connect the kernel lower bound to the exponential-distance sum.
standard math Dehn's theorem parameterizes multicurves and simple arcs by Z^{6g-6+2n}, with annular twisting numbers.
Used in Section 4 to enumerate twisting families of arcs crossing collars and to extract the ℓ³ factor in Proposition 3.
domain assumption Uniform mean value inequality for holomorphic n-differentials and the cusp bound in equations (1) and (2).
Used in Section 5 to control products of Beltrami differentials; constants depend on topological type and injectivity radius.
ad hoc to paper The modified cusp regions can be chosen so that ∫_{cusps} |φ(ds²)^{-2}| dA ≤ 1/8 ||φ||₁.
Asserted in the proof of Theorem 5 after modifying cusp regions; it is load-bearing for the constants 7/8 and 41/64 in inequality (7) and is not fully expanded.
domain assumption Mumford's compactness theorem: subsets of moduli space with systole bounded below are compact.
Used in the proof of Theorem 5 to reduce to surfaces with sufficiently small geodesic lengths.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The vanishing rate of Weil-Petersson sectional curvatures." pith.science (2026). https://pith.science/paper/SGTMEK26

@misc{pith2026190809859,
  author       = {Pith},
  title        = {Pith review of: The vanishing rate of Weil-Petersson sectional curvatures},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SGTMEK26}},
  note         = {Machine review of arXiv:1908.09859}
}

read the original abstract

The Weil-Petersson metric for the moduli space of Riemann surfaces has negative sectional curvature. Surfaces represented in the complement of a compact set in the moduli space have short geodesics. At such surfaces the Weil-Petersson metric is approximately a product metric. An almost product metric has sections with almost vanishing curvature. We bound the sectional curvature away from zero in terms of the product of lengths of short geodesics on Riemann surfaces. We give examples and an expectation for the actual vanishing rate.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

17 extracted references · 15 canonical work pages

[1]

Burns, H

K. Burns, H. Masur, and A. Wilkinson. The W eil- P etersson geodesic flow is ergodic. Ann. of Math. (2) , 175(2):835--908, 2012

work page 2012
[2]

S. Bochner. Curvature in H ermitian metric. Bull. Amer. Math. Soc. , 53:179--195, 1947

work page 1947
[3]

Geometry and spectra of compact R iemann surfaces , volume 106 of Progress in Mathematics

Peter Buser. Geometry and spectra of compact R iemann surfaces , volume 106 of Progress in Mathematics . Birkh\"auser Boston Inc., Boston, MA, 1992

work page 1992
[4]

Uniform bounds on harmonic Beltrami differentials and Weil-Petersson curvatures

Martin Bridgeman and Yunhui Wu. Uniform bounds on harmonic B eltrami differentials and W eil- P etersson curvatures. Arxiv:math/1908.02535, 2019

work page Pith review arXiv 1908
[5]

Growth of the W eil- P etersson diameter of moduli space

William Cavendish and Hugo Parlier. Growth of the W eil- P etersson diameter of moduli space. Duke Math. J. , 161(1):139--171, 2012

work page 2012
[6]

Classification of W eil- P etersson isometries

Georgios Daskalopoulos and Richard Wentworth. Classification of W eil- P etersson isometries. Amer. J. Math. , 125(4):941--975, 2003

work page 2003
[7]

John D. Fay. Fourier coefficients of the resolvent for a F uchsian group. J. Reine Angew. Math. , 293/294:143--203, 1977

work page 1977
[8]

Trudinger

David Gilbarg and Neil S. Trudinger. Elliptic partial differential equations of second order . Classics in Mathematics. Springer-Verlag, Berlin, 2001. Reprint of the 1998 edition

2001

Show all 17 references

[9]

Kerckhoff

Steven P. Kerckhoff. The N ielsen realization problem. Ann. of Math. (2) , 117(2):235--265, 1983

1983
[10]

R. C. Penner and J. L. Harer. Combinatorics of train tracks . Princeton University Press, Princeton, NJ, 1992

1992
[11]

uller theory in R iemannian geometry . Birkh\

Anthony J. Tromba. Teichm\"uller theory in R iemannian geometry . Birkh\"auser Verlag, Basel, 1992. Lecture notes prepared by Jochen Denzler

1992
[12]

Wells, Jr

Raymond O. Wells, Jr. Differential analysis on complex manifolds , volume 65 of Graduate Texts in Mathematics . Springer, New York, third edition, 2008. With a new appendix by Oscar Garcia-Prada

2008
[13]

Scott A. Wolpert . Chern forms and the R iemann tensor for the moduli space of curves. Invent. Math. , 85(1):119--145, 1986

1986
[14]

Scott A. Wolpert . Geodesic length functions and the N ielsen problem. J. Differential Geom. , 25(2):275--296, 1987

1987
[15]

Scott A. Wolpert . Behavior of geodesic-length functions on T eichm\"uller space. J. Differential Geom. , 79(2):277--334, 2008

2008
[16]

Scott A. Wolpert . Families of R iemann surfaces and W eil- P etersson G eometry , volume 113 of CBMS Regional Conference Series in Mathematics . Published for the Conference Board of the Mathematical Sciences, Washington, DC, 2010

2010
[17]

Scott A. Wolpert . Geodesic-length functions and the W eil- P etersson curvature tensor. J. Differential Geom. , 91(2):321--359, 2012

2012

Pith tools

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

[1] [1]

Burns, H

K. Burns, H. Masur, and A. Wilkinson. The W eil- P etersson geodesic flow is ergodic. Ann. of Math. (2) , 175(2):835--908, 2012

work page 2012

[2] [2]

S. Bochner. Curvature in H ermitian metric. Bull. Amer. Math. Soc. , 53:179--195, 1947

work page 1947

[3] [3]

Geometry and spectra of compact R iemann surfaces , volume 106 of Progress in Mathematics

Peter Buser. Geometry and spectra of compact R iemann surfaces , volume 106 of Progress in Mathematics . Birkh\"auser Boston Inc., Boston, MA, 1992

work page 1992

[4] [4]

Uniform bounds on harmonic Beltrami differentials and Weil-Petersson curvatures

Martin Bridgeman and Yunhui Wu. Uniform bounds on harmonic B eltrami differentials and W eil- P etersson curvatures. Arxiv:math/1908.02535, 2019

work page Pith review arXiv 1908

[5] [5]

Growth of the W eil- P etersson diameter of moduli space

William Cavendish and Hugo Parlier. Growth of the W eil- P etersson diameter of moduli space. Duke Math. J. , 161(1):139--171, 2012

work page 2012

[6] [6]

Classification of W eil- P etersson isometries

Georgios Daskalopoulos and Richard Wentworth. Classification of W eil- P etersson isometries. Amer. J. Math. , 125(4):941--975, 2003

work page 2003

[7] [7]

John D. Fay. Fourier coefficients of the resolvent for a F uchsian group. J. Reine Angew. Math. , 293/294:143--203, 1977

work page 1977

[8] [8]

Trudinger

David Gilbarg and Neil S. Trudinger. Elliptic partial differential equations of second order . Classics in Mathematics. Springer-Verlag, Berlin, 2001. Reprint of the 1998 edition

2001